| > There's really no reason console games of that era needed, or even could benefit from 6 cores First: One problem of the PS3 was that in a late stage of its development a (rather weak) GPU was "tacked on" (originally, the Cell was supposed to replace the GPU). Now for your point: Just have a look what scientists did on PS3 clusters using the SPUs of the Cell, and considering this imagine what creative things would have been possible for games on the PS3. The problem was that at this time, game development got much less specific to one platform (in opposite to how it was for basically all the previous game console generations, though I am willing to admit that the original Xbox being very similar to a PC did give a hint concerning the direction the game industry was heading), so you wanted a much more "standardized" (typically meaning: similar to PC) architecture. The Xbox 360 was simply much more similar to PCs, so it was much easier to port engines from one platform to another. > the people I've talked to who have worked for it said the architecture was a nightmare to work with. For previous console generations this was less of a problem (basically every console was quirky). It is easy to find statements of developers on the internet how the PS2 was magnitudes more of a nightmare to develop for than the PS3. So, before, being difficult to develop for a console was much more accepted in consideration of the unique features each console brought to the table. What did change was rather the expectation among developers to be able to easily port engines between platforms. |