Skip to main content

Full text of "Computer Demos—What Makes Them Tick?"

See other formats


AALTO UNIVERSITY 

School of Science and Technology 

Faculty of Information and Natural Sciences 
Department of Media Technology 


Markku Reunanen 


Computer Demos—What Makes Them Tick? 


Licentiate Thesis 


Helsinki, April 23, 2010 


Supervisor: Professor Tapio Takala 





AALTO UNIVERSITY ABSTRACT OF LICENTIATE THESIS 
School of Science and Technology 
Faculty of Information and Natural Sciences 


Department of Media Technology 


























Author Date 
Markku Reunanen April 23, 2010 
Pages 
134 
Title of thesis 
Computer Demos—What Makes Them Tick? 
Professorship Professorship code 
Contents Production T013Z 
Supervisor 
Professor Tapio Takala 
Instructor 





This licentiate thesis deals with a worldwide community of hobbyists called the demoscene. 
The activities of the community in guestion revolve around real-time multimedia 
demonstrations known as demos. The historical frame of the study spans from the late 1970s, 
and the advent of affordable home computers, up to 2009. So far little academic research has 
been conducted on the topic and the number of other publications is almost egually low. The 
work done by other researchers is discussed and additional connections are made to other 


related fields of study such as computer history and media research. 


The material of the study consists principally of demos, contemporary disk magazines and 
online sources such as community websites and archives. A general overview of the demoscene 
and its practices is provided to the reader as a foundation for understanding the more in-depth 
topics. One chapter is dedicated to the analysis of the artifacts produced by the community and 
another to the discussion of the computer hardware in relation to the creative aspirations of the 


community members. 


The purpose of the thesis is the documentation of the demoscene and its numerous practices. In 
the current void of demo-related research the study can serve as a stepping stone for other 
researchers. Among the most important findings are the highly self-reflective nature of the 
community, the connections between technology and expression, and the positioning of the 
underground activities in a wider historical context. A large part of the community and its 


artifacts still remain uncharted, suggesting several possibilities for further studies. 








Keywords 


Computer demos, Digital Culture, Home Computers, Multimedia 











AALTO-YLIOPISTO LISENSIAATINTUTKIMUKSEN TIIVISTELMA 
Teknillinen korkeakoulu 
Informaatio- ja luonnontieteiden tiedekunta 


Mediatekniikan laitos 

















Tekijä Päiväys 
Markku Reunanen 23.4.2010 
Sivumäärä 
134 
Työn nimi 
Computer Demos—What Makes Them Tick? (Tietokonedemot — mikä saa ne hyrräämään?) 
Professuuri Koodi 
Sisällöntuotanto T013Z 








Työn valvoja 


Professori Tapio Takala 





Työn ohjaaja 





Tämä lisensiaatintutkimus käsittelee kansainvälistä harrastajayhteisöä, joka tunnetaan nimellä 
demoskene. Yhteisön toiminta keskittyy reaaliaikaisten multimediaesitysten eli demojen 
tekemiseen. Tutkimus kattaa aikakauden, joka alkaa 1970-luvun lopussa edullisten 
kotitietokoneiden ilmestyessä ja jatkuu näihin päiviin saakka. Toistaiseksi aiheesta on tehty 
vähän tieteellistä tutkimusta ja muunkin kirjallisuuden määrä on lähes yhtä vähäinen. Muiden 
tutkijoiden tekemä työ kartoitetaan työn alussa ja tukena käytetään muuta materiaalia mm. 


tietokonehistorian ja mediatutkimuksen aloilta. 


Tutkimuksen materiaali koostuu pääosin demoista, aikakauden levykelehdistä sekä 
verkkomateriaalista, kuten yhteisön verkkosivuista ja demoarkistoista. Yleinen johdatus 
aihepiiriin auttaa lukijaa muodostamaan kokonaiskuvan ennen syvällisempiin aiheisiin 
siirtymistä. Kokonainen luku on omistettu yhteisön tuottamille artefakteille ja toinen luku 


tietokonelaitteiston suhteelle yhteisön luovaan ilmaisuun. 


Tutkimuksen tavoitteena on kuvata demoskeneä ja sen lukuisia käytäntöjä. Vastaavien 
julkaisujen vähäisyyden vuoksi tässä tehty työ voi toimia pohjana muille aiheesta tehtäville 
tutkimuksille. Tärkeimpiä tehtyjä havaintoja ovat yhteisön vahva itsereflektointi, teknologian ja 
ilmaisun väliset yhteydet, sekä alakulttuurin toiminnan sijoittaminen laajempaan historialliseen 
kontekstiin. Suuri osa yhteisöstä ja sen artefakteista on edelleen kartoittamatonta aluetta, mikä 


tarjoaa lukuisia mahdollisuuksia jatkotutkimukselle. 








Avainsanat 


Tietokonedemot, digitaalinen kulttuuri, kotitietokoneet, multimedia 





il 





Foreword 


The background work for this thesis started back in 2004 as an attempt to collect together 
all the academic pieces of text dealing with the demoscene. Little by little the hobby project 
evolved into more serious research, which eventually led to a desire to write something 
about demos myself. Most of the text was written during the spring of 2009 in Mexico 


during my half-a-year leave of absence from The Helsinki University of Art and Design. 


Being a rather active demoscene member since early 1991 and participating in dozens of 
productions have had a significant impact on many aspects of my life. As a byproduct of a 
hobby I had the opportunity to learn valuable skills such as programming and groupwork. 
Largely because of demos I found the interest to study software science and new media, 
which eventually have led me to my current venues. Last but not least the friendships 
originally formed in the scene circles have lasted for years and still do. So, in addition to 
an academic piece of work, the thesis can be considered a testament to those nineteen years 


spent as a member of the community we call the scene. 


The thesis was completely produced using free tools such as LyX, [ATEX and The Gimp. The 
efforts of the creators of these tools are highly appreciated. To encourage the same culture 
of sharing the text of the thesis is published under The Creative Commons Attribution 3.0 
License (see creativecommons.org). The screenshots of software remain property of the 


original artists. 


Pd like to dedicate some thank yous to the people who have helped me along the way. 
Thanks to my busy supervisor Prof. Tassu Takala for his encouragement, and to my former 
workmate Tommi Ilmonen for helping me to get started in the art of academic writing. A 
collective thank you to all the people who commented on my work or helped me to find 
material, especially Antti Silvast, Mikko Heinonen, Petri Lankoski, Daniel Botz, Marko 
Ohra-aho, Doreen Hartmann and Petri Isomáki. Finally Pd like to thank Delia for her 
support in the good old non-digital world. 


Helsinki, April 23, 2010, Markku Reunanen 


111 


Contents 


1 Introduction 


2 Material and Methods of the Study 


2.1 


2.2 


2.3 


2.4 
25 
2.6 


Existing: Works: Lit a AA ee Kn 
2.1.1 Research Publications «ss. 
2.1.2 Other Publications on Demos .................04. 
2:1:3 «Related Research isn s r o ni RA eones 
2.1.4 Diffusion of Innovations Theory ..................., 
Contemporary Textual Resources . ..... o... e 
22l; TExtfles e od La a ed a 
2:2:22 Disk Magazine S i ol a eGo eda tana 
Online Resources «e 
2.3.1 Community Websites . ........ e 
2:3:2 “DemovArchives “i 2005 o e aa 
2:3:3. Usenet NEWS» casudo o Ee nó AA o 
Selected Demos e So ee bad tee ey Roa E aa 
Secondary Artifacts «sk KK ee ee 
Content. Analysis: 200: Gwin A a MA Sy dwt el ee 


3 Demoscene Characteristics 


3.1 


3.2 
3.3 


3.4 


Historical Frame «se 
3.1.1 Microcomputer Revolution . . ss e. 
3.1.2 Models of Scene History . . < «ss e. 
Demographics esn ca e e ee 
Relationship to other Digital Communities . . . «s. 
3.3.1 Hackers or Not? «LVV 
3.3.2 New Media Art ... nn 
3.3.3 Computer and Video Games . . «sn. 
Handles and Groups . «e 


iv 


3.4.1 TakingaName KVL. 31 


3.4.2 Group Dynamics ......... e... 32 
3.5 Competition and Fame esse 33 
3.5.1 Striving for Fame . . se 34 
3.5.2 Elite and Lamers 0. sole. 34 
3.6 Social Networking «e 36 
3.6.1 Parties = 2 224 2 IA ia kka kia Kuutoset A 37 
3.6.2 Swapping, Trading and Spreading .................. 39 
3.6.3 Online Interaction .. . oaa 41 
3.6.4... Länguäpö <a be ek PS NN KON PS es 42 
3.7 Self-Reflectivity + covers. ee ee 43 
Demoscene Artifacts 45 
Al: DEMOS anata 4 A etapa ear ROS 46 
4.1.1 Megademos and Trackmos ......... o... o... . +... .. 47 
41.2 Demo narratives. ui a AR deca 48 
4:1:3; TOMES a eel ad 48 
4.1.4 Development of Effects . .......... 0... oo... +... .. 49 
4.2 Size-LimitedIntros .................. e... 52 
421: “G4K MGS: imss ar li da anes 52 
4:27:27 "AK INEOS: iisa a o Ek taa Eat 54 
4.2.3 1kIntros and Beyond . . «hans 56 
4:37 ~Demo-Aesthetics: i marsit oi so Bk SB ah a emma! lee e 57 
4.3.1 Hardware-Dictated Early Years ............. o... .. 58 
4.3.2 Appearance of Demo Design .......... o... o... ... 59 
4:3.3. Frames of Reference. = sak. sary a ee N Road kunta 60 
4:47 Misual Artifacts ssi v2 Sa EMS a Aa a Sk omal VN s 62 
4.41: Stl images msn tandem eaa Kaa leven: Gone ayaa 62 
4:4:27 ID ODJECE oy eee Soe bab ee ek ee Oe ee a 63 
4.4.3 ASCII and ANSI Art ......... o... e... e... e... 64 
44-4 Disc Cover AT sk sn eine Se LA na as hae weed et 8 65 
A Scene MUSIC a A eines 66 
4:3:1 Cp MUSIC os ss A A KR a 67 
43225 - Modüles ye) soe ase ee a ay NEN 4 BORA Rida a ea 68 
4:5:3- Sample MUSIC- oia aves Be Oe A TN eS da 69 
4.5.4 Software Sound Synthesis «se. 70 
46. Disk: Magazines: aeo o KOET Pa A Ja kA ees a ai a a 71 


4.6.1 Selected Disk Magazines ....... o... .. e... . +... .. 71 


4.6.2 Topics Of Interest .......... o... e... e... +... .. 76 
Hardware and Software Platforms 79 
5.1 8-Bit Computers .......... 0.0... 0000000002 eee eee 80 

SAL” Commodore 64 «20k a tes a WA eb eae Ää asta 81 

5.1.2 Less Popular 8-Bit Computers ..................-.. 83 
5.2 32/16-Bit Computers .............. 0000200000 00004 84 

5.2.1 Commodore Amiga. ...............2. 02000000045 85 

5:22 Atari ST familys 20s ed A ware A san a 86 

5.2.3 Early PC Compatibles .............. o... ...... 88 
5.3 True 32-Bit Computers ............... 0020000000004 90 

5.3.1 Amiga 1200 and 4000 .......... o... o... ...... 90 

5:3:2 The Improving PC: oe ss vsk sc da oe Ree as 92 

5.3.3 Other 32-bit Demo Platforms. .......... oo... o... .. 94 

534 ID Acceleration: ca e ie Phe RP Hon 95 
5.4 Roles of Software «e 97 

AL TOOLS R. eon A RA OS NÓ Rime 97 

54:27 “Algorithms: +. 008 s A a a dd Naa 99 
5.5 Effects of Changing Technology ........... o... ..... e... 100 

5.5.1 Patterns of Diffusion «se. 100 

5.5.2 Digital Heritance of the Scene 0. «Kn. 102 
Conclusion 104 
References 106 
Demo Content Analysis Form 116 
List of Demos Analyzed 117 
Themes of Disk Magazine Articles 121 


vi 


Glossary 


API Application Programming Interface 
Artpack A collection of images and/or music, sometimes containing an interactive viewer 


BBS Bulletin Board System, also known as a board. A system offering services such as 


messaging and file retrieval using a modem. 


Blitter A co-processor whose purpose is to assist the main processor in graphics generation 


by copying image data blocks and filling regions 
C-64 Commodore 64 


C2p Chunky to planar conversion: the reordering of byte-per-pixel data to separate bit- 


planes 
Charts Ranking lists of demos, programmers, tunes etc. 
Chippack A collection of chiptunes 


Chiptune A musical style that resembles the sound produced by 8-bit computers. Also 


refers to the small size of a sound file. 


Chunky Graphics data stored so that each pixel corresponds to one or multiple consecutive 


bytes 
Coding Programming 


Color clash Bleeding, an unwanted effect on 8-bit computers where colors would be dis- 


played incorrectly due to character block attribute limitations 
Compo A competition 
Copyparty A party focused on copying the latest productions and illegal software 


Crack intro A welcome screen containing messages to other groups, shown at the begin- 


ning of cracked programs 


Cracking The removal of copy protection from commercial software 


vii 


Crowdpleaser A production that specifically tries to appeal to the audience at a party 
Demo A program showcasing the programming and artistic skills of the author(s) 


Demomaker A program that enables one to create demos with little or no programming 
skills 


Disk image A single file containing all the information stored on a medium such as a 
diskette 


Diskmag A disk magazine: an electronic magazine originally published on diskettes. Also 


known as ”mag” and ”maggy”. 


Emulator A program that lets a computer run software originally written for other plat- 


forms 


Faking May refer to one of the following: cheating in voting, reusing stamps in mailswap- 
ping, or programming a demo effect so that it looks more advanced than it actually 


1S. 


Fast compo A competition where a production (demo, music or picture) has to be created 


in limited time from scratch 
Flyby A flight across a 3D scene 
Fps Frames per second 
Fuckings The derogatory opposite of greetings 
Gfx Grafix, graphics 
Graphician A graphic artist 
Greets Greetings 
Infofile A text file that describes the content of an archive 
Intro A size-limited demo, for example four or sixty-four kilobytes 
Invitro Invtro, an invitation demo made for a party or a competition 
Joke group A group that creates mostly low-quality tongue-in-cheek productions 
Lamer A derogatory term used to describe non-skilled people 
Leet Elite, at times written with numbers: ”1337” 


Leetspeak A style of writing where words are twisted and letters replaced by numbers and 


other symbols 


Vili 


Metaballs Polygonal approximation of an implicit surface, spheres melting together when 


they are close to each other 


Mod A module: a piece of music containing both the score and the instruments in the same 


file. In the context of computer games refers to game modifications. 
Musicdisk An interactive collection of tunes 
Music syne The synchronization of music and visual effects 
Mzx Muzax, musax, music 


Newbie A newcomer, other forms like ”newb” or ”noob” also exist 





Nuskool New school, modern hardware and style 


Object show A demo consisting mainly of 3D objects that are displayed (usually rotated) 


on-screen 

Oldskool Old hardware and software or audiovisual style 

Otaku The Japanese counterpart of geek/nerd, an overdevoted hobbyist 
Pixeling Drawing an image laboriously at pixel level 

Prod A production 


Reset demo A hidden demo screen that appears when the user resets the machine in the 


middle of a demo 
Ribbon An effect involving a polygonal stripe or tube 
Ripping Using material made by others without permission 
Rotozoomer Rotating and zooming an image 
Scene The demo community, demoscene 
Scroller Text moving across the screen 
Shader A user-programmable unit in a graphics processor 
Slideshow A demo displaying still images, often with music 
Soft synth A software sound synthesizer 
Song A tracker music file containing only the notes but no samples 
Spreading The publishing process of a production 


Sprite A small image moving independently on top of the background 


1X 


Swapping Interchange of demos, pictures, music and other productions among community 


members 
SysOp BBS system operator 


Tracker A program for music composition based on the concept of placing notes on indi- 


vidual vertical tracks 


Trackmo A demo that loads its content directly from the diskette tracks. Also used to 


describe the related continuous style. 
Trading Modem-based swapping 
Trainer A game modification that provides features such as unlimited lives or ammunition 


Trolling Posting of controversial statements to heat up online discussion, also know as 


”flamebaiting” 
Votedisk A diskette used for voting at a party 
Warez Illegal files 


Wild compo A competition allowing almost any kind of entries without limits on the plat- 


form or size 
Wobbler An effect based on deforming a 2D image, typically by sine curves 


Writer An effect where text is written on the screen 


Chapter 1 


Introduction 


The demoscene—or simply the scene, as it is known by its members—is a worldwide com- 
munity of hobbyists interested in computer demos. A demo in this context can be defined as 
a short, most often non-interactive program that displays audiovisual content in real-time. 
The demo community has its roots in the late-1970s home computer revolution that made 
the technology widely accessible to households and hobbyists for the first time in history. 
Pirate groups that spread copies of illegal software (mostly games) attached screens known 
as crack intros with their messages to the programs, which ultimately lead to the forma- 
tion of a different community focusing on the programming of such demonstrations alone 
(Polgar, 2005, pp.40-61; Saarikoski, 2004, p.192; Tasajarvi et al., 2004, pp.12-15). 


The motivation for this study springs from the current lack of existing research on demos 
and the demoscene. The shortage can be explained in many ways, but the bottom line 
is that a diverse hobbyist culture has largely remained an unstudied piece of history for 
over twenty years. Introductions to the phenomenon have appeared in a small number of 
academic publications and some individual aspects have received further attention (Section 
2.1.1), but in general such works are scarce. Outside the academic world the number of 
publications is somewhat higher (see Section 2.1.2), but the writings often lack either width, 


depth or reliability. 


The research problem can be crystallized into a question: what are the forms and practices 
of the community known as the demoscene? The aim is to document the community and link 
it to a wider historical scope by utilizing different sources such as community discussions, 
artifacts and research literature. The time frame of the study starts from the late 1970s with 
the appearance of the first crack intros and popular microcomputers, and extends to cover 
the past thirty years. The geographical point of view is Finnish and Nordic, but all effort 
is taken to impartially discuss works from all the countries where significant activities have 
taken place. While the demoscene should not be considered a monoculture, its international 


nature makes the results easier to generalize. 


It became evident early on that a study on a community, such as the demoscene, could 
be placed in numerous contexts. Certainly it is part of the history of computing in gen- 
eral and the domestication of technology, but such an approach alone was not considered 
satisfactory—the contexts of art or media research would be equally valid. Moreover, the 
strong social aspect of the community would warrant a youth culture, subculture, or gender 
study approach as well. Finally, due to the recurring theme of technological change and its 
interplay with the community, the main context of the study turned out to be the history of 


home computers and their use. 


My personal relationship with the subject deserves some discussion, since it has inevitably 
affected the approach taken, as well as the analysis. The positive sides of a lengthy in- 
volvement with the community are a large amount of first-hand experience on the matters, 
an understanding of the practices and the language, and existing connections inside the 
community. Achieving the same level of expertise starting from scratch would require a 
significant amount of time. However, the very same familiarity also introduces challenges 
to the study. How to step outside of the community and question one’s learned points of 
view? How to investigate the things you ”already know”, so that the criteria of credible 
research are met? How to treat people and artifacts neutrally without the bias of likes and 
dislikes formed during the years of personal involvement? It is impossible to completely 
discard one’s personality from the process, but the use of established research methods and 
the focus on the large-scale phenomena instead of details, such as individual persons or their 


works, help to alleviate the bias. 


The source material and the methods used in the analysis are discussed in chapter 2. The 
chapter starts with a survey of the existing written works, after which the primary sources 
are presented. Chapter 3 serves as a broad introduction to the scene and its cultural practices, 
providing a frame of reference for the later chapters. Various artifacts such as demos, music 
and visual works produced by the scene are discussed in Chapter 4. The following chapter, 
number 5, walks the reader through hardware generations from the first 8-bit machines to 
the modern multimedia computers. In addition, the major shifts that have taken place in 
the software domain are presented. Examples of contemporary demo effects are provided 
to illustrate the relationship of technology and the creative aspirations. The final chapter 


provides concluding remarks about the study and some future directions. 


Chapter 2 


Material and Methods of the Study 


Due to the multi-faceted approach of the study, no single primary source was sufficient to 
provide answers to the research problem. Various materials such as disk magazines, online 
forums, demos and text files were chosen in order to document different aspects. Each of 
these primary sources required a specific approach to the analysis, since they differ from 
each other considerably. The most important method was content analysis, which was ap- 
plied to the material in different ways, depending on the type of source. Additionally, exist- 
ing work conducted by other writers was examined in order to reflect on their observations, 


and to strengthen the theoretical backbone of the study. 


2.1 Existing Works 


During the five years of collecting demoscene-related written material it soon became ev- 
ident that it hardly exists. The bibliography page of Demoscene Research (Reunanen & 
Silvast, 2004) contains all the directly related works encountered, excluding occasional 
newsflashes found in newspapers and magazines. Currently the bibliography contains close 
to forty printed works such as articles, theses or books. Not all the publications can be con- 
sidered analytical or research-oriented, which brings the amount of academic works found 


down to about twenty. 


It is an intriguing question why exactly the community has attracted so little public in- 
terest. In comparison, a variety of publications is available on other contemporary digital 
phenomena. To elaborate the situation we should consider the probable, tightly intertwined 


explanations to this invisibility: 


e Underground nature of the scene. Partly due to its illegal roots and partly due to its 
closed youth culture nature, the community has intentionally kept itself away from 


mainstream visibility. 


e Small market segment. While several hobbies such as video games and music feed 
multi-billion dollar industries, the demoscene is hardly a significant market that could 


be specifically targeted. 


e Low visible effect on society. Demoscene is not affiliated with negative phenomena 
(such as violence, vandalism or computer break-ins) that would attract interest. The 


positive effects do not necessarily differ from those of any computer hobby. 


e Low commercial significance of productions. The community creates productions for 


its own uses that typically do not have immediate commercial value. 


e Technological nature of the scene. A good understanding of computers, programming 
and digital media is required to analyze the artifacts and the related practices of the 


community. 


e Geographical location of the scene. Concentrated mostly in North and Central Eu- 
rope (see Section 3.2 for further discussion) the demoscene has been out-of-sight of 


American media researchers as well as away from the roots of computer history. 


These generalizations obviously do not come without exceptions. As an example, the un- 
derground nature of the scene has been on the decline. A small number of books has been 
published lately (Polgar, 2005; Tasajarvi et al., 2004; Vigh & Polgar, 2006), demos have 
appeared in art exhibitions (Digitalcraft, 2002; Tasajárvi, 2003), and there even exists an 
advocacy group with the name Demoscene Outreach Group (Scheib et al., 2002). Likewise, 
the historical link to software piracy, described by Polgar (2005, pp.40-62) among others, 


can be taken as a counter-example in the case of the social effect. 


2.1.1 Research Publications 


The most relevant publications on demos so far have been a few theses, conference papers, 
and chapters in books dealing with a related topic. The majority of the publications originate 
from the Nordic Countries, which appears to be a natural consequence of the strong local 


demo community in relation to the overall size of the population. 


Among the earliest scene-related articles published, and therefore among the most refer- 
enced, is the demoscene overview The Hacker Demo Scene and lts Cultural Artifacts by 
Borzyskowski (1996). The study took place in 1992-1994, so it already represents the sit- 
uation fifteen years ago. Notable in the title alone is the recognition of demos as cultural 
artifacts. The most interesting part of the article is the 21-item list of demo characteris- 
tics that provides observations on the total of 743 demos viewed. Borzyskowski presents 
statistics on the origins of the demos, thus providing material for the analysis on the geo- 
graphical distribution of the groups of that time. Numerous references to cyberpunk more 


likely reflect the discourse of the time than the contemporary scene reality. 


4 


So far, the most extensive study on computer demos, their aesthetics and development ap- 
pears to be the unpublished (as of September 2009) doctoral thesis by Botz (2008). With an 
art/media history approach Botz walks through the different eras of computer demos with 
the intention of building a big picture of their aesthetic development in relation to the tech- 
nological and cultural currents of the time. Unfortunately, the themes and claims presented 


in the thesis are hardly represented here due to the language barrier. 


Both Lónnblad (1998) and Roininen (1998) have written a master’s thesis on a demo-related 
topic (available in Finnish only). Lónnblad approaches the scene from the point of view of 
musicology and discusses the structure of demo music, based on a few examples of the time. 
Together with her other published article (Lónnblad, 1997), the thesis still remains the only 
study of demo music in such depth. Roininen’s thesis provides an equally rare approach: 
her focus is on the social dynamics of the scene. Her outsider view to the community is both 
a benefit and a hindrance: she makes sharp observations on the social dynamics but lacks 
the technical mindset that would be required to understand the phenomenon completely. 


Both theses also provide a typical overview of the community. 


As part of a gender study oriented doctoral thesis The Net Is not Enough: Searching for the 
Female Hacker Nordli (2003a, pp.71-91) analyzes the Norwegian demo party The Gather- 
ing’99. In addition to scene-related content, her thesis provides valuable tools for under- 
standing the role of women in the chiefly male-dominated computer hobbyist circles. The 
same theme also appears in another article by her (Nordli, 2003b). Another relevant doctoral 
thesis was written by Saarikoski (2004). Koneen lumo (The Lure of The Machine), includes 
a section on the demoscene. Saarikoski provides an overview of the phenomenon from a 
Finnish standpoint and connects it to the developments of the time. The general topic of 
the thesis is the history of the Finnish home computer culture starting from the 1970s, and 
demos are treated as part of that larger framework. The thesis is a continuation of his earlier 
licentiate thesis (Saarikoski, 2001b), which contains similar analysis of the demoscene in 


chapters nine and ten. 


The bachelor’s thesis by Kurki (2002) provides a comparison between the practices of the 
demoscene (a boy culture) and the decoscene (a girl culture). The thesis treats both com- 
munities as postmodern tribes and imagined communities that exist only through their man- 
ifestations. Another comparison between two youth cultures was written by Faler (2001), 


who used the graffiti scene as his reference. 


2.1.2 Other Publications on Demos 


The first book published about the demoscene was Demoscene: The Art of Real-Time by 
Tasajarvi et al. (2004). The book was closely connected to the demo art exhibition held at 
Kiasma, The Finnish Museum of Contemporary Art (Tasajárvi, 2003). The book contains 


an introduction, articles and an illustrated list of works exhibited at the museum. The most 


interesting contribution is the proposed tripartite model of scene history, which will be 
discussed in more detail in Section 3.1.2. Overall the amount of content is limited and the 


selection of demos represents only a small fraction of Finnish groups. 


So far, the most ambitious demo book project was realized by Tamas Polgar, the author of 
Freax: The Brief History of the Demoscene (Polgar, 2005). The first volume contains an 
introductory part and a history of the Commodore 64 scene, followed by a history of the 
Amiga scene. It is evident that the target audience of the book is the scene members them- 
selves, since the text requires a good understanding of related terminology and practices to 
be understood. Certainly the amount of material is massive, but the content should not be 
taken as is without criticism: at places factual content and scene rumors are blended heavily 
to create narratives. Polgar’s work is still going on and the next volume, dealing with the 
IBM PC and alternative platforms, is expected. Between the two volumes the writer partici- 
pated in another Freax book project dealing with the visual art of the scene (Vigh & Polgar, 
2006). The other author, David Vigh, had already published an online collection on the 
same topic at an earlier date (Vigh, 2003). While both art books are visually rich and serve 
as a cross-section to the styles and themes of a number of scene artists, they struggle with 
the problem of selecting a balanced set of works that would represent the whole timespan 


and different genres. 


Articles discussing demo-related issues have appeared in a few books [see Carlsson (2008); 
Inkinen & Salmi (1996); Shatz (1993, 1994)] and numerous newspapers. A typical news- 
paper article is a short newsflash about a local party, as observed by Saarikoski (2001a) as 
well. Usually the point of view is that of an outsider, and the article reveals more about 
the attitudes towards the hobbyists than about the actual people in question. In addition 
to newspapers, media-related magazines have occasionally covered demos on their pages. 
Two examples of comprehensive introductions published in magazines are Green (1995) 
and Saarikoski (2001a). Also the game magazine Edge covered PlayStation 3 demos in 
an article (Edge, 2008). The most active publication in this respect is SCEEN, which has 
already in its first two issues featured several articles describing demo parties, groups and 
exotic platforms (Barbat, 2005; Cruz, 2005; Scholz, 2007a,b). 


2.1.3 Related Research 


This section deals with the publications that were used for widening the theoretical frame- 
work of this study. Among the most important sources of inspiration and points of compar- 


ison were media research, digital art and computer history. 


Computer history related studies were a fundamental source when trying to position the de- 
moscene in a larger temporal context. Factual information helped to answer questions such 
as ”Why did microcomputers appear at homes in the early 1980s?”, and ”Why did Com- 


modore dominate the home computer market?” The doctoral thesis of Saarikoski (2004), 


which describes the Finnish computer hobbyists starting from the 1970s, was used in par- 
ticular for finding out about the early phenomena of the home computer age. The central 
theme of the thesis is closely related to the research problem as well: Saarikoski focused on 
the different forms the domestication of computers took in youth culture and mainstream 


media. 


Basic facts concerning technical specifications and release dates of game consoles and home 
computers were mostly collected from the book Game.Machines by Forster (2005), which 
provides a comprehensive, although brief, overview of popular gadgets, coupled with high- 
quality illustrations. The Chronology of Personal Computers, a detailed timeline collected 
by Polsson (1995), was another reference used when connecting scene activities to the his- 
tory of computing. The history of Commodore, arguably the most influential company in 
the 1980s home computer business, was studied by Bagnall (2005). While the book is writ- 
ten more like a story and appears at times rather opinionated, the amount of insight into the 


company history alone makes it a valuable source. 


The work of Sherry Turkle on the psychology of computer users has certainly influenced 
this study. Her observations on different types of computer users made in The Second Self 
(Turkle, 1984) sheds light to the intimate relationship between the machine and the user, 
also apparent in many demoscene-related phenomena. Turkle divides the computer users 
into two categories: hard masters who approach problems on the basis of their technical 
skills, and soft masters who are creatively oriented and use technology as a tool. While 
the division is not likely to be as clear-cut in practice, the concept could be utilized when 
studying the creative processes of the demosceners. Turkle’s more recent book, Life on 
the Screen (Turkle, 1997), continues the earlier work and focuses on the identities people 
assume in the Internet age. Quite a lot of material is shared with the former book, and 
the Internet technologies of the time seem archaic from today’s perspective; one could say 
that the increased focus on technology has made the content age more rapidly. The most 
useful findings in the latter book are related to the complex interactions between the real 


and assumed personalities of computer users. 


The artistic uses of computers were studied in order to proportionate scene activities with 
mainstream ones. The earliest history of computer-based art was documented by Franke 
(1971), a pioneer of the field himself. The crude minimalism dictated by technological 
limits and the lack of established practices in the works of the 1960s may seem outdated 
to the contemporary reader, but, on the other hand, there are interesting similarities to size- 
limited intros (see Section 4.2) that are based on the algorithmic creation of visuals. Modern 
day media art classics have been discussed by for example Wands (2006) and Tribe & Jana 
(2007). In practice both books are mainly annotated collections of artworks with photos 
and screenshots. A more research-oriented approach is available in Media Art Histories 
(Grau, 2007), with articles covering a wide range of topics from historical milestones to 


philosophical essays on the essence of new media. 


The observations of Manovich (2001) on the connections of new and old media can be 
applied to demo research as well: the narrative structures and visual language found in 
demos most likely owe a lot to methods originally developed in the context of cinema. As 
an example, the section on compositing (pp.136-160) discusses the layering and combining 


of different image sources, a technique also used in numerous demos. 


Youth culture, subculture and gender studies would be realistic frames of reference for de- 
moscene research. In the scope of this thesis, however, such approaches are largely omitted. 
So far, the only demoscene-oriented publications conducted in relation to those disciplines 
appear to be the writings by Nordli (2003a), Kurki (2002) and Roininen (1998). The hacker 
culture books by Levy (1994) and Thomas (1991) were used for comparison between the 
Europe-centered demoscene and the hacker culture originating from the United States. The 
historical content of the two books also provide insight to the very birth of computer use 
as a hobby and a way of life, also discussed to some extent by Turkle (1984, pp.165-195, 
202-207). The closely related cracker culture was documented by Rehn (2004) and Vuori- 
nen (2007). The cracker/warez scene publications provide opportunities for reflection with 


the characteristics of the demoscene. 


