Today is the eighth lesson of the Introduction to the demoscene, and this time I hope to answer the question posed at the end of the last lesson: why do the core players in the chip music scene oppose the idea of chip music as some sort of video game derivatives?
In the end of last lesson, we found that both Bryan Chan, the founder of Vancouver Chip Music Society, and Sulumi, one of the most influential chip musicians in China, the leaders of the chip music community, were all expressing one view: "chip music is not video game music." One of the most important reasons for this is that after the game industry phased out the use of programmable sound generators (PSG) based chip music in the 1990s, the core group of chip music creators has shifted significantly: from professional game developers to an underground community of computer enthusiasts centered on the demoscene. With this shift, the mainstream values of the chip music community have also shifted towards a "hacker ethic" and a clear divergence from the commercialized video game business.
Chip musician and researcher Anders Carlsson (GOTO80) writes in "Chip music: low-tech data music sharing"
Since today's chip music scene is essentially noncommercial, the commercial industry of video games is out of the scope of this text.
The primary question we hope to answer in this lesson is both how did the chip music scene, which is closely linked to video games in terms of production methods and performance equipment, come to be? And why does this scene emphasize its independence from video game music?
In the overseas chip music scene, there is a consensus among almost all participants, whether researchers, musicians, or community leaders, that they intentionally delineate their activities in the chip music scene from the video game realm, which clearly runs counter to previous descriptions of chip music in China that link chip music to nostalgic games.
why is this, though? A footnote in GOTO80's article caught my attention when the author mentioned that his definition of "scenes" came from Alan O'Connor's 2002 paper Local Scenes and Dangerous Crossroads: Punk and theories of Cultural Hybridity (2002), which states.
When punks use the term 'scene' they mean the active creation of infrastructure to support punk bands and other forms of creative activity.
However, it is clear that the game industry has not moved to actively build the infrastructure for chip music in the way described above, but instead a series of industry changes brought about by the widespread use of CD tracks and pre-recorded audio in games has actually suppressed the continued development of chip music in the game industry.
In the article "Chiptune, Ownership and the Digital Underground" included in last year's issue of The Cambridge Companion to Video Game Music (2021), quoted Mark Knight (TDK), author of the music for the Amiga version of Wing Commander (1990), argues that the advent of CD tracks has taken away the character of game music:
In my opinion ... those new formats killed computer game music. It started with the PlayStation, when instead of being stuck by limitations which forced [composers] to create music in a certain style, in a certain way and using certain instrumentations, suddenly you could go into a recording studio, you could record an orchestra or a rock band or whatever you wanted, really, and then plonk it on a CD as Red Book audio. Suddenly game music didn't sound distinctive any more. It sounded like everything else.
In this paper, Colin Anderson, then head of audio at DMA Designs and responsible for the music for games such as Lemmings (1991) and Grand Theft Auto (1997), is quoted in an interview that likewise emphasizes the impact of CD audio on game music. The interview was conducted in Abertay, UK. This interview is part of the Video Game Design and Development online course at Abertay University, UK: Interview with Colin Anderson
Probably themost significant change was the fidelity of the audio that you could create. [Sampling and CD audio] gave you access to the same resources that the film and television industries would use ... and that meant for the first time you could use real recordings of real instruments, of real sound effects ... instead of having to synthesise them.
While acknowledging the improved audio fidelity that CD tracks bring, Colin Anderson also mentions that CD tracks are harder to integrate into the interactivity of the game than music synthesized by sound chips:.
On the downside, we lost interactivity for a while. The synth chips were particularly good because they were being coded at quite a low level. They were really good at responding to gameplay as it moved, and that went away when we started using CD and things like that ...
In addition, Colin Anderson points to Grand Theft Auto 2 as an example of an important change that CD tracks brought to game music: in some cases, game companies began working with traditional record labels to license popular music from the labels as built-in tracks. This has somewhat weakened the originality of game music.
For GTA 1, [all of the soundtrack material] was 100 per cent completely original material that we recorded in-house ... In GTA 2 that started to change. As soon as the game became successful, suddenly people turned up wanting their favourite tracks to be licensed, and [that commercial pressure] increased [with each subsequent release].
