HappySqurriel said:
| Squilliam said:
I didn't say it was less than the HD consoles. It really depends on where you set par as to whether you think the Wii U is sub-par. If you set par at say 3-4* more powerful than the Xbox 360 then you'd have to say the Wii U is likely to be sub par.
There is no good reason to come out with cherry-picked examples of chips which are far more expensive than the Wii U GPU and are considered some of the best chips in a large batch tuned for a mobile platform. There are no salvage parts and voltages/clock speeds must be set at an average which allows for the best yields with respect to cost, speed and cooling requirements. There will always be chips that perform better than spec, however spec must be set at a reasonable level in order to have a more cost-effective console.
|
My expectations for the Wii U is that it will have a GPU that is (around) 3 to 6 times as powerful as what is available for the HD consoles, includes hardware support for tessellation and occlusion culling, has high quality anti-aliasing (roughly 8xAA) "for free", and other customizations to increase real-world performance; resulting in a moderately powerful GPU that is highly efficient. Most games will not be that graphically intensive and will be rendered at 1080p@60fps, but more impressive looking games will probably be rendered at 720p (possibly at a steady 30fps).
This leaves Sony and Microsoft tons of room to win a figurative "Dick measuring contest", but not much room to make meaningful improvements. After all, they will have to produce something several times as powerful to get past a point where developers use the additional processing power for something beyond increased resolution and frame-rate; and this would be problematic because it would push the sale price of their systems far beyond what consumers would be willing to pay.
|
Do the math. How many transistors do you realistically expect that they put in the Wii U given a 45nm process and relatively normal die sizes? The Xbox 360 CGPU is 168mm^2. Even doubling that with better transistor scaling would likely only yield ~1.3B transistors @ 1.5* greater density on the same process.