2.1.4 Diffusion of Innovations Theory 


Diffusion theory, originally formulated by Everett M. Rogers in his 1962 book Diffusion of 
Innovations (Rogers, 2003) is the framework for the analysis of the technological changes 
that have affected the demoscene throughout the years. The theory provides a wide range of 
tools for understanding the different phases and mechanisms of innovation adoption, mak- 
ing it a lucrative choice when discussing the effects of new hardware or software on the 
community. Innovation and diffusion are defined by Rogers as follows: ”An Innovation is 
an idea, practice, or object perceived as new by an individual or other unit of adoption” 
(p.36), and "Diffusion is the process in which an innovation is communicated through cer- 


tain channels over time among the members of a social system” (p.5). 


Rogers divides the diffusion process into five phases: knowledge, persuasion, decision, 
implementation and confirmation. In the first phase the individual first learns about the 
innovation and its functionality. In the persuasion phase the individual forms an attitude 
towards the innovation, which leads to the decision phase, where the innovation is either 
accepted or rejected. In the implementation phase the individual starts to use the innovation. 
Finally, in the confirmation phase the individual evaluates whether the decision made was 
correct and may still revert back to the old practice. A related concept is re-invention, 
referring to the unexpected ways of adapting an innovation to a practical situation. (Rogers, 
2003, pp.168-218) 


The adopters, likewise, are divided into five categories based on the time of adoption: in- 


novators (2.5% of all the adopters), early adopters (13.5%), early majority (34%), late 


majority (34%) and laggards (16%). Each type has its own characteristics based on socioe- 
conomic factors, personality and communication behavior. The innovators are described as 
venturesome persons who have the most connections outside of the local community, but 
who may not be held in high regard by the community. The early adopters are the respected 
opinion leaders of the community and therefore crucial to the adoption of the majority. The 
first big group, the early majority, has the tendency to adopt, but not hastily and not with- 
out the consultation of their peers. The late majority is skeptical about the innovation and 
adapts because of necessity and/or peer pressure. Another characteristic of the group is the 
conformance to the community’s norms. The last group, the laggards, are the last ones to 
adopt, the most skeptic, and have their frame of reference in the past. The diffusion process 
and the positions of the adopter categories is often presented as a bell curve (p.281) or the 
cumulative S-curve (p.11). (Rogers, 2003, pp.267—299) 


Other important concepts of the diffusion theory are the change agent, properties of inno- 
vations and diffusion networks. The change agent refers to the external instance that wants 
to diffuse an innovation to the community. The methods and connections chosen by the 
change agent are seen as a crucial factor in the success of the process. (pp.365—+401) The 
properties of innovations such as their relative advantage, compatibility with the community 
practices and complexity are discussed in chapter six of the book. Together with the type of 
innovation decision, communication channels, social system and change agent’s efforts they 
determine the total rate of adoption (p.222). The concept of a diffusion network comprises 
a variety of themes ranging from interpersonal links to the homo/heterophily of clusters of 
people, which all affect the diffusion process to a certain degree (pp.300-364). (Rogers, 
2003) 


The diffusion theory has also attracted criticism, part of which is presented in the book, 
together with advice on how to overcome the particular problem (Rogers, 2003, pp.105— 
130). To avoid the possible pitfalls the criticisms need to be taken into account in this study 


as well. 
1. Pro-Innovation Bias. The implication that an innovation should be diffused rapidly 
and completely, as-is without re-invention. 


2. Individual-Blame Bias. The tendency of researchers to side with the change agencies 


and blame the failures on the adopters. 


3. Recall Problem. The researchers rely on the inaccurate data reported by the respon- 
dents of the study. 


4. Issue of Equality. The omission of the negative effects of an innovation, especially 


the widening of the socioeconomic gap in the developing nations. 


The first two issues are not significant to the study, since they largely follow from the link 


between the researcher and the body funding the research. No such link exists in this study, 


9 


so a neutral point of view can be more easily maintained. The issue of equality is not highly 
relevant either, since this study is not about economy or developing countries. However, if 
we consider honor and fame as the scene ”currency”, issue number four appears to be more 
relevant: does a new innovation favor the already respected and hurt the lowest ranks? The 
recall problem is clearly the most important in the context of this study: how to ensure that 
the data is not skewed? Certain factors of the material and methods chosen help alleviate the 
problem. First of all, the data collected is based on sources such as contemporary texts of 
the time instead of interviews. It should be noted that the recall problem can affect textual 
sources as well, especially if they were written years after the actual events. In general, the 
focus on contemporary sources largely removes human factors such as nostalgia and bad 


recall, with the downside that it does not allow for valuable personal reflection. 


2.2 Contemporary Textual Resources 


The self-documenting nature of the demoscene is a great benefit for a researcher trying to 
find material on the past. Already from the late 1980s, there exists quite a rich variety 
of textual sources describing contemporary events, artifacts, and other topics that were of 
interest to the community. This section deals with traditional textual sources such as disk 
magazines and various types of text files, whereas more modern on-line sources appear in 
Section 2.3. 


2.2.1 Text files 


Among the most useful text files are the documents related to demo parties (see 3.6.1 for 
more discussion on parties). In pre-Internet days, it was a common practice to send out an 
invitation to a party as a text file that was distributed by all possible means to get participants 
to the event. Typical themes of an invitation were the date, the location, the facilities, the 
entrance fee, and the competitions of the party. After the party, equally important files 
were the compo results, containing the rankings and respective points of the works that took 
part in the competitions. These two types of text files are basically informative by nature, 
whereas the third related type, party report, serves a completely different purpose. In party 


reports we can find first-hand experiences and subjective opinions of the writers. 


When demos are distributed by a means or another they are usually accompanied by a 
short piece of text, an infofile, containing at least the name(s) of the author(s) plus contact 
information. A more condensed form of info files is the file_id.diz file, intended for Bulletin 
Board Systems (BBS) where they were displayed in the file listings. Another category of 


text files are the tutorials that contain help on programming tricks for beginners. 


10 


The text files described here are mostly used as a secondary source, for checking the facts. 
Questions such as What was the winning demo of the Motorola Inside party in 2004?” are 
quickly answered by the respective result file. The only exception to the sporadic use is the 
use of competition results to gather information about the relative popularities of each demo 


platform, discussed in Chapter 5. 


2.2.2 Disk Magazines 


Disk magazines—or diskmags for short—are the counterparts of printed magazines in the 
scene circles. Both share some features such as the division of content into separate articles, 
and the concept of an issue. These similarities exist due to the heritage from the Gutenberg 
domain, but there are also numerous properties that are unique to disk magazines. For an 
example of what the magazines look like see Fig. 2.1. A deeper description of the technical 
aspects and contents is provided in Section 4.6. The focus here is on the relevance of disk 


magazines to the study. 


So far, diskmags have attracted extremely little attention: even directly demo-related publi- 
cations fail to mention them. Articles on diskmags have appeared mostly in other diskmags. 
Claus-Dieter Volko is among the most active writers on the topic, as exemplified by the col- 
lection of his articles (Volko, 2009). A historical review of the first years of diskmags was 
written by Jacobsson (2006), a former editor of Propaganda himself. Haavisto (2001) pub- 


lished an informal introduction to diskmags in the Finnish print magazine Enter. 


RUBRICS 


REVIENS 





Figure 2.1: R.A.W. #6 (1993) table of contents 


Disk magazines contain a variety of material ranging from demo/game reviews to program- 


ming tricks, and even short stories. What makes them an invaluable source for research is 


11 


their contemporary nature: the themes discussed in the articles were fresh and interesting at 
the time of the writing. The points of view represent the attitudes of the era, and since the 
target audience was the scene members themselves, the opinions stated were not watered 
down (actually quite the contrary). Moreover, the contents of the magazines are usually 


well organized into sections which makes it easier to concentrate on the topics of interest. 


In this study diskmags were used for numerous purposes. They served as sources of con- 
temporary thoughts, topics of interest, and trends. To represent the different eras and 
platforms, five diskmags were chosen: Sex’n’Crime (Commodore 64, 1989-1990), Zine 
(Amiga, 1989-1991), R.A.W. (Amiga, 1991-1996), Imphobia (PC classic, 1992-1996) and 
Hugi (PC modern, 1996-2008). Due to the lack of actual hardware the magazines were 
viewed using emulators (software that lets a computer run other computers” programs). The 
analysis continued the earlier work by Reunanen & Silvast (2009) and was conducted using 
content analysis as the main method (see Section 2.6). In the first phase the articles were 
sorted into categories to get an overall view on the themes of interest and types of articles 
published. In the second phase, the articles of interest were examined further to reveal reoc- 
curring patterns of the language: what kind of expressions were used to convince the readers 
and to describe positive or negative issues? The focus was especially on controversial topics 


such as new hardware and software platforms. 


2.3 Online Resources 


The rise of the Internet has not gone unnoticed in the demoscene. These days a great deal of 
demo-related resources is available online at the disposal of the researcher. To examine all 
the websites available alone would warrant a separate thesis, so only general observations 


are provided on each type of resource. 


2.3.1 Community Websites 


Community websites host a variety of different contents: for example demo-related discus- 
sion boards, party information, news, reviews, demo archives, articles, group homepages, 
photos and videos can be found on the web. Demo archives are discussed separately in Sec- 
tion 2.3.2. In the context of the study the most important sources were discussion boards, 
demo archives and video sites that provide video captures of demos in an easily accessible 


format. 


The most important website examined is the popular pouet.net, which hosts a discussion 
board, news and most importantly a large database of demos and groups. In addition to 
the discussion board (or ”BBS”) it is possible to comment on the individual productions 


and leave a positive/neutral/negative rating. The rating system is a useful meter of the 


12 


= Account " Custom " Prods = Random " Groups * Parties " Boards " Users " Search = BBS = Lists = Faq = Submit = 


ARTE by sanity [ +.nfo ] 
platform : € Amiga ECS 
type : demo 
release date : december 1993 
release party : The Party 1993 
compo : amiga demo 
ranked : 
169 
(55 E] 





Figure 2.2: The production view of pouet.net displaying the details of Arte by Sanity. 


popularity of a particular demo, also revealing works considered as classics. The demo 
database is maintained by the global community on a voluntary basis and it contains, as 
of May 2009, an impressive number of 51,858 productions and 448,108 comments, with 
more added daily. The details listed for each production in the database are as follows (see 
Fig. 2.2 for a screenshot): 


Name of the production and its author (usually a group) 

e Screenshot 

e Platform the production runs on 

e Type (e.g. demo, game or diskmag) 

e Release date, party, competition and the ranking at the competition 

e Date when the production was added and by whom 

e Popularity figures such as the amount of positive, neutral and negative ratings 

e Download link(s) to the production and its info file 

e Comments and individual ratings 
Not every detail applies to all the included productions: for example, many old demos were 
released outside of parties. The screenshots are limited to a size of 400x300 pixels, which 


somewhat limits their usefulness from a research perspective, since most modern demos 


run in far higher resolutions. The pouet.net system of classification of productions is both 


13 


created and validated by the community, which makes it a useful resource when devising 
similar taxonomies. The taxonomy used in Chapter 4 mostly conforms to the pouet.net 


classification. 


Another content-filled community website is ExoticA (Lunder, 1996). Especially useful for 
a researcher are the lists of Amiga and Commodore 64 demogroups that contain member and 
country information with a description for over two thousand groups. Such material could 
serve as the basis for different studies on the demographics and the history of the two scenes. 
On the site, there is also additional content on Amiga games, demo parties and Amiga/C-64 
music. Numerous other community websites exist with different foci: for example, a variety 
of activities take place on the scene.org servers (pouet.net is also hosted there), maintained 
by the non-profit Scene.org foundation whose aim is to support the community by offering 
a forum for communication and sharing (Scene.org, 2009). Among the services offered are 
a demo archive (described in Section 2.3.2), mirrors of other archive sites, IRC (Internet 


Relay Chat) servers, and web space for demogroups or demo-related sites. 


In her doctoral thesis Goryunova (2007, pp.12-20) examined three active artistic websites, 
such as runme.org, and came up with the term art platform. According to her an art platform 
”is a terminological solution for describing an online platform that enables the crystalliza- 
tion of a cultural phenomenon through the use of of platform’s mechanisms or one, which 
significantly contributes to such a process.” When comparing demo-related websites to her 
definition one can instantly notice strong similarities: art platforms are widely available, 
grass-root, hobbyist sites with an autonomous nature and built-in distinction mechanisms. 
Community and offline meetings can be compared to demo parties and thus it would seem 
at first that demo sites could be considered art platforms. Upon closer inspection, how- 
ever, it becomes apparent that demo sites rather follow already existing practices instead 
of dynamically forming completely new ones, which ultimately sets them apart from art 


platforms. 


2.3.2 Demo Archives 


In the first place, Internet demo archives are a forum for product distribution but, in the 
case of old demos and platforms, they have also become sites for the preservation of digital 
heritage. Among the first and best-known projects was the Hornet archive, known best 
in the 1990s sources by its location at that time, ftp.cdrom.com. The Hornet effort started 
already in 1992, and it is currently part of the larger scene.org demo archive. In addition 
to generic demo sites there are numerous platform-specific archives: The C64 Demo Portal 
(www.c64.ch), Crack-intro specific intros.c64.org, Sinclair Spectrum site zxdemos.org and 


the Amiga oriented ada.untergrund.net, to mention just a few. 


From a research standpoint, such archives could be used in many ways: to retrieve needed 


files, to calculate different statistics on the productions, and to browse for possible material 


14 


of interest. The majority of the sites mentioned feature a web-based interface that displays 
textual information about the productions and a screenshot to get an idea of how the item 
looks like. In the course of this study the archives were simply used for downloading demos 


and to a lesser extent for finding out specific details about them. 


2.3.3 Usenet News 


Usenet newsgroups are an international Internet forum for discussion. Among numerous 
other interest groups, the demoscene has used the newsgroups for active communication. 
The two newsgroups examined in the study are alt. sys.amiga.demos and comp.sys.ibm.pc.de- 
mos, at times referred to by the acronym CSIPD. Both can be viewed in the Google Groups 
archive (Google, 2001), including statistics about the activity during the years. While very 
active in the 1990s (at times over a thousand messages per month in CSIPD), the groups 
seemed to cool down around 2002, and during the last few years the discussion has been 
only sporadic. The downturn can be explained by the growing importance of community 


websites, providing easy access and a wider variety of content in addition to mere forums. 


In the newsgroups, there is such an abundance of topics and messages (tens of thousands 
in CSIPD alone) that something had to be left out. This study focuses only on the dis- 
cussions concerning times of change: How did the community react to emerging hardware 
and software platforms? Messages containing clear opinions for or against the change were 
selected and analyzed in order to reveal the contemporary attitudes and the rhetoric used 
when trying to convince other participants (see Section 2.6 and Chapter 5). The time of 
each discussion in relation to the introduction of the innovations provides some additional 
insight to the rate and phases of adoption among the community. The majority of the work 
had already been done for an earlier publication (Reunanen & Silvast, 2009), so little extra 


effort was required to incorporate the results into the study at hand. 


2.4 Selected Demos 


The choice of demos for a study like this is a highly delicate matter. How to represent all the 
years and platforms fairly? One possible pitfall is "history written by the winners”, meaning 
that the famous party winner productions do not necessarily represent the community as a 
whole. After all, most of the demos created are not party winners nor made by the narrow 
elite. To counter these problems the decision was ultimately left to the community itself. 
Based on the popularity rankings found on pouet.net plus various party results and charts, 
a total of 117 demos and intros were chosen to be analyzed. To alleviate the strong bias 
towards top entries, a lower-ranking entry for the same platform was chosen for every 3-4 
highly regarded productions to represent the normal” level of the time, which also allowed 


for some comparison between the two categories. The aim was to represent the whole 


15 


timespan and the most important platforms evenly. The complete list of the works viewed 
is presented in Appendix B. 


DIGITAL ` 
MEMORIES 


The Best of Commodore 64 [Vol. 1] 


MINDCANDY 


Votume =: amiga demos 





Figure 2.3: Demo DVDs: MindCandy 1 (PC), MindCandy 2 (Amiga) and Digital Memories 
Vol. 1 (Commodore 64) 


Because of the lack of real hardware, some compromises had to be made in the viewing 
process. 8-bit, Amiga, Atari ST and MS-DOS productions were mainly run using emulators. 
It should be noted that even the access to contemporary computers does not automatically 
guarantee that the software would work, due to differences in hardware and software (e.g. 
device drivers). Another workaround was the use of easily viewable demo DVDs (Digital 
Memories, 2006; MindCandy, 2002, 2006) and video clips of demos, provided by sites such 
as capped.tv and demoscene.tv. The DVDs are portrayed in Fig. 2.3. 


The research method for the analysis was content analysis (discussed in Section 2.6). The 
form used in the study can be found in Appendix A. Basic information such as the year, 
platform and country of origin were collected for statistical use, but the main focus was 
on the content of the productions. No pre-assumptions were made on the types of parts 
or effects, so the categories and trends that emerged are purely based on the observations 
made. The results of the study are presented throughout the thesis, especially in Chapters 4 


and 5 in the respective contexts. 


2.5 Secondary Artifacts 


A number of other artifacts, not discussed above, are created in various scene activities. 
However, in the scope of this thesis they are considered secondary artifacts that will not 
be investigated in depth. The purpose is not to downplay their importance, but to maintain 
the focus on the most important types of digital artifacts. For the sake of completeness, 
an overview is presented here to provide the reader a wider view on the subject. The list 


16 


below is mainly based on the categories of pouet.net, with additions found in other sources 


throughout the research. 





Figure 2.4: Party items: badges, wristbands and a t-shirt 


e Computer games. Despite the traditionally problematic attitudes towards gaming 
(see Section 3.3.3), major parties these days feature game development competitions 


among others. 


e Demo tools: pieces of software made for demo development. Examples of such tools 
are converters, packers, trackers, demomakers and programming libraries. Sections 


4.5.2 and 5.4.1 contain some general discussion about these tools. 


e Group emblems like stickers, t-shirts and banners, useful for getting visibility at par- 


ties. 


e Party-related artifacts such as badges, wristbands, t-shirts, posters and votedisks 


(diskettes used for voting at a party). See Fig. 2.4 and Section 3.6.1. 
e Text files such as BBS ads, voting sheets for charts, and tutorials targeted at beginners. 


e Videos. Since the emergence of the so-called wild compos (competitions with flexible 
rules) at parties, the demoscene has increasingly produced video clips, with 3D ani- 
mations and amateur short films being among the most common types. These days 


demos are also often rendered into video clips to facilitate easy viewing. 


A detailed investigation of such plenitude is obviously outside the scope of this study. Many, 


if not most, of the artifacts mentioned have been equally omitted by other demo researchers 


17 


so far. Such omission can be attributed to the intentional focus on demos and the community, 


and partly to the relatively low visibility of such artifacts. 


2.6 Content Analysis 


As defined by Krippendorff (2004, p.xviii), ”[...] content analysts examine data, printed 
matter, images, or sounds—texts—in order to understand what they mean to people, what 
they enable or prevent, and what the information conveyed by them does.” Traditionally, the 
method has been used to analyze textual content, such as newspapers, but as the definition 
suggests, content analysis can also be applied to a wider set of media types. An important 
property of the method is its indirect nature: phenomena are studied through their manifes- 
tations in communication, taking the context of the communication into account. According 
to Krippendorff (2004, pp.29-30), the fundamental components of content analysis are as 


follows: 


A body of text, the data that a content analyst has available to begin an analytical 
effort 


e A research question that the analyst seeks to answer by examining the body of text 
e A context of the analyst’s choice within which to make sense of the body of text 
e An analytical construct that operationalizes what the analyst knows about the context 


e Inferences that are intended to answer the research question, which constitute the 


basic accomplishment of the content analysis 


e Validating evidence, which is the ultimate justification of the content analysis 


A body of text in the context of this study can be real text, such as a selection of diskmagazine 
articles or newsgroup discussions, or a set of computer demos, as discussed in the previous 
sections. The research question and the context are defined in Chapter 1. The analytical 
construct consists of interpretively reading and viewing the research material, bringing in 
my own knowledge about the practices of the community. Together with the inferences, 
that part of the study is the most dependent on the personal qualities of the researcher. Con- 
text analysis, as well as any empirical method, should be repeatable and produce reliable 
results instead of opinions, and that is why validating evidence is required. The use of 
different sources in this study can be seen as one aspect of validation: for example, news- 
group discussions often deal with the same topics as disk magazines and therefore support 
or contradict the inferences made. Other points of reference for comparison are the history 
of home computers at large (Section 3.1 and Chapter 5), and the empirical work of other 


researchers (as presented in Section 2.1). 


18 


In this study about half of the research material is textual. For such material, context anal- 
ysis is an easily applicable and a relevant method, especially since we’re interested in the 
meanings of the texts to the community. In addition, the nature of content analysis helps 
to circumvent certain problems that could be more pronounced when using other methods: 
the effects of bad recall, dealing with large bodies of text, and covering a large time span. 
For computer demos, however, the suitability can be questioned—is it feasible to apply the 
method to artifacts? To justify the use, demos must be understood as communication ob- 
jects of the community. They are not created for the sake of mere experimentation: they are 
meant to be seen, heard, and reacted to. Similarly to texts, demos contain narratives, written 


in an audiovisual language. 


19 


Chapter 3 


Demoscene Characteristics 


This chapter serves as a background for comprehending the topics that will be discussed 
later on. In addition to an overview of the phenomenon, the goal is to reveal some of 
the practices of the scene, formed over a period of more than twenty years. The artifacts 
created by the demoscene cannot be sufficiently explained on purely technical grounds, 
since various practices of the community shape them in multiple ways. An overview of the 
home computer era and comparisons to other digital communities are provided in order to 


place the scene into a wider historical frame. 


With no knowledge about the controversial and highly competitive practices of the com- 
munity, an “outsider” may easily regard demos simply as any tinkering of young com- 
puter enthusiasts. The following ironic quote by Grant Smith eloquently illustrates the gap 
(Leonard, 1994): 


Jonny looks around, confused, his train of thought disrupted. He collects him- 
self, and stares at the teacher with a steady eye. "I want to code demos," he 
says, his words becoming stronger and more confident as he speaks. "I want 
to write something that will change people’s perception of reality. I want them 
to walk away from the computer dazed, unsure of their footing and eyesight. I 
want to write something that will reach out of the screen and grab them, mak- 
ing heartbeats and breathing slow to almost a halt. I want to write something 
that, when it is finished, they are reluctant to leave, knowing that nothing they 
experience that day will be quite as real, as insightful, as good. I want to write 


demos." 


Silence. The class and the teacher stare at Jonny, stunned. It is the teachers turn 
to be confused. Jonny blushes, feeling that something more is required. "Either 


that or I want to be a fireman." 


20 


Similar divide between young computer hackers and the society can be observed in the 
frustration of the Hacker Manifesto aka. The Conscience of a Hacker written by The Mentor 
(1986). Parents, teachers and society as a whole seem ignorant and condemning in the eyes 
of an enthusiast devoted to his community. Not willing to play by their artificial rules the 


hacker finds an escape in the networks where he feels free and empowered: 


Yes, I am a criminal. My crime is that of curiosity. My crime is that of judging 
people by what they say and think, not what they look like. My crime is that of 


outsmarting you, something that you will never forgive me for. 


The attitudes expressed on demoscene forums such as message boards and disk magazines 
bear some resemblance to those of the Hacker Manifesto. In general scene members are 
evaluated on the basis of their actions and merits—not unlike the early hackers of the late 
1950s and early 1960s described by Levy (1994). Similar attitudes inside other digital 
communities were observed by Turkle (1997, pp.177—209, 234-269). 


3.1 Historical Frame 


According to a commonly held view the first demos appeared in the mid-1980s. Such tim- 
ing has been proposed by Gruetzmacher (2004), Polgar (2005, p.57) and Saarikoski (2004, 
p.193). Thus, we are dealing with a timespan of twenty-five years. To include the first 
widely available home computers, such as Apple II (1977), Commodore PET (1977) and 
TRS-80 (1977) that paved the way, we need to add almost ten years more. In the context of 
computer history such an amount of time is a considerable span: several generations of hard- 
ware and software have appeared, become popular and ultimately disappeared from com- 
mon use. Likewise, when talking about the demoscene there is quite little in common with 
the first insecure steps and the current self-conscious community with its well-established 
practices. Therefore it should be considered at all times that we are not dealing with a 
uniform ”scene”—a static community that could be frozen and dissected—but a dynamic 


phenomenon instead. 


3.1.1 Microcomputer Revolution 


Back in the mid-1970s, following the developments in the field of semiconductor chip de- 
sign, the very first commercially available 8-bit home computers appeared: MITS Altair 
8800 was released in 1975, IMSAI 8080 the same year and Apple I in 1976 (Polsson, 1995). 
The first widely available models, such as Apple II and TRS-80 (1977), became popular 
in the late 1970s. The popular Atari VCS also appeared in 1977, followed by other game 
consoles, such as Philips G7000 (1978) and Mattel Intellivision (1980) (Forster, 2005). By 


21 


the early 1980s, other companies had also realized the possibilities of the home computer 
market, and a great variety of colorful 8-bit computers started to emerge on the market: 
Commodore VIC-20 (1981) and 64 (1982), Sinclair Spectrum (1982), MSX compatibles 
(1983) from numerous manufacturers, and Amstrad CPC (1984), to name a few. For the 


first time in history, computing was available to the masses. 


Gaming was a popular hobby already on the early home computers, and illegal game copy- 
ing was a common practice. Software developers tried to fight back by using different copy 
protection schemes, but hobbyists circumvented the protections, which became known as 
cracking. The interview of Mitch/Eagle Soft in Illegal #30 (1988) reveals that organized 
cracking groups existed as early as in 1982 on the Commodore machines. The archive 
known as Apple II Crack Screens estimates that the first crack intros date back to 1981 
(Scott, 2003). The site includes a screenshot gallery of crack intros, and reveals that ini- 
tially they were no more than simple text screens displaying the handle of the cracker and 
the name of the game. Likewise, pouet.net contains a few Apple II crack intros from 1981 
and 1982. 


On the Commodore 64, crack intros became more advanced and contained even simple 
effects and music (see Fig. 3.1). The C-64 oriented intros.c64.org features over 4000 crack 
intros with screenshots in its extensive collection. Other popular types of game modification 
were trainers that provide extra help such as unlimited lives or ammunition. A plus sign 


after the game name means that it contains a trainer (Polgar, 2005, p.51). 








PRESETS : 


nr A | | 


1567... Ou CAH TALE 


TAtRO BY OOO 





Figure 3.1: A crack intro by Fairlight (1987) 


A commonly held view among demo researchers is that when crack intros started to evolve 
they became a specialized field on their own and eventually split from the cracker/warez 


culture to become what we now call the demoscene (Gruetzmacher, 2004; Saarikoski, 2004, 


22 


p.192; Tasajárvi et al., 2004, p.15). However, the actual story might be more complicated, 
since many groups continued the legal and illegal activities in parallel, cracking games and 
making legal demos at the same time. The same swappers that would distribute illegal 
software would be involved in demoscene activities as well, as can be seen for example in 
the contact advertisements of early disk magazines. The separation of the two communities 
did not happen overnight, and the ties between demosceners and crackers most likely existed 
at least until the early 1990s. 


On the whole, phenomena such as cracking and the demoscene can be seen as examples of 
unexpected use of home computers. While manufacturers had definitely expected computer 
games, BASIC programming, and productivity to be among the most popular uses, they 
could not possibly have anticipated the complex communities that emerged around their 
machines. Nor were they designing user experiences—they were building commercially 
viable technological platforms. In the diffusion of innovations theory this phenomenon 
is called re-invention, which is defined as the degree to which an innovation is changed 
or modified by a user in the process of its adoption and implementation (Rogers, 2003, 
p.180). The relationship goes both ways: new platforms are used in unexpected ways but 
the community also needs to adjust its practices to the new devices, as claimed by Lehtonen 


(2003), who saw the adoption process as a set of trials that need to be passed. 


3.1.2 Models of Scene History 


In the writings concerning demoscene history, three rather overlapping models have been 
proposed so far. An obvious first choice would be to proceed chronologically and use time 
as the main variable. Such a format is apparent in the demo histories written by Polgar 
(2005) and Tasajarvi et al. (2004, 11-26). There is nothing wrong with such an approach: 
it is precise and lets one compare the contemporary events between each other. However, a 
mere timeline of events does not illustrate the larger trends in an easily observable format, 


as can be seen, for example, in the extensive computer history project by Polsson (1995). 


Another model of scene history is presented by Tasajárvi et al. (2004) at the end of the book 
Demoscene: The Art of Real-Time. The timespan is cut into three eras: Oldskool (1980— 
1991), Middleskool (1992-1996) and Nuskool (1997—). Both oldskool and nuskool are 
terms that appear in demoscene sources too, oldskool ("old school”) referring to old com- 
puters or style and nuskool ("new school”) likewise to the contemporary style. According 
to Tasajärvi, the oldskool era was the era of Commodore 64 and Amiga 500, characterized 
by simple effects, small parties, and young sceners. The middleskool era took place on 
the Amigas and MS-DOS and was marked by design, growing parties and techno, with de- 
mosceners moving to institutes of higher education. Finally, the nuskool post-modern era is 
characterized by PCs with 3D acceleration, huge parties filled with gamers, the Internet, and 


sceners at work. To criticize the model, one could note that the threefold division is rather 


23 


arbitrary and that the model does not age well: the nuskool era would eventually need to be 
split again, because nuskool always represents the current status quo. On the other hand, 
Tasajárvi makes relevant observations on the parallel events of each era, and the humoristic 


remarks reveal that the presented model is not to be taken overly seriously either. 


While Polgar (2005) deals with individual platforms strictly chronologically, another type 
of structure is also revealed in the book, namely the hardware-based model. Simply put, 
first there is the Commodore 64, then the Amiga and then the PC. On one hand, such 
structure is clearly defined and lets one discuss the complete lifespan of a single platform, 
but on the other hand, the temporal interactions and overlaps between platforms cannot be 
revealed easily, and they would cause redundant repetition of the same details. In this thesis, 
hardware is categorized similarly in Chapter 5, but the discussion of historical trends is not 


tied to a certain computer only. 


Other thematic approaches would be equally realistic, but no such histories have been found. 
A timeline based on demo effects and styles would provide an interesting view to the de- 
velopments in artistic expression. Coupled with a technological timeline, such an approach 
would reveal hidden connections between the two domains (although it can be argued that 
the two are inseparable to start with). Technology, too, could be treated in more depth than 
mere hardware specifications: development tools and algorithms constitute another layer, 


which is at least as important and which has both constrained and facilitated self-expression. 


3.2 Demographics 


The demoscene has most often been categorized as a dominantly European community of 
young men. Here, the focus is on the factors of nationality, age and gender, in order to 
explain some of the reasons behind the demographic distribution of the scene. Addition- 
ally, some observations are made to highlight the effects such distribution has had on the 


community. 


According to two scener maps, Frappr (2005) and Andry Joos’ Weltkarte der Demoscene, 
as seen in Hartmann (2008), the core of the demoscene is indeed located in Northern and 
Central Europe. The topic has not been extensively studied by any researcher so far, but 
the numbers provided by Borzyskowski (1996) also mostly confirm the claim. The most 
active countries among the 743 Amiga demos that he analyzed were Germany (64), Finland 
(47), Australia (45), Sweden (38), Norway (38) and Denmark (26). Only the high number 
or Australian demos is rather surprising, since the country does not appear active on the two 
maps or in any other source. The number can most likely be attributed to the origin of the 
paper. The activity of the Nordic countries is observable in Borzyskowski’s study as well, 
especially when considering their relatively small population. The most common countries 


among the productions listed in Appendix B are Finland and Sweden (19 demos both), 


24 


Germany (17), France (9) and The Netherlands (9). Another indicator of the geographical 
distribution is the location of the biggest past or present demo parties: Assembly (Finland), 
Breakpoint (Germany), Evoke (Germany), The Gathering (Norway), Mekka & Symposium 
(Germany) and The Party (Denmark). 


Two notable origins of digital culture, Japan and the United States, appear underrepresented 
in the examples above. Surely both countries have had the economic and technological re- 
sources for the development of a demoscene, even more so than their European counterparts, 
but rather little has ever emerged. Examples of Japanese demoscene are anecdotal at best, 
and while the US has had some recognized groups such as Renaissance and Hornet, plus 
cracker activity (Polgar, 2005, p.48), the amount of activity has traditionally been low. The 
most believable explanation to this disparity is simply that the hobbyist culture has taken 
other forms in the two countries. The US hobbyists have a strong frame of reference in 
their long-standing hacker culture (Levy, 1994; Thomas, 1991), which can be seen as the 
forefather of the free software movement (Raymond, 2006), whereas Japanese hobbyism 
has taken the form of otaku culture marked by anime, information hoarding and equally 
deep devotion (Lamarre, 2004; McNicol, 2006). The geographical distance, especially in 
the pre-Internet era, has hindered communication between the continents, and additionally, 
technical details have restrained the interchange: for example, the Commodore 64, Atari 
ST, and Commodore Amiga were all closely tied to the television standards. The European 
PAL (50 Hz) and the American NTSC (60 Hz) systems could not necessarily run the same 


software without modifications. 