We will see that once music production in the game industry entered the era of pre-recorded tracks and CD audio as the primary medium, there was a significant change for developers of game music, both in the production process and in the expectations of customers for the music. It can be said that the era of chip music for games came to an end, after which the demoscene became virtually the only refuge for chip music, and chip music gradually positioned itself in opposition to the game industry. The article "Chiptune, Ownership and the Digital Underground" reads:
The age of PSG video game music, then, was brought to an end as much by the commercial realities of video game production as it was by the increasing technical capacity of home consoles. However, while chiptune might have disappeared from games, reports of its demise were greatly exaggerated. Chiptune was about to develop an edge, one that would set it in direct opposition to the corporate world of professional game development.
The merging of chip music with the demoscene stems in large part from the use of chip music by the warez community. The demos at the heart of the demoscene originated as cracktros created by the warez community to show off their technical prowess.
The size of cracktros had to be small enough to fit on the boot sector of a floppy disk and to run and animate before loading the game, which is the root of the size-coding tradition in the demoscene. And this quest for minimizing program capacity makes chip music the most suitable music format for cracking headers naturally, as written in the article "Chiptune, Ownership and the Digital Underground".
The code to execute cracktros had to be compact and efficient to fit in the boot sectors of floppy disks, so that the cracked game could be uploaded and downloaded easily from bulletin board services via dialup and rewritten to new floppies. The simple waveforms and sequences of PSG music, which could be stored as space-efficient single-cycle samples and tracker sequences for playback on sample-based systems, lent itself perfectly to this end. Chiptune became the sound of the digital underground.
The paper also mentions that in the warez community, many times the impact of well-produced cracktros surpasses even the cracked game itself.
Over time, the competition to demonstrate both coding virtuosity and graphical and musical creativity became more important than the cracks themselves. End users would actively seek out cracked software for the cracktros, rendering the game an almost insignificant by-product of the cracking process.
These "cracktros", which are not involved in cracking activities, have gradually become an independent form of folk computer art, namely demos, and the community that produces them is the subject of this series of courses: the demoscene.
The demoscene gradually diverged from the warez community in the early 1990s, focusing its activities on creating demos with complex audio and visual effects, rather than engaging in software cracking activities, at about the same time game companies abandoned the use of sound chips. As the demoscene's use of sound chips became more widespread, especially with the development of Tracker (sample sequencer) type music production software, the demoscene surpassed the game industry as the core group for chip music production, both in terms of the abundance of creative tools and the quantity of their work. With this change, the creative ethics and values of the chip music community have also moved closer to the non-commercial demoscene and are clearly differentiated from the game industry. Chiptune, Ownership and the Digital Underground states:
The production and sharing of cracktros became an end in itself, and evolved into the demoscene, a distributed online community of digital arts practice dedicated to the production of complex real-time audiovisual displays. That combination of anti-commercialism, a distinctive sense of community and a culture of sharing marks a definite point of departure of chiptune, as a constituent part of the crackscene and demoscene, from the increasingly professionalized and corporate approach to video game music production.
It also points to a difference in mindsets. On one side sits the corporate perspective, which recognizes that there is value - and cost - in the production of professional content, be that music or software, and that it is therefore justifiable for a company to protect its investment by using a combination of digital rights management (DRM) and litigation to ensure that only legitimate copies are in circulation.
Set against this, the Hacker Ethic, the core philosophy of hacking culture, which originated from Massachusetts Institute of Technology in the 1950s and 1960s,18 sets out the intellectual counterpoint to this enterprise- driven process of making intellectual property out of everything, namely the ‘belief that information sharing is a powerful good and that it is an ethical duty . . . to share . . . expertise by writing free software and facilitating access to information . . . whenever possible’.
Hacker ethics are the key to explaining many of the digital cultural phenomena that are difficult to understand in China, but are prevalent overseas, especially in Europe. Take for example the warez community, which we detailed in lesson 3. Even with software piracy being illegal in most countries since the mid-1990s, people still took the risk of doing such an illegal thing non-profit and voluntarily.
