I happen to notice nobody has mentioned multi threaded rendering which has also proven to double FPS / performance on the few games it has been implimented on http://forums.anandtech.com/showpost.php?p=31520674
read this it showwed double or more performance increase over standard single / dual threaded rendering because it doubles the call / draw throughput so it is very relevant and offers a nice increase in performance. Still till this day only a handfull of games have used multi threaded rendering the reason being till next gen games were not designed with full dx11 ability or optimizations. The reason present and past gen games didnt use dx11 full capabilities was it was not possible or feesible to do so in the game development phase between console and PC for port overs and such.
Next gen games bring this and more including bypassing the direct X API altogether and also possibly using direct compute functions and maybe even AVX instructions so with just these alone will give a very large performance boost look at what the AVX performance on present AMD CPU's is on linux it is staggering the increase in the performance increase over the present instructions used.
I just wanted to link this to give an idea of the raw processing power a ATI / AMD GPU has https://en.bitcoin.it/wiki/Mining_hardware_comparison
literally you compare ATI vs Nvidia at same price or performance lvl in games to the raw processing power they can yeild in say bitcoin mining and such. The ATI / AMD cards will double and sometimes tripple the performance of the Nvidia cards so if ATI / AMD raw grunt processing power is used in next gen games to do some of the processing there is a chance for some huge increases in overall performance.
A GPU has a lot of cores (hundreds). Each core is basically able to compute one 32-bit arithmetic operation per clock cycle -- as a pipeline. Indeed, GPU's work well with extreme parallelism: when there are many identical work units to perform, actually many more than actual cores ("identical" meaning "same instructions", but not "same data").
GPU's achieve great performance by using heavy parallelism, with hundreds (if not thousands) of cores. This is made possible by pipelining (each individual operation takes many cycles to run, but successive operations can be launched like trucks on a highway) and sharing instruction decoding (since many cores will run the same instructions at the same time).
THIS IS FROM ONE OF MY PREVIOUS POSTS DIRECTLY FROM SONY
The system is also set up to run graphics and computational code synchronously, without suspending one to run the other. Norden says that Sony has worked to carefully balance the two processors to provide maximum graphics power of 1.843 teraFLOPS at an 800Mhz clock speed while still leaving enough room for computational tasks. The GPU will also be able to run arbitrary code, allowing developers to run hundreds or thousands of parallelized tasks.
Sony is building its CPU ( x86 same as PC AMD 8core ) on what it's calling an extended DirectX 11.1+ feature set ( ATI HD 7850 equivalent directX11 APU ), including extra debugging support that is not available on PC platforms ( presently ) . This system will also give developers more direct access to the shader pipeline than they had on the PS3 or through DirectX itself. "This is access you're not used to getting on the PC, and as a result you can do a lot more cool things and have a lot more access to the power of the system," Norden said. A low-level API will also let coders talk directly with the hardware in a way that's "much lower-level than DirectX and OpenGL," but still not quite at the driver level.
almost all of this stuff can move over into the PC realm of games easily thru tweaks and adjustments / optimizations so AMD CPU's and GPU's / APU's could get a very nice performance boost in next gen games.Edited by rickcooperjr - 9/21/13 at 9:46am