Capcom To Use Emulation In Upcoming Products

JayBonci writes: "The emulation and classic gaming world has been a large grey area for a while. The subject of reverse engineering other products for the sake of emulating has come up numerous times (see also the Bleem and Connectix lawsuits), but what about emulating your own stuff? It seems that Capcom is going to use emulation as a standard way to make more games cross platform. The blurb is short, but an interesting though. How viable a solution is this? If the performance hit is fairly minimal, how will this affect the future of cross platform game development? It's an interesting solution, IMO."

Comments

Big Machine, Little Game (Zork)

[fm6]

Actually, the Z machine (which was actually invented for Zork, though I imagine the Sierra pseudo-machine was probably similar) wasn't invented for portability -- that was just a bonus. Infocom was faced with porting Zork from a PDP-10 (1 million 36-bit words, and strained even so) to an Apple II (64 Kbytes). An obvious way to reduce the footprint is to target a pseudo machine that's optimized for your application. You lose something in performance, but that hardly matters for a non-realtime, non-graphics game.

Speaking of Zork, I would dearly love to finish playing the original Zork. (I made it through a few puzzles when it was on ARPANET.) You can still buy PDP-10 compatibles, but the power drain is horrendous. Does CompuServ still use them? Anybody done an emulator?

(I know about the 16-bit FORTRAN vesion. Played it. Parser not as smart, not all puzzles.)

For that matter, I'd love to see a new adventure game -- graphic or non-graphic -- done the Zork way. You start with a few rooms and puzzles, and let people from all over the word bang on it. Then you grow the thing organically. Commercial adventure games were rarely as good.

__________

Re:Seems unlikely to be true emulation

[tuffy]

They have source code, why would they use performance-hogging emulation?

Because a lot of their arcade games are written with a lot of 68000 assembly in order to milk the most out of a very old processor. In order to get their 2D fighers onto other systems, all that coding needs to be re-done - which is a pain. It's much easier just to emulate that old hardware (which they're still writing games for) and have the whole arcade game running on the home systems at a fraction of the effort.

Is this really such an innovative concept ?

[FauxPasIII]

Just to play devil's advocate, this doesn't sound like anything more than Capcom throwing out a buzz word to generate publicity... so they develop a midlayer that exports a generic API on several different platforms, and then code their games to that API. That's *NOT* emulation. That's what SDL does, that's what OpenGL and OpenSound do. It's basic tiered design, and it's a good idea. Hell, the point would be moot if all the game console manufacturers would get together and implement a unified API anyway, which they ought to do. But using the word 'emulation' just to stir up interest is pretty weak, IMHO.