It is a relevant question whether the scene can be treated as a monoculture or not. At first, 
the practices of the community seem identical regardless of the location: parties, demos 
and groups are practically the same in all the countries involved. The international nature of 
the scene is further exemplified by international groups and communication forums. Some 
nationalism can be observed in demos in the form of flags, and national communities are 
referred to by their name, such as ”the Finnish scene”, but on the whole nationality does not 
seem to play a big role, nor is it a source of controversy. Social, economical and political 
differences between demo countries imply that the origins of the scene are different in each 
of them, but on the other hand, the established community with its well-defined practices 


seems to function as an attractor that largely hides national differences. 


The age of demosceners has been estimated by several authors, but no conclusive work 
exists. Roininen (1998, p.82) proposes a range as wide as 14 to 30 years and mentions that 
the oldest demosceners were born in the mid-1970s (representing the situation of 1998). 
As an interesting sidenote, she also mentions that the demosceners she interviewed came 
from middle-class families (p.123). Saarikoski (2004, p.191) suggests a range of 15 to 21 
years, but also mentions that "scene veterans” of over 30 years of age may still be active. 
Hugi #16 (1999) features the statistics of a study conducted among 224 readers: the biggest 
age groups were people born in 1980-1981 (33%), 1978-1979 (25%), 1976-1977 (13%) 


25 


and 1982-1983 (13%). Only a few respondents were born before 1970 or after 1983. That 
study, too, represents the situation already ten years ago, but the low number of people 
in their mid-teens is apparent, hinting at a low number of young newcomers joining the 
community. The same trend was revealed in a small-scale study of the ages of 26 randomly 
chosen scene members appearing on the Assembly'08 party photos of slengpung.com: most 


of them were already in their late 20s or early 30s. 


There is a unanimous agreement among the researchers that demoscene members are dom- 
inantly male (Kurki, 2002, p.11; Roininen, 1998, p.69; Saarikoski, 2001a). Such a phe- 
nomenon is nothing uncommon among other digital communities either: Levy (1994) ob- 
served the same among the early hackers, Thomas (1991, p.xvi-xvii) among network hack- 
ers, and Hapnes (1996) among Norwegian computer enthusiasts. All those communities 
also share the characteristics of deep devotion and high regard for technical skill. It would 
seem that computer enthusiasm is a male domain and has been so from the very early days 


of computing. But why is it so? 


Saarikoski (2004, pp.167-186) discusses the genderedness of the computer hobby from a 
Finnish point of view and notes that hobbyists reflect the overall trend of the computer 
industry: from the very early days experts have been men, although women have usually 
conducted the mechanistic work. In the 1980s families, a home computer was chiefly con- 
sidered a toy for the boys and the visual imagery of advertisements further strengthened 
the traditional setting. Nordli (2001) states that, even if the home computer was initially 
bought for all the children, the active boys would soon displace girls and ultimately end 
up as the sole users of the machine. She also mentions image reasons as one factor affect- 
ing girls’ attitudes. Kurki (2002, pp.33—46) treats the scene as a gendered community and 
claims that its competitive, rational and hierarchical nature in general does not invite female 
participants. Her study reveals prejudices towards female community members: they easily 
get categorized as mascots or “nagging bitches”, and their capability is often questioned. 
According to Roininen (1998, p.67—68), the few girls who visited parties at that time were 


mainly girlfriends of demosceners, not members themselves. 


3.3 Relationship to other Digital Communities 


Demoscene is only one community among many whose activities revolve around digital 
technology. Hackers, crackers, otakus, media artists, net gamers, game modders and free 
software advocates co-habit the digital domain and its networks. Some of them, for example 
crackers and otakus, share the exclusive underground nature of the scene, while some are 
geared towards mainstream visibility, like media artists and the free software movement. 
Here we will briefly look into the features of certain other communities in order to discuss 


what the demoscene is and what it is not. 


26 


3.3.1 Hackers or Not? 


It would be straightforward to categorize demosceners as hackers, since they share the same 
interest in computers and the underground nature. This was the approach taken by, for 
example, Borzyskowski (1996). Saarikoski (2004, pp.190—210) uses the term multimedia 
hackerism, however, consciously acknowledging the problematic definition of the word. 
The problem with the word “hacker” are its multiple meanings that vary greatly depend- 
ing on the context. In The Jargon File (Raymond, 2003) alone, there are eight different 


definitions for the word. 


The origins of computer hackers have been documented by Levy (1994) in his well-known 
book Hackers. Similar histories are presented by Thomas (1991, pp.10—15) in Hacker Cul- 
ture, and Raymond (2006) in his open source oriented The Cathedral and the Bazaar. 
Emerging at the first computer labs of American universities in the early 1960s, the first 
hackers created a lifestyle composed of technical competence and openness, which was in 
dire contrast with the exclusive professionalism of IBM and other corporations of the time. 
Levy describes three hacker types in total: true hackers, hardware hackers, and game hack- 
ers. Thomas’ book complements the timeline by focusing on network hackers from the 
1980s to early 2000s. 


Levy (1994, pp.39-49) describes what he calls the hacker ethic, meaning the philosophy of 
the first hacker community. A similar philosophy is still, after forty years, echoed in the 
contemporary open source movement, as can be easily observed in The Cathedral and the 


Bazaar (Raymond, 2006). The characteristics of hacker ethic according to Levy are: 


1. Access to computers—and anything which might teach you something about the way 
the world works—should be unlimited and total. Always yield to the Hands-On Im- 


perative! 
2. All information should be free. 
3. Mistrust Authority—Promote decentralization. 


4. Hackers should be judged by their hacking, not bogus criteria such as degrees, age, 


race, or position. 
5. You can create art and beauty on a computer. 
6. Computers can change your life for the better. 
The principles of accessibility, art, meritocracy and positive change are in line with the 
scene practices, but the rest are not shared as unequivocally. For example, the freedom of 


information in the form of collaboratively sharing program code is extremely rare. The 


dominating culture has been one of secrecy: not revealing one’s tricks and secrets to others. 


27 


To illustrate the rarity, we may consider the list of available demo source codes, published 
in Hugi #35 (2008), containing 112 items. Compared to the total amount of demos (see 
Section 4.1) we get an estimate in the order of magnitude of per mil. Also, the principle of 
decentralization is partly in contrast with the scene’s tendency to follow the commercially 


dominant hardware and software platforms (further discussed in 5.5.1). 


According to Thomas (1991, pp.xiii-xv), in mainstream culture the word ”hacker” is often 
linked to network criminals. To distance itself from illegal activities the hacker commu- 
nity has started to use another word, cracker, to describe the system breakers (Raymond, 
2003), whereas in the scene context the word has traditionally referred to copy protection 
removal, cracking. Interestingly, in the cracker study written by Vuorinen (2007), we find 
explanations to why the scene and the hacker ethics are so distinct in some aspects. Vuori- 
nen describes three software distribution models: the proprietary, the open and the cracker 
model, and claims that crackers are a product of proprietary software distribution. In con- 
trast to this, the open model springs from a different paradigm, which is why it's inevitably 
in conflict with the proprietary system. Based on Vuorinen's analysis on crackers, it is 
easier to understand the demoscene’s (with its roots in software piracy) affiliation with the 


commercial world. 


3.3.2 New Media Art 


By definition, demos are new media art: creative multimedia made with digital tools. How- 
ever, there exists a gap between the demo community and the different genres of media art. 
Even a quick glance at media art publications (Grau, 2007; Tribe & Jana, 2007; Wands, 
2006) reveals that demos are not part of the same discourse. What are the reasons for such 


separation, and are the domains drifting towards each other or rather further apart? 


One explanation to the situation can be found by inspecting the origins of the communities. 
While the demoscene is a child of 8-bit home computers and games, and at least originally a 
youth culture, other genres of media art have their roots in other domains. The technological 
forefather of the contemporary movements, computer art, established itself little by little 
in the 1960s and was initially restricted to large companies and educational institutions 
that could afford the expensive hardware, as documented by Franke (1971). According to 
Wands (2006, p.184), many digital artforms have their roots in traditional media. Among 
others, the genres of installation and video art, sculpture and performance all have their 
counterparts in the field of new media art—existing practices evolved into something new 
when augmented by digital technology. Probably the closest relatives of the demoscene in 
this respect are visually-oriented VJ (Visual Jockey) art, game art (the creative use of game 


engines), and algorithmic art (also known as software art). 


With so many genres and constant evolution it is difficult to talk about "new media art” as a 


single entity. However, for the purpose of comparison with the demoscene some generaliza- 


28 


tions need to be made. The first notable difference between demos and other forms of digital 
art is the limited scope of forms demos can assume: there are strictly defined categories of 
artifacts, pedantic competition rules and platform restrictions. The audience also has clear 
expectations concerning the content, acceptable frame rate and the style: what is demo-like 
and what is not. Another defining characteristic of computer demos in comparison to, for 
example, net art, game art and digital installations, is that they are hardly ever interactive, 
not even participative. Demos are meant to be watched, not touched, which connects them 


conceptually to digital video. 


Thematic differences, too, separate computer demos from other forms of media art to some 
extent. The demo community is markedly non-political (with a few rare exceptions dis- 
cussed in Section 4.1.3) and mainly focuses on audiovisual show-off instead of activism. 
In dire contrast with this, a great number of media art works try to convey a message and 
engage the the audience both emotionally and intellectually [for some examples see Wands 
(2006, p.26, 56, 57, 68, 106, 180)], which of course is not to say that technologically im- 


pressive audiovisual perfection would not be an integral part of other genres. 


The persistently underground nature of the demoscene separates it from many forms of 
digital art discussed above. The exclusive nature of the community tends to keep its arti- 
facts inside the borders, out of sight of the outsiders. There have been some attempts to 
bring more mainstream visibility to demos during the last few years (see 2.1), which can be 
interpreted as a growing need to receive recognition outside the community as well. Nev- 
ertheless, such activism has been scarce and, based on the frequency of the attempts, the 
trend does not seem to be on the increase either. While other forms of media art are fre- 
quently exhibited at high-profile galleries and museums, or reach a wide audience through 


the Internet, the demoscene remains both the main producer and consumer of its artifacts. 


3.3.3 Computer and Video Games 


The history of the demoscene is inseparably connected to computer and video games. Early 
home computers, often used for playing games, and the related software piracy, as argued 
above, were the necessary prerequisites for the existence of the demoscene of today. Tech- 
nological developments driven by the multibillion-dollar game industry have facilitated de- 
moscene activities in the form of increasingly multimedia-capable hardware and content 
production tools. Naturally, demosceners also play digital games. Despite all of this, the 
attitudes towards computer and video games have been ambivalent and problematic for a 


number of years. 


Starting from the early 1990s, the demoscene has clearly distanced itself from computer 
game players. The change is apparent in, for example, disk magazines, which used to con- 
tain game reviews and hints until the early nineties. Later on, the number of game-related 


articles dropped close to zero (see Section 4.6). The change took place in parallel with the 


29 


growth of the self-consciousness of the community, and can be explained by the increased 
need to emphasize the uniqueness of the scene, and thus separate it from other communities. 
According to a popular attitude, reflected in several diskmags, the sceners were considered 
as skillful and creative, whereas game enthusiasts could not do anything else than play with 
their computers. To complete the ambivalence, the amount and quality of games avail- 
able have been repeatedly used as arguments—both for and against—in the opinionated 
diskmag articles and Internet discussions about emerging platforms. The following com- 
ment by StyX/HeadcrasH taken from Hugi #23 (2001) recaps the scene mentality rather 


well: 


I’m talking about our beloved species of gamers. Well, the hardcore-gamers, 


those people we usually call more exactly lamers. 


A number of demo-related publications mention the connection between the demoscene 
and game companies (Kauppinen, 2005; Saarikoski, 2004, p.205; Saarikoski & Suomi- 
nen, 2009; Scheib et al., 2002; Tasajárvi et al., 2004, p.23). In the articles, the scene is 
sometimes even described as ”a pre-school for the game industry”. The article about game 
companies employing demosceners, written by Beck (2008), mentions twelve high-profile 
game houses, so it would indeed seem that real-time multimedia-oriented programming, 
pixeling and composing skills required in the demoscene are valuable assets in game devel- 
opment. Saarikoski (2004, p.205) also mentions three game companies (Digital Illusions, 
Housemarque and Remedy) founded by demosceners, but goes on to say that to most of 
its members, the scene has been a serious hobby and a way of life, not an investment in a 
future career. The source material of my study does not reveal such tendencies either, so 
the connection to the game industry could be better considered as a byproduct instead of a 


direct motivator. 


The complex nature of the attitudes towards gaming is clearly visible in the source material: 
while game players were criticized harshly on many occasions, game development hardly 
ever received negative attention. At times there was even notable excitement over the games 
developed by former scene members—once a member, always a member. Another related 
factor is that gamers and sceners co-habit the same hobbyist domain, whereas the game 
industry operates in a very different dimension altogether, and thus will not be considered a 


competing community. 


3.4 Handles and Groups 


The vast majority of all demos are created by groups. As an example, almost all the demos 
viewed for the study were published by a group (Appendix B). On the lowest level, an 


individual is a member of a small tightly-knit community of the group, and in a wider 


30 


scope, a member of a much larger community such as "the Amiga scene”, "the PC scene”, 


and ultimately the combination of all of them, the scene. 


Most demo-related publications acknowledge the existence and structure of demo groups. 
However, hardly any of them have tried to go beyond the easily observable surface, or to 
explain what actually are the meanings and reasons behind such hierarchy. Usually, we 
only get to read about the division of labor: there are coders, musicians and graphic artists 
who create demos together. As of now, the most notable works dealing with the identity of 
the demoscene members are the theses by Kurki (2002) and Roininen (1998). Kurki uses 
the concept of a micro community to describe demo groups, referring to the small, devoted 
communities that are formed inside the larger postmodern tribe, in this case the demoscene. 
According to Kurki, the members of such exclusive micro communities tend to share the 
same interests and strongly identify themselves with the community (pp.26-29). Roininen 
(1998, pp.38-44) emphasizes the organized nature of the scene and the tight group work 


involved in the demo creation process. 


3.4.1 Taking a Name 


Demoscene members are best known in the community through their handles, also known 
as aliases. Examples of people using their real name in productions have popped up at 
times during the study, but in general such cases have been scarce. Likewise, groups al- 
ways have a name. The demoscene inherited its naming conventions from its forefather, the 
cracker/warez scene. According to Polgar (2005, p.40), crackers used handles to distract 
law enforcement and hide their real identity. Sex’n’Crime #20 (1990) supports the claim by 
discussing the reasons why people changed their handles: fear of law enforcement, suspi- 
cion of the post office and a tarnished name. Nevertheless, such reasons do not sufficientl y 
explain why the more legally-oriented scene would still retain the practice, hence there must 


be more to it. 


On a practical level handles are frequently represented in relation to the group: for exam- 
ple, Crust of Appendix. Another way of expressing the same relation is with a slash like 
Crust/Appendix, still pronounced ”of”. Long group names may be abbreviated to two or 
three letters for the sake of saving time or space (for example in filenames), so the exam- 
ple handle might end up looking like Crust/APX as well. An fundamental property to note 
here is the strong tie between the individual and the group: the notation itself reveals that a 


member is ”of” the group, a part of a bigger, clearly defined whole. 


Kurki (2002, p.48) observed that handles (or nicks as she calls them) are often related to sci- 
ence fiction or fantasy literature and movies, genres that are of interest to young males. The 
same orientation is apparent in the visual art of the scene as well: especially in the 1990s 
fantasy was among the most popular themes (see Section 4.4). Another category of handles 


mentioned by her are the variants based on the real name of the person. The international 


31 


nature of the scene is illustrated by the language of the handles, which is mostly English 
(Roininen, 1998, p.18). The topic has appeared a couple of times in disk magazines: Im- 
phobia issues #8 (1994) and #12 (1996). In the former, Zeb/Zuul Design rather humorously 
categorizes aliases in the following way: first names, shorts (like abbreviations), lamer- 
aliases (stupid names), function-handles (that describe the person’s activity), gloomy/scary 
and with imagination (exceptional names). In the latter, Phoenix/Hornet sheds some more 
light on the concept of handles from a scener’s perspective. He seconds Polgar when dis- 
cussing the origins of handles by mentioning the privacy concerns of early crackers. He also 
presents arguments for/against the use of handles and, interestingly, commercial reasons are 
brought up as a possible benefit of using real names: companies are more likely to contact 


people who go by their real name. 


Taking a name serves two major purposes that are not practical but highly personal instead: 
self-expression and power. A name represents the individual and his identity in the com- 
munity. Instead of hiding from the police behind an alias, a demoscener brings himself 
forward and makes a statement, no matter how small, to the community by selecting a han- 
dle. The observations of Kurki (2002, p.48) and Roininen (1998, p.90,106) also suggest 


that the expression of one’s identity is a fundamental reason for choosing a nickname. 


Similar practices exist in various other contexts (e.g. online communities, the graffiti scene 
and hacker culture). Faler (2001) compared the demoscene with the graffiti scene and found 
a number of reminiscent properties: handles and groups were present, but also other prac- 
tices such as phases of entering the community and the networked nature of the hobbies. 
Roininen (1998, pp.107—110) used hip-hop culture for her comparison. Handles and groups 
were present in the hip-hop circles as well. Network hackers, too, use handles and form 


groups with names, as documented by Thomas (1991, p.58, pp.90-91). 


3.4.2 Group Dynamics 


The three most visible roles in demo groups, apparent in artifacts such as demos and intros, 
are the coder (programmer), the musician and the graphician (graphic artist), each repre- 
senting a certain type of activity needed for creating demos and other artifacts. Each role and 
its respective duties are well defined, although there is also room for movement. Numerous 
other roles can be found in source material such as ExoticA! (Lunder, 1996), each repre- 
senting a different task related to the activities of a group: ASCII artists, crackers, design- 
ers, editors, leaders, organizers, support/hangaround members, swappers/traders, raytrace 
artists, suppliers (of software), SysOps (BBS operators), and webmasters. A similar—even 
if much more limited—taxonomy of duties is apparent practically in all the writings on the 
topic: Burger et al. (2002), Gruetzmacher (2004), and Roininen (1998, pp.15—17). 


Figure 3.2 was compiled using the productions listed in Appendix B in order to approx- 


imate the number of people involved in the creation of a demo. It should be noted that 


32 


demos 
A 


20 


10 








= 
authors 





Figure 3.2: Number of authors in demos 


the extreme ends are somewhat emphasized by crack intros made by one person only, and 
large megademos that were collaborations between multiple groups. While groups at their 
peak can consist of dozens of members, it seems that demos and intros are typically created 
by only a few authors. Such disparity suggests that in the demoscene, small workgroups 
are dynamic and productive: they do not require extensive communication or coordination. 
In the domain of software production a similar effect was observed by Brooks (1975), who 
claimed in his well-known essay The Mythical Man-Month that adding new programmers to 
a project running late will actually slow down the progress even more, due to the increased 
amount of communication and the time lost in introducing the project to new participants. 


3.5 Competition and Fame 


The demoscene is in a constant state of competition, which is exemplified by numerous 
practices of the community. To get to the top and acguire fame one needs to impress, 
win, and be connected. Kurki (2002) discusses the same phenomenon and compares the 
scene (a male-dominated culture) with deco swapping (a female-dominated culture) from 
the standpoint of competition. According to her, both communities are competitive, but 
in a different way: in the scene competition is made visible and accepted, whereas deco 
swappers compete under the hood and do not criticize others’ works directly. 


33 


3.5.1 Striving for Fame 


Probably the most visible manifestations of the constant state of competition are the compos 
(competitions) held at parties. The winning productions, typically the best three, receive 
prizes and, more importantly, visibility inside the community. Winning competitions at big 
parties is a fundamental way of growing the fame of the individual and the group. Other 
equally explicit manifestation of competition are the charts that used to be an important 
part of disk magazines. In the very first diskmags (Sex’n’Crime as an example) the rankings 
were decided by the editors, but soon popular vote established itself as the standard. The 
public wrote down their favorites from different categories on voting sheets, sent in the 
sheets, and ranking lists were compiled based on the votes. A high position on the charts 
was both a result and a source of fame: on one hand, popular groups and artists got more 


votes, and on the other hand, the visibility contributed to their fame. 


Fame can be acquired through several means. All sorts of participation in community ac- 
tivities, be it discussion on disk magazines or online forums, attending parties, or swapping 
productions, increase the visibility of the group. Technical skills are admired, as we have 
observed already, but the most known and highest-ranking groups have one more thing in 
common in addition to mere skills: they are well connected. In demos it is common to send 
greetings (greets) to other respected groups that are connected to the authors in some way. 
It is hardly a surprise that, in the demos observed, the top groups know and greet each other, 
thus mutually building their prestige in the community. For an example of greetings see 
Fig. 3.3. In the example, the greetings have almost become a mechanistic tool: the long 
list of groups is sorted alphabetically, spans multiple pages, and the high-contrast names are 


displayed for quite a while. 


Competition on the whole is a typical feature of male-dominated youth cultures. Similar 
patterns have been observed in other computer enthusiast communities by Hapnes (1996, 
p.136), Levy (1994, p.115-118), Thomas (1991, p.xvi), and Turkle (1984, p.210, 231). An- 
other common denominator is that skill is held in high regard. Already the first hackers tried 
to improve existing code snippets by inventing ingenious programming tricks, rewarded by 
the approval of the community (Levy, 1994, p.44). Both Kurki (2002, p.43) and Roininen 


(1998, pp.65—71) name direct competition as a factor in the genderedness of the demoscene. 


3.5.2 Elite and Lamers 


The social system of the demoscene can be described as a form of meritocracy: recogni- 
tion is distributed based on the achievements of the individual and the group, not evenly. 
According to Levy (1994, pp.115-118), a similar ranking existed already among the MIT 
hackers, who divided computer users to winners and losers. Accordingly, the demoscene 


has two categories that describe the high performers (elite, ”leet”) and the lowest ranks 


34 





Figure 3.3: The greetings part of Eclipse by Electromotive Force 


(lamer). Already the disk magazines of the late 1980s, such as Sex’n’Crime, featured such 
wording. An illustrative example of the past exclusive nature of the elite circles was written 
by Kauppinen (1991). To call somebody a lamer was a serious insult because of all the nega- 
tive connotations associated with the term. Quoting the editor’s reactions from Sex’n’Crime 
#21 (1990): 


Dear Roy of Dynamics, let me say this from the bottom of my heart: you are 


lame! 


Numerous articles in diskmags have been dedicated to discussions about lameness: who is 
lame and what are the characteristics of a lamer? On the other hand, the essence of elite has 
not received even nearly as much attention. It can be assumed that ”elite” is something that 
does not require much defining and, additionally, the concept of lamer is more interesting 
since it clearly represents a lower rank. Nobody would want to be a lamer and therefore 
it is important to distance oneself from them, by putting them down and defining them as 
something else than oneself. Some quotes illustrate the efforts the scene members have 
taken to define a lamer: 


The most typical lamer type is the guy in a group nobody knows because he 
has no contacts, moreover no coders, musicians, gfx... But this kind of lamers 
would like to be famous. But he isn’t because he can’t do anything. (Brain- 
wawe, Zine #02, 1989) 


The first is the lamer who rip all, because he can’t do something alone. The 


first is a very dirty race of lamers because they injure the scene. The other is 


35 


a person who use his computer only to play games. (Zorlock, Imphobia #1, 
1992) 


The people with no interest in the scene and in demos, scene music and graph- 
ics whatsoever — the, I write and mean the word honestly, LAMERS. (Curt 
Cool/Depth, Hugi #19, 2000) 


While there seems to be a general consensus that lamers exist and are bad for the scene, 
the opinions towards newcomers (newbies) are more divided. Some sceners, when describ- 
ing a lamer, actually refer to the characteristics of a newcomer who has not yet learnt the 
necessary practices and skills. Some argue that the scene should actually support newbies 
to help them develop themselves and that everybody has been a lamer or newbie at some 
point before improving their rank. All in all, it is apparent that the lamer/elite division has 
lost quite a lot of its importance during the years. In early diskmags the words appeared 
constantly and little by little they became less common. The change can be attributed to at 
least two factors: firstly the rise of the friendship attitude in the early nineties as counter- 
action against the harsh atmosphere of the scene, and secondly, to the increasing age of the 


community members (see Section 3.2). 


3.6 Social Networking 


Even though the social dimension is largely omitted in this study, some aspects need to be 
discussed due to their fundamental relevance: after all, the demoscene is a social network 
more than anything else. The sections above have already outlined demographic factors and 
revealed some facets of the dynamics and hierarchies of the scene. In this section, the focus 
is on the different communication channels that the scene members use for keeping in touch 


and discussing topics of interest with others. 


The communication channels of the demoscene can be compared to the diffusion networks 
in Rogers’ innovation diffusion theory. Such networks are structured, somewhat stable 
and link individuals by patterned flows of information. A central concept is is the ho- 
mophily/heterophily of the network: the degree of similarity of a pair of individuals in 
terms of beliefs, education and socioeconomic status. Generalization 8-12 states: ”Indi- 
viduals tend to be linked to others who are close to them in physical distance and who are 
relatively homophilous in social characteristics”. Homophilous communication is likely to 
be more effective, but on the other hand heterophily is needed to introduce innovations to 
the system. Opinion leaders also play an important role in the system, since their credibility 
helps to diffuse innovations to the less adventurous adopters. (Rogers, 2003, pp.300-364) 
Demoscene networks such as swappers’ networks, interpersonal connections, and commu- 


nication forums can all be thought of as layers of a large diffusion network spanning the en- 


36 


tire community. In this context, innovations are new effects, programming methods, styles, 


attitudes or practices that diffuse through the communication channels. 


3.6.1 Parties 


A party is a meeting place for demosceners and a venue for various activities. Parties have 
existed in some form throughout the entire history of the scene, although their purpose, 
form and magnitude have changed significantly during that time. Most researchers who 
have dealt with the history of demo parties agree that they were preceded by the piracy- 
oriented copyparties, where cracker groups got together to swap software and meet each 
other (Polgar, 2005, p.60; Saarikoski, 2001a; Stamnes, 2004, p.45). When studying disk 
magazines, the origins and importance of parties were confirmed: already the Commodore 
64 disk magazine Sex’n’Crime contained several references to the Venlo copyparty held in 
Germany in the late 1980s, and other diskmags discussed parties a great deal as well. While 
parties are often mentioned in demo-related publications, they have not been investigated 
in depth by other authors than Hege Nordli, who published an article written from a gen- 
der study standpoint and whose dissertation contains a chapter on The Gathering, a large 
Norwegian party (Nordli, 2003a,b). 


In the early 1990s, parties increased considerably in size. Large meetings rapidly expanded 
into yearly happenings with thousands of visitors. For example, Stamnes (2004) mentions 
that in 1994 almost 1800 people visited The Gathering. According to the party database of 
pouet.net, other big parties of the early 1990s were the Finnish Assembly, Swedish Com- 
puter Crossroads, and Danish The Party. The Party eventually faded away in 2002, and 
Computer Crossroads did not continue past 1994. The other two still exist, even though in a 
rather different form. Another large party appeared in 1997 when the earlier German Mekka 
and Symposium meetings joined forces to become Mekka & Symposium. In 2003 the party 
ceased operation and was replaced by the yearly Breakpoint, also in Germany and still in 
operation. A notable difference between Breakpoint and other big happenings is its dedi- 
cation to pure demoscene activities: "No gamers, no script kiddies, only creative people” 
(Breakpoint, 2009). 


To categorize demo parties by their size the following taxonomy was devised: 
e Meetings. A small-scale informal get-together of local sceners and their affiliates. 


No competitions or formal program. Up to tens of visitors, might even have a name. 


Often organized at private premises. 


e Small parties. A formally organized happening with competitions, schedule and a 
group of organizers. May attract up to hundreds of visitors, even participants from 


other countries. Nowadays parties of this magnitude also have a website. A nominal 


37 


entrance fee is collected. Typical premises for small parties are for example schools 


and youth centres. 


e Big parties. A large-scale international happening with up to thousands of visitors, 
taking place in a building as large as an ice hall or a fair center. Needs a large group of 
organizers with specialized hierarchies like security and network teams. Considerably 
high entrance fee and high-profile sponsors. Extras such as posters and t-shirts are 


sold. Fig. 3.4 depicts how the main hall looks like at a big party. 





Figure 3.4: Assembly’05 main hall 


When parties grow as big as the last category their nature changes considerably. Advertis- 
ing, budgeting, planning, and sponsorship negotiations require a considerable investment of 
time and effort. A controversial theme appearing in numerous disk magazines has been the 
tendency of large demo parties to drift towards game-oriented LAN parties. From an eco- 
nomical standpoint such development is understandable: computer games bring in sponsor 
money, and game enthusiasts are a large audience. It is questionable if the current de- 
moscene alone could even provide for several large happenings a year. On the other hand, 
the trend has been a constant source of contempt. For example, as seen above, Breakpoint 
clearly distances itself from the game parties. Already in 1995, Diesel8 observed the devel- 
opment in R.A.W. #9: 


Last year’s Assembly was probably the definite break-through for the PC- 
scene. Hardly anyone had brought their Amigas, but a lot of people were 
playing Doom on PC. This trend was even stronger this year — the handful 


Amigas were either left alone playing some old modules, or turned off. Com- 


38 


pared to Party IV, Assembly is definitely less scene-party and more game-geek- 


gathering. 


Competitions (compos) are a fundamental part of the party culture and a good example 
of the competitive nature of the community. Different competitions let the groups plus 
individual programmers, musicians and graphic artists show their skills to the audience. 
The competition winners also receive prizes in the form of money, software and hardware. 
The most important prize, however, is the added fame and visibility of the group and the 


artist. Some of the most common types of competitions are: 


e Demo compo. Traditionally the most important competition with the highest prizes. 
e Intro compos. Size-limited intros, such as 64k and 4k intros. 


e Music compos. Different categories such as instrumental, freestyle and four-chan- 


nel/multichannel music exist. 


e Graphics compos. Different categories such as raytraced, pixeled, ANSI, thematic 


and freestyle graphics exist. 


e Wild compo. A competition with loose rules allowing exotic hardware, videos and 


even per formances. 


At times, competitions are split into smaller ones based on the hardware/software platform. 
Numerous other types of competitions can be found on party results and invitations: banner, 
fast, game development/playing, short movie and even sports competitions have taken place 
at parties. A fast compo refers to the tight time limit set for the creation of a complete work 


from scratch. For detailed discussion on the different types of artifacts see Chapter 4. 


Competition results are decided by a public vote. A jury often reviews the entries before 
they are shown to ensure that they conform to the competition rules. Originally, the voting 
was conducted using voting sheets or diskettes (”votedisks”) and more recently it has been 
moved to the party network. Figure 3.5 portrays the user interface of The Party 1994 voting 
diskette. The voting process is not strictly controlled, which has at times lead to accusations 
of so-called faking or fake voting, where the results have been skewed. Charts have also 
been faked as seen already in Sex’n’Crime #21 (1990). Such accusations and discontent 
with crowdpleasers (technically unimpressive productions that appeal to the audience) are 


somewhat common after parties. 


3.6.2 Swapping, Trading and Spreading 


When new productions are created, they must be spread to the rest of the community. Tradi- 


tionally, a swapper was responsible for the process and sent diskettes by mail to his contacts 


39 





Figure 3.5: Voting diskette by Nuts from The Party 1994 


in order to distribute the production through the interpersonal network. Swapping evolved 
into its own specific and demanding field in the 1980s, featuring factors like advertisements 
(found on most diskmags), stamp faking to save on postage costs (Polgar, 2005, p.52), and 
competition for most contacts and the newest ”stuff” (Tasajárvi et al., 2004, pp.13-14). 
At times, the word trading has been used separately from swapping to denote modem- 
based distribution. Since the mid-1990s, the Internet started to increase its importance as 
a distribution channel and by 2000 it had practically made the earlier channels obsolete. 
Swapping/trading was international right from its earliest days, following the tradition set 
by cracker groups. 


Demos are distributed for free: a fact so simple and obvious that it is almost too easy 
to omit it without further thought. With the exception of possible competition prizes, no 
money whatsoever is made out of the productions. In this sense demoscene encourages free 
and equal sharing, although the line is drawn at the source code of the demos, which usually 
are not published (see Section 3.3.1). The models of distribution have obviously not been 
inherited from the commercial world nor the hacker/free software paradigm. The closest 
equivalent and a direct predecessor is the cracker model that is based on free (although 
illegal) software trading, with honor being the main currency (Rehn, 2004). 