Today, in many European countries, the Pirate Party, whose main political advocacy is against the existing intellectual property system, is very active, and in Germany, Sweden, Austria, and the Czech Republic, the Pirate Party has gained some seats in parliament. Echoing this, European courts are also very cautious about protecting software functions with patent rights.
I think we must be equally mindful, in the context of today's boom in "knowledge payment" in China, where "copyright defenders" seem to occupy the moral high ground, that the notion of intellectual property rights promoted by the commercial sector is not naturally in the public interest, and that moral norms and laws should not favor the commercial sector, nor should ethics and legal regulations suppress competition and innovation in favor of the strongest in the business world. Europe's thriving computer underground culture and its challenge to the commercial world gives us an excellent frame of reference, as noted in Chip Music, Ownership, and the Digital Underground:
History has demonstrated beyond doubt that Bill Gates and Paul Allen were right about the commercial potential of software as a commodity, but the continued growth of the underground warez scene, and the legitimate adoption of freeware, Creative Commons and open source as models for publishing and distribution suggest that attitudes around the sharing of digital content remain as strong as ever. (The Cambridge Companion to Video Game Music, p. 42)
That anti-commercial ethos, however, and the pervasive culture of sharing, not just within the chipscene, but more broadly within the different subcultural groups that lurk beneath the increasingly corporate digital mainstream, certainly poses a challenge: as Bill Gates noted, who will create professional content – be that music or games or productivity software – if nobody is prepared to pay for it? History, however, suggests that content will still be produced both commercially and – to a very high standard – within deprofessionalized communities like the chip music scene. (The Cambridge Companion to Video Game Music, p. 49)
We may have seen countless product plans in China to sell the idea of "creator economy", "knowledge payment" and other ideas to integrate creators into the "platform economy", almost all of these plans presuppose some kind of "no payment, no creation" logic, but we can indeed see that both open source communities, demoscene, chip music communities, and even domestic translation groups are non-profit while creating a lot of content.
It is worth noting that not all demoscene creations are driven by a hacker ethic, and Anders Carlsson, author of Chip music: low-tech data music sharing, sees a simpler and more direct reason for the involvement of teenage crackers.
Parts of the cracker community have ideals similar to those of hackers and phreakers, but it seems likely that a lot of early crackers were kids cracking games for fun, for intellectual challenge and for ‘street credibility’.
I think a comparable community in China is the community of electronic dictionary enthusiasts, teenagers who hid under their desks to write games for Wenquixing back in the day, whose motivations were extremely similar to those of teenage software crackers in the West.
Before Tracker came along, the environment for developers and regular users to make music for their computers was very different. As we have already mentioned in the previous lesson, for game developers in the 80s, modifying the program code directly or using MML (Music Macro Language) were the common ways to write music for games, but both were quite difficult and time-consuming, and such a high threshold was almost exclusively passed by developers employed by game companies or hobbyists proficient in programming to write chip music in this way. However, non-developers are not completely without options. As early as the early 1980s, some commercial software on the market has tried to provide ordinary users with the ability to write or play music using the sound chip in their home computers. The 1982 Commodore Music Maker, for example, offered a music keyboard layer for the Commodore 64 that converted the Commodore 64's computer keyboard into a piano-like music keyboard with accompanying music performance and editing software.
If you look at it from the perspective of a music toy, the Commodore Music Maker is largely passable, and it shows some of the features of the Commodore SID chip. Another early music production software that was once widely used was Will Harvey's Music Construction Set, which was one of the early staples of Electronic Arts (EA), and more feature-rich than Commodore Music Maker.
We will find that commercial music software for non-developers almost always tries to mimic the common 5-line staff notation on the screen. While it is intuitive for users with some musical knowledge, for the sound chip as a miniaturized synthesizer, this design hides the functional details of the chip music and weakens the sound design part of electronic music.
In 1986, Chris Hülsbeck developed SoundMonitor for the creation of Commodore 64 music, starting with the idea of simplifying the music development process on the Commodore 64, which was still based on the hexadecimal code-based Commodore 64 music editing process, but made significant improvements to the previous machine code editor-based workflow. Many of the software's practices were inherited by the Trackers. The origins and influence of SoundMonitor are reviewed in the 2018 book Bits and Pieces: A History of Chiptunes, published by composer Kenneth B. McAlpine's, a researcher at the University of Melbourne.
