I’ve noted here at MiT7 (Media in Transition 7) that we’re now achieved some very reasoned discussion and understanding of the virtues of different approaches to preserving and accessing computer programs. Not that we’ve solved the underlying problem, of course, but I’ve been pleased to see how our overall approach has evolved.
Instead of simply dismissing emulation, migration, or the preservation of old hardware, we’ve had some good comments about the ways in which these different techniques have proven to work well and about what their limitations are. We saw this in the plenary discussion on archives and cultural memory late this morning – audio of that conversation will be coming online. Update: Here it is.
Clara Fernandez-Vara’s presentation “Emulation as a Tool to Study Videogame History” [Abstract] developed this discussion extremely well with regard to one important tool, the emulator. She presented the idea of a game in emulation as facsimile – not the original edition, but also not the Cliff Notes that we’d have to consult otherwise. She showed us a range of work with emulators that gives reserach, teaching, and casual access to older games, which would otherwise be neglected. Saving state, speeding games up, and even playing several of them at once with the same inputs are all facilitated by emulators.
Fernandez-Vara went on to note some limitations of emulation, such as that the physical controller, often significant to play, cannot be replicated in hardware; nor can particular hardware features such as those of the Dreamcast’s VMU or of a C64 floppy drive, which would whirr when something interesting was about to happen in a text adventure. Boxes and manuals are often very important and can’t always be effectively digitized; with online games and worlds, keeping the context is even harder. Finally, emulators have to be updated for new contemporary platforms every few years.
Much of the work of building emulators, Fernandez-Vara also noted, is done by fans who work as volunteers – institutional support can help them and can allow libraries to accumulate holdings. It would be nice if current platforms (the PS3!) were more backwards-compatible, too. “Abandonware” could be officially made available for use, to clear up legal questions.
The only thing I’d add to Fernandez-Vara’s excellent discussion is a slightly different framing perspective on the emulator. The emulated game may be usefully understood as a facsimile, but I see a different way to understand the emulator itself.
My suggestion is that an emulator can be conceptualized as an edition of a computer.
The first edition would be the original piece of hardware – for the Commodore 64, the August 1982 beige keyboard-with-CPU. Actually, in the case of the Commodore 64, even keeping to the United States there are at least three different “hardware” editions, since there are three ROM revisions, one used in very early machines, one in 1982 and 1983 computers, and one that was used in Commodore 64s and in the C64 mode of the Commodore 128. The three ROM revisions are not the only things that changed during the time the Commodore 64 was manufactured and sold, but they do change the behavior of the system. I suppose these better understood as being different “printings,” since the changes are limited to the ROM. That would be an interesting discussion to pursue. Either way, though, printing or edition, there are three different sets of hardware, three hardware Commodore 64s.
When the creators of VICE (the emulator I use) produce a program that operates like a Commodore 64, I understand this as being an edition of the Commodore 64. Yes, it’s a software edition. It isn’t an official or authorized edition – only being a product of Commodore would allow for that. (There are official, authorized emulators, but this is not one.) It’s not, of course, the original and canonical edition. But it’s nevertheless an attempt to produce a system that functions like a Commodore 64, one which took a great deal of effort and is effective in many ways. Thinking of this an edition of the system seems to be a useful way to frame emulation, as it allows me to compare editions and usefully understand differences and similarities.
Thanks for the post!
Understanding emulation as accessing the facsimile of a work was more a metaphor than a definition, to evoke the attempt at being faithful to the source and yet still working under many constraints. Your post really makes me think that there’s a series of really interesting parallels with studying literary history and its sources.
The idea of emulators as a further edition of the machine can be even more complex than that. In the case of MSX machines, it was not a sole machine, but a computer standard complied by many different manufacturers. All ran the same operating system, but then had different hardware (from differently sized RAM / ROM to very different keyboards, for example). I don’t think it’s possible to find a canonical edition of the MSX system, since the key was its multiplicity. What’s more, there are different MSX emulators, which focus on different aspects of the platform–fMSX was the first emulator for multiple platforms, which others have made improvements and expansions on (such as better sound emulation, or loading games like you would in a console emulator). Tracing the history of these editions could be as fascinating as the study of the different editions of Hamlet, for example.
Clara, are there slides or some such of the presentation lurking on the Internet?
Clara, I think your metaphor is very useful, since it shows how games in emulation preserve certain important aspects and are useful in many situations – without being exactly the same as play in original hardware.
I just wanted to push that a but further, since thinking of editions seemed helpful in one way to me. Namely, it helps me to see that hardware also comes in editions.
The MSX is a fascinating case of a platform that isn’t one machine made by one manufacturer, similar in some ways to the Windows MPC specification and the 3D0. It would be a great topic for a Platform Studies book!
Nick, thanks for posting this. I really like the idea of emulators as “editions” of a computer system. Mapping software emulators onto textual studies reveals some aspects of emulators that we might otherwise overlook. Just as there are Bowdlerized versions of texts, for example, we can begin thinking about Bowdlerized versions of computer platforms, which leave out certain characteristics deemed to be undesirable to the new audience.
Seen in this light, I might even go so far to say that the Mac/PC version of Ian Bogost’s A Slow Year, which runs in a modified—and clamped down—Stella emulator is a Bowdlerized version of the Atari VCS. (Though, of course, Ian edited Stella, itself an edition of the VCS, for reasons other than the typical morally-instructive Bowdlerization.)
Thinking of emulators as editions reminds me of another analogue: the release of movies. How many “cuts” of Blade Runner are there? The studio release, the extended release, the director’s cut, the extended director’s cut, and so on. Which one is the “authentic” or “definitive” Blade Runner? Ridley Scott would probably have a different answer than Harrison Ford, who might have a different answer than the estate of Philip K. Dick, who likely has a different answer than Warner Brothers. In this case, it is not at all clear that the originary edition is the “authentic” edition. I wonder if someday we will reach a similar point in platform studies—in which VICE (for example) is seen as a more definitive version of the C64 than the original release…