Some curious counterexamples do exist, however. While some demo groups proceeded to 
game programming (Section 3.3.3), some actually made commercial demos for companies. 
Saarikoski (2004, p.210) noted that the Finnish demogroup Future Crew made demos for 
SSI and Creative Labs. Gruetzmacher (2004) mentions another demo from the same group, 
made for Waite Group Press and also an Afri Cola demo by the German group Farbrausch. 


40 


The homepage of the Dutch Ultra Force demogroup lists as many as five commercial demos 
(Ultra Force, 2007). Other examples can be found on pouet.net: Swedish Triton created a 
commercial presentation for Gravis UltraSound cards, and OS/2 promotional demos exist 
at least from three groups. Obviously, some of the top democrews had the intention of using 
their skills for commercial demo production, too. An oddity among the various subscenes 
is the isolated Dutch MSX2 scene that actually followed a commercial model: according 
to comp.sys.msx discussions and Szarafinski (1995), demos were sold to other groups at 
fairs, which indicates that the scene in question was in some respects more aligned with the 


commercial MSX2 world than with the worldwide demo community. 


3.6.3 Online Interaction 


Disk magazines, despite their digital nature, are still a physical medium that needs to be 
handed over to the next reader, and thus interactive communication between people is 
slow at best. Electronic communication media such as BBSs (Bulletin Board Systems), 
Usenet newsgroups and, more recently, web-based discussion forums offer a faster cycle 
and thus enable almost real-time conversation on topics of interest. Another difference is 
that diskmags are edited, require full articles to be written and therefore are at least some- 


what moderated, whereas the online media let almost anything through. 


Demo-related BBSs (boards) would be an interesting source of early online discussion, but 
due to their scattered nature their message bases are hard to access these days, and a signifi- 
cant amount of material may have been completely lost already. An early effort of bringing 
the demo-related boards closer to each other was the formation of Creativity Demo Network 
(Imphobia #12, 1992), which synchronized the member BBSs with each other. Soon there- 
after, Usenet newsgroups like comp.sys.ibm.pc.demos started to gain ground as important 
forums (see Section 2.3.3). Old newsgroup discussions are significantly easier to access in 
comparison to bulletin boards, thanks to the Google Groups archive (Google, 2001). Fi- 
nally, in the 2000s, newsgroups were chiefly replaced by general web-based forums such as 


pouet.net and platform-specific or thematic sites. 


When comparing Usenet groups, such as comp.sys.ibm.pc.demos or alt.amiga.demos, to the 
pouet.net forum, it is easy to notice differences in the tone and topics of the conversations. 
The newsgroups are somewhat public and official forums, accessible to anybody, whereas 
Pouet can be thought of as a private corner of the demoscene. Coupled with the more serious 
nature of the community in the 1990s, it is easier to understand the differences. While 
newsgroup discussions chiefly consist of serious topics and debate (even fights), the Pouet 
forums are more geared towards pure pastime (which is not to say that real topics would 
not be debated as well). Examples of the laid-back discussion are the numerous "Random 


x threads”, where x can vary from anything between ”girl with hardware” to ”line of code”. 


41 


Even serious threads are usually marked by humoristic remarks and trolling (controversial 


statements intended to heat up the discussion). 


As can be observed above, the communication media are in a state of constant change. 
Disk swapping was challenged by modem trading, hand-written letters by BBSs, BBSs 
and disk magazines by the Internet and so on. Following Marshall McLuhan’s thinking, 
new media inevitably and irreversibly shape the society and its practices (McLuhan, 1994). 
The developments in communication technology have certainly shaped the demoscene, too, 
bridging its geographical gaps and creating a sense of immediacy not possible twenty years 
ago. As an example of this change, a demo may appear on a website the same day it was 
finalized and it will immediately be available to any community member with an Internet 
connection. Debate on the merits of the production starts at once. Demos have been spread 
and debated since the beginning of the demoscene, but such immediacy has made the world 


a lot smaller in the spirit of the global village, as called by McLuhan. 


3.6.4 Language 


English is the lingua franca of the scene, as can be noted in practically all the source material 
of the study. Such a situation is hardly surprising due to the dominant status of the English 
language in the field of computing, popular culture and education. The international demo 
community (originally the cracker community) needed a shared language and chose the 
common denominator that more or less everyone understood. Even magazines such as 
Illegal and Hugi that started as 100% German ended up being published in English to better 


serve the international community. 


Like any other established community, the demoscene too has its slang. The glossary on 
page vii, although far from exhaustive, contains some of the most common slang words 
used by the community. Some words are demoscene-specific, while some originate from 
common computing slang or the hacker community, as documented in The Jargon File 
(Raymond, 2003). Learning the shared language is part of the initiation process a new 
member needs to undergo to be accepted by the community. While English slang is a 
common denominator of the community as a whole, the national subscenes also have their 


own words. 


The so-called leetspeak, common in the hacker culture (Thomas, 1991, pp.56-61), has ap- 
peared in the scene circles at times. The system basically consists of replacing letters with 
numbers and other symbols. Additionally the words may be twisted to resemble their pro- 
nunciation and their case may be changed arbitrarily. For a more detailed description see 
Mitchell (2005). Elite hacker wares might end up looking like 1337 h4x0r w4r3Z. In ad- 
dition to mere wordplay and style, such language is a way of consciously distancing the 
community from the outsiders. Mitchell suggests that it was also a practical method for 


circumventing BBS rules that banned certain words. Polgar (2005, p.41) also claims that 


42 


such language was used by software pirates to evade law enforcement. In the demoscene 


context leetspeak has appeared for example in infofiles. 


3.7 Self-Reflectivity 


Self-consciousness of the demoscene is revealed by the high amount of reflection that takes 
place in its discussions (see Section 4.6.2). The community is clearly aware of its own 
existence, borders and dynamics. It has also taken a name for itself, the scene, to emphasize 
its uniqueness. External factors, such as changing society and technologies, constantly 
impose new challenges on the demoscene, which must react in a way or another. Numerous 
debates on diskmags, web forums and newsgroups have been dedicated to the future of 
the scene in a changing world. Such discussions already feature a meta level: the scene is 
referring to itself as a high-order entity instead of only its concrete manifestations such as 


productions and parties. 


Reoccurring themes during almost twenty years have been the looming death of the scene, 
platform wars, increasing hardware requirements, lameness, gamers and competition rules. 
Most of them can be reduced to the binary juxtapositioning of good/bad such as Amiga/PC, 
skilled/beginner, elite/lame, demos/games. The elite/lame distinction is discussed further 
in another article (Reunanen & Silvast, 2009). This way—at times almost violently—the 
community collectively evaluates its attitudes towards external changes that threaten the 
established and safe status quo. In a sense the process is democratic: anybody can try to 
affect the public opinion by participating in the debates, although the views of high-ranking 
members are likely to weigh more than the others. The following quotes illustrate the 


controversial nature of the process: 


So realize it— wake up! The future you are dreaming of does not exist! Stop do- 
ing AGA stuff and the splitting of our beloved Amiga scene! (Rufferto/Covert 
Action Team, R.A.W. #6, 1993) 


Also to all coders: Do not try to make more than you can handle! I really 
hate to see some simple vectors or other effects requiring a 66MHz Pentium or 
similar!! (Thomas/S!P, Imphobia #8, 1994) 


Nothing that Windows95 does can’t be done (much better) with dos, and what is 
more important, I get the complete control over the machine (with no overhead 
even): a thing that Windows’ Direct-xxx will never really give. (Fabio Bizzetti, 


comp.sys.ibm.pc.demos, 1996) 


Even though the opinions stated in a debate might be completely opposite, both sides still 
share a common goal: they try to support the continuity and blossom of the community. 


For a nostalgic member changes may seem like threats to the very existence of the scene 


43 


(frame of reference in the past), whereas progressive members may want to ensure the 
longevity by keeping the scene accessible to next generations (frame of reference in the 
future). Interestingly enough, nostalgic attitudes can be found in disk magazines as early 
as in the early 1990s. Young demosceners who had started their career perhaps a few years 
earlier complained how the scene had changed for the worse and how "back in the day” 
things were better. On one hand, such tendency illustrates the power of tradition, and on the 


other hand, it serves as an example of how the scene constantly evaluates and defines itself. 


44 


Chapter 4 


Demoscene Artifacts 


The activities of the demoscene revolve to a great extent around its various artifacts. How- 
ever important, the artifacts are not created merely for the sake of experimentation or self- 
expression: they are, probably more than anything else, objects of communication meant to 
be seen by an audience. Most of the works are digital by nature and thus infinitely copy- 
able, which has several implications on their creation and distribution. This chapter deals 
with various objects produced by the scene from a historical and technological standpoint, 


ultimately trying to reveal the reasons why the artifacts are created as they are. 


The categories presented here are based primarily on the taxonomy used in the ”Prods” sec- 
tion of pouet.net. On one hand, the power of such taxonomy is that it is constantly evaluated 
by the community, and thus represents its views on the artifacts, but on the other hand, for 
an outsider it may not mean anything at all. Very similar kinds of taxonomies were present 
in other sources such as disk magazines, charts and competition results, which confirms that 
the community shares the naming conventions at any given time, while historically the same 
claim might not hold true. The overall impression one gets is that the scene is attracted to 
precise categories. An artifact is a demo or an intro but nothing in between. Such a ten- 
dency shapes the outcomes more than the other way around: instead of flexibly adapting the 
categories to accommodate any kind of work, the artifacts are made so that they readily Fit 


in the existing model. 


Over time the plurality of artifacts produced has grown little by little, which can be noticed, 
for example, in disk magazines and party rules/results. While early sceners were happy (to 
simplify a little) with cracks, diskmags and demos, the current variety includes different 
types of music and visual works, demos of multiple kinds, diskmags, packs, videos and so 
on. This development is by no means surprising: the community itself has grown more 
sophisticated and appropriated or invented new forms of expression, supported on a lower 


level by the concurrent developments in software and hardware. In this case, the word 


45 


appropriation refers to the absorption of external practices from other contexts such as the 


popular culture, which too is under constant change. 


4.1 Demos 


To distill the essence of demos, we can define them as real-time multimedia presentations. 
Sometimes they have been likened to music videos, too, in order to explain the phenomenon 
to outsiders (Lónnblad, 1998, p.4; Saarikoski, 2001a). Demos can be considered the most 
integral artifact of the community, as illustrated by its very name: the demoscene. The 
most defining property of demo-like artifacts is their size. Smaller types such as 64k or 4k 
intros limit the means of self-expression, since they do not allow the authors to use pre- 
rendered music, video clips, or good-quality still images. Full-blown demos, on the other 
hand, allow for such content and therefore the outcome is restricted only by the creativity 
and the skills of the authors (as opposed to a tight technical limit). The size limit for a demo 
has increased over the years due to many reasons, such as increased screen resolutions, type 
of music, larger storage media and faster data communication networks. In general, it could 
be stated that the size limit has always been so comfortable as to let the artists focus on the 
content instead of size optimization. To give a few examples, some of the size limits at the 
yearly Assembly party have been as follows: 4 MB (1994), 6 MB (1998), 20 MB (2006) and 
64 MB (2008) (Assembly Organizing, 1994, 1998, 2006, 2008). 


The amount of existing demos is hard to determine. The estimate proposed here is based 
on the pouet.net website, since it represents all the demoscene hardware and software plat- 
forms. As of May 2009, the site contains 51,858 productions, which serves as a starting 
point for the estimate. The figure includes other types of artifacts too, such as disk maga- 
zines, videos and games, but it can rather safely be assumed that the majority of the pro- 
ductions are demos of some sort (demos, 64k intros, 4k intros etc.) since they represent the 
most common productions published at parties (still images and tunes are not listed). Surely 
not all the demos in existence are listed on the site due to numerous reasons: unknown pro- 
ductions may not be interesting enough to be added, some of them have never been officially 
published, some may have disappeared altogether, and simply because the community has 
not yet taken the trouble of adding them. If we assume— very conservatively—that for ev- 
ery demo on the site there exists another that is still missing we get a rough approximation 
of 100000 demos. As the actual number may be twice or three times as high, the order of 
magnitude alone reveals the vast amount of artifacts we are dealing with and the effort that 


has been invested into the demoscene activities. 


46 


4.1.1 Megademos and Trackmos 


A megademo is a large multi-part demo, as opposed to early one-screen intros. According 
to pouet.net, the majority of productions called ”megademo” were released in the late 1980s 
or early 1990s, especially for Amiga computers, after which the word practically fell out of 
use. The newer appearances use the word either to emphasize their retro style, or ironically, 
when the production is something else in reality. Regardless of such change, even many 
modern demos could actually be defined as megademos, since they feature several separate 


parts and are large in size. 





Figure 4.1: Red Sector Megademo (1989) 


The original purpose of a megademo was to be a massive tour de force of a group, displaying 
its skills in different areas like programming, music and graphics. When considering a tra- 
ditional megademo, it most often consists of a sequence of unconnected parts each showing 
one important effect or other content such as a still image or animation. A loading screen 
between the parts is another common property of old megademos. The newer trackmo style 
demos of the early 1990s loaded the next part already while showing the current one and did 
not need loading screens. The continuous trackmo style became standard and still remains 
so, although modern day productions do not necessarily need to load anything from the disk 
during the execution. Technically, ”trackmo” refers to loading data directly from diskette 


tracks using a trackloader subroutine, bypassing the slow operating system file handling. 


Most modern demos do not require (or even allow) any user interaction, but classic megade- 
mos often required a key press or mouse click in order to advance to the next screen. Some 
of them even had an interactive selector where the viewer could choose the desired part. 


Such practice was especially popular on some platforms, most notably the Atari ST. Figure 


47 


4.1 shows the loader screen of the well-known Red Sector Megademo (1989), which pre- 
cisely fits the format of an oldschool Amiga megademo described above (see (Botz, 2008, 
pp. 100-106) for an in-depth analysis of the demo in question). Botz (2008, p.135) links 
the disappearance of interactive demos to the growing importance of demo parties: a linear 


show filled with impressive effects was suitable for demo competitions. 


4.1.2 Demo narratives 


The most common structure of a demo is a predefined linear sequence of effects. In some 
rare cases, the sequence is varied with dynamic elements that change every time the produc- 
tion is run. A notable exception to the rule is the interactive megademo type (see Sections 
4.1.1 and 5.2.2) that does not follow a rigid structure, but lets the user freely decide the 
order of effect screens instead. The individual screens, however, typically follow a linear 


narrative, progressively improving the effect to impress the audience. 


While any temporal sequence ultimately forms a narrative, intentional storytelling in de- 
moscene productions is a rare phenomenon. Productions known as story demos are exam- 
ples of the opposite. Story demos highly resemble short films with elements like narrative 
text or narrator voice, soundtrack, sound effects, cuts and a storyline. Such resemblance is 
hardly a surprise, according to Manovich (2001), who claimed that many narrative struc- 
tures of new media originate from cinema. In addition to pure story demos, lightweight 
storytelling can be found in demo texts that at times constitute stories on their own. The 
role of text in demos has not been studied so far, with the exception of a seminar paper 


written by Hartmann (2008), who analyzed five demos from a literary point of view. 


4.1.3 Themes 


Demos, unlike many other forms of media art, seldom try to convey political, religious, 
or social messages to the audience. Anecdotal evidence of political commentary in demos 
does exist, but in general such themes are scarce (Fig. 4.2). In addition, it is questionable 
whether such content is meant to be taken as a statement, because symbols loaded with 
meanings and opinionated rhetoric styles may equally well serve the purpose of stylistic 


experimentation. 


When compared to "serious topics”, humor is a much more common theme in demoscene 
productions. Different kinds of joke productions exist for the sole purpose of giving the au- 
dience a good laugh. Such productions are often made by joke groups that operate outside 
the traditional lame-elite dimension (see also 3.5). Joke productions range from intention- 
ally badly designed and implemented demos to laboriously crafted megademos that use 
humor as their main theme. Humor in demos is a broad topic, ranging from adolescent 


jokes to references to popular culture or other demos. Even some parodies of well-known 


48 


JUSE DAE A 3 8211) 
SAMA 


GUA FUCKIAS teitä ei maailma voi 


GOUCDLBY penne YoU’ KE vihata, mutta minua 


se vihaa, sillä miia 
Il SERLRANS GULE OF 
HERE todistan siitä; e tä 


sen teotiexit palkat 


BLEED Y Joh 73 vä 
ti op 





Figure 4.2: Political and religious themes. Red Storm by Triad and Paimen by Coma. 


demos exist: Sqrt(2) Reality, Real Reality video (parodies of Second Reality) and Rope (par- 
ody of Dope), to give three examples. Roininen (1998, p.64) also mentions joke demos in 
her thesis, describing them as intentionally badly made demos that amused some and bored 
the others. 


4.1.4 Development of Effects 


The basic format of a demo is a collection of real-time effects, possibly but not necessarily 
tied together by transitions and a consistent visual style. The fundamental property of an 
effect is its visual fidelity: it needs to look impressive. A high-quality effect is visually 
impressive, original and technically advanced at the same time. Especially in the era of 


limited computing power, the technical execution was highly regarded (see Section 4.3). 


When analyzing the demos listed in Appendix B, their effects were first collected into lists. 
Next, the data were combined so that for each year, a list of effects and their frequen- 
cies (number of demos where it appeared) could be calculated. For a single year there 
were possibly as little as three demos, which was not sufficient to reveal large-scale trends. 
Therefore, the time span was broken into five-year periods, and the respective lists were 
combined. Dividing the effect frequency by the demo count yielded the percentage of de- 
mos that featured a certain effect, and let us compare their relative popularity during each 
period. Another possible approach would have been to count all the occurences of all the 
effects and compare them to the amount of all the effects. However, such an approach 
would have required significantly more effort, and on top of that it would have introduced 
an additional problem: what to do, when an effect is a combination of two or more? Does 


it count as multiple effects? Counting the demos instead circumvented the problem. 


The first period, from 1985 to 1989, is presented in Table 4.1. The most obvious finding 
is that practically all the demos contained scrollers in some form. Text moving across 
the screen was a fundamental element in crack intros, featuring greetings to other groups, 


information on the crack itself, and credits. The young demoscene of the late eighties 


49 





Effect % Effect % 


scroller | 95 wobbler 26 
colorbars | 37 starfield 26 
sprites | 32 | 2D animation | 21 
































Table 4.1: Most popular effects from 1985 to 1989. Percentage of demos featuring a certain 
effect. 


largely inherited the effects of the crack intros and, thus, early demos also feature a great 
number of different scrollers. Colorbars, sprites and wobblers can be directly mapped to 


the capabilities of hardware at the time (see Sections 5.1 and 5.2). 












































Effect % Effect % Effect % 
scroller 78 writer 25 | dotvectors | 22 
flat shaded vectors | 72 zoomer 25 plasma 22 
2D/3D animation | 44 | vector balls | 22 | image morph | 22 
starfield 28 | landscape | 22 tunnel 19 
sprites 28 | rotozoomer | 22 | line vectors | 19 





Table 4.2: Most popular effects from 1990 to 1994. Percentage of demos featuring a certain 
effect. 


As can be seen in Table 4.2, scrollers were still fashionable in the early nineties. Almost as 
popular were flat shaded 3D vector graphics, such as cubes and space ships. Commodore 
Amiga, the dominant demo platform of the period, was well suited for such graphics with 
its hardware-assisted polygon filling and relatively powerful 32/16-bit CPU required for 
the mathematics involved (see 5.2.1 for further discussion). 3D graphics had definitely 
made their breakthrough, since line- and dot-based vector graphics plus vector balls were 
frequently seen in demos. Such a development also reflects the increasing sophistication of 
demo programming. Another notable trend is the appearance of image-based effects such 
as different zoomers. Tunnels appeared already in the early nineties, but became much more 
popular a few years later, as can be seen below. Overall, the growing variety of effects is 


apparent when compared to the very first period of demoscene productions. 























Effect % Effect % Effect % 

texture mapped vectors | 61 scroller 27 metaballs 21 
radial distortion 55 shaded vectors 27 particles 21 
tunnel 39 bump 27 flyby 21 

image distortion 33 | flat shaded vectors | 24 rotozoomer 21 
line vectors 30 floor 24 | 2D/3D animation | 18 


























Table 4.3: Most popular effects from 1995 to 1999. Percentage of demos featuring a certain 
effect. 


The period between 1995 and 1999 could be described as the golden age of software- 
rendered 3D graphics and image distortions. Texture mapped vector graphics rapidly gained 


popularity in the late nineties, as can be seen in Table 4.3. The development was supported 


50 


by the increasing processing power and graphics modes better suited for texture mapping 
(Section 5.3.1). At that time the PC had already taken over the home computer market and 
become a major player in the demoscene, too. Most importantly, however, texture mapping 
can be seen as an example of scientific knowledge entering the community: in the field 
of computer graphics the technique had been used for years. Flat shaded vector graphics 
had already become old-fashioned and vector balls disappeared almost completely. Bump 
mapping, metaballs (3D balls melting together when they come close to each other) and 
particles can be taken as additional examples of programming tricks absorbed from the 


field of computer graphics. 


Various image-based effects, such as radial distortions, distortions by curves and roto- 
zoomers, frequently appeared in demos. In addition, some of the tunnels can be counted 
as radial distortions, since they are based on the same method of mapping a 2D image into 
polar coordinates. Some effects, once popular, had already lost their prestige. For example, 
scrollers could be seen in only 27% of the analyzed demos, and moving 2D sprites had 


disappeared altogether. 























Effect % Effect % 

texture mapped vectors | 63 | image distortion | 32 
2D/3D animation 47 particles 32 
flyby 42 ribbons 21 

tunnel 42 blur 21 

shaded vectors 32 floor 21 




















Table 4.4: Most popular effects from 2000 to 2004. Percentage of demos featuring a certain 
effect. 






































Effect % Effect % 

texture mapped vectors | 57 particles 29 
shaded vectors 43 scroller 29 
landscape 43 | flat shaded vectors | 29 
ribbons 36 tunnel 29 

flyby 36 | 2D/3D animation | 29 





Table 4.5: Most popular effects from 2005 to 2009. Percentage of demos featuring a certain 
effect. 


Affordable powerful 3D accelerators started to gain support in demoscene productions from 
2000 onwards. High polygon counts and increasing resolutions were the biggest differences 
compared to the previous software-rendered 3D effects. The polygon became the principal 
graphical primitive, and most of the popular effects of the last ten years clearly reflect the 
development (see Tables 4.4 and 4.5). Animations and tunnels that used to be 2D effects 
turned into 3D animations and vector tunnels. Notable newcomers in the figures are rib- 
bons, moving polygonal stripes and tubes. Interestingly, there seems to be little difference 
between the two last periods. During the first fifteen years of demoscene productions the 


development was rapid, but lately the pace seems to have slowed down to a certain extent. 


51 


There are several possible explanations to this, but one factor certainly is that computer 
hardware has reached such a level that it does not effectively limit the creative aspirations. 
Creativity, tools and the complexity of high-end programming are likely to be the current 


limiting factors as opposed to mere processing power. 


4.2 Size-Limited Intros 


While demos have grown in size all the time, the size limits of intros have remained fixed 
for a number of years. The most typical sizes at competitions are 64 kB and 4 kB, but 
several other sizes such as 1 kB, 256 B and even 64 B exist. The word intro refers to the 
cracker roots of the demoscene: little screens with effects called crack intros were placed at 
the beginning of illegal software to display the name of the crackers and to send messages to 
other groups. The name became so rooted in the culture that it still remains in use, although 


there is very little in common between crack intros and the intros of today. 


Why impose such completely arbitrary limits on productions? Crack intros needed to fit 
in small size and the early home computers often had only 64 kilobytes of memory, but 
such limitations have had little foundation since the 32/16-bit generation. Tradition may 
be counted as one, although insufficient, reason. More credible explanations lie in the de- 
moscene’s aim to break borders and do the impossible. To create a high-quality audiovisual 
presentation in small space is considered more valuable than doing the same without limits, 
although to the viewer both might look exactly the same. Even a video clip would pro- 
vide the same experience, but it is in dire contrast with the real-time practices of the scene. 
Creative process, when constrained by almost impossible limits, forces the artist to push 
the boundaries of the medium, to circumvent the limitations and to constantly rethink his 
approach. For examples of the same phenomenon in other domains, one might consider 


artforms such as calligraphy, haiku poems, miniature paintings and ornaments. 


Although a few kilobytes of storage space have hardly been of importance during the last 
twenty years, there is one counterexample as well. So-called BBS intros used to accompany 
demos in the same compressed archive in the 1990s. Their purpose was to advertise a 
bulletin board system in an interesting manner. The intro would be added to the archives 
distributed at the site and therefore it needed to be small: a large file would have increased 


download times to the discontent of the users. Fig. 4.3 portrays an example of a BBS into. 


4.2.1 64k Intros 
64-kilobyte (65536 B) intros have been established as a common category of demoscene 
productions, although somewhat equal categories such as 40k and 96k intros have existed 


in the past as well. The size limit contains a reference to binary numbers (16 bits or 219) 


52 


** This file was downloaded from the best denoboard in Finland ** 
Triplex BBS + Running with PCBoard(r) vt5.21 and Novell Netware v3.12 


* 268 » 1200+ Demos - 1800+ Modules - 500+ Sources - 350+ Scene-GFX « 26B * 
* Internet E-Mail address and demoscene releated newsgroups for FREE * 


« DG: 68:100/11 - GSN: 864/350/83 - Internet: Personal E-Mail & newsgroups » 
+350-0-506 2277 æ Three V.34 (28.600bps) nodes are waiting for your call! 





Figure 4.3: A BBS intro for Triplex BBS 


and to early home computers such as the Commodore 64 that had 64 kilobytes of RAM. A 
typical 64k intro is one executable file containing all the required code and data. The limit 
is an absolute maximum—it may not be exceeded even by one byte. Figures 4.4 and 4.5 
contain two examples of this category: a 40k intro from the year 1994 featuring flat shaded 
vector graphics and a 64k from the year 2000 with ray tracing. Figure 4.4 also serves as an 
example of the early 1990s Amiga style, and 4.5 features the artistic white texts typical of 


its time. 


Such limited space requires its own approach from the authors. A common technique is the 
use of data compression: all the required code and data are compiled into one executable, 
which is then fed to an executable compressor such as Pklite (MS-DOS 16-bit), pmwlite 
(MS-DOS 32-bit), StoneCracker (Amiga) or UPX (multiplatform). The resulting program 
contains a stub that decompresses and executes the original code transparently. 


However, data compression alone does not solve the size dilemma. In comparison to pre- 
rendered music and still images, program code tends to be relatively small, which inevitably 
leads to the use of algorithmic methods to create the audiovisual content. Small still images, 
such as logos, can be fitted in as seen in some 64k’s, but high color and pixel resolutions are 
out of reach even when dealing with compressed data. Likewise, audio needs to be either 
minimalistic or generated through sound synthesis. More discussion on sound synthesis is 
provided in Section 4.5.4. In all size-limited intro categories, the role of the programmer is 


emphasized in contrast to demos that allow a richer variety of media types. 


53 





Figure 4.4: Falu red color by Razor 1911 (40k 1994) 


4.2.2 4k Intros 


In comparison with 64k intros, the 4 kB (4096 bytes) intros are an even more challenging 
category, being only th of the size of the former. Such a constraint introduces completely 
new obstacles and limits the available media types in practice to program code: even the 
smallest sound sample or still image would be big in this context. The role of the program- 
mer is emphasized even further, since most of the content needs to be generated algorith- 
mically. The techniques and tools presented here are chiefly based on the IN4k website 
(IN4K, 2005), dedicated to sharing knowledge among programmers interested in 4k intro 


programming. 


As can be observed on IN4K (2005), the 4k category has given rise to a number of methods 
to ease the size limit. Specific tools have been created for compression (a compressor 
adequate for 64k intros may be unusably big) and executable tweaking. The properties of 
the executable file format, dynamic loading of libraries and compilers are all exploited to 
free bytes for the actual content. Special programming styles are employed to minimize 
the overhead of compiled code and even minimal virtual machines have been proposed to 
reduce the precious size. 


According to pouet.net, the first categorical 4k intros date back to the STNICC 1990 com- 
petition that featured 3.5k entries. By the mid-1990s, big parties too started to accept them 
in competitions with Assembly’94 being among the first (Assembly Organizing, 1994). The 
technology of 4k intros has undergone a dramatic change since those times (in this discus- 
sion we do not consider, for example, crack intros that could fit in the definition too). While 


54 





Figure 4.6: Two 4k intros. Pure Spirit by Spirit New Style (1996) and Nucleophile by Portal 
Process and TBC (2008). 


early attempts ran on platforms such as MS-DOS and thus had to be self-contained (due 
to the simplicity of the operating system), new 4k intros have a variety of APIs such as 
OpenGL, DirectX and sound/speech available through the operating system. The same gen- 
eral trend from hardware pushing to system-friendly programming is discussed in Chapter 
5. The development has also been a source of controversy: what is the actual role of the 
programmer when using external libraries that do complex things for free (Scholz, 2007b)? 
The requirements for the intros have grown equally: in the 1990s sound was not even al- 
lowed in 4k competitions (see Assembly Organizing (1998) for an example), while these 
days, a high-ranking intro needs to feature quality sound in addition to the visuals. The 
switch from home-grown 3D vector routines to hardware accelerated APIs is easily notable 
in the technical and aesthetic properties of 4k intros: low resolutions and simple vector 
graphics have evolved into polished 3D shows, as can be observed in Figure 4.6. 


55 





Figure 4.7: Ixaleno, a 4k procedural graphics entry by RGBA 


Rather recent newcomers among size-limited intros are the procedural graphics. The first 
ones listed on pouet.net are from the Buenzli party of the year 2005. Such programs do not 
feature real-time effects, animation or sound. The only aim in the category is to create a 
good-looking image by using all means available. Technically, procedural graphics can be 
categorized as 4k intros, as they have multiple features like compression and programming 
approaches in common with the more traditional counterpart. However, without sound 
and animation there is more space for the image generation code and, thus, better visual 
fidelity can be achieved (see Fig. 4.7). Procedural graphics most often appear in a separate 
competition instead of the real-time category. 


4.2.3 1k Intros and Beyond 


The minimalism does not end at 4096 bytes. Even smaller intros are present on pouet.net, 
down to a mere 32 bytes. The next category after the 4k intro is the 1k intro (1024 bytes). 
Cutting down another three kilobytes limits the expression to more or less one single effect 
that can be parameterized to make the visuals more dynamic. Due to the executable file 
overhead, little space is left for the actual program code. Code compression is still possible 
and the operating system APIs can be accessed, but in general 1k intros tend to look like 


simplified 4k intros. 


To go beyond one kilobyte, down to 512, 256, 128, 64 and 32 bytes, effectively rules out 
the use of APIs, high-level languages, code compression and ultimately even the executable 
formats of advanced operating systems. Based on the production lists of pouet.net, the most 
popular platform for such tiny intros is MS-DOS, chiefly due to its executable format that 


56 





Figure 4.8: Tube by 3SC, a 256-byte intro for MS-DOS 


does not contain anything but the program code itself. In addition, graphics can be output 
with very little overhead and the computing resources are higher than on 8-bit machines 
that have equally simple executable file formats. Tiny intros usually contain one single 
effect with no sound (see Fig. 4.8 for an example). Such pushing of completely artificial 
limits could on one hand be considered the culmination point of the demoscene’s striving 
for technical perfection, and on the other hand, a representation of the scene’s tendency to 
create and adhere to its rigid taxonomies. 


4.3 Demo Aesthetics 


The aesthetic style of demos is affected by the hardware capabilities, production tools, 
skills, computer games, community likes/dislikes and the popular culture, all of which have 
changed rapidly during the last 20 years. Productions from the late 1980s are completely un- 
like their counterparts these days, both technologically and aesthetically. Currently, the most 
significant contribution to the discussion is the doctoral thesis by Botz (2008). Lónnblad 
(1997, 1998) analyzed demoscene music, Kurki (2002, p.57—62) used Moppi Productions as 
a case example when discussing developing visual styles, and Simmonds (2001) described 
some of the most common aesthetic traits of demos. Scholz (2007a) also includes some 
discusson of aesthetics in his article about the demo group ASD. 


