tepples wrote:
Yet the 6502 itself draws plenty of Atari 2600, Commodore 64, and NES homebrew development, even with all the weirdness of the TIA, VIC-II, and PPU.
In general, the scope of game design for those systems is not in the same league as the SNES or systems of that generation. There's also this community that has grown up around these systems (mainly the small home computer scene) with the mentality of pushing these systems as far as they will go. There's also more merit and recognition in that achievement than say the "16bit generation" - in which everything
pretty much as been explored and done on a commercial level, already.
I see the complex convoluted design and interface of sPPU, coupled with the mirroring attributes of the cpu, to be the primary reason why high level stuff isn't really suited for the SNES. And in that respect, especially now for some reason, that makes it a very active deterrent for people wanting to delve into snes dev. People are afraid to write their own libraries. They want this ready made, flexible, fast lib with high level support - with maybe a dash of ASM here and there. And that's not really applicable on the snes, unless you significantly reduce the scope of your game design - limiting yourself to an overall reduction capable power/resource (getting things done). Speed/power in relation to capability, comes with customized routines specifically for said games. The more flexible and generalized a routine is, the less efficient is becomes in relation to speed.
As far as the asm coders go, the sPPU and cpu shouldn't be a problem though. Yes, it might have an initial steeper learning curve - but once that learning curve is out of the way (as with all systems), you're able to move forward with your project design. The 65x cpu design isn't holding you back. The convoluted sPPU interface/design isn't hold you back. What's holding you back, is the lack of will to learn the system's in and outs. Stef's post is a testament to that.
The Genesis requires little investment to start producing an acceptable product, but in all reality if you want a highly polished product - you're still going to have to invest roughly the same amount of time as you would on the SNES. That initial investment of becoming familiar with the SNES hardware and structure, the time involved and set aside for that, becomes a relatively small portion of what's required to put out a decent product. And I think the biggest problem is that people don't see this. Most people really don't have an accurate idea of how much time and effort goes into making a game to the standards of the commercial soft for era of that system. And it's this inaccurate perception that's, in my opinion, the reason for the lack homebrew on the snes.