When Hülsbeck wrote Shades, he was still writing music by entering note data as hexadecimal values directly into a machine code monitor, a piece of software that allows users to view and modify the data stored in computer memory. It was a perfectly work-able approach, but only because Hülsbeck knew exactly what values to write to which memory locations. For anyone else, it was completely opaque. Hülsbeck knew that the process could be improved, and started working on a dedicated music editor. He worked throughout the summer of 1986 and submitted the result, SoundMonitor, to 64er, who published the code as their listing of the month.
SoundMonitor is, to all intents and purposes, a prototracker, and many of the elements that would come to typify the tracker experience can be seen in its interface. Figure 5.3 illustrates the interface: three vertically scrolling tracks, one per channel, and dedicated control commands to trigger portamento effects and arpeggiated chords. In terms of user-friendliness, it was a massive improvement over coding music directly or modi-fying data within a machine code monitor, and it provided a platform for very detailed SID programming for those musicians who understood the basics of the technology. (Bits and Pieces: A History of Chiptunes, p.131)
Because SoundMonitor source was printed in the 64er magazine as hexadecimal machine code without any anti reverse engineering techniques, SoundMonitor was not only used to develop music programs for the Commodore 64, but was also adapted itself and used by many developers as part of their demos.
In the Commodore 64 enthusiast scene, reverse engineering and modification of exist works and tools are widely accepted. This is very different from the general music scene, which is very concerned with originality, and even goes to court over the similarity of a few bars. In the article "Bits and Pieces: A History of Chiptunes" the collaboration between German musician Thomas Detert and the X-Ample demo team is used as an example of how they improved and used SoundMonitor and incorporated it into their own compositions.
SoundMonitor provided an avenue into C64 music for many musicians and a platform for development for other coders, particularly in Hülsbeck's native Germany. Thomas Detert, a noted C64 video game composer in his own right, started using SoundMonitor soon after, scoring demos for the Omega 8 group. But the more he tried to push it, the more he found that the CPU load needed by the driver routine was just too great; the music he was writing in SoundMonitor was pulling resources that the group needed for the graphics routines.
To improve things, Detert turned to his friends at X-Ample Architectures, another German demo group. Together, they combined the core of SoundMonitor with a new driver written by Markus Schneider and a front-end interface by Joachim Fraeder to create the X-Ample Music Player...
This approach, of developing bespoke routines by hacking and adapting the work of others, was fairly commonplace and perhaps goes some way to explain the coherence of the C64 sound..Schneider notes that '90% of the music players were based on Rob Hubbard's techniques.... Most people just optimized it and invented more tricks'. (Bits and Pieces: A History of Chiptunes p. 132)
The Amiga, introduced in 1985, was the first multimedia computer in the modern sense, with audio and video capabilities far ahead of its contemporaries. One of the most powerful features was the Amiga computer's audio system, which could playback four channels of PCM audio samples simultaneously.
The Amiga 500, introduced in 1987, was even cheaper, selling for just $699 or £499 at launch, which made it quite popular in Europe. The Amiga computer came with an audio output interface by default, but did not have audio sampling and recording capabilities. To make better use of the Amiga computer's audio capabilities, a series of audio sampling peripherals designed specifically for the Amiga computer were developed. They were usually small boxes attached to the Amiga computer's printer interface that converted the external input analog audio signal into a digital audio signal that could be processed in software. In the late 1980s, a set of Amiga computers and sampling expansion cards as the core of home recording equipment cost no more than £1,000.
The combination of Amiga computers and sampling capabilities greatly expanded the options available to computer music enthusiasts in the late 1980s, allowing them to sample and collage music from common home recording media, such as cassette tapes, vinyl records and radio. In the past, this often required the use of expensive studio equipment, such as Fairlight CMI, which costs tens of thousands of dollars.
The first music production utilities to hit the Amiga platform were still notation-based, including Will Harvey's Music Construction Set, which was ported to the Amiga computer as Deluxe Music Construction Set, and Aegis Sonix. More innovative music production software emerged during the same period, such as EA Instant Music, which presented musical contours in horizontal scrolls and automatically matched chords algorithmically.