57 


4.3.1 Hardware-Dictated Early Years 


For quite many years demos were marked by hardware pushing, doing the impossible. The 
newborn demoscene did not have traditions, so the audiovisual language had to be con- 
structed from scratch by absorbing visual traits from other domains. At first, the most 
important frame of reference was computer games: early demos and crack intros look like 
opening screens of games, with effects on top. Visually thinking, game names were re- 
placed by group logos and game credits/instructions with demo credits or greetings. Figure 
4.9 displays the opening screen of Bubble Bobble side-by-side with a crack intro by Ikari 
to illustrate the relationship. The lack of music authoring tools forced groups to rip game 


music to their productions, which made them even more game-like. 


BUBBLE BoBBLE 


GRAPHICS BY ANDREW THRELFALL. 


PAL VS HD JUSI 


FOR THE BEST 
* PRESS FIRE FOR BUBBLE BOBBLE ¢ TREDE MITA We REST! 





Figure 4.9: Bubble Bobble opening screen by Software Creations (1987) and Ikari crack 
intro from 1988 


The properties of 8-bit hardware heavily affected the aesthetics. The first effects were 
scrollers, colorbars and moving sprites—things the machines were directly capable of with 
low-level programming. Most early home computers were based on character graphics: the 
screen was divided, for example, into 40x25 characters that could be changed and copied. 
The character block size, typically 8x8 pixels, is easily observable in the visual style of de- 
mos made for 8-bit computers. Likewise, the aesthetics of demo music were deeply marked 
by the properties of the rather simple sound chips until the Commodore Amiga with its 
module music changed the course of demo music (Section 4.5.2). More insight on the re- 
lationship of computer games, 8-bit hardware properties and early demo aesthetics can be 
found in the dissertation by Botz (2008, pp.53—69). 


Distinctive early demo aesthetics have already become a source of nostalgia. Retro or old- 
skool productions mimic the early style on much more capable platforms. Old effects im- 
plemented using hardware accelerated graphics and CD-quality sound do not share anything 
else with old demos than the audiovisual references. Similar trends can be observed outside 
the demoscene too, in the form of retro-inspired new media art, retro style music and re- 
makes of old video game classics (see Section 4.5.1 and Heinonen & Reunanen (2009) for 


more discussion on the retro phenomenon). 


58 


4.3.2 Appearance of Demo Design 


Specific design discussions appeared on disk magazines in 1993 (Imphobia #7, R.A.W. 
#5). Sex’n’Crime already had some rare references to design in 1990, but the in-depth 
treatment of the subject can be considered a thing of the early or mid-1990s. Naturally, 
productions had always been designed in one way or another, but the concept of design had 
not been considered a separate entity. The first "design demos” appeared on the Amiga line 
of computers, whose established demoscene had passed the peak of pure technical show-off 
in the early 1990s. Increasing visual sophistication can be observed in Figure 4.10, depicting 
the design demo Groovy! by Lemon. (Lemon Point). What is understood by ”design” in 
this context notably differs from other fields of study, such as user interface or industrial 
design. According to Mount/Parasite (RAW #5, 1993), good demo design consists of the 


following ingredients: 





Figure 4.10: Groovy! by Lemon. (1993) 


Color scheme 


Composition 


Speed 
e Homogeneous appearance 


e Never letting the viewer get bored. The writer suggested that an ordinary effect should 
be shown for 5—10 seconds only. 


59 


The article represents the views of the Amiga scene. On the PC side, the same topic was 
dealt with by Darkness/Imphobia who admitted that PC demos were usually badly designed, 
but also noted that the tide was turning (Imphobia #7, 1993). He even went so far as to sug- 
gest the use of storyboards for demo design—a method used in numerous other fields of 
art and design. The article seemed to spark interest, since the following issue of Imphobia 
featured four articles on design alone. Baldric/Extreme brought up the importance of tran- 
sitions between effects and synchronization with music ("music sync”). The other articles 


also reflected the same definition of design and acknowledged its high importance. 


Visually the demoscene productions have increasingly started to contain postmodern fea- 
tures. Old demos and discussions reveal a strong concept of authorship: one needed to 
make one’s productions from scratch. Such attitude bears resemblance to the heroic artist- 
as-genius trend of western art. As Kiia Kallio summed up in comp.sys.ibm.pc.demos in 
1993: 


ANYONE WHO COPIES PICTURES IS A LAMER!!! As an gfx-artist I am 
very well aware of how difficult it is to make a good picture. The reason why 
paintings are copied is that IT IS SO MUCH EASIER. 


The Internet provided a wide variety of easily accessible graphical material, such as pho- 
tos, starting from the mid-1990s, which was coupled with the introduction of the popu- 
lar Photoshop software, designed for the editing of existing material. The development is 
immediately noticeable in the demos of the era: hand-drawn graphics were replaced by 
overlays consisting of edited images. The programmed content, too, was increasingly pre- 
sented as filtered layers. Similar general trends in new media were observed by Manovich 
(2001, pp.129—160), who claims that a postmodern author bases his/her works on composit- 
ing pieces of existing media objects. Manovich also draws parallels to experimental cinema, 
such as Sergei Eisenstein, and painters like Pablo Picasso and Georges Braque, and portrays 


them as the forerunners of postmodern composition/montage in media. 


4.3.3 Frames of Reference 


When seeking audiovisual inspiration, the demoscene both reuses its existing trends and 
reaches out to other domains. When a new style, originating from an external frame of 
reference, is adopted by influential authors it gets diffused inside the community and may 
end up being a new dominant trend. So strong is this following that demos published ap- 
proximately at the same time typically resemble each other in many respects. New effects 
are diffused the same way and affect the demo aesthetics on their behalf. To mention two 
notable frames of reference, the early 1990s graphical style was fantasy-oriented (Section 


4.4.1), and was replaced by collage in the late nineties (Section 4.3.2 above). 


60 


VIRTUAL DREAMS 





FAIRLIGHT 


Figure 4.11: Full Moon by Virtual Dreams from 1993, displaying an effect resembling 
Wolfenstein 3D 


Games, as discussed in Section 4.3.1, were among the first frames of reference in demo 
aesthetics. Their influence has never completely disappeared and audiovisual references to 
games can still be found in numerous demos. Examples of direct references are the popular 
Wolfenstein 3D (published by id Software 1992) and Doom (id Software 1993) parts, where 
the viewer is taken through a maze of texture-mapped walls. In this particular case, the 
adoption was extremely rapid, as can be seen in Figure 4.11, which depicts Full Moon from 
1993. The appearance of real-time 3D graphics in games, later enhanced by techniques 
like texture mapping and shading, was also reflected in the countless object shows of the 


nineties. 


The popular and even the high culture of each era are reflected in the demoscene produc- 
tions. For example, the layered collage style was most likely adapted from television com- 
mercials and music videos. This assumption is supported by Roininen (1998, pp.95—98), 
whose interviewees named commercials, movies and rock videos as their sources of inspira- 
tion. Demos have been likened to music videos (Lónnblad, 1997, p.24; Saarikoski, 2001a), 
and the two formats indeed share some properties: duration of a few minutes, synchronized 
combination of visuals and audio, montage and other cinematic techniques like cuts and 
camera rides. The real-time nature of demos and the low use of video footage technically 
separate the two, even if the end result might at times look similar. To exaggerate a little, 
the visual content in music videos is subjected to the audio: it is a video track for the song, 
whereas music in demos could be called a soundtrack for the visual effects. Figure 4.12 por- 
trays State of the Art, a pioneering techno demo from 1992. The audiovisual style closely 
resembles music videos, but the visuals are drawn in real-time. Two clear-cut examples of 


popular culture being absorbed by the scene are the popularity of techno music (see Section 


61 





Figure 4.12: State of the Art by Spaceballs (1992) 


4.5), and the themes and styles found in still images (Section 4.4) in the early and mid- 
nineties. More recently, the hip-hop and graffiti cultures have also become popular frames 
of reference, which was revealed when studying demos produced since the late 1990s. 


4.4 Visual Artifacts 


Visual artifacts such as artistic still images, group logos, 3D models and ASCII art constitute 
a major part of the visual language of the demoscene. They are created for several different 
purposes: demos, intros, competitions, artpacks, slide shows, disk magazines and infofiles 
are the most common uses. These artifacts are static by nature, but may also become live 
when animated and effected through programming. Like all the other artifacts of the scene, 
they have changed significantly over time, due to factors such as improved pixel and color 
resolution, development of authoring tools and new visual styles in the popular culture. 


4.4.1 Still Images 


Still images can be considered the counterpart of paintings or drawings of the non-digital 
world: they are static, two-dimensional and, in theory, offer an unrestricted medium for 
artistic self-expression (theoretically thinking, with 24 bits per pixel and a resolution of 640 
by 480, the possible amount of different combinations is as high as 27378). At times, still 
images are created for specific competitions at parties, and at times they are used to support 
other content, such as in demos and disk magazines. 


62 


As demo effects represent the skill of the programmer, still images are the showcase of 
the graphic artist, "the graphician”. In some sense the art of the demoscene is in line with 
very old traditions of visual arts, such as renaissance painting: in addition to an interesting 
theme and good composition, the technical execution of the work has to be perfect to be 
approved. The traditional way of creating still images has been pixeling, the drawing and 
editing of an image at pixel level. With low color and pixel resolution, the challenge was 
to create good-looking pictures with such limited means. As an example, a typical image 
on the Commodore Amiga was 320x256 pixels with 32 colors. For years, the fundamental 
tool for pixeling was Electronic Arts’ Deluxe Paint (1985-1994), observable for example 
in the amount of iff/lbm images in competitions and in the interviews of GFXZone (2000). 
Higher resolutions and improved software (e.g. Photoshop) eventually diminished the need 
to tweak images at pixel level and, together with new visual styles, changed the face of 
demoscene graphics. The development of still images can be observed, for example, in the 
image collections of GFXZone (2000), Vigh (2003) and Vigh & Polgar (2006). 





Figure 4.13: Examples of fantasy style by Titan, Cougar and Suny 


Two popular styles of demoscene art have been fantasy (Fig. 4.13) and collage. The con- 
troversial collection of demoscene still images and the original works called No Copy? 
(Parssinen et al., 1998) reveals the sources of inspiration of many recognized scene artists. 
The works of fantasy artists such as Frank Frazetta, Luis Royo, Hajime Sorayama, and es- 
pecially Boris Vallejo have been pixeled to still images over and over again. Interestingly, 
one can find references to several other genres as well: the paintings of Salvador Dalí, H.R. 
Giger, Odd Nerdrum and even Ferdinand von Wright have been reproduced among others. 
Such references reveal that scene artists have been aware of the so-called ’ gallery art” and 
looked for inspiration in it too. The methods and values of the 1990s scene art were further 
elaborated by Geurtsen (2000), who regarded mere scanning as taking a shortcut and re- 
drawing as a way of learning. The older images of the ”No Copy?” gallery represent the age 
of fantasy/Deluxe Paint, whereas the newer ones can be categorized as collage/Photoshop. 
A similar trend can be observed in the collections of Vigh (2003) and Vigh & Polgar (2006). 


4.4.2 3D Objects 


Realtime 3D vector graphics appeared in demos in the late 1980s in the form of simple 
wireframe vectors, flat-shaded objects and balls rotated in 3D space (see Fig. 4.14). Even 


63 


such basic vector graphics require rather advanced knowledge about mathematics, such as 
transformations and projection. While the first 3D objects were extremely simple, the level 
of sophistication improved quickly, and already in the mid-1990s, demos contained consid- 
erably more advanced techniques as can be observed in Figure 4.14 on the right. The demo 
portrayed already featured texture mapping, combined with bump and environment map- 
ping. Further technological developments, such as 3D acceleration (Section 5.3.4), both 


provided new possibilities and increased demands on the quality of the graphical represen- 


tation. 


Kr 





Figure 4.14: Vectormania by Phenomena (1990) and Solstice by Valhalla (1995) 


Simple models can be, for example, sketched on paper and typed in by hand, so early on 
there was no demand for extensive 3D modeling. Neither was the hardware capable of 
displaying complex objects. An algorithmic approach enables the creation of models of 
higher polygon count, as seen in numerous demos displaying regular forms, such as spheres 
and tori. But creating realistic objects, such as a texture-mapped human figure, ultimately 
requires special software for the manipulation of the polygon mesh. Modeled objects started 
to appear as early as in the mid-1990s, and have thereafter become a standard component 
of demos, further enhanced by keyframe animation for the movements. Another use for 3D 
modeling in the demoscene has been raytracing: raytraced still images have appeared in 


demos, graphics competitions and in wild competitions in the form of animations. 


4.4.3 ASCII and ANSI Art 


ASCII (American Standard Code for Information Interchange) and ANSI (American Na- 
tional Standards Institute) are two well-known standards for representing textual informa- 
tion. ASCII dates back to 1963 when it was established as a national standard for text 
interchange (ASA, 1963). The ANSI code contains a richer set of characters, control codes 
for terminal output and, additionally, the option to change the attributes of the text such as 
brightness or color. The first available ANSI standard dates back to 1977 (ANSI, 1977). 
Both codes have been used for artistic purposes in various different domains, including the 


demoscene. Already the first computer artists in the 1960s experimented with character- 


64 


based expression, as documented by Franke (1971). Typewriter art can be considered as the 


conceptual predecessor of the digital experiments. 




















NN TT 
“e ansaan We NZ NM fessos? 808 Lol Zoco 
/ N /N /_\ / 
\ / \// TS NON f 
[N N \ ss Wevevenes 
| \------ / — \------- / — N---------—— / la N 
on a ee D YN / \ LN / M / 
loc... | =, BN À \/ /A \ /| 
| / MY / | 
.. ----- / \------ / | / /----- 
esse o MN.) aos a ‘ay ds ue, cet az0!/1124 
JIN *\Newcsscsens enano man e. N 
f IN NN 
eh A. N [a p pen d i x] 


Figure 4.15: ASCII logo for Appendix 


In the context of the demoscene, text logos have appeared as part of infofiles (see Fig. 4.15), 
ANSI/ASCII collections have been published, and there is even a long-standing competition 
for creating demos that display their visuals in text format (tAAt, 2002). The limitations of 
such output are obvious: the pixel and color resolution are extremely low (for example 
80x50 characters with sixteen fixed colors). However, such limitations create a distinctive 
retro style and provide the programmers and graphic artists yet another opportunity to prove 


themselves by ingeniously overcoming the challeng. 


4.4.4 Disc Cover Art 


Decorating 5.25” diskette covers is an old form of visual art that was born in the Com- 
modore 64 cracker and demo circles. What makes disc covers unique among the visual 
artifacts is their physical nature: they accompanied diskettes sent out by swappers and were 
photocopied again and again when they progressed in the swapping networks. The cover 
art was chiefly drawn by hand for the sake of quality, and the colorscale was limited to 
high-contrast black and white to accommodate the photocopy process. The electronic dis- 
tribution of productions and the switch to smaller 3.5” diskettes eventually diminished the 
artform, although decorated labels were still made for some demos (Polgar, 2005, p.155) 
and voting diskettes used at parties. For an example of how the disc covers looked like, see 
Figure 4.16. More examples can be found in the Freax Art Album (Vigh & Polgar, 2006). 


The themes and styles of disc cover art were analyzed briefly in the course of this study, 
based on the scene paper art collection of Jazzcat/Onslaught containing over 1600 works 
(mostly disc covers and voting sheets) as of 3 April 2009. Unfortunately, the collection can- 
not be accessed online any longer. Even a concise study like this reveals the great variety of 


disc cover art: while obvious adolescent themes like women, drinking and fantasy definitely 


65 





TRN 4:4 


20» 1 8009 
ona] kesä 303 INN 


A Ai + 





be Anaa] 





TOHON WHIP ME WITH YOUR WRES!" 





Figure 4.16: Disc cover art by Electric/Extend. Image from the collection of Jazzcat/On- 
slaught. 


are present, it would be an oversimplification to discuss the art form solely on that basis. 
Visual references to different genres such as graffiti, comic book art (especially similar in 
its black and white expression), and even surrealism can be found in numerous works, and 
some of the authors have certainly developed a distinctive personal style. The fidelity of 
the covers varies notably as well, ranging from beginners’ quick sketches to thought-out 
compositions. 


4.5 Scene Music 


Scene music is created for different purposes: demos, disk magazines, musicdisks, artpacks 
and competitions. Accordingly, the role of the music varies from a central artifact to mere 
background noise. As noted by Lónnblad (1998, p.14), a scene musician typically com- 
poses more music than the group can actually use in its productions. During its existence, 
scene music has changed significantly, reflecting both the technological and artistic currents 
of the time. Developments in hardware and software have provided for a widening scale of 
expression, and the emergence of new underground and mainstream musical styles (espe- 
cially techno with its multiple subgenres) have given inspiration to the musicians. In this 
section, the main focus is on the technological side of things: how did each generation of 


hardware manifest itself in the demo music? 


66 


So far, among the most notable written works on demo music have been a master’s thesis and 
a journal article by Lönnblad (1997, 1998). In her article, Lönnblad systematically analyzes 
the soundtracks of two well-known PC demos of the time, Second Reality by The Future 
Crew (1993) and Caero by Electromotive Force and Plant (1995), from a musicological 
point of view. The most central content is the side-by-side annotation of the music together 
with the demo effects. The master’s thesis contains the same annotation plus the results of 
two surveys and a description of tracker software. Another published work, dealing with 
scene-related chip-music, was written by Carlsson (2008). Polgar (2005, pp.20-22) also 
provides some background information on the roots of the scene music in Freax. Some 
more, although surprisingly scarce, material was found in disk magazines (see Section 4.6). 
The sources specific to a certain type of music are discussed in their respective contexts 


below. 


4.5.1 Chip Music 


Chip music, also know as chiptunes, can be defined in two ways. By the first definition, it 
means music composed for the early home computers and their simple sound chips, hence 
the name. The second use of the term refers to both a musical style reminiscent of the 
old sound chips and the small size of the files, most typically modules (see 4.5.2 for the 
definition). A study on chip music in the context of computer and video games was written 
by Dittbrenner (2007) and published by the University of Osnabriick. 


Technical limitations of the early sound chips, such as the low amount of channels (typi- 
cally three), the lack of sample playback capability (partially overcome by CPU-intensive 
programming tricks), few waveforms and limited modulation capabilities produced a dis- 
tinctive kind of sound. Carlsson (2008) describes some of the current retro activities revolv- 
ing around chip music, such as artists producing music with old hardware. Other examples 
of related acts are Desert Planet and PRESS PLAY ON TAPE (Desert Planet, 2001; PPOT, 
2001). The growing interest in 8-bit music has also given rise to websites such as the 
Commodore 64 music archive The High Voltage SID Collection (HVSC, 2000) and chip- 
tune.com, and soundchip emulators like Sidplay 2 that plays C-64 music (White, 2001). 


On more capable platforms, chiptunes are created by using short looping samples, remi- 
niscent of wavetable synthesis used in early synthesizers and sound chips. According to 
Carlsson (2008), chip music as a style emerged on the Amiga already in 1989. After that, 
chip music has frequently appeared in intros, musicdisks (also known as chippacks) and 
competitions. The popularity can partly be attributed to the distinctive sound carrying a 
reference to the early days of computing, and partly to the need to save space in 64k or 40k 
intros. Probably as big a factor is the mere challenge of keeping the music files tiny by using 
clever tricks, which can be taken as another case of limits embracing creativity, comparable 


to size-limited intros discussed in Section 4.2. 


67 


4.5.2 Modules 


Modules, or alternatively mods, in the context of digital music can be defined as files con- 
taining both the score and the sample data needed for playing back a tune. The counterpart 
of a module is a song, which only contains the notes in order to save disk space: the sam- 
ples are loaded separately when needed. Modules were the dominant kind of demo music 
and also a popular music format in computer games starting from the late 1980s, until new 
types such as MP3 and software synthesizers started to gain ground around the year 2000. 
Nowadays, modules still appear in competitions, musicdisks and demos, especially on plat- 
forms that have limited processing power or memory. As of May 2009, the Amiga Music 
Preservation project (AMP, 2002) lists close to 100,000 modules on their database, and ac- 
cording to the ExoticA project page, there are over 360,000 modules in the Modland archive 
(Lunder, 1996): figures that readily confirm the magnitude of the phenomenon. 


Pos J] Sese PLAV. | ST = PLST | 





PINETONE | E] EA] MNE 5 | zuer 3 ad] SAMPER E 
zuu = |eeea 7 | protracker 02:50 


POCA E a a 

LENGTH 7 e RESOURCE & HOH Pirx: 
n H = = [Uk ¿Patron . 
SPINIT |S KA asis E Pix: 


Figure 4.17: ProTracker 2.3d (1994) 


The fundamental concepts of a module are patterns, tracks, rows, notes, instruments, effects 
and song order. The editors used for creating modules were called trackers: Figure 4.17 
depicts the user interface of ProTracker 2.3d, released in 1994 for the Amiga. Visible 
in the screenshot are the horizontally organized tracks and vertical rows (most often 64) 
that together constitute a pattern. On the rows and tracks the composer can place notes, 
instruments and effects. An instrument consists of a sample and additional information such 
as the volume, name, length and loop parameters. Various effects change the parameters of 
the music in real-time. Examples of the most common effects are vibrato, volume slide and 
arpeggio; a full list of the ProTracker effects with their descriptions can be found in The 
ProTracker Guide by Pedersen (2004). The overall structure of the tune is dictated by the 


song order, which is a list of pattern numbers to be played. 


68 


According to both Polgar (2005, pp.101-103) and Carlsson (2008) the first tracker was 
The Ultimate SoundTracker programmed by Karsten Obarski for the Amiga computers in 
1987. The paradigm was soon accepted among the demoscene and extended by software 
such as NoiseTracker (1989), StarTrekker (1990) and ProTracker (1990), each of which 
added new features to the basic functionality. The availability of tracker source code let 
multiple groups build on the work done by others. For some years, trackers remained the 
sole property of Amiga users, but the appearance of affordable sound cards and the growing 
processing power eventually let PC compatibles play modules as well. The first trackers 
available for PC were Scream Tracker 2 (1990), MODEdit (1991), Whacker Tracker (1992) 
and FastTracker (1992), all of them rather modest in comparison with their established 
Amiga counterparts. The introduction of Scream Tracker 3 (1994) and FastTracker 2 (1995) 
brought PC tools to a much more mature level. Both programs allowed the use of multiple 
tracks, far exceeding the four hardware channels available on the Amiga. Atari ST also had 
its own trackers such as ST Soundtracker (1990), Esion XLI (1991) and Protracker ST v2.1 
(1993). 


The concept of a tracker became so deeply rooted in the demoscene that, even in the 2000s, 
new tracker programs have been created. Some examples of contemporary projects are 
MilkyTracker (2005), ProTrekkr (2008) and Schism Tracker (ca. 2004), a free reimplemen- 
tation of the earlier Impulse Tracker from 1997. Modern-day trackers again extend the 
traditional tracker paradigm with features like filters and real-time sound synthesis. Even 
commercial trackers have been published at times: Oktalyzer (1988) and DigiBooster Pro 
(ca. 1997) for Amiga computers, and more recently the multiplatform digital audio work- 
station Renoise (2002). 


4.5.3 Sample Music 


The use of sample music in demos dates back to the 1980s. None of the early 8-bit home 
computers (Commodore 64, Sinclair ZX Spectrum, MSX Compatibles etc.) had hardware- 
assisted sample playback capability, which made digitized sounds such as speech a curios- 
ity requiring clever programming and heavy CPU load in order to be implemented. C-64 
”digidemos” date back to at least 1985, when General Demo by The Professionals 2010 
was released, according to the pouet.net database. Even the more powerful Atari ST line of 
computers, first available in 1985, did not have the feature before year 1989 when the STE 
model was released. However, thanks to its powerful CPU, the ST was significantly more 
capable of sample mixing and playing than its 8-bit contemporaries. The PC compatibles 
of the time had even more limited sound capabilities, but simple speech could be found 


already in the 1989 demo Summer Holiday made by The Sorcerers. 


The appearance of the Amiga, with its four digital channels and trackers, changed demo 


music in the late 1980s (see Section 4.5.2 above). Technically, tracker music also consists 


69 


of samples, but the independently controllable channels and the possibility to effect and play 
the instruments at different pitches separate it from pre-rendered music. The high memory 
load of completely pre-rendered music was unacceptable at a time when the computers 
typically had only 512 kilobytes of memory. As an example, two minutes of 8-bit music 


played back at 22 kHz requires 2.5 megabytes of space. 


Samples were reintroduced in demos in the 2000s by the arrival of lossy compressed file 
formats, MP3 and Ogg Vorbis. Early signs of the change were the MP3 competition held 
already at Assembly’99, and demos such as State of Mind by Bomb! (1998) and Yume 2 by 
Inf (1998). At that time, the PC had become the dominant platform and the computational 
resources were very different to the 1980s. Tens of megabytes of available memory and 
CPU speeds in the order of magnitude of hundreds of MHz enabled the real-time decom- 
pression of lossy audio formats, ultimately diminishing the role of the modules. Because 
of the pre-rendering, any digital audio tool could be used for making demo music, which 
opened up new possibilities for expression and also brought scene practices closer to the 


conventions of commercial audio production. 


4.5.4 Software Sound Synthesis 


The increasing demands on 4k and 64k intros created the need to improve the music as 
well. All through the 1990s, music was actually banned in 4k intro competitions, sometimes 
justified by the claim that it should be a competition between programmers only (Assembly 
Organizing, 1998). From a technical standpoint, a chiptune module with a player routine is 
way too big for a 4k intro, and even with 64 kilobytes it's only possible to fit in very simple 


samples as instruments. 


Software sound synthesizers (soft synths), also common in numerous commercial packages, 
became the preferred solution. The fundamental idea is to generate the waveforms and 
filters algorithmically in order to save space. The code still needs to be extremely tight, 
especially in the case of 4k intros. Some basic principles of tiny synthesizers are explained 
at the 4k intro-related site IN4K (2005). A more detailed series of 64k soft synth related 
articles called The Workings of FR-08's Sound System was written by Hinrichs (2001). The 
latter article describes the technical challenges with their solutions and additionally touches 
on an important topic: composing. A sound routine alone is not music. For more detailed 
discussion on different software sound synthesis methods, see the report by Tolonen et al. 
(1998). 


70 


4.6 Disk Magazines 


Disk magazines (diskmags) can be described as scene journalism. The most defining prop- 
erties when compared to other scene artifacts are interaction, collaboration and communi- 
cation. A diskmag is an interactive application that contains articles about topics that are 
of interest to the community. The word ”diskmag” springs from the historical practice that 
such magazines were indeed distributed on physical diskettes when they first appeared in 
the late 1980s. Later on, other distribution channels such as BBSs and the Internet became 


common, and diskettes lost their importance, but the name still remained in use. 


Disk magazines did not appear from nowhere: both Polgar (2005, p.49) and Jacobsson 
(2006) mention an earlier, paper-based cracker magazine called /llegal that was published 
starting from 1986. At first, the articles were in German only, but starting from 1987 also 
partly in English to serve the international audience. Illegal was the creation of Triad, an 
active cracking group at the time. A quick inspection of the first English issues (22-24) 
confirms that the magazine already shared several properties with the early diskmags: an 
editorial, news, charts and interviews. Other historical predecessors were the commercial 
disk magazines, most notably Softdisk, which appeared in 1981 for Apple II computers 
(Softdisk, 1995). 


To publish an issue with over a hundred articles requires a collaborative effort of multiple 
persons. Typically a diskmag has a number of editors: an editor-in-chief coordinating all 
the efforts, and additional editors in charge of a specific section. Active readers submitting 
articles, news and ads to be published play an important role in the content creation. The 
efforts of programmers, graphic artists and musicians are required for the implementation 


of the interactive piece of software. 


4.6.1 Selected Disk Magazines 


The diskmags reviewed for this study are Sex’n’Crime, Zine, R.A.W., Imphobia and Hugi. 
The selection is based on their perceived importance in various scene sources plus the times- 
pan of the publication. The magazines cover the years from 1989 to 2008 and represent 
three different hardware platforms: the Commodore 64, the Amiga and the PC. Due to the 
MS-DOS-Windows migration of the PC scene, Hugi also represents two different software 
platforms. This section deals with the mentioned diskmags and their development during 


their lifespan. 


Sex’n’Crime was a Commodore 64 diskmag that dealt with the demo and cracker scenes 
of 1989 and 1990. According to Jacobsson (2006), it was the first scene-related diskmag 
to exist. During two years, an impressive amount of 26 issues were published, which is 


more than once a month on average. The high count can be explained by the low amount 


71 


APRIL 1990 
6. .GAHES AND HOWIES 
T CHART 





Figure 4.18: Sex’n’Crime #14 (1990) table of contents 


of content: a typical issue contained only ten articles. The development between the first 
and the last magazines was not as radical as with other diskmags observed. Only the worst 
usability problems were fixed, while the format and the style remained almost unchanged 


(see Fig. 4.18 for an example). 


one who swaps with Moskwa TV. 


The first thing the reader notices when reading Sex’n’ Crime is the hostile atmosphere of 
the scene, as exemplified by the above quote from the news section of Sex’n’Crime #07. 
Numerous articles and letters were dedicated solely to the constantly ongoing clashes be- 
tween persons or groups, complete with appropriate accusations of lameness and stealing. 
The editors did not even try to calm down the discussion, but actively participated in the 
fights instead. According to the articles, the worst wars had even led to physical fighting 
at copyparties. Another instantly observable property was the still vague division between 
the pirate and the demo groups: crackers were listed on the same charts as the legal groups. 
In contrast with the other magazines reviewed, computer games received a high coverage 
(almost 10% of the articles). A more complete list of topics found in Sex’n’Crime can be 
found in Appendix C, Table C.1. 


Zine represents the first generation of Amiga diskmags, spanning the years from 1989 to 
1991. The geographical origins of Zine are highlighted by the fact that, in the first issues, 
some of the articles were still written in German. The distinction between crackers and 
demosceners was not yet well established, and the contents ranged from demo reviews 


to news of busted swappers. An exceptional property of Zine, when compared to other 


72 


J EHS al TROS C 


fill this sucking news and info was By the way, if something is 

found on a toiletpaper by Chester incorrect about your group, please 

Of BRAINSTORM, If you also want to send me a message, It will be 

say the scene what has happened to published in the next issue (the 

you, send your blasting news to the eighth one) in the rubric ’ANNEX’, 

following address: I can’t check all thöse stuff... 
(let's begin now?) 


ZINE/NEHS 
P,0, BOX 2 Mr, Frost and Sixpack left 
CH-4148 PFEFF INGEN BRAINSTORM, — and” three new members 
SWITZERLAND Came; 
The Fly, Axel, Peace, 
50, for this time you can enjoy the 
following messages to all the folks 
out there in the endless universe 
of AMIGA-users, US-original-supliers of = 5c00pex, 


Y NEWS ’N* INFOS (1) 


Figure 4.19: Zine #7 (1990) news page 





magazines included in the study, was the focus on interviews: 14% of all the articles were 
interviews (see Table C.2 in Appendix C for a more detailed list of the topics). Graphically 
the magazine was mostly text-based and did not yet reach the visual sophistication of later 
Amiga publications of the early and mid-1990s, although the final issues already hinted at 
things to come with their improved layout and navigation. Figure 4.19 illustrates how the 
magazine looked like in the middle of its lifespan. 


Already in Sex’n’Crime, contact advertisements had been an important part of the content. 
In Zine, even more so: the last issue (#11) sported as many as 34 pages of such ads, revolv- 
ing around the theme of new swapping contacts. The wide geographical distribution of the 
ads serves as an example of the international nature of the demoscene already in its early 
days. Rather exceptionally, advertising was not free but there was a small charge involved. 
Even commercial ads could be found on the pages, although not in significant numbers. 


R.A.W., the other Amiga diskmag examined, serves as a good example of how early sim- 
ple magazines evolved into audiovisually rich and systematically edited publications. Fig- 
ure 4.20, a screenshot from issue #2 (1992), already reveals a thought-out color scheme and 
a sophisticated user interface when compared to the earlier diskmags. A screenshot taken 
from R.A.W. #6 can be seen in Fig. 2.1. 


R.A.W. was published from 1991 to 1996. The timespan coincides with the golden age of 
Amiga demos, and thus the articles reveal several facets of the development of that particular 
scene. Towards 1996, the large-scale migration to the PC became evident, which eventually 
led to the decline of R.A.W. A notable difference to early publications was the amount of 


73 


Germany, and they presently getting 
a lot of new nenbers: 


Before 1 Jet you lose on the news. of Elicna 
I nust thank the following people of Trisanist 
Por suppling us with neus. Hope you of Freaggles 
vill continue! ! 


