Mario Kart 8 may be able to run on PS360, whether framerate will hold up is questionable. Graphically it's similar to Sonic All Stars, but MK8 game runs in a higher resolution (SART runs sub 720p I think), at 720p, and SART runs in 30fps while Mario Kart 8 runs at 60fps, with 12 racers, and even with 2 player split screen with absolutely no framerate drops. Is there another game on PS360 that can compare to MK8 (artstyle-wise)?
Bayonetta could probably run on PS360, with some framerate issues and some tearing. I've played the first game, and there was some very noticable tearing in it.
X would probably need to have certain things reduced, like perhaps a shorter draw distance
I guess I'll have to add something regarding this whole "160SP / 176GFlops, 320SP / 352GFLOPS" idea. Many have used ports from PS360 games as evidence to support the whole "160SP" hypothesis, but why don't those people consider the fact that the system has a completely different architecture and that the dev tools before launch were terrible? The CPU is likely worse at SIMD (based on what I've heard), but excels in general processing when compared to Cell and Xenon, so obviously running SIMD-heavy games would not work very nicely using PS360 techniques. Other techniques will need to be used, like what Criterion stated several months ago about how it was unfair to compare the architectures since they are different. They understood where the CPU was "weaker" and where it was "stronger", so they leveraged that, and with help from Nintendo, they seemed to have succeeded. Wonder how many other 3rd party devs have asked Nintendo for some aid....
Anyway, why wouldn't it be able to be 320SPs again? Because the ported games currently don't show evidence of it, right? The extra GFLOPS would definitely give Wii U a resolution boost and even a graphical boost. But..... wouldn't seeing these graphical improvements from a 352GFLOP GPU imply that the GPU would be nearly maxed out at launch? I mean, no console is maxed out at launch, and those in favor of the 160SP hypothesis seem to get that idea that devs are either close to maxing out or have reached the GPU's potential already. Given the fact that documentation for devs prior to launch was terrible, given the fact that the aid wasn't available until after launch, and given the fact that the architecture is very different, how in the world could devs have possibly maxed out anything at launch?
Regarding poor-performing Wii U games, or games that only appear on par, the graphics, AA, resolution etc will NOT magically increase just because of an increase in GFLOPS. It's not as easy as flipping a switch, and it will cost some extra money to make those changes happen. It will cost even more considering the different architecture.
I'm not saying the 160SP hypothesis is wrong, but using PS360 ports as evidence to support the idea is not very good evidence considering the situation we have. It could very well be a 176GFLOP GPU and still be more powerful than the 240GFLOP Xenos, simply because Latte is likely far more efficient since it is more modern. It may not even be 176 nor 352 at all, how can we even be sure it has such a straightforward SP count like 20 or 40 per shader block? We assume Nintendo did nothing to modify the standard. It could be something more than 20, less than 20, more than 40 or inbetween 20 and 40. And with some unexplained logic there (or even with explained logic) total performance cannot be judged just based on FLOPS alone. Latte has a tessellator, and it's guarranteed to be more efficient than Xenos's tessellator (at its minimum, it's an R700 tessellator, which is still more efficient than Xenos), it has compute shader support, a huge EDRAM pool, and we have yet to see any of the OpenGL 4.x stuff be used (last I checked, it's based on OpenGL)....too much stuff is being ignored when figuring out how capable Latte is. I doubt any of the current gen ports are taking advantage of the newer features Latte has, and we know for sure it has them.
Also, to ask about DICE and their reasoning, have they confirmed when they did their "test" for frostbite 3 on Wii U? Was it during the terrible documentation period or after?