Fact is that eventually the optimizers for the given compiler get good enough for the task at hand and people move on.
Most intensive graphics and physics are already being done on the GPU anyway and there are quite a few ways to target them as well.
Also, GPU is a temporary technology. Wait for some time, and see CPUs getting more GPU like.
Yes they do when shared memory is used, not all GPUs have dedicated memory.
Even the PS4 will have shared memory between CPU and GPU systems.
> Also, GPU is a temporary technology. Wait for some time, and see CPUs getting more GPU like.
This is Intel's wish because their GPUs suck, I don't think it will happen.
Game devs head for it too, so that they don't have to manage a shader language.
It is inevitable. This will happen.
How exactly would CPU be more like GPU? What would happen to branching and out of order execution pipelines? Do you see today's complex superscalar cores so cheap and small in the future that you could have thousands of them on single die? What other way is there get vector computations to work as well as scalar computations on same hardware? Would you rather have few superscalar cores and a lot of SIMD's around it? Or do you envision a completely new architecture?
I'm just trying to get a picture of how this would be done. Any reading material would be welcome.