Thanks Lo: i 
| sold his Aniga and left the} 
/ PIC scene. 
/ Tech 
/ findroneda — ' ' is a new board in Sueden. "g 
f 
A new nodentrader called | 
Joined. | 


left and joined D-Tect. | | 





Figure 4.20: R.A.W. #2 news page 


meta-level discussion on the scene, its practices and future (see Table C.3 in Appendix C). 
The amount of self-reflection indicates the growing consciousness of the community. No 
more was it just a hobby, but a phenomenon you were part of. The tone of the discussions 
was still highly controversial at times, as illustrated by the following excerpt written by 
Captain James T. Kirk/Starship Enterprise Designs/TEK in his article Why Friendship Sucks 
(R.A.W. #4): 


In today’s demo scene there is no place for compromises, so do not bore others 


with your lameness, please. Work harder, or drop it. Friendship sucks! 


Imphobia, published in 1992-1996, was among the most influential diskmags on the PC. 
The period was marked by the rise of the PC scene from obscurity to a major player. The 
same change was also reflected in the audiovisual quality of Imphobia: the first issues 
looked crude, but in two years the design had started catching up with the Amiga counter- 
parts, obviously inspired by them. The main screen of issue 8, as seen in Figure 4.21, serves 


as an example of the development. 


The amount of content in the magazine was overwhelming at times: the total number of 
articles peaked at 254 in the last issue published (Imphobia #12). Thematically, Imphobia 
resembled R.A.W. with some notable exceptions (see Table C.3 in Appendix C for a list of 
themes). The high visibility of parties (with 15% of all the articles categorized as directly 
related) reveals the growing importance of parties as forums for releasing new works. An- 


other significant difference to R.A.W. is the high ratio of articles related to programming 


74 


N - A 


(| e A pra 
| AÑ Pa A = ES y A 
mprolra 





Figure 4.21: Imphobia #8 main screen 


(6%). The growing self-consciousness of the community is clearly visible in Imphobia as 


well, especially towards the last issues. 


Contemporary technological developments of the time were reflected in the contents: the 
Internet received some attention already in 1993, and in the last issues, the uncertain future 
of MS-DOS sparked speculation on the scene platforms to come. The attitudes towards 
Windows were chiefly negative at the time, and Linux was proposed as a possible alter- 
native, together with a community-based operating system Demo OS (also referred to as 
DemOS) which never progressed beyond the planning stage (Imphobia #12). 


Hugi is the newest magazine reviewed for this study. The first issue was published already 
in 1996 under the name Hugendubelexpress. At first, the diskmag was mainly written in 
German, but starting from issue 10 (1998) the articles started appearing predominantly in 
English. The German content is not considered in the statistics. The latest published issue, 
as of now, is from 2010. 


During the twelve years of its existence, the magazine has seen two significant technological 
changes: the switch from MS-DOS to Windows and the growth of the Internet. The changes 
were reflected not only in the articles but also in the implementation of the magazine itself. 
The first issues published were MS-DOS only, and after a transitional period the diskmag 
become Windows-only in 1999 (#18). In the mid-1990s the attitudes towards Windows 
95 were still mainly negative, with fears of being constrained by the slowness of the new 
system, but in the late 1990s the attitudes turned more favorable. The advent of the Internet 
is exemplified by the online version of Hugi, starting from issue 22 in 2001. The decline 


of the magazine, marked by the decreasing number of articles and yearly issues, in the 


75 


Djam(m) has been active in the demo Wcene as a musician since the 


old Atari ST times. Currently the 26-year-old French is a member of 
his own group Sunlikamelo-D and the Kosmic Free Music Foundation. 
Besides, he does freelance work for other groups such as Bomb and 
Nomad. He, is one of the main characters in the tracker scene in 
France. His new homepage will open soon. It is going to be located 
at http? //uuu .error-464 nm 


Djam: No, | made my Little modules at home with my Atari ST but | was not in 
any group. Just Like one of these Kids admiring the ST's megademos! The guys 





Figure 4.22: An interview in Hugi #15 


2000s can be attributed to the growing importance of websites in the communication of the 


demoscene. 


Visually Hugi continues the tradition of earlier diskmags, with only minor changes to the 
paradigm (see Fig. 4.22 for an example). An exceptional feature in the user interface is 
the addition of hyperlinks; an innovation most likely influenced by the Web. The themes 
do not stray far from the standards set by the earlier publications either (see Table C.5 in 
Appendix C). Some details do, however, stand out: the ratio of the content related to pro- 
gramming (almost 22%) is higher than in any other magazine of the study. The growing 
technical sophistication of demo effects is obvious in the programming articles, often dis- 
cussing university-level mathematics and algorithms. Another peculiarity of Hugi is the 
exceptionally high number of articles about other diskmags. 


4.6.2 Topics of Interest 


To get a general overview of the themes of interest, the articles of the chosen diskmags 
were categorized according to their main topic, one theme for each article. Thus, the unit of 
analysis was an article. To cover the entire lifespan of each diskmag, issues were selected 
evenly from the whole range of years, unless it was possible to read all of them. Another 
parameter in the selection was the aim to cover the period of time from 1989 to 2009. The 
sum of individual issues examined was 54 and the number of articles examined was 2,957. 
The results by magazine are presented in Appendix C. 


76 


In order to study the phenomenon as a whole, the figures for each diskmag were normalized 
to counter the effect of different amount of issues and articles, after which the occurrences of 
each type were averaged between the magazines. The chosen method somewhat favors the 
magazines with a low article count, since after the normalization their weights are equal to 
the others. It is an acceptable skew, because otherwise Imphobia with its high article count 
and sampling would have dictated the results. The most popular themes stand markedly out 


as can be observed in Table 4.6. 






































Description % Description % 
editorial content 9.3 diskmag-related discussion 3.2 

party invitation/report/results | 9.3 humor 3.0 
interview 8.4 | presentation of / discussion about a person | 2.4 

scene meta-level discussion | 8.4 game 2.4 
news 6.7 | presentation of / discussion about a group | 2.1 
programming 5.9 hardware/software 1.6 

letter to the editor / response | 5.1 story 1.6 
review 4.8 politics 1.5 

chart 4.5 swapping 1.4 

ad 4.2 non-scene music 1.4 




















Table 4.6: The twenty most common types of articles found in disk magazines based on the 
number of articles 


Obviously, each magazine needs to have some sort of editorial and since it is one of the few 
things in common with all of the publications it is rather natural that it should be among the 
top themes. The importance of parties can be anticipated as well, since the events are an 
integral part of the culture. Generally speaking, the relative number of party-related articles 
appears to grow over time: in the earliest issues, they were actually rather rare. The self- 
reflective nature of the community is again highlighted by the high amount of meta-level 
discussion of the scene, coupled with the comments sent by active readers. Interviews, 
news, charts and contact ads appear in practically all of the issues: they could be considered 


the bread and butter of the magazines, so to speak. 


The high ranking of programming is somewhat surprising, since the C-64 and Amiga diskmags 
hardly discussed any programming tricks. However, in the PC magazines the topic was pop- 
ular (see Tables C.4 and C.5 in Appendix C). Another unexpected finding is the relatively 
high amount of game-related articles: for example, in R.A.W. and Imphobia games seemed 
to be an almost banned topic. The occurrence can be attributed to Sex’n’Crime that fre- 
quently featured reviews of new computer games on its pages. Further discussion on the 


complicated relationship between the scene and games can be found in Section 3.3.3. 


The study presented in this section should be considered as an initial attempt at categorizing 
disk magazines and their contents. The main goal, the creation of a general overview of the 
topics that have been of interest to the community, was fulfilled. However, numerous facets 


of the study should still be improved to obtain more reliable data. Firstly, the magazines 


77 


should be chosen so that they represent the timespan in question more evenly. While the 
amount of issues and articles examined is already high, it should be even higher in order to 
say anything conclusive about diskmags as a whole. Secondly, the simplistic approach taken 
in the categorization is not adequate in all respects. Is a review of a musicdisk a review or 
is it an article about music? In the future, a more detailed and flexible taxonomy should be 
devised to better accommodate such overlaps. Another property worth further consideration 
is the choice of the variables, since the number of articles alone tends to underemphasize, 
for example, news sections that might span tens of pages but only count as one article in the 


figures. 


78 


Chapter 5 


Hardware and Software Platforms 


This chapter deals with the different hardware platforms that have been popular among the 
demoscene. Software like operating systems and demo authoring tools are discussed briefly, 
since they are such a fundamental part of the equation. The aim is to make explicit the 
interplay of hardware and self-expression: how did the machines limit or support artistic 
aspirations, and how were the hardware properties visible in contemporary productions? 
Depending on the point of view, the scene has already existed during either three or four 
generations of microcomputer hardware, which in the context of technological development 


is a vast timespan. 


For many years, the demoscene was largely focused on pushing the limits of the hardware. 
Even if early computers were not designed for things like real-time vector graphics or play- 
back of digitized sound, such feats were achieved to impress the community. An important 
factor was that roughly until the early 1990s, there was a deep uniformity inside platforms. 
For example, the Commodore 64, Atari ST and Amiga 500 were monolithic computers 
manufactured by one company only. Computers representing the same platform chiefly 
shared the same CPU speed, chipset features and system software. Such unity guaranteed 
compatibility and allowed the use of methods closely tied to hardware properties. Eventu- 
ally "hardware banging” was rendered unnecessary and impossible by the increasing level 
of abstraction in operating systems and the growing processing power, leading to system- 
friendly programming through standard APIs (Application Programming Interfaces) and 


detachment from the hardware details. 


The line chart in Figure 5.1, published in an earlier article by Reunanen & Silvast (2009), 
depicts the relative popularities of different platforms at the four largest yearly demo par- 
ties (Assembly, The Gathering, The Party, and Mekka & Symposium) during the period of 
1992-2002. This preliminary study reveals some interesting developments, most notably 
about the relative popularities of the Amiga and PC. In 1995 the two were practically even, 


after which the PC little by little started to take over as the dominant hardware platform. The 


79 


120 





A, 
100 os % 

Dn Π64 demo 
S ® Amiga demo 
5 Y Amiga intro 
5 "Y PC demo 

2 PC intro 

a 








1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 


year 
Figure 5.1: Productions at major parties 1992-2002 (Reunanen & Silvast, 2009) 


bankruptcy of Commodore in 1994 and the increasing number of PC compatible computers 
appearing in homes are reflected in the chart as well. The increasing popularity of Com- 
modore 64 is somewhat harder to explain: why would an aging home computer become 
more popular fifteen years after its introduction? Probable reasons for the phenomenon are 
the increased popularity of retro computing and the migration of active C-64 sceners to large 


mainstream Amiga/PC parties. 


5.1 8-Bit Computers 


The first 8-bit computers were hardly capable of any multimedia, but by the early 1980s, 
several computer manufacturers had already started mass producing models that could be 
used for games, which eventually allowed the demoscene to emerge, as argued in Section 
3.1.1. The early 1980s home computer market was characterized by the great variety of 
mutually incompatible 8-bit machines built out of colorful plastic. Even the models made 
by the same company were not necessarily compatible with each other, as was the case with 
Commodore's VIC-20 and C-64. Encouraged by the perceived market potential, a high 


number of electronics manufacturers entered the market, only to pull out a few years later. 


8-bit computers were largely built out of off-the-shelf components—few companies had the 
will or capability to start designing processors or special chips—which made them tech- 
nically close to each other. The main processing unit was typically Zilog’s Z80 (used in 
Amstrad CPC, MSX compatibles and Sinclair ZX Spectrum among others), MOS Tech- 
nology’s 6502 (used in Apple II, VIC-20 and Atari’s 8-bit computers), or Motorola’s 6809 


(used in Dragon and TRS-80 computers). The processing power of each chip was somewhat 


80 


equal, even though Z80 ran at significantly higher clock speeds (3.5-4MHz) compared to 
6502 or 6809 which typically operated at 1 MHz only. Common sound chips, such as the 
General Instrument/Yamaha manufactured PSG (Programmable Sound Generator), could 
produce three-channel audio with square waves. Sprite graphics were available on some 
models to support the hardware drawing of game objects. To keep the computers affordable 
the use of a TV set as monitor and audio tapes as storage media were common features. A 
comparison of the technical specifications of the most popular 8-bit home computers can be 
found in The Encyclopedia of Game.Machines (Forster, 2005, pp.202-203). 


Video chips had somewhat more variety, but a fundamental property found in most of them 
was the characted-based nature of graphics generation. The screen consisted of a character 
matrix (for example, 40x25 on the Commodore 64 and 32x24 on the MSX compatibles) 
that contained indices to a character attribute table. Typical attributes of characters were 
its bitmap and foreground/background color. Initially the character attributes were reset 
to a readable font such as the ASCII table, but by changing the attributes, any kind of 
graphics could be produced. The base size for a character was commonly eight by eight 
pixels. A quick glance at almost any computer game or demo instantly reveals the character 
block locations on the screen. The color resolution was most often less than the bitmap 
resolution: for example, Sinclair ZX Spectrum only allowed two colors inside a character. 
Such a limitation led to a situation known as color clash or bleeding, where unwanted 
colors would show on objects because of their location. The ZX Spectrum graphics found 
in Freax Art Album (Vigh & Polgar, 2006, p.171-184) show how talented artists were able to 
circumvent the color clash by careful placement of objects, dithering and protective regions 


between characters. 


Even though the 8-bits disappeared from the mass market around the year 1990, the de- 
moscene still keeps making productions for them. Pouet.net reveals that in 2008 at least 
124 demos or intros were released for the Commodore 64, 58 for ZX Spectrum, 18 for 
Atari XL/XE, 12 for Amstrad CPC, 5 for MSX compatibles and 4 for VIC-20. One signif- 
icant reason why the 8-bit machines have been able to continue for so long are the various 
good-quality emulators available. Coupled with cross-development tools, emulators can be 


used for relatively easy software development for old hardware. 


5.1.1 Commodore 64 


The Commodore 64 was released in 1982 as a successor to the popular VIC-20. The ag- 
gressively priced and technically capable computer soon became a top seller both in the US 
and Europe. According to Bagnall (2005, p.303), at its peak in 1984 Commodore sold five 
thousand units a day. Forster (2005, p.62) estimates that the number of machines sold is 
around 20 million units, which is far beyond other comparable 8-bits such as the Amstrad 
CPC (2.5 million) and Sinclair ZX Spectrum (5 million). In 1985, Commodore revised the 


81 


product line with C-128 that contained Z80 for running CP/M programs, but its arrival was 
already being shadowed by the 32/16-bit generation. For more background history on C-64 
and its marketing, see Bagnall (2005, pp.243—303). 


The main processing unit of C-64 is MOS Technology’s 6510, a slightly revised version 
of 6502, running at 1 MHz. As the computer’s name suggests, there are 64 kilobytes of 
RAM available. The VIC II (Video Interface Chip) was advanced for its time, featuring 
sixteen colors, hardware scrolling and eight sprites. The graphics are based on the typical 
character generation paradigm with 40x25 characters, yielding 320x200 pixels in the two- 
color high-resolution mode and 160x200 in the multi-color mode. Many of the parameters 
of the video chip can be modified during the screen refresh to facilitate programming tricks, 
which gave C-64 an additional edge over its competitors. (Bauer, 1996) Also, the sound 
chip, SID (Sound Interface Device), was above average at its time with three channels, four 
waveforms, amplitude modulation and a hardware envelope generator. Especially advanced 
features, like ring modulation and programmable filtering, were absent from simpler sound 
chips. (Kubarth, 2006) The characteristic sound of SID still inspires musicians, as discussed 
in Section 4.5.1. For storage, C-64 used either cassettes or 5.25” floppies (the disc drive was 


sold separately). 





Figure 5.2: A shaded torus from Mathematica by Reflex (1995) 


As discussed in Section 4.3.1, the first demo and crack intro effects were based on the 
properties of the hardware: colorbars, scrollers, moving sprites, writers and wobblers being 
among the most common ones. A completely different era started in the 1990s, when first 
the Amiga and later also PC demos started to feature advanced effects like vector graphics. 
The hardware of Commodore 64 is not suitable for polygonal graphics nor the real-time 


calculations involved, but the programmers did their best to bridge the gap. This game of 


82 


catch-up was played for several years, and feats such as texture mapping and shading were 
achieved by careful optimization, use of low blocky resolutions and precalculation (Fig. 
5.2). Such devotion earned the C-64 programmers some reputation in the other scenes, 
observable in contemporary diskmag and newsgroup discussions, as exemplified by the 


following comments from Imphobia #5 (1993): 


It’s amazing, what these guys get out of the C64, but I think that it’s kinda fool- 


ish to keep programming such an old computer. (Rick Dangerous/Surprise!Productions) 


But no computer will ever beat the C64 in being explored and used even beyond 


its limit, I think. (Marvin/Sirius Cybernetics) 


Copying effects from more powerful machines became an important, and still ongoing, 
trend in C-64 demos. Another direction was to create new hardware-assisted high-resolution 
effects and to improve old ones further. The modern C-64 demo style could be described as 
a mixture of the two trends. Design also made its way to the platform, and concept demos 
or carefully designed visuals can be found among the productions (the demoscene’s concept 
of design is discussed in Section 4.3.2). The continuing popularity of the Commodore 64 
is illustrated by the fact that in 2008 it still attracted notable demoscene interest with its at 
least 124 demos and intros. In contrast only 18 were released for the once-popular Amiga 


500, according to pouet.net. 


5.1.2 Less Popular 8-Bit Computers 


Other 8-bit platforms, too, have their respective scenes, although significantly smaller than 
the C-64. According to pouet.net, demos have been systematically created for at least Am- 
strad CPC, Atari XL/XE, Commodore VIC-20, MSX compatibles, SAM Coupé, and Sin- 
clair ZX Spectrum, most of which were among the best-selling machines of the time. Out 
of these six platforms, the ZX Spectrum has been the most active, despite its limited graph- 
ics capabilities. Almost all the game-capable 8-bit computers have attracted at least some 
demo-related activity. While Commodore 64 enjoyed wide international acclaim, the other 
machines were often popular only in certain countries, which is also reflected in their de- 


moscenes. 


The comparison written by Forster reveals the similarities and differences between the most 
popular 8-bit home computers. The MSX line of computers, released in 1983, is used here 
to represent a typical middle-ground home computer, not excellent nor abominable in com- 
parison with its competitors. MSX is based on the Zilog Z80 CPU running at 3.58 MHz and 
features 64 kilobytes of main memory in most configurations, even if the standard allows 
for less. A BASIC interpreter is stored in ROM and serves as the main user interface of the 
system as well. Graphics are character-based with 32x24 locations, yielding a maximum 


resolution of 256x192 pixels with 15 colors. The number of single-color sprites is as high 


83 


as 32, which was above average, since many video chips did not even support sprites at all. 
Audio is produced by the Yamaha PSG chip (AY-3-8910), which features three independent 
channels of square wave sound, volume control and simple envelopes. Out of the box MSX 
supports cassettes for data storage, but also a diskette drive was available as an add-on. 
(Forster, 2005, pp.202—203) 


The demoscenes of the alternative 8-bit platforms mostly share the same properties as the C- 
64 scene described above. Early hardware pushing, the effect of more powerful platforms, 
and stylistic design experiments can be recognized, even if the scale is considerably smaller. 
At large demo parties, the Commodore 64 might still have a separate demo competition, but 
other classic machines have traditionally participated either in the wild or generic oldschool 


compos. 


5.2 32/16-Bit Computers 


The saturated home computer market, badly placed product lines, and Commodore’s ag- 
gressive tactics forced most manufacturers to drop out of the game after the mid-1980s. 
The next generation was thus marked not only by the increasing computing power, but also 
the decreasing number of different platforms. The 8-bit MSX, released in 1983, was the 
first (although commercially unsuccessful) attempt to standardize components across dif- 
ferent manufacturers, but it was not until the IBM PC clones” arrival in the mid-1980s that 


proprietary platforms controlled by a single company started to lose ground. 


Traditionally, the hardware platforms following 8-bit computers have been called the /6- 
bit generation. The name is rather misleading, since the Motorola 68000 CPU used in 
the Atari ST, Commodore Amiga, and Apple Macintosh actually has full 32-bit general- 
purpose registers and addressing modes, and only the data bus is 16 bits wide. Such naming 
is most likely used in order to gather the early PC compatibles (which were 16-bit) and 
other contemporaries under the same umbrella term, but here, the term 32/16-bit is used for 


the sake of exactness. 


The shift from 8-bit to 32/16-bit machines meant that computing resources increased in 
numerous aspects. While 8-bit computers had typically had 64 kilobytes of main memory, 
the next generation sported at least 512 kilobytes: eight times as much. Processor speeds 
were increased to 7-8 MHz from the previous 1-4. CPUs like the 68000 had improved 
features missing from their 8-bit counterparts, such as 32-bit registers, hardware for mul- 
tiplication/division and complex addressing modes (Motorola, 1992). Maximum pixel and 
color resolutions were improved, and in the domain of audio, the Commodore Amiga re- 
defined the sound capabilities of home computers in 1985. Gone was the character-based 
generation of graphics: the new machines operated with bitplanes only, with the exception 


of the PC that also retained its text mode. 


84 


Workbench Screen (Lal! 


[ey Poets seme a 
a RAM DISK 
= ER N 


EES CopyPrefs al 
Preferences PY Workbench! .3 


[e]Jäonkbenchi.3 : , 
MEME 


Utilities Systen Expansion Enpty 


on] a 
ie Prefs 


Trashcan 


CL Anigashel 1 
1,5Y5:> 





Figure 5.3: Workbench 1.3, the graphical user interface of the Amiga computers 


The development affected software as much as hardware. The user interfaces of 32/16-bit 
computers were very different than the BASIC interpreter found on most 8-bit machines. 
Apple Lisa (followed by the popular Macintosh), Atari ST with its GEM, and Amiga with 
its Workbench all featured a mouse-based Graphical User Interface (GUI). Figure 5.3 is 
a screenshot of Workbench 1.3, the GUI of the Amiga computers. The familiar WIMP 
(Windows, Icons, Menus, Pointing) paradigm is recognizably similar to its more modern 
counterparts. For further discussion on the history of graphical user interfaces see Bag- 
nall (2005, pp.416-417, 449-451), Johnson (1999), Lineback (2001), Myers (1998) and 
Manovich (2001, pp.63-115). 


5.2.1 Commodore Amiga 


The Commodore Amiga 1000, released in 1985, was an advanced multimedia machine for 
its time, featuring hardware capabilities not seen in other home computers. According to 
Bagnall (2005, pp.394-481) the machine, originally codenamed Lorraine, was designed 
as a powerful game console, which explains its audiovisual orientation. At one phase the 
project was funded by Atari, but ultimately it was Commodore who brought the final prod- 
uct to the market. While the Amiga 1000 and Amiga 2000 (1987) were well received, main- 
stream commercial success only started with the introduction of the cost-reduced Amiga 500 
released in 1987. A500 built on the success of the Commodore 64, being a convenient up- 
grade for the C-64 owning masses. Its general popularity and good multimedia capabilities 
made A500 the most important demo platform for a period of about six years. 


85 


The Amiga hardware is based on the parallel operation of the main processor and custom 
co-processors. The CPU is Motorola’s 68000 running at 7.14 MHz and it shares the main 
memory with the custom chips. The original Amiga 1000 has only 256 kilobytes of memory, 
but A2000 and A500 feature 512 kilobytes. Media-wise, the most important co-processors 
are the Blitter, Copper and Paula. Blitter makes it possible to draw lines, fill areas and copy 
pixels from one location to another without burdening the main processor. Copper is another 
chip dealing with graphics, and executes its own list of simple instructions that enable the 
programmer to change various display properties in sync with the screen refresh. Paula 
deals with audio output, enabling the playback of four independent 8-bit sample channels 
in stereo with a frequency up to 28.8 kHz and 6-bit volume control. (Commodore-Amiga, 
1991) The features of the audio chip led to, and are directly reflected on the module music 
format (Section 4.5.2). 


Often-quoted graphics modes such as 320x256 with 32 colors and 640x512 interlaced with 
16 colors (Forster, 2005, p.207) are only some of the possible modes available, since the 
number of bitplanes can be chosen from one to six in low-resolution modes and from one 
to four in high resolution. The whole area of the screen can directly be used for displaying 
graphics, by setting a wider or higher mode, meaning that the border area does not have 
any special importance like on other platforms. Other special hardware features also exist: 
Extra Halfbrite allowing the use of 64 colors in the low resolution modes, HAM (Hold-and- 
Modify) allowing the use of the whole palette of 4096 colors in the same time, and Dual 
playfield letting the graphics move on two independent layers, a feature useful for action 


games. (Commodore-Amiga, 1991) 


Amiga demos of the 1980s mainly focused on numerous hardware tricks, and featured ef- 
fects such as colorbars aka copperbars, various kinds of scrollers and moving sprites, not too 
far from the C-64 roots. Wireframe and flat-filled vectors appeared already in the eighties 
and were perfected in the early nineties with features like pattern filling and transparency. 
Fast vector graphics almost became an obsession, and the programmers competed with the 
amount of polygons that could be displayed at a steady 50 fps, also known as ”in one frame”. 
The so-called design demos (see Section 4.3.2) appeared halfway along the A500’s lifecy- 
cle, increasing the audiovisual sophistication of demoscene productions considerably in the 
form of music sync, impressive still graphics, transitions between effects, and thought-out 


composition of visual elements. 


5.2.2 Atari ST family 


Atari entered the 32/16-bit home computer market rapidly with its ST line of computers 
in 1985 after losing the Amiga platform to Commodore. Different models such as the 
520STM, which contained a TV modulator, and the 520STF with a built-in floppy drive, 


soon followed to offer features missing from the original version. The 1040ST line (1986) 


86 


was enhanced with a full megabyte of memory, and the STE models from 1989 introduced 
Amiga-like features, such as a blitter, a palette of 4096 colors (only 16 on-screen at a time), 
and sample playback hardware, to the consumer lineup. Atari also manufactured a more 
professionally-oriented product line known as the Mega ST. (Forster, 2005, pp. 104—105) 


520ST is the baseline Atari, based on the same Motorola 68000 processor as the Amiga, 
running at 8 MHz. The machine has 512 kilobytes of RAM and a 3.5” floppy drive for 
storage. Apart from the relatively powerful processor, the hardware is not highly suitable 
for arcade games: the sound chip is practically the same Yamaha PSG (YM2149) that was 
used in many 8-bit computers, there are neither hardware sprites nor a blitter, and the video 
chip can only display 16 different colors from a palette of 512. The basic graphics modes 
are the 320x200 pixel lores with 16 colors (four bitplanes), the 640x200 medres with 4 
colors (two bitplanes), and the 640x400 two-color hires that requires a special monitor. The 
modest sound capabilities are somewhat compensated by the built-in MIDI ports that made 
Atari ST popular among musicians. (Forster, 2005, p.105, 206) 


GRAO AA 


5 i "4 i lw 





Figure 5.4: Cuddly Demos screen selector by The Carebears 


The Atari ST demoscene was marked by its direct competition with the Amiga. The same 
war was fought in mainstream media: which one is the best home computer (Saarikoski, 
2004, pp.136-137)? Apart from its faster CPU, the Atari ST was clearly in the under- 
dog position with its modest graphics and sound hardware. Effects from the Amiga were 
mimicked and improved through ingenious programming tricks, but the flow was mostly 
one-way. Some specialties existed, however. The extensive use of interactive selection 
screens in Atari megademos is a property virtually unique to ST demos. Significant effort 
was invested in the selection screens alone: at times they contained game-like features (Fig. 


5.4). In general, the amount of interactivity in demos is minimal, which makes the phe- 


87 


nomenon even more unusual. Another curiosity were the reset demos that appeared when 
the user pressed the reset button at the back of the computer. Atari megademos frequently 
featured parts from multiple groups, which is a rather uncommon feature in the scene as 
well. Atari ST demo history from 1987 to 1999 has been documented by Brandt (2000). In 
addition, three demo compilations have been produced in DVD format for easy viewing of 
Atari productions (Atariscene, 2005a,b, 2006). 


Before leaving the home computer market in the 1990s, Atari released the last member of 
the ST family: the Atari Falcon030 (1992). The machine contained several improvements 
over the STE line, such as better graphics modes (up to 640x480 in 16-bit color), a Motorola 
68030 processor running at 16 MHz, a DSP chip for audio, and 4 MB of RAM in the basic 
configuration (Forster, 2005, p.105). Falcon faced harsh competition from the Amiga 1200, 
and the PC compatibles that had become a major player by that time, and it was soon 
withdrawn from the market when Atari decided to focus on its game consoles (Atari Jaguar 
was released in 1993). Despite its low sales, the Falcon, too, has attracted some demoscene 


attention, as exemplified by pouet.net which contains 652 productions for the platform. 


5.2.3 Early PC Compatibles 


The IBM PC first appeared in 1981, as the big iron company’s attempt on the microcom- 
puter market. The machine featured the Intel 8088 processor running at 4.77 MHz, 16 
kilobytes of memory (expandable to 256), a 5.25” floppy drive and the MDA text display 
adapter or the CGA (Color Graphics Adapter). From a multimedia standpoint the PC was 
severely lacking and rather expensive too, costing $1565 at launch. IBM tried to fix the 
shortcomings by introducing the PCjr in 1983 with improved graphics and sound capabil- 
ities, but the machine did not sell well. The PCjr was also rather expensive with its $1300 
price tag. (Forster, 2005; Polsson, 1995, p.99) In comparison, the Commodore 64 was ini- 
tially sold at $595, in early 1983 already at $399 and finally, in mid-1983, at $200 (Bagnall, 
2005, pp.293-294). The prices were brought down by inexpensive PC clones that started 
appearing shortly afterwards, and ultimately turned IBM into a minor player in the market 
it had created. 


The PC has always been relatively strong CPU-wise. The IBM PC/AT (released in 1984) 
featured the fully 16-bit capable 80286 processor, roughly comparable to the Motorola chips 
of the time. The later Intel chips such as the 32-bit 80386 in 1985, and 80486 in 1989, kept 


the PC compatibles competitive with their rivals. 


The expandability of the PC was a key factor when it slowly started becoming a multimedia 
capable platform. The original CGA was only capable of four colors at 320x200 pixels, but 
the EGA (Enhanced Graphics Adapter) introduced in 1984 could already display sixteen 
colors at resolutions up to 640x350, somewhat reminiscent of the Atari ST. Finally in 1987, 
the VGA (Video Graphics Array) was released, and brought the PC in line with the Amiga 


88 


computers in terms of color and pixel resolution, if not in power. VGA featured a maxi- 
mum resolution of 640x480 pixels with sixteen colors and a lower resolution of 320x200 
pixels with 256 colors. (Polsson, 1995) The latter mode was not based on the bitplane 
paradigm popular on the competing platforms, but a chunky framebuffer, where each byte 
corresponded to a pixel. Due to its enhanced visual quality and easy programming, VGA 
quickly established itself as the demoscene standard, and only few CGA or EGA demos 


were ever made. 


The basic PC speaker sound was not suitable for games, and a number of sound cards 
started to emerge when PC clones became more popular at homes. AdLib set the standard 
for several years in 1987 with its FM synthesis, and was followed by the Creative Labs’ 
Sound Blaster line in 1989. Sound Blaster retained AdLib compatibility and in addition 
provided one sound channel for 8-bit sample playback. Gravis UltraSound, capable of 
playing 32 independent 16-bit samples at the same time, was introduced in 1991. (Gohler, 
2003) In addition to the expensive sound cards, some early PC demos also support home- 
made Covox clones, which are simple 8-bit D/A converters plugged into the parallel port. 





Figure 5.5: Second Reality by Future Crew (1993) 


The 1980s PC demoscene was small due to several reasons: the PC had not yet made its 
breakthrough as a home computer in Europe, games were scarce, multimedia capabilities 
modest, and there was no significant scene tradition on the platform yet. The situation 
started to improve in the early nineties, when PCs, too, got the capability of playing module 
music (Section 4.5.2), and the number of active groups increased. An often-mentioned 
milestone was Second Reality, a demo released in 1993 by the Future Crew (Fig. 5.5), which 
showed that PC productions could match the Amiga ones, although the level in general still 


was lower. 


89 


5.3 True 32-Bit Computers 


32-bit computers became mainstream in the 1990s and their constantly growing perfor- 
mance left its mark in the demoscene. Processor speeds in excess of one hundred megahertz, 
megabytes of available RAM, hard drives, and improved data networks made microcomput- 
ers completely different to their 1980s’ counterparts. IBM PC compatibles arrived in homes 
in increasing numbers, which eventually blurred the boundary between business and home 
computers. After Atari pulled out of the market and Commodore went bankrupt in 1994, 
the eighties’ colorful variety of incompatible platforms finally disappeared from the main- 
stream computing and was replaced by industry standards. Operating systems saw equally 
radical changes in the nineties, when Windows 95 and NT, OS/2 and Linux emerged as 
alternatives to the aging MS-DOS. 