In 1987, Karsten Obarski developed a music production application called Ultimate Soundtracker, which combined the sampling capability of the Amiga computer with the vertical scrolling interface of SoundMonitor, and improved the operation of SoundMonitor in a way that was more suitable for music production.
And then, in 1987, Karsten Obarski, a programmer and musician with the German software house EAS, was asked by his friend, Guido Bartels, to produce a C64-style soundtrack for an Arkanoid-style game he was developing. Obarski had bought him self an Amiga 1000 a few weeks earlier and had already written some simple sample-triggering music routines. He had also worked with Chris Hülsbeck's SoundMonitor on the C64 and liked its vertical, column-based layout. As figure 5.6 shows, his Ultimate Soundtracker combined the two.
Obarski simplified the interface of SoundMonitor, losing some of the complexity of the C64 program by bringing all four of the Amiga's four channels together into a single editable 64-step grid called a pattern. Each pattern displays the four channels as vertically scrolling columns of cells, headed Melody, Accompany, Bass, and Percussions, with a horizontal time bar that stretches across all four tracks to show the current play position. These patterns give a simple overview of the order and structure of the music, something like a musical spreadsheet, and can be sequenced and arranged in different ways to produce a song. (Bits and Pieces: A History of Chiptunes p. 138)
To some extent, the Tracker's interface and mode of operation is even easier than notation-based music editing software, especially for users who can use Trackers like a drum machine without knowledge of traditional music theory such as staff notation.
This no-nonsense interface removed graphical metaphors altogether, instead focusing on a concise textual representation that gave a complete overview of the musical 'source code'. That move made it fairly straightforward to learn the screen layout and simplified the process of reading and editing music for those who weren't fluent in traditional score notation. The software's simplified layout and the keystroke interface turned the process of composition into something that more closely resembled musical touch-typing. (Bits and Pieces: A History of Chiptunes p. 139)
Also, you've got the best elements of a drum machine, where you can string patterns together, and you can see the position of the drums and things relative to each other in time, and then you can fit your melody around that. So it's kind of like all of the advantages of a drum machine with the advantages of something like a classical notation. It also relates very closely to the hardware, so you know there's really nothing separating you from your instrument and you feel you're really getting the best out of it as well. (Bits and Pieces: A History of Chiptunes p. 140)
Ultimate Soundtracker was the first music production software to use the sampled track sequence mode, and is the origin of the name "Tracker". It was compatible with a number of subsequent Tracker programs. In fact, StarTrekker, DOC Soundtracker, NoiseTracker, ProTracker, etc. were all reverse-engineered from Ultimate Soundtracker, fixing some of the bugs in the original version of Ultimate Soundtracker and improving the stability of the program. The entry on the Amiga game data site ExoticA records details has details on the various adaptations of Ultimate Soundtracker.
In the process, Karsten Obarski's copyright was ignored by the warez scene, and Ultimate Soundtracker, sold as commercial software, was not a widespread success. However, SoundTracker's production process was widely adopted by demoscene in the process, and the MOD format became the de facto standard for tracker file formats on Amiga computers. Many times "MOD music" is also used as a generic term for sample-based music files produced with Tracker software.
Compared to PSG music, which relies heavily on hardware features, sample-based Tracker music keeps file sizes small enough to play relatively consistently across devices. Before MP3 music became popular in the early 90s, Tracker/MOD music was also the most active music sharing community on FidoNet, Usenet and the Internet, and distributed "music albums" on disk images: disks containing only music were called "music disk", while a disk containing a mix of music, images and programs is called a "pack disk". The community that formed around the production and distribution of Tracker music was also known as the "MOD scene".
In the days before the ubiquity of the Internet, file sharing was nowhere near as straightforward as it is now. Music files were often shared on indi¬vidual bulletin board systems (BBSs), remote computer servers that hosted software and services. By the mid-1980s, Fidonet, a global network that connects together different BBSs, had begun to emerge, making it a little easier to post and swap MODs.
