My theory on PS3 graphics

[gamerdtr]

The PS1 was the easiest to program for in its generation since it had a straight forward CPU whilst N64 was a little harder with its co-processors and Saturn was the hardest with its two separate (not on the same chip as some have mistaken in forums) CPUs battling for IO. The PS1 also was not the most powerful of its generation.

The PS2 was the hardest to program for in its generation because it liked ASM code. GameCube had more straight forward hardware and XBOX was entirely a PC. PS2 was the weakest console of its generation.

The PS3 is the hardest to program for because the programmers must specify themselves how to allocate work to the SPUs. It is more automatic for the XBOX 360 and the Wii has straight forward upgraded GameCube hardware. PS3 has the most powerful processor of the lot.

But the thing is like everybody programmers are lazy and want things to be easy, but the only way to get more performance out of processors is going the multi-core way and specifying work more "manually". This means it is harder to code since running tasks in parallel as efficient as possible requires the programmer to carefully think about how he assigns jobs.

I read an University lecture on the future of processors and the way I understand it is the PS3's Cell is like a manual car. If you have ever driven a automatic car I am sure you noticed how it battles to find the right gear if you go up hill whilst with a manual car it is no big deal in getting the best performance. I am not sure that I described it clearly enough here but that is what came to mind when I read the lecture notes.

So developers are just going to have to get used to the fact that it is going to get harder to code (gulp...that means me too).

So it makes sense developing for PS3 first and then XBOX 360 and we have already seen the positive results from taking that approach.

But deveopers will surely take the easy (lazy) way out for years to come, so oh well.