Even before the disappearance of the two formerly big players, the era was already char- 
acterized by unification. In demos, chunky screens and effects rapidly replaced bitplane- 
oriented graphics, modules could be played on any platform with little overhead, and even 
the hardcore assembler programming gave way to effective C/C++ compilers. Ultimately, 
the demoscene had to give up its historical hardware pushing as well and start relying on 
industrial APIs. The level of abstraction increased and hardware properties had less im- 
portance, which on one hand directly contradicted with the traditional scene mentality, and 
on the other hand, let the authors focus more on the actual content. In the big picture, the 
developments emphasized the role of software in contrast to hardware, until 3D accelerators 


brought specialized graphics hardware back to focus. 


5.3.1 Amiga 1200 and 4000 


The AGA (Advanced Graphics Architecture) Amigas, A4000 and A1200 from 1992, were 
the last home computers released by Commodore before it filed for bankruptcy in 1994. At 
that time, harsh competition from the PC compatibles had already become apparent, and 
moreover, the new machines were not as exceptional as the Amiga 1000 had been in 1985. 
In 1992 the Amiga demoscene was still strong and especially the A1200 was well (if not 
unanimously) received as the next hardware platform that would let Amiga keep up with 
the PC hardware-wise. Several emotional diskmag and newsgroup discussions of the early 
1990s revolved around the new machines and their potential future, as illustrated by the 


following quotes: 


Seriously, when and this is a definate WHEN a4000 demos will be released ex- 
pect a lot of HOT demos... just think of what we can do with an 7Mhz Amiga.... 
and then *4 = one hot production... (Rat/Steel Design, alt.sys.amiga.demos, 
1992) 


90 


Some people even promised to code ONLY for AGA in the future (e.g. Lax- 
ity/Kefrens or Argon & Crash/Complex). Probably they want to be innovative 
and advanced — I call this stupidity! I bet you think, that in a year or two the 
whole scene has moved to the AGA standard, but I think you are completely 
wrong! (Rufferto/Covert Action Team, R.A.W. #6, 1993) 


In my opinion, competitions should always run on the best compatible standard 
configuration available, means on an A4000 at 25Mhz! (RokDaZone/INFECT, 
R.A.W. #7, 1994) 


The Amiga 1200 was not a complete redesign, but rather an incremental upgrade to the 
existing consumer models. The sound capabilities remained practically the same, graphics 
were still bitmap-based, processor speed increased to a modest 14 MHz (fully 32-bit with 
Motorola 68020) and the main memory grew to two megabytes. The machine also had 
an internal IDE controller for hard drives and a PCMCIA card slot for industry-standard 
expansion. Graphics modes were improved the most and the new AGA chipset could display 
256 colors using eight bitplanes with a 24-bit palette. Maximum resolutions were improved 
too, up to 1280x512 pixels interlaced, but from a demoscene perspective such development 
was rather insignificant, since contemporary effects could not have been possible in real- 


time. 


Commodore’s last attempt, the CD32 console, was released in 1993. After the bankruptcy, 
the assets were bounced from one company to another: Escom, Gateway 2000 and finally 
a spin-off known as Amiga Inc. (Knight, 1997). Amigas could still be purchased at stores 
for some years, but no new models were being developed. In the mid-nineties, a 14 MHz 
CPU and two megabytes of memory were already outdated, but for several years, third- 
party companies such as Phase 5 kept the hardware up-to-date by releasing turbo cards with 
Motorola’s 680x0 processors up to 68060 and memory expansion sockets (Chapman, 1999). 
The accelerators became popular in the demoscene circles, and the Amiga productions made 
after the mid-1990s usually require one to run acceptably. Even PPC processor boards based 
on the Motorola 603/604 chips and 3D accelerated display adapters became available over 
time (Chapman, 1999). 


PC and Amiga demos started to resemble each other in 1995 the latest. Big multiplatform 
parties, old Amiga groups moving to the PC, and similar hardware specifications can be 
recognized as some of the factors behind the development. Popular demo effects of the 
time, such as shaded vector objects, texture mapping, tunnels, various image distortions, 
and voxel landscapes were chunky by nature, meaning they were easier to implement on a 
byte-per-pixel display than on bitplanes, which gave the PC an advantage. On the Amiga, 
different solutions were tried to remedy the situation. Finally, a sufficient compromise be- 
tween speed and resolution was devised in the form of c2p (chunky-to-planar) routines that 
converted a byte-per-pixel image to a planar one. The conversion was still a computation- 


ally demanding operation, which can be noted in the reduced resolutions and drawing areas 


91 





Figure 5.6: Two chunky-type effects from Tint by The Black Lotus (1996): a tunnel and a 
morphing environment-mapped vector object. 


of contemporary demo screens (Fig. 5.6). Mikael Kalms, the author of one of the most 
known converters, has documented his approach and mathematical basis in detail (Kalms, 
1997). 


5.3.2 The Improving PC 


It is problematic to discuss the PC compatibles as one hardware platform only, since a 1981 
IBM PC shares very little with contemporary computers. A proprietary, business-oriented 
16-bit computer has turned into a multi-purpose 32- or 64-bit multimedia platform with a 
great variety of manufacturers producing their own implementations. Software-wise "the 
PC” is not a single entity either: MS-DOS was the dominant operating system in the PC 
demoscene until the late 1990s, at which point a viable alternative finally had to be cho- 
sen. In hindsight, it seems obvious that the scene would follow mainstream computing and 
eventually choose 32-bit Windows (95/98, NT and alike), but at the time, the choice was not 
as self-evident. Alternatives such as OS/2, BeOS, Linux, and even the community project 
DemOS were proposed and seriously discussed in diskmags and newsgroups. Initially, there 
was strong opposition against Microsoft that was seen as a monopoly producing slow and 


low-quality software that would not let the programmer control the machine any longer: 


And I can’t believe how the PC owners always do that what some companies 
(intel, microsoft) want: pay pay pay and get no adequate power. (FREAK, 
alt.sys.amiga.demos, 1994) 


The demo scene is a totally different world than all the microsoft business win- 


dows crap. (R. Eijkelhof, comp.sys.ibm.pc.demos, 1994) 


If someone starts to make Win95 demos, I dont have anything against it... But 
— please — don't call it "demoscene”. It would dishonourable. (Markus Aurala, 


comp.sys.ibm.pc.demos, 1996) 


92 


Positive attitudes can be found too, usually based on practical issues like improved demo 
compatibility and accessibility, which would keep the scene alive in the long run. OS/2 
and BeOS never saw any significant demoscene activity, but Linux has attracted some pro- 
grammers, and currently pouet.net contains 433 different productions for it, many of which 
multi-platform, i.e. available for other platforms as well. Conceptually, the Linux de- 


moscene serves as a link to the free software and hacker cultures. 


Hardware development in the 1990s followed the same incremental pattern that was appar- 
ent in the eighties already. Graphics modes were improved by the Super VGA (SVGA) 
cards, bringing high resolutions like 800x600 and 1024x768 pixels available to consumers 
and increasing the amount of colors to 16 and 24 bits (65,536 and 16.8 million colors). The 
aging ISA expansion bus was replaced by high-speed alternatives such as the VESA local 
bus (486 only), PCI (Peripheral Component Interconnect), and AGP (Accelerated Graphics 
Port) that allowed for significantly better throughput to the graphics adapter. The demoscene 
welcomed faster graphics, and since the mid-1990s several productions started supporting, 
or even requiring, graphics cards beyond the basic VGA standard. Processor speeds climbed 
from tens to hundreds of megahertz, with Intel Pentium being a notable milestone in 1993. 
The increasing processing power was not received unanimously, since it conflicted with the 


demoscene tradition of optimizing: 


Where does it stop then? If it goes on like this, the future demos will support 
pentium only (or run decently on a pentium)! I think the limit MUST be drawn 
by a 486DX33 for now and in the near future. If your demo runs smooth on 
a 486DX2-66 and not on a 486DX33 i advise you to quit coding and learn 
to optimize (or learn assembly ;) ). (Scout/Success, comp.sys.ibm.pc.demos, 
1994) 


While the PC scene had played catch-up with Amiga until the mid-nineties, after that the 
situation was slowly reversed and it was the PC’s turn to take the technical lead. Especially 
3D vector graphics, well-suited for the hardware, received considerable attention. Texture, 
environment and bump mapped objects could be seen in numerous demos. The term object 
show refers to demos where 3D objects would be rotated on the screen without much other 
content. Another related concept is the flyby, a simulated flight in a vector world, again 
with little other content. The next revolution in demo graphics took place when affordable 
3D accelerators became available in the late 1990s (discussed in Section 5.3.4). The avail- 
ability of true-color graphics modes made demos more colorful than before and enabled the 
improved shading and blending of colors in effects. On the other hand, the improving pixel 
and color resolutions were not as suitable for completely hand-drawn graphics as the old 
low resolutions had been. The same phenomenon was taking place in the field of new me- 
dia on a larger scale as well and pixeling started to be replaced by true-color photo editing 
and composition (Sections 4.3.2 and 4.4.1). An example of the late nineties’ demo style is 


portrayed in Figure 5.7. 


93 





Figure 5.7: Inside by CNCD (1996): shaded texture mapping and an overlay with a popular 
culture reference 


5.3.3 Other 32-bit Demo Platforms 


Some other 32-bit platforms have received their share of demos as well. Rather interestingly, 
the Macintosh line of computers, often regarded as art and media oriented, has attracted 
little productions so far. The Macintosh has not been a significant platform for computer 
games either, which reflects its different use. However, the hardware differences (CPU 
family, external connectors and expansion cards), that once clearly separated the Mac and 
PC computers became less important after the year 2000 and practically disappeared in 
2006, when Apple too moved to the Intel x86 processors. After that, the only meaningful 
difference has been the operating system. Observation of pouet.net reveals that it was only 
after the introduction of Mac OS X in 2001 that the demoscene became interested in the 
platform. The state of Macintosh demoscene five years ago was documented in the Mac 
Life magazine by (Beck, 2004). 


The traditionally closed development model has hindered the attempts to create demos for 
game consoles. According to pouet.net, demos for popular platforms like Sony’s PlaySta- 
tion 2, Microsoft’s Xbox line and Nintendo’s GameCube or Wii are rare experiments as of 
now. Sega’s Dreamcast has seen a little more activity, but nothing that could be called a full- 
blown demoscene. The most active contemporary consoles are the handhelds: Nintendo’s 
Game Boy Advance, Sony’s PlayStation Portable (Fig. 5.8), and the Korean open-source 
handheld GP2X each have a selection of tens of demos. 


Other 32-bit platforms with demos are mobile phones and PDAs. They are usually based 
on the ARM line of RISC processors, common in many portable battery-operated devices. 


94 





Figure 5.8: Suicide Barbie by The Black Lotus running on PlayStation Portable 


Mobile demos can be taken as just one platform experiment more, but their existence also 
hints at the professional lives of some of the community members, who most likely have 


ended up as developers in the large mobile hardware and software industry. 


The increasing level of abstraction in software development has effectively hidden the un- 
derlying hardware and, in many cases, even the operating system. Such a trend can be 
observed in demoscene productions as well. While the demos of the eighties and early 
nineties were tied to the hardware and exploited it directly, API-oriented programming has 
largely detached demo programming from a particular platform. The existence of multi- 
platform productions, unimaginable in the eighties, proves the observation. Cross-platform 
APIs such as OpenGL (Woo et al., 1999) are available on Windows, Linux and Mac OS X, 
and thus it has become possible to recompile demos across platforms with comparably little 
extra effort. The web browser is another example of a hardware-neutral application plat- 
form, and demos written in Java, JavaScript or Flash are readily viewable on any modern 


computer with no extra porting effort. 


5.3.4 3D Acceleration 


The Voodoo line of graphics cards, released in 1996 by 3dfx Interactive, introduced acceler- 
ated 3D vector graphics to the mass market. Until that point 3D acceleration had practically 
been restricted to expensive professional equipment. Other contemporary contestants in the 
market were Matrox and ATI. A bit later, Nvidia joined the competition with its popular 
Riva and GeForce series. Demoscene productions continued to be software rendered until 
the end of the 1990s, when big parties like Assembly started to feature competitions allow- 
ing 3D acceleration (Assembly Organizing, 1999). Figure 5.9 is a screenshot of Virhe, an 


early accelerated demo from Assembly’99. 


95 





Figure 5.9: An early 3D demo using the Glide API: Virhe by MatureFurk (1999) 


The migration to 3D accelerators was an important milestone on the path from low-level 
hacking to system-friendly programming. In practice, it was impossible to utilize the accel- 
erator chips without using the vendor-provided API. The Voodoo cards featured their own 
proprietary Glide interface, but also OpenGL soon became popular, when it was adapted 
from the Silicon Graphics workstations to provide increased functionality and compatibility 
with professional graphics. OpenGL and its newer variants still remain the only low-level 
cross-platform 3D APIs, but their popularity has been somewhat shadowed by the wide 
use of Microsoft’s Direct 3D in Windows and Xbox games. As of 2009, Windows demos 
are still released for both interfaces. Twenty of the demos listed in Appendix B are 3D 


accelerated. 


From a programmer’s point of view, the first accelerators not only offered new possibilities, 
but also set new limits. Improved resolutions, high polygon counts and good frame rates 
were the indisputably positive effects, but the fixed 3D pipeline (Woo et al., 1999, p.11) 
was rigid, and some flexibility was lost in contrast to software-based rendering. Individual 
pixels could no longer be controlled directly, and visual output had to be constructed of the 
few primitives offered by the API. The programmable pipeline only reappeared in the early 
2000s in the form of shaders that let the programmer control the rendering process in more 
detail. Demo aesthetics did not change as rapidly as the technology did, and in practice, 
the first accelerated demos were like any other productions of the time, albeit with a high 


resolution and polygon count. 


96 


5.4 Roles of Software 


Quite obviously, hardware alone does not constitute a platform. Various technically ad- 
vanced, but commercially failed hardware designs found in the history of computing high- 
light the importance of application software—and marketing. So far we have discussed the 
role of hardware and operating systems in relation to the demoscene activities. Here the 
focus is on the concepts of authoring tools and algorithms: what kind of tools does demo 


making require? What kinds of connections exist between demo programming and science? 


5.4.1 Tools 


Software development tools are the essential foundation for demo production. The real-time 
nature of scene productions has, until recently, required the use of low-level programming 
languages to achieve sufficient performance. There are certain signs of change, such as the 
use of browser-based technologies (Section 5.3.3). The use of assembly language, in the 
form of machine language monitors and assemblers, was a common demoscene practice 
until the mid-1990s, reflecting also the aim to control the machine completely. The emer- 
gence of high-performance C/C++ compilers rapidly changed demo programming in the 
later half of the nineties. Once again the change of paradigm did not go completely with- 
out opposition, as illustrated by the comment of Shadow Weaver in comp.sys.ibm.pc.demos 
(1994): 


All the majorly cool ones always were coded in complete assembler for com- 
plete speed, if not almost ALL Amiga demos were in assembler. I think the 
most of the flames are coming from people that JUST DON’T KNOW assem- 
bly language well enough to do all the stuff in, so they stick to their C++ and 


figure it works well enough. 


The so-called demomakers are interesting curiosities among demo development tools. With 
little or no programming experience, it is possible to put together a production with effects 
and music. Figure 5.10 is a screenshot of the RSI Demomaker for the Amiga from 1991. 
Seemingly, the purpose of demomakers is to make demo creating easy and accessible to 
anybody. However, in reality the situation is not as straightforward, since productions made 
with ”too easy” tools are frequently frowned upon in the discussions. Taking shortcuts is not 
in line with the demoscene concepts of authorship, skill and control. The same phenomenon 
is illustrated by the low amount of Flash productions, even if the program itself would seem 


fit for creating demos at first sight: 


Anyway, flash demos are not real demos, at my opinion, except if they’re made 
fully using a SWF ActionScript editor, since a real demo is a coder demo. 
(BadSector, pouet.net discussion ”"TEH FLASH DEMO - SCENE”, 2002). 


97 


RED SECTOR DEMOMAKER V1 R ttt 
¿ML rata li PAIKAN AML oe ee la 


-PREFERENZLIST- 


-- GRAPHICS -- 


LITTLELOGD i 


FONT lexic: 
FONT 8x8; 

ONT 16%16 1PL 
i aa | 


ITEOBJECT 8: 
SPRITEOBIECT 1: 





Figure 5.10: RSI Demomaker by Red Sector Inc. and TCC Design (1991) 


The discussion thread ”Your Software for musicproductions, GFX, Coding is ?” (2007— 
2009) found on pouet.net with its 166 messages offers a general overview of the tools used 
in the most recent productions, even if not all the tools reported were used for strictly demo- 
related purposes. While such a sample is far from conclusive, some trends do appear. Table 


5.1 contains the most freguently mentioned tools. 















































Category and Tool | # | Category and Tool | # 
Programming 2D Graphics 
Visual Studio 48 Photoshop 38 
GCC 36 GIMP 16 
NASM 12 Deluxe Paint 8 
Devpac 5 Paint Shop Pro 8 
FASM 5 MS Paint 5 
Music Tools 3D Graphics 
Renoise 19 Blender 14 
Modplug Tracker | 13 3ds Max 11 
Reason 12 SoftImage XSI 3 
MilkyTracker 11 LightWave 3 
Ableton Live 11 Cinema 4D 3 

















Table 5.1: Popularity of demo-related authoring tools (2007-2009) 


C/C++ compilers, such as Visual Studio and GCC, are the most common demo program- 
ming tools at the moment, due to the powerful code they generate. The GCC tool chain 
is often used as a cross-compiler for developing productions for exotic platforms, where it 
might be the only option available. Adobe’s popular Photoshop is equally popular among 


the demoscene, and its free alternative GIMP has found its user base as well. Rather in- 


98 


terestingly, the aging Deluxe Paint, popular in the nineties, is still used to some extent. 
3D graphics are most commonly produced with the free Blender, and the commercial 3ds 
Max that is widely used for realtime applications, such as video games. Other 3D packages 
mentioned represent commercial feature-rich software, comparable to 3ds Max. The strong 
tracker tradition of the scene is reflected in its music tools: three out of five most popular 
music applications are trackers. Reason is a professional sequencer, and Ableton Live a 


realtime DJ tool that can also be used as a sequencer. 


If a suitable commercial tool does not exist one will be created. Perhaps the best examples of 
such activity are the various trackers that have existed since the late 1980s. The demoscene 
largely adopted the tracker paradigm, and since trackers, aside from few exceptions (see 
Section 4.5.2), were not commercially available, the solution was to program new ones. In 
addition, assemblers, executable compressors, and even complete paint programs have been 


created in the scope of demoscene activities, according to the demo tool list of pouet.net. 


5.4.2 Algorithms 


Demo-related programming does not fundamentally differ from any other software devel- 
opment, and thus the use of algorithms and data structures is part of the process. 3D graph- 
ics, sound programming and data compression are examples of advanced topics, requiring 
university-level knowledge of mathematics and software science. The focus on impressive 
audiovisual output, instead of factors like mathematical correctness, lets the programmer 
cut corners and use ad hoc solutions, as long as the end result appears interesting, but some 


topics are ultimately so complex that external reference material is required. 


It appears that the connection between demos and science has grown stronger by time, when 
the complexity of effects has increased. Early effects such as scrollers, colorbars and mov- 
ing sprites could be achieved with clever ad hoc programming and rudimentary trigonom- 
etry. The increasing popularity of 3D graphics in the nineties raised the bar, with topics 
such as transformation matrices, shading algorithms, texture mapping and depth sorting. 
Fractal-based graphics, most commonly Mandelbrot or Julia sets, can be found in numerous 
productions. Software sound synthesis (Section 4.5.4) involves various methods of signal 
processing. These examples propose that the scene is aware of many scientific findings that 
support its activities. At times, the links are even directly emphasized, as seen in Figure 
5.11. 


The diffusion of scientific knowledge inside the demoscene is a complex topic, so the dis- 
cussion here is intentionally brief, especially since no interviews were conducted as part 
of the study. In general, it seems that new effects and techniques are first introduced by a 
few innovators, popularized by high-profile productions and then rapidly diffused to other 
groups, well in line with the diffusion of innovations theory by Rogers (2003). Forerun- 


ners function as mediators between the scene and the scientific world, and later adopters 


99 


ong 


mapping 





Figure 5.11: Dope by Complex (1995) 


learn about new methods in the context of the demoscene discourse. When compared to the 
diffusion process of new hardware and software platforms [see Section 5.5.1 and Reuna- 
nen & Silvast (2009)], new effects appear to diffuse painlessly with little or no opposition: 
platforms have traditionally been a source of constant controversy in diskmags and news- 
groups, whereas new effects have mostly been embraced. One likely explanation to this 
phenomenon is that new effects and methods only serve as neutral instruments when trying 


to create impressive works, as opposed to platforms that engage the user on several levels. 


5.5 Effects of Changing Technology 


The demoscene has little influence on the constantly changing landscape of information 
technology. New hardware and software platforms are introduced each year, and existing 
ones eventually fade into obscurity. From a scene perspective, the changes appear external, 
uncontrollable and inevitable by nature. The community cannot affect such developments 
to any significant degree, but it can react to them through discussion and reinvention. At the 
other end of the scope, old productions and obsolete platforms constitute a large part of the 


history of the scene. 


5.5.1 Patterns of Diffusion 


As suggested by Turkle (1984, 1997) throughout her work, computers involve people on 
an intimate and personal level, serving as mirrors of the human spirit. Several discussions 
observed in the course of this study were marked by their highly emotional tone. Mere resis- 
tance to change is not a sufficient explanation to why new, seemingly beneficial innovations 
have been so fiercely opposed. After all, we are dealing with young proficient computer 
users that have few problems when learning to use new technology. It would seem that 


computers indeed operate beyond an instrumental level, and that an emotional bond is cre- 


100 


ated between the user and the machine. Suggesting that a platform should be abandoned 


directly violates the bond and calls for counteraction. 


Despite the initial opposition, the demoscene eventually adapts to mainstream computing 
trends. The inconvenience of going with abandoned or alternative platforms is too high to be 
tolerated for an extended period. Sceners in general are not among the first adopters. As an 
example, the 32-bit Microsoft Windows did not become a widely accepted demo platform 
before the late 1990s, almost five years after its introduction. Demoscene practices had to be 
adapted to the new situation (no more direct hardware access), and the new platform had to 
offer something worth migrating (3D acceleration and a unified multimedia API). Another 
relevant factor was the increasing commercial pressure: new computers were not available 
with the old operating system any longer. Sceners, aside from being artists and enthusiasts, 
are consumers too. Seen from a utilitarian point of view, the skills acquired with MS-DOS 
were becoming obsolete in the working life, even though the importance of such a factor 


might actually be rather negligible in this context. 


The general concepts of the diffusion of innovations theory (Rogers, 2003) can be applied 
to the demoscene’s adoption processes. When a new hardware or software platform ap- 
pears, the innovators try them out and release experimental productions, but the majority 
needs more time to adopt, as can be observed for example in the production catalog of 
pouet.net. It is only after the opinion leaders adapt that the new innovations start to gain 
momentum. High-profile productions, made by famous groups (early adopters) for a new 
platform, awaken the interest of the community, and when critical mass is achieved the ma- 
jority will eventually follow. Some sceners will never adopt or require considerably more 
time, comparable to the laggards of the diffusion theory. Large parties, the most important 
channel for publishing new works, both reflect and set trends by allowing the use of new 
platforms in their competitions. As noted above and in Section 2.1.4, an innovation needs 
to be compatible with the community practices to ensure rapid adoption. Another factor in 
the process is the complexity of the innovation. For example, to switch from the Amiga 500 
to the Amiga 1200 was a small step, but to switch from MS-DOS to Windows meant that 
completely new tools had to be learned. Such added complexity partly explains why the 


former migration took place with relative ease when compared to the latter. 


Aune (1996) discusses the adoption process from a domestication standpoint: how do new 
technologies such as computers become parts of everyday life? Aune’s third definition of 
domestication contains an interesting connection to the demoscene activities: To tame or 
bring under control”. The striving for control over the machine is a common theme in the 
scene discourse. The arrival of a new platform means loss of control and increased uncer- 
tainty, which can only be conquered by mastering the new tool. Another relevant obser- 
vation made by Aune is that domestication happens on many levels (individual, household 
and society). Likewise, a demoscene member is part of a group and the community as a 


whole when making his personal decisions. The individual does not operate in a void, but 


101 


constructs his relationship to new technology through negotiation with others. The twofold 
process of adapting an artifact to already existing routines, and adjusting the routines to the 
artifact, can be observed in the demoscene as well. Intrestingly, Aune even mentions de- 
mos in her article, although she uses imprecise wording such as Amiga clubs” and ” game 


freaks” when referring to demo groups. 


Another study from a domestication point of view was conducted by Lehtonen (2003), who 
used mobile phones and digital television as his case examples. Lehtonen claims that new 
technologies are domesticated little by little, through various knowledge-producing trials 
where both the user and the technology are tested (another example of influences going 
both ways). Rogers’ straightforward diffusion model is contested, but overall, the article 
can be seen as part of the diffusion discourse with a different approach and terminology. 
Lehtonen, too, notes that from a user’s perspective technology may appear uncontrollable, 
something that can only be reacted to. The concept of warm experts, used by Lehtonen, is 
well in line with the work of Aune (1996) and Rogers (2003): individuals seek help from 


their trusted peers when making decisions about adopting or discarding an innovation. 


5.5.2 Digital Heritance of the Scene 


The constantly changing hardware and software platforms pose a significant challenge to 
the accessibility of old demos. During the first two hardware generations the problem could 
be omitted, but now old software, computers and their accessories are increasingly becom- 
ing rarities. The original manufacturers like Commodore and Atari Computers are out of 
business, and even if they were not, supporting the old models would not be commercially 
viable. A notable property of the digital artifacts is their fragility: on one hand they can be 
infinitely copied but the loss of media, or even a small fraction of it, can make it impossible 
to successfully recover the artifact. The Charter on the Preservation of the Digital Heritage 
published by UNESCO (2003) recognizes the following challenges: 

e Making digital heritage accessible 

e Rapid obsolescence of hardware and software 

e Lack of resources for preservation efforts 

e Lack of preservation strategies and methods 

e Inadequate legislation 

e Low public awareness 


e Creation of digital objects that last 


e Selecting what should be preserved 


102 


e Ensuring the authenticity of digital heritage 
e Preserving not only locally but globally 


e Collaboration between the different stakeholders 


By examining the list, it can be noticed that the challenges are not only technological or eco- 
nomical. The lack of strategies, methods, skills and awareness equally hinder the preser- 
vation efforts. While UNESCO’s pamphlet is mostly concerned with governments, large 
institutions and companies, the principles have utility for the grassroots level as well. As 
an example of this Heinonen & Reunanen (2009) discuss the challenges faced by Finnish 


private collectors in relation to required national efforts. 


Media art is a domain that has already been struck by the dilemma of rapidly aging hardware 
and software. Both Paul (2007) and Mark Tribe [as quoted by Wands (2006, p.206)], the 
founder of the pioneering net art community rhizome.org, have proposed similar measures 
to ensure the continuous availability of works of digital art: migration to newer platforms 
and emulation of old platforms on new ones. Tribe mentions the importance of documenta- 
tion and the possibility of a complete recreation of a work. Although somewhat opposed to 
the idea (in fear of turning art institutions into ”computer museums”), Paul also discusses 


the possibility of collecting old hardware and software to ensure continuity. 


Preservation efforts of the demo community have been strikingly similar to the ones pro- 
posed above. Emulation is widely utilized to run old demos and, because of its already 
good accuracy and availability, even for developing software for aging platforms. The re- 
markable amount and quality of emulators can be attributed to the increasing worldwide 
popularity of retro phenomena. Disk images are conversions of physical diskettes into files, 
intended to be used with emulators and suitable for long-term archival. Documentation 
in the form of screen shots and descriptions takes place on various community websites 
(see 2.3). Demo archives preserve software and make it accessible to the community. It is 
becoming increasingly commonplace to make video captures of demos, which can be con- 
sidered as a form of migration. Dedicated sites such as capped.tv and demoscene.tv offer the 
videos for online viewing. The popular youtube.com is also utilized for distribution. Some 
notable demo DVDs have been published to enable the viewing of classic works in good 
quality (Atariscene, 2005a,b, 2006; Demo or Die!, 2005a,b,c, 2006a,b; Digital Memories, 
2006; MindCandy, 2002, 2006). Another form of migration is the porting of productions to 
other platforms, as discussed already in 5.3.3. Some productions have been fixed to run on 
improved hardware (a typical example being from Amiga 500 to the AGA Amigas). The 
complete recreation of demos is rare, because other forms of preservation have rendered the 


laborious task of recreation unnecessary. 


103 


Chapter 6 


Conclusion 


The study revealed a devoted community that continuously re-evaluates itself through dis- 
cussion. Constant competition between groups and individuals has left its mark in many 
practices of the community. One could even say that recognition is the only currency of the 
scene. The great variety of artifacts became evident as well: several thousands of demos, 
intros of different types, pictures, tunes, disk magazines, group emblems, text files, and 
party-related items have been created during a time span of twenty-five years. The change 
in the audiovisual quality of demos was something that stood out when analyzing the arti- 
facts. What started as single-screen intros with moving text later evolved into a complex 


form of media art, requiring years of learning to master. 


Two main observations made about the relationship between technology and creativity are 
that the capabilities of the hardware have always been directly reflected in the works of scene 
art, and that the demoscene’s aspirations have increasingly shifted towards self-expression 
from technically-oriented hardware pushing, when the tools available have allowed such 
development. Necessary tools have also been created by the community to fill in the gaps. 
Mastery of both hardware and software was a reoccurring theme: in accordance to its mind 
over matter mentality, the scene has encouraged its members to do the impossible and 


develop their personal skills further. 


The most important contribution of this thesis is the broad overview of different topics 
related to the demoscene. While several other introductions to the phenomenon do exist, 
they have been brief and remained on the easily observable surface of the community. As 
such, the thesis serves as a stepping stone for researchers interested in demos: efforts can 
be focused on advanced topics instead of the trivial mapping of the domain. The different 


types of demoscene-related artifacts had not been documented in detail previously. 


The study largely focused on the two content analyses. The analysis of disk magazines can 
be considered a unique approach. So far disk magazines have gone largely unnoticed by 


researchers, in spite of their rich content that sheds light on contemporary topics. Likewise, 


104 


the content analysis of demo effects—even if rudimentary—revealed clear trends that had 
only been estimated in earlier publications: what were the dominant effects of each era, how 
they reflected the capabilities of the hardware, and how they eventually disappeared after 
being made obsolete by new inventions. Even more connections could be made between 


demos and other domains in order to build a richer history of the events. 


The breadth of the presentation does not come without problems. While the variety of topics 
covered is high, many important themes worth further attention had to be discussed briefly 
in only a few paragraphs. After constructing a general overview of the phenomenon the next 
logical step would be to study certain topics in depth. Some examples of such topics could 
be the composition of demo effects, the use of tools, or the diffusion of different innovations 
in the scope of the demoscene. Disk magazines still have more to offer and would constitute 
a worthy theme as well. A largely uncharted territory such as the demoscene would provide 
possibilities for many related fields of study, far beyond my own scope: youth culture or 
subculture research, art history, gender studies, and software science, to name just a few, 


could each provide their novel approach to the topic and add missing pieces to the puzzle. 


105 


References 


Printed Publications 


ANSI, 1977. ANSI (American National Standards Institute) Standard X3.64—1977. 
ASA, 1963. American Standard Code for Information Interchange, ASA X3.4—1963. 


Aune, M., 1996. The Computer in Everyday Life. Patterns of Domestication of a New 
Technology. In Lie, M. & Sgrensen, K.H. (eds.), Making Technology Our Own: Domesti- 
cating Technology into Everyday Life. Scandinavian University Press, pp.91—120. 


Bagnall, B., 2005. On the Edge: The Spectacular Rise and Fall of Commodore. Variant 


Press. 


Barbat, N., 2005. You Can't Do This on Xbox! SHizZLE—Rumbling with 4 MHz. 
SCEEN #1, pp.42-44. 


Beck, J., 2004. Lebenszeichen aus dem Untergrund: Mac-Demoscene. Mac Life 03.2004, 
pp.99-100. 


Borzyskowski, G., 1996. The Hacker Demo Scene and Its Cultural Artifacts. Curtin Uni- 
versity of Technology. 


Botz, D., 2008. Hacker-Asthetik. Die geschichtlichen, technischen und gestalterischen 
Voraussetzungen zum Verständnis des Mediums Computerdemo (Hacker Aesthetics. His- 
torical, technological and artistic premises of Computer Demos). Ludwig-Maximilians- 


Universität Miinchen, Institut fiir Kunstpädagogik. (Unpublished manuscript). 