That online sharing took place mainly in North America and western Europe, and from it emerged an active community of practitioners—not just musicians but coders and graphics artists—who developed custom fonts and graphics for musicdisks and packdisks, packaged distributable collections of music that were like digital concept albums. (Bits and Pieces: A History of Chiptunes p. 144)
After the 1990s, digital audio playback and recording became a basic feature of PC sound cards, and Tracker/MOD music became widely available on IBM PC-compatible machines. In particular, warez scene, which were closely associated with demoscene, often had Tracker music data embedded in the Keygen, so that outside of North America and Western Europe, where demoscene had less of a presence, Tracker music was often also referred to as "Keygen Music".
In China, due to the unfamiliarity of the demo and MOD scenes, Tracker music is often incorrectly described as "MIDI music". This is very common in texts about "Keygen music". However, overseas, MIDI has a much weaker presence in discussions about chip music and demoscene.
One important reason for this is that MIDI is much more expensive to use than Tracker. Although the Atari ST, introduced almost simultaneously with the Amiga in 1985, had a built-in MIDI interface and was much less expensive, it had much less built-in audio capability. The original version of the Atari ST only had a built-in Yamaha YM2149 sound chip, the version of the AY-3-8910 manufactured by Yamaha under license, which was basically the same as the Amstrad CPC, MSX and other 8-bit computers.
Furthermore, to fully play the role of the Atari ST built-in MIDI interface, it must be used together with a synthesizer, multi-track recorder, and other equipment that supports MIDI interface, and this will obviously increase the cost of use. Likewise, support for MIDI sampling equipment is more expensive, in the late 1980s, small recording studios can afford to buy the Akai S900 sampler at a price of £1,699, or about £5,300 today, which is still quite expensive for the average household.
Another reason that limited the popularity of MIDI in the demoscene and chip music community was the difficulty of balancing creative flexibility and broad compatibility with limited budgets in the late 80s and early 90s as the standard for recording industry design.
For recording studios, MIDI protocols served as communication protocols between synthesizers, samplers, and computers, allowing for a wide variety of sound effects to be generated by connecting a large number of devices, with little regard for the compatibility of the resulting data in different environments. When using different brands of sound modules, the same MIDI data could produce different sounds.
Standards such as General MIDI(GM) and Standard MIDI File(SMF) in 1991 attempted to standardize MIDI by specifying 128 basic instruments. This however clearly limited creative flexibility, especially by failing to include the vocal and natural ambient sound samples common in Tracker music. At the same time, PC sound card manufacturers rarely implemented the General MIDI standard according to the specification, and most sound cards use the built-in FM synthesis chip to emulate the sound of the instruments in the standard, which is obviously less straightforward for participants in the demoscene seeking to demonstrate their technical prowess than making music directly for the OPL2/OPL3 FM synthesis chip used in PC sound cards.
It can be said that PSG-based game music and Tracker-based demoscene music are the two main sources of "chip music", and "chiptune" also originated from the demoscene to describe those Tracker works imitating PSG sounds:
As more and more of these sound chip-style musicdisks were circulated, MOD fans found themselves in need of a term to differentiate them from other forms of tracked music. The Amiga, of course, could sound like a PSG, but it could also sound like a rock band or an orchestra. By 1990, chip had started to appear as a stylistic designator in MOD filenames, and in 1991 the term chiptune, meaning a piece of chip music, appeared as a title on Nuke's mod.chiptune-12k on an Anarchy packdisk. (Bits and Pieces: A History of Chiptunes" pp. 145-146)
Discussions of chip music in today's mass media often omit the contributions and influences of the demo scene on chip music and instead cater to the nostalgic sentiment of game lovers, which completely ignores the anti-commercial spiritual core of chip music as an underground cultural scene and is not fair and objective enough.
However, both the excessive association of chip music with commercialized game music and the controversy surrounding "fake chips" have their objective reasons and historical origins to a certain extent.
In the next session, we hope to answer these questions by tracing the development of chip music from around the turn of the millennium to the present, and to explain the characteristics of the chip music tools and the "new school" music scene such as LSDJ and Nanoloop, which generally existed outside the demoscene in the mid to late 90s.