Brooks, F.P., 1975. The Mythical Man-Month: Essays on Software Engineering. Addison— 
Wesley. 


Burger, B., Paulovic, O. & Hasan, M., 2002. Realtime Visualization Methods in the De- 
moscene. In Proceedings of the Central European Seminar on Computer Graphics 2002, 
pp.205—218. 


Carlsson, A., 2008. Chip Music: Low-tech Data Music Sharing. In Collins, K. (ed.), From 
Pac-Man to Pop Music. Ashgate, pp.153—162. 


106 


Commodore-Amiga, 1991. Amiga Hardware Reference Manual. Third edition. Addison- 
Wesley. 


Cruz, F., 2005. Mobile Mayhem. Demo Platforms of the 21st Century: Cell Phones. 
SCEEN #1, pp.52-56. 


Dittbrenner, N., 2007. Chip-Musik: Computer- und Videospielmusik von 1977-1994 (Chip 
Music: Computer and Video Game music from 1977 to 1994). Universitát Osnabriick. 


Faler, M., 2001. Johdatus demosceneen (Introduction to the Demoscene). In Sihvonen, T. 
(ed.), Sähköä, Säpinää, Wapinaa: Risteilyjä teknologian kulttuurihistoriassa (Electricity, 
Action, Shaking: Cruises in the Cultural History of Technology). Turun yliopiston historian 
laitoksen julkaisuja nr. 59, Turun yliopisto (University of Turku), pp.7—23. 


Forster, W., 2005. The Encyclopedia of Game.Machines: Consoles, Handhelds & Home 
Computers 1972—2005. Gameplan. 


Franke, H.W., 1971. Computer Graphics—Computer Art. Phaidon Press. 


Goryunova, O., 2007. Art Platforms. The Constitution of Cultural and Artistic Currents on 
the Internet. Media Lab, University of Art and Design Helsinki. 


Grau, O. (ed.), 2007. Media Art Histories. MIT Press. 


Haavisto, M., 2001. Diskmagit haastavat lukijan (Diskmags Challenge the Reader). Enter 
7/2001, p.72. 


Heinonen, M. & Reunanen, M., 2009. Preserving Our Digital Heritage: Experiences from 


the Pelikonepeijoonit Project. In History of Nordic Computing 2. Springer, pp.55—64. 


Hapnes, T., 1996. Not in Their Machines. How Hackers Transform Computers into Sub- 
cultural Artefacts. In Lie, M. & Sgrensen, K.H. (eds.), Making Technology Our Own: Do- 
mesticating Technology into Everyday Life. Scandinavian University Press, pp.121-150. 


Inkinen, S. & Salmi, M., 1996. Media aseena ja työkaluna — hakkereita, teknohippejá 
ja koneromantiikkaa uuden median verkoissa (Media as a Weapon and a Tool—Hackers, 
Techno Hippies and Machine Romanticism in New Media Networks). In Tarkka, M., Hin- 
tikka, K.A. & Mákelá, A. (eds.), Johdatus uuteen mediaan (Introduction to New Media). 
Edita, pp.90-91. 


Johnson, S.A., 1999. Interface Culture. Basic Books. 


Kauppinen, J., 2005. Demoskenen alakulttuurista nousee suomalaisen it:n kárki (Top 
names of the Finnish IT Industry Rise from the Demoscene Subculture). Aamulehti 7 July, 


pp.20. 


Krippendorff, K., 2004. Content Analysis: An Introduction to Its Methodology. Second 
Edition. Sage Publications. 


107 


Kurki, R., 2002. WE ARE! Tutkimus postmodernista identiteetistä sukupuolittuneissa 
yhteisöissä (A Study on the Postmodern Identity in Gendered Communities). Lahden am- 
mattikorkeakoulu (Lahti Polytechnic). 


Lehtonen, T-K., 2003. The Domestication of New Technologies as a Set of Trials. Journal 
of Consumer Culture, 3(3), pp. 363-385. 


Levy, S., 1994. Hackers. Dell Publishing. 


Lönnblad, H., 1997. Kahden tietokonedemon vertaileva analyysi (A Comparative Analysis 
of Two Computer Demos). Musiikin Suunta (Musical Currents), 19(2), pp.28—34. 


Lönnblad, H., 1998. Tietokonedemot kulttuurina ja musiikin harrastuksen muotona (Com- 


puter Demos as Culture and a Form of Musical Hobby). University of Helsinki. 
Manovich, L., 2001. The Language of New Media. MIT Press. 


McLuhan, M., 1994. Understanding Media: The Extensions of Man. 1st MIT Press ed. 
MIT Press. 


Motorola, 1992. M68000 Family Programmer's Reference Manual. Motorola. 





Myers, B.A, 1998. A Brief History of Human-Computer Interaction Technology. Interac- 
tions, 5(2), pp. 44-54. 


Nordli, H., 2001. From Spice Girls’ to Cybergirls: The Role of Multimedia in the Con- 
struction of Young Girls’ Fascination for and Interest in Computers. In van Lieshout, M., 
Egyedi, T. & Bijker, W. (eds.) Social Learning Technologies. The introduction of multime- 
dia in education. Ashgate Publishing, pp.110-133. 


Nordli, H., 2003a. The Net Is not Enough: Searching for the Female Hacker. Norwegian 


University of Science and Technology. 


Nordli, H., 2003b. The Gathering—Computer Parties as Means for Gender Inclusion. IST- 
2000-26329 SIGIS, Deliverable Number D04. 


Paul, C., 2007. The Myth of Immateriality: Presenting and Preserving New Media. In 
Grau, O. (ed.), Media Art Histories. MIT Press, pp.251-274. 


Polgar, T., 2005. Freax. The Brief History of the Demoscene. Volume 1. CSW Verlag. 


Rehn, A., 2004. The Politics of Contraband—The Honor Economies of the Warez Scene. 
Journal of Socio-Economics, 33, pp.359-374. 


Reunanen, M. & Silvast, A., 2009. Demoscene Platforms: A Case Study on the Adoption 
of Home Computers. In History of Nordic Computing 2. Springer, pp.289-301. 


Rogers, E.M., 2003. Diffusion of Innovations. 5th ed. Free Press. 


108 


Roininen, T., 1998. Demoscene: että tietyt aineettomat arvot ja tavat yhdistävät joukkoa 
ihmisiä (Demoscene: That Certain Immaterial Values and Manners Unite a Group of 


People). Lapin yliopisto (University of Lapland). 


Saarikoski, P., 2001a. Valtavirtaa vastaan — Demoscene suomalaisen kotimikroilun histo- 
riassa (Against the Mainstream—Demoscene in the History of Finnish Home Computers). 
Lähikuva 3/2001, pp.54—65. 


Saarikoski, P., 2001b. Pioneerien leluista kulutuselektroniikaksi. Suomalainen kotimikro- 
harrastus tietotekniikan murroksessa 1980-luvun alusta 1990-luvun puoliväliin (From Pio- 
neer Toys to Consumer Electronics. The Finnish Home Computer Hobby During the Com- 
puter Revolution from Early 1980s to Mid-1990s.). Turun yliopisto (University of Turku). 


Saarikoski, P., 2004. Koneen lumo. Mikrotietokoneharrastus Suomessa 1970-luvulta 1990- 
luvun puoliväliin (The Lure of the Machine. Finnish Microcomputer Hobby from 1970s to 
mid-1990s). Nykykulttuurin tutkimuskeskuksen julkaisusarja, Jyväskylän yliopisto (Uni- 
versity of Jyväskylä). 


Saarikoski, P. & Suominen, J., 2009. Pelinautintoja, ohjelmointiharrastusta ja liiketoimin- 
taa. Tietokoneharrastuksen ja peliteollisuuden suhde Suomessa toisen maailmansodan jäl- 
keen (Enjoyment of Games, Hobbyist Programming and Business. The Relationship of 
Computer Hobbyism and the Game Industry in Finland after the Second World War). In 
Suominen, J., Koskimaa, R., Mäyrä, F. & Sotamaa, O. (eds.), Pelitutkimuksen vuosikirja 
2009 (Yearbook of Game Research 2009). Tampereen yliopisto (University of Tampere), 
pp.16-33. 


Scholz, A., 2007a. Iconoclash. Opium for the masses. SCEEN #2, pp.50-56. 


Scholz, A., 2007b. Teenyweenyfantastikum. A Whole World in a 4 kilobyte Nutshell. 
SCEEN #2, pp.72-75. 


Scheib, V., Engell-Nielsen, T., Lehtinen, S., Haines, E. & Taylor, P., 2002. The Demo 
Scene. In ACM SIGGRAPH 2002 Conference Abstracts and Applications. ACM Press, 
pp.96-97. 


Stamnes, B., 2004. Computer Parties. Case: The Gathering. In Tasajárvi, L. (ed.), Stamnes, 
B. & Schustin, M., Demoscene: The Art of Real-Time. Even Lake Studios & katastro.fi, 
pp.45—52. 


Tasajárvi, L., 2003. demoskene katastro.fi —The Exhibition. Arsis, 1/2003, p.9. 


Tasajärvi, L. (ed.), Stamnes, B. & Schustin, M., 2004. Demoscene: The Art of Real-Time. 
Even Lake Studios & katastro.fi. 


Thomas, D., 2003. Hacker Culture. University of Minnesota Press. 


109 


Tolonen, T., Välimäki, V. & Karjalainen, M., 1998. Evaluation of Modern Sound Synthe- 
sis Methods. Report 48. Helsinki University of Technology, Department of Electrical and 


Communications Engineering, Laboratory of Acoustics and Audio Signal Processing. 
Tribe, M. & Jana, R., 2007. New Media Art. Taschen. 
Turkle, S., 1984. The Second Self: Computers and the Human Spirit. Simon & Schuster. 


Turkle, S., 1997. Life on the Screen: Identity in the Age of the Internet. First Touchstone 


edition. Touchstone. 
Vigh, D. & Polgar, T., 2006. Freax Art Album. CSW Verlag. 


Vuorinen, J., 2007. Ethical Codes in the Digital World: Comparisons of the Proprietary, the 
Open/Free and the Cracker System. Ethics and Information Technology, 9(1), pp.27-38. 


Wands, B., 2006. Art of the Digital Age. Thames & Hudson. 


Woo, M., Neider, J., Davis, T. & Shreiner, D., 1999. OpenGL Programming Guide: The 
Official Guide to Learning OpenGL, Version 1.2. Third Edition. Addison—Wesley. 


Online Material 


AMP, 2002. Amiga Music Preservation — Home. [Online] Available at: 
http://amp.dascene.net/ (updated 12 May 2009) [Accessed 28 May 2009]. 





Assembly Organizing, 1994. ASSEMBLY *94—THE THIRD PHASE. [Online] Available 
at: ftp://ftp.scene.org/pub/parties/1994/assembly94/info/asm-94.zip [Accessed 13 June 
2009]. 





Assembly Organizing, 1998. ASSEMBLY ’98 OFFICIAL INVITATION TEXT. [Online] 
Available at:  ftp://ftp.scene.org/pub/parties/1998/assembly98/info/asm98inv.txt [Ac- 
cessed 3 June 2009]. 





Assembly Organizing, 1999. ASM99 >> COMPOS >> RESULTS. [Online] Available at: 
ftp://ftp.scene.org/pub/parties/1999/assembly99/results.txt [Accessed 26 June 2009]. 





Assembly Organizing, 2006. Demo Compos—Assembly. [Online] Available at: 
http://www.assembly.org/2006/compos/realtime/demo [Accessed 13 June 2009]. 





Assembly Organizing, 2008. Demo Compos—Assembly. [Online] Available at: 





http://www.assembly.org/summer08/compos/realtime/demo [Accessed 13 June 2009]. 


Bauer, C., 1996. The MOS 6567/6569 video controller (VIC-II) and its application in the 
Commodore 64. [Online] Available at: http://www.cebix.net/VIC-Article.txt [Accessed 23 
June 2009]. 





110 


Beck, J. & Sobiraj, L., 2008. The Demoscene and modern games. [Online] Available at: 
http://zine.bitfellas.org/article.php?zine=13 &id=19 [Accessed 2 June 2009]. 





Brandt, F., 2000. Atari ST demo history. [Online] Available at: 
http://www7.in.tum.de/-brandtf/ataridemos.html (updated in 2007) [Accessed 24 June 
2009]. 





Breakpoint Organizing, 2009. Breakpoint 2009 — Everything Is Under Control. [Online] 
Available at: http://breakpoint.untergrund.net/ (updated 25 April 2009) [Accessed June 16 
2009]. 





Chapman, I., 1999. Big Book of Amiga Hardware. [Online] Available at: 
http://www.bboah.com/ (updated 22 March 2009) [Accessed 25 June 2009]. 





Desert Planet, 2001. Desertplanet.com — Videogame influenced electro-pop and space 
punk! — Videopelihenkistä elektropoppia ja avaruuspunkkia. [Online] Available at: 
http://www.desertplanet.com/ (Updated in 2009) [Accessed 27 May 2009]. 





Digitalcraft, 2002. origami digital -Demos without restrictions. [Online] Available at: 
http://www.digitalcraft.org/index.php?artikel_id=411 [Accessed 17 May 2009]. 





Edge Staff, 2008. Inside the PS3 Demoscene. [Online] Available at: 


http://www.edge-online.com/magazine/inside-ps3-demoscene [Accessed 1 June 2009]. 





Frappr, 2005. Frappr Maps — My Maps — demoscene. [Online] Available at: 
http://www.frappr.com/demoscene [Accessed 15 June 2009]. 





Geurtsen, D., 2000. Disintegration of the Old Graphics Scene. [Online] Available at: 
http://www.kameli.net/nocopy/disint.htm [Accessed 9 June 2009]. 





GFXZone, 2000. GFXZone—the ultimate demo scene graphics site! [Online] Available 
at: http://gfxzone.planet-d.net/ (updated in 2007) [Accessed 9 June 2009]. 








Google, 2001. Google Groups. [Online] Available at: http://groups.google.com/ (updated 
in 2009) [Accessed 29 May 2009]. 


Green, D., 1995. Demo or Die!. Wired, 3(7). [Online] Available at: 
http://www.wired.com/wired/archive/3.07/democoders.html [Accessed 18 May 2009]. 





Gruetzmacher, T., 2004. PC Demoscene FAO. Plain text version 0.83. [Online] Available 
at: http://tomaes.32x.de/text/pcdemoscene_faq.txt (updated 16 June 2004) [Accessed 5 
June 2009]. 





Göhler, S., 2003. Phonomenal! ...a retrospective view on soundcard history. [Online] 
Available at: http://www.crossfire-designs.de/index.php?lang=en&what=articles&name= 
showarticle.ntm&article=soundcards (updated 30 September 2003) [Accessed 25 June 
2009]. 








111 


Hartmann, A., 2008. ”Type-Demos”: Die Rolle der Schrift in der Demoscene (The Role 
of Writing in the Demoscene). [Online] Available at: 





http://www.4players.de/4players.php/download start/Downloads/Download/50943.html 
[Accessed 15 June 2009]. 


Hinrichs, T., 2001. The Workings of FR-08's Sound System. [Online] Available at: 
http://www.kebby.org/articles/fr08sndl.html (updated 12 June 2007) [Accessed 3 June 
2009]. 





HVSC Administration, 2000. The High Voltage SID Collection (HVSC) — Commodore 64 
music for the masses. [Online] Available at: http://www.hvsc.c64.org/ (updated 1 May 
2009) [Accessed 27 May 2009]. 





IN4K, 2005. Main Page—IN4K. [Online] Available at: http://in4k.untergrund.net/ 
(updated 9 October 2007) [Accessed 16 May 2009]. 





Jacobsson, P., 2006. Differentiate or Die. A study of the evolution of magazines in scene 
town. Recollection 1. [Online] Available at: 
http://www.atlantis-prophecy.org/recollection/?load=online&issue=0&sub=article&id=10 
[Accessed 21 May 2009]. 





Kalms, M., 1997. C2P Tutorial. [Online] Available at: 
http://www.lysator.liu.se/-mikaelk/doc/c2ptut/ [Accessed 25 June 2009]. 





Kauppinen, J., 1991. Finnish Scene-history, 1986 to 1991. [Online] Available at: 
http://www.byterapers.scene.org/scene-sfscene86-91.htm [Accessed 16 June 2009]. 





Knight, G., 1997. Amiga history guide. [Online] Available at: 
http://www.amigahistory.co.uk/ahistory.html (updated 23 February 2003) [Accessed 25 
June 2009]. 





Kubarth, D., 2006. SID in-depth information site. [Online] Available at: 
http://sid.kubarth.com/ (updated in 2008) [Accessed 23 June 2009]. 





Lamarre, T., 2004. An Introduction to Otaku Movement. EnterText, 4(1), pp.151—187. 
[Online] Available at: http://arts.brunel.ac.uk/gate/entertext/4 1/lamarre.pdf [Accessed 15 
May 2009]. 





Leonard, J., 1994. PC Demos Explained. [Online] Available at: 
http://www.oldskool.org/demos/explained/explained/demos.html (updated 21 September 
1998) [Accessed 5 June 2009]. 





Lineback, N., 2001. Graphical User Interface Gallery. [Online] Available at: 
http://toastytech.com/guis/ (updated 16 June 2009) [Accessed 23 June 2009]. 





Lunder, G., 1996. ExoticA! Scenery Project. [Online] Available at: 
http://www.exotica.org.uk/wiki/Scenery (updated 29 April 2009) [Accessed 28 May 
2009]. 





112 


McNicol, T., 2006. Meet the Geek Elite. Wired, 14(7). [Online] Available at: 
http://www.wired.com/wired/archive/14.07/posts.html?pg=5 [Accessed 16 May 2009]. 





The Mentor, 1986. The Conscience of a Hacker (”Hacker's Manifesto”). Phrack, 
1(7). [Online] Available at: http://www.phrack.com/issues.html?issue=7 &id=3#article 
[Accessed 5 June 2009]. 





Mitchell, A., 2005. A Leet Primer. [Online] Available at: 
http://www.technewsworld.com/story/47607.html [Accessed 17 June 2009]. 





Pedersen, H., 2004. The Protracker Guide. [Online] Available at: 
http://www.silent-dreams.de/handbook.html (updated 22 Jan 2008) [Accessed 28 May 
2009]. 





Polsson, K., 1995. Chronology of Personal Computers. [Online] Available at: 
http://www.islandnet.com/~kpolsson/comphist/ (updated 24 March 2009) [Accessed 29 
May 2009]. 





Press Play on Tape, 2001. PRESS PLAY ON TAPE—The Commodore 64 Revival Band. 
[Online] Available at: http://www.pressplayontape.com/ [Accessed 27 May 2009]. 





Párssinen, M., Reunanen, M., Juntunen, I. & Isomáki, A., 1998. No Copy? [Online] 
Available at: http://www.kameli.net/nocopy/ (updated 7 January 2006) [Accessed 30 May 
2009]. 





Raymond, E.S., 2003. The Jargon File, version 4.4.7. [Online] Available at: 
http://catb.org/jargon/ [Accessed 15 May 2009]. 





Raymond, E.S., 2006. The Cathedral and the Bazaar. [Online] Available at: 
http://catb.org/esr/writings/cathedral-bazaar/ [Accessed 15 May 2009]. 





Reunanen, M. & Silvast, A., 2004. Demoscene Research. [Online] Available at: 
http://www.kameli.net/demoresearch2/ (updated 3 April 2010) [Accessed 3 April 2010]. 





Scene.org, 2009. scene.org—news. [Online] Available at: http://www.scene.org/ [Ac- 
cessed 21 May 2009]. 





Scott, J., 2003. APPLE IT CRACK SCREENS. [Online] Available at: 
http://artscene.textfiles.com/intros/A PPLEII/ (updated in 2008) [Accessed 17 June 2009]. 





Simmonds, A., 2001. Decoding Art. A Critical Analysis of the Demo Scene. [Online] 


Available at: http://www.hugi.scene.org/online/remix/hugi 23 - demoscene science ashton 





simmonds decoding art a critical analysis of the demo scene.htm [Accessed 18 June 
2009]. 





Softdisk, 1995. About us | Softdisk. [Online] Available at: http://www.softdisk.com/about/ 
(updated in 2004) [Accessed 26 May 2009]. 





113 


Szarafinski, S., 1955. Tilburg 1995. [Online] Available at: 
http://fms.komkon.org/MSX/Docs/Tilburg1995.txt [Accessed 17 June 2009]. 





tAAt, 2002. Text Mode Demo Competition. [Online] Available at: http://taat.fi/tmdc/ (up- 
dated November 2008) [Accessed 9 June 2009]. 





Ultra Force, 2007. Ultra Force Development — Demogroup. [Online] Available at: 
http://www.ultraforce.com/demogroup.php [Accessed 17 June 2009]. 





UNESCO, 2003. Charter on the Preservation of the Digital Heritage. [Online] Available 
at: http://portal.unesco.org/ci/en/ev.php-URL_ID=13367 [Accessed 15 May 2009]. 





Vigh, D., 2003. Pixelstorm. [Online] Available at: http://www.gr4ss.com/pixelstorm.html 
[Accessed 18 May 2009]. 





Volko, C-D., 2009. Diskmags: Some of my articles about Diskmags. [Online] 
http://www.students. meduniwien.ac.at/-n0102122/index_articles1.htm (alternative URL: 
http://www.hugi.scene.org/adok/) [Accessed 20 May 2009]. 








White, S., 2001. SIDPLAY2 Home Page. [Online] Available at: 
http://sidplay2.sourceforge.net/ (updated 12 June 2004) [Accessed 27 May 2009]. 





Research Material 


CD-ROMs and DVDs 


Atariscene 2003, 2005. [DVD] Dead Hackers Society. 

Atariscene 2004, 2005. [DVD] Dead Hackers Society. 

Atariscene 1985-1988, 2006. [DVD] Dead Hackers Society. 

Digital Memories: The Best of Commodore 64, 2006. [DVD] CSW Verlag. 
Demo or Die! Vol. 1., 2005a. [DVD] Abyss. 

Demo or Die! Vol. 2., 2005b. [DVD] Abyss. 

Demo or Die! Vol. 3., 2005c. [DVD] Abyss. 

Demo or Die! Vol. 4., 2006a. [DVD] Abyss. 

Demo or Die! Vol. 5., 2006b. [DVD] Abyss. 

MindCandy Volume 1: PC Demos, 2002. [DVD] Fusecon. 


MindCandy Volume 2: Amiga Demos, 2006. [DVD] Fusecon. 


114 


Shatz, P., 1993. Walkthroughs and Flybys CD. [CD-ROM] Waite Group Press. 


Shatz, P., 1994. Modeling the Dream CD. Walkthroughs and Flybys II. [CD-ROM] Waite 


Group Press. 


Disk Magazines 


Hugi, 1998-2008. Issues #11, #15, #16, #19, #23, #27, #31 and #35. [Online] Available 
at: http://www.hugi.scene.org/fate.php?page=hugi [Accessed 21 May 2009]. 





Illegal, 1987—1988. Issues #22, #23, #24 and #30. 
Imphobia, 1992-1996. Issues #1—#12. 

R.A.W., 1991-1995. Issues H1—H9. 

Sex’n’Crime, 1989-1990. Issues #1—#21. 


Zine, 1989-1991. Issues #01, #02, #07 and #11. 


115 


Appendix A 


Demo Content Analysis Form 


Name: 

Author(s) and roles: 

Type: 

Platform(s): 

3D accelerated: yes/no 
Year of publication: 
Country/countries of origin: 
Release party and ranking: 


Parts and effects: 


Music type: 


References (popular culture, other demos etc.): 


Other notes: 


116 


Appendix B 


List of Demos Analyzed 


Commodore 64 


Omega Demo 3, Teeside Cracking Service, demo, 1985 
Aliens, Scoop, demo, 1986 

Infiltrator I, The 1001 Crew, crack intro, 1986 
Judge Dredd, Triad, crack intro, 1986 

The Soldiers, Radwar Enterprises, demo, 1986 
Combat School, Fairlight, crack intro, 1987 
Electric Cafe, Ash & Dave, demo, 1987 

Graffiti, The Robots, demo, 1987 

Phantom+, Beastie Boys, crack intro, 1987 
Return of the Jedi+, Ikari, crack intro, 1988 
Brainbuster, Eltronic, demo, 1989 

Operation Thunderbolt, Optical, crack intro, 1990 
St. Dragon+, Exodus, crack intro, 1990 

Dutch Breeze, Blackmail, demo, 1991 

Red Storm, Triad, demo, 1992 

Seal of Focalor, Megastyle, demo, 1992 

Tower Power, Camelot, demo, 1993 

Torture 5, Padua, demo, 1994 

Mathematica, Reflex, demo, 1995 

Parts, Oxyron, demo, 1995 

Unsound Minds (Follow the Sign 3), Byterapers, demo, 1996 
Altered States 50%, Taboo, demo, 1997 

Triage 3, Smash Designs, demo, 1998 

Contact, The Ultimate Mayas, demo, 1999 


117 


Deus Ex Machina, Crest & Oxyron, demo, 2000 
Edge of Disgrace, Booze Design, demo, 2008 


Amiga 


TechTech, Sodan & Magician 42, demo, 1987 
Megademo, Scientific Cracking Team, demo, 1988 
Red Sector Megademo, Red Sector Inc., demo, 1989 
Budbrain Megademo, Budbrain Productions, demo, 1990 
Mental Hangover, Scoopex, demo, 1990 

Enigma, Phenomena, demo, 1991 

Odyssey, Alcatraz, demo, 1991 

Puuro, Complex, 40k intro, 1992 

State of the Art, Spaceballs, demo, 1992 

9 Fingers, Spaceballs, demo, 1993 

Dream Trippin’, Digital, demo, 1993 

Full Moon, Virtual Dreams/Fairlight, demo, 1993 
Falu Red Color, Razor 1911, 40k intro, 1994 
Nexus 7, Andromeda, demo, 1994 

Baygon, Melon Dezign, demo, 1995 

Closer, CNCD, demo, 1995 

Tint, The Black Lotus, demo, 1996 

Phase, Phase Truce, 64k intro, 1997 

Torque, Scoopex, 40k intro, 1998 

Eraser Head, Floppy, 64k intro, 1999 

Klone, DCS, demo, 1999 

Gift, Potion, 64k intro, 2000 

Lapsuus, Maturefurk, demo, 2001 

Humus 3, Push Entertainment, 4k intro, 2002 
Fistpig, Ephidrena & Spaceballs, 4k intro, 2003 
Silkcut, The Black Lotus, 2004 


Atari ST 


README.PRG, The Exceptions, demo, 1987 

The B.I.G. Demo, The Exceptions, demo/musicdisk, 1988 
Cuddly Demos, The Carebears, demo, 1989 

The Union Demo, The Union, demo, 1989 

Galtan 6 Demo, Galtan 6, demo, 1990 


118 


Dark Side of the Spoon, Unlimited Matricks, demo, 1991 
Grotesque, Omega, demo, 1992 

Flip-O-Demo, Oxygene & Diamond Design, demo, 1993 
Necrosys, Hemoroids, demo, 1994 

Magique, Wildfire, demo, 1995 

Reanimation, Syntax, demo, 1995 

True Lies, Antic, demo, 1996 

Lasse Reinbóng, The Naughty Bytes, 96k intro, 1997 
Virtual Escape, Equinox, demo, 1999 

Fantasia, Dune & Sector One, demo, 2003 

Suretrip II — Dopecode, Checkpoint, demo, 2009 


PC/MS-DOS 


Summer Holiday, Sorcerers, demo, 1989 
Megademo, The Space Pigs, demo, 1990 
Cronologia, Cascada, demo, 1991 

Dragnet, Dutch Computer Enterprise, demo, 1991 
Vectordemo, UltraForce Development, demo, 1991 
Amnesia, Renaissance, demo, 1992 

Crystal Dream 2, Triton, demo, 1993 

Plan-B, Sonic PC, 100k intro, 1993 

Second Reality, Future Crew, demo, 1993 

Show, Majic 12, demo, 1994 

Animate, Schwartz, 4k intro, 1995 

Dope, Complex, demo, 1995 

Stars: Wonders of the World, NoooN, demo, 1995 
Jade, Shock!, 64k intro, 1996 

Megablast, Orange, demo, 1996 

303, Acme, demo, 1997 

Brighten the Corners, Valhalla, 64k intro, 1997 
Poor guy!, Sunset Design & Procreation, 4k intro, 1997 
te-2rb, TPOLM, demo, 1998 

State of Mind, Bomb!, demo, 1998 

The Fulcrum, Matrix, demo, 1998 

Moral Hard Candy, Blasphemy, demo, 1999 
Viagra, Mewlers, 64k intro, 1999 

Heaven 7, Exceed, 64k intro, 2000 

Tube, 3SC, 256B intro, 2001 

Another Soul Lost, Traction, 4k intro, 2003 


119 


PC/Windows 


Virhe, Maturefurk, demo, 1999 

Mikrostrange, Haujobb, demo, 2000 

604, AND & Sly & SynSUN, demo, 2001 

Le Petit Prince, Kolor, demo, 2001 

32 Degrees in the Shade, Yodel, demo, 2002 

IV — Racer, The Lost Souls, demo, 2002 

fr-025: The Popular Demo, Farbrausch, demo, 2003 
Subversive, Bypass & Black Maiden, 64k intro, 2003 
Arise, Stravaganza, demo, 2004 

State of the Art 2004 Invitation, Equinox, 64k intro, 2004 
195/95, Plastic, demo, 2005 

Aether, mfx, demo, 2005 

Fascination, Brainstorm & Traction, demo, 2006 

Meet the Family, Fairlight, 64k intro, 2006 

Lifeforce, ASD, demo, 2007 

Candystall, Pittsburgh Stallers & Loonies, 4k intro, 2007 
Germ, Atomic Destruction, 64k intro, 2007 

Falling Down, UkScene Allstars, demo, 2008 

Nazca, Cocoon, demo, 2008 

Elevated, RGBA & TBC, 4k intro, 2009 


Other 


Your Song is Quiet, Inward & CyberPunks Unity, demo, 2007 (Spectrum) 
Syntax Infinity, Traktor & Tulou, demo, 2009 (MSX2) 


120 


Appendix C 


Themes of Disk Magazine Articles 









































Keyword # | Keyword # 
news/rumors 57 | interview 14 

letter to the editor/reaction | 29 movie 8 

editorial 22 | party/fair 7 

ad 21 | scene meta | 5 

chart 21 review 1 

game 20 politics 1 
Total 206 








Table C.1: Keywords of Sex’n’ Crime articles 















































Keyword # Keyword # 
interview 30 drug 5 
editorial 25 politics 5 
scene meta 18 group 4 
party/fair 17 diskmag 4 
swapping 11 | letter to the editor/reaction | 4 
review 10 philosophy/religion 4 
humor 8 game 3 
non-scene music 8 programming 3 
ad 7 news 3 
law enforcement | 6 chart 3 
hardware/software | 6 other 23 
Total 207 














Table C.2: Keywords of Zine articles 


121 





























































































































Keyword # Keyword # 
person 68 poll 11 
scene meta 68 chart 9 
editorial 56 swapping 7 
humor 48 history 6 
party/fair 41 politics 6 
letter to the editor/reaction | 40 programming 5 
group 39 | hardware/software | 4 
review 39 Internet 4 
interview 36 literature 4 
diskmag 18 movie 4 

ad 14 real life 4 
news/rumors 13 religion 4 
non-scene music 12 other 32 

Total 560 
Table C.3: Keywords of R.A.W. articles 

Keyword # Keyword # 
party/fair 182 politics 24 
scene meta 132 | disk magazine 20 
interview 85 humor 20 
chart 84 real life 19 
editorial 75 group 18 
programming 72 Internet 16 
ad 56 news 14 
review 55 greetings 12 
poem 52 game 11 
hardware/software 36 | non-scene music | 11 
story 35 design 10 
scene music 32 UFO 10 

graphics 30 swapping 6 

letter to the editor/reaction | 28 | law enforcement 5 
BBS 25 other 66 

Total 1241 














Table C.4: Keywords of Imphobia articles 


122 





















































Keyword # Keyword # 
programming | 172 hardware/software 13 
party 102 politics 11 
disk magazine | 72 group 10 
scene meta 63 news 9 
editorial 61 Internet 7 
interview 60 | letter to the editor/reaction | 7 
review 59 humor 6 
story 42 ad 5 
chart 21 graphics 5 
scene music 15 game 4 
poem 14 other 39 
Total 797 














Table C.5: Keywords of Hugi articles 


123