| HappySqurriel said: The TEV is quite a bit different from a conventional shader that is used in the XBox (or XBox 360 or PS3) but it can perform many of the same operations ... "Julian Eggebrecht: He was probably referring to the TEV pipeline. Imagine it like an elaborate switchboard that makes the wildest combinations of textures and materials possible. The TEV pipeline combines up to 8 textures in up to 16 stages in one go. Each stage can apply a multitude of functions to the texture - obvious examples of what you do with the TEV stages would be bump-mapping or cel-shading. The TEV pipeline is completely under programmer control, so the more time you spend on writing elaborate shaders for it, the more effects you can achieve. We just used the obvious effects in Rogue Leader with the targeting computer and the volumetric fog variations being the most unusual usage of TEV. In a second generation game we’ll obviously focus on more complicated applications." Anyways, just to add more to my initial argument for Kwaad: "Julian Eggebrecht: Maybe without going into too much detail, we don’t think there is anything visually you could do on X-Box (or PS2) which can’t be done on GameCube. I have read theories on the net about Flipper not being able to do cube-mapped environment maps, fur shading, self-shadowing etc... That’s all plain wrong. Rogue does extensive self-shadowing and both cube-maps and fur shading are not anymore complicated to implement on GameCube than on X-Box. You might be doing it differently, but the results are the same. When I said that X-Box and GameCube are on par power-wise I really meant it. "
|
Here's how the TEV works:
There is one pipeline that can handle up to 16 texels. This is a significant departure from the standard pipeline configuration of most nVidia and ATI chips.
The XBox1 gpu had 4 pipelines that could render 4 texels each. If a pixel had 4 layers, one pipeline could process it all in one clock cycle. If it had 5, it needed two clock cycles. If you had 4 pixels, each with 4 layers, it could render that in one cycle. If you had 1 pixel with 2 layers, 1 with 4 layers, one with 2 layers and another with 5, it would do the first 3 in one cycle, but have to process the last one in two. If you had 16 pixels with only 1 layer, it would take 4 clock cycles to process them all.
Here's a breakdown:
Scenario A) Pixel 1(4 layers) + Pixel 2(4 layers) + Pixel 3(4 layers) + Pixel 4(4 layers) = 1 clock cycle (16 layers)
Scenario B) Pixel 1(2 layers) + Pixel 2(4 layers) + Pixel 3(2 layers) + Pixel 4(5 layers) = 1 clock cycle for first 3 pixels, 2 for the 4th pixel (11 layers)
Scenario C) 16 Pixels (1 layer each) = 4 clock cycles
Let's look at the same scenarios with the TEV.
Scenario A) Pixel 1(4 layers) + Pixel 2(4 layers) + Pixel 3(4 layers) + Pixel 4(4 layers) = 1 clock cycle (16 layers)
Scenario B) Pixel 1(2 layers) + Pixel 2(4 layers) + Pixel 3(2 layers) + Pixel 4(5 layers) = 1 clock cycle (11 layers + room for 5 more layers)
Scenario C) 16 Pixels (1 layer each) = 1 clock cycle
TEV can do 16 layers regardless of how many pixels they are in.
The only advantage the XBox1 gpu had was a better geometry engine and programmable shaders. Now, from what I have heard, Wii's gpu doubles this number, and has a higher clock speed. Not to mention increased memory. Sure, Wii doesn't have programmable shaders, but that doesn't mean it's less powerful than the original xbox. A NASCAR vehicle doesn't have air-conditioning or power door locks, but does that mean it's less powerful than a fully loaded Camry? Of course not. You have to look at the whole picture.
---------------------------------------------------------------
Legend11:
The reason that most multiplatform games looked better on the xbox is because the shader routines were included in the SDK and didn't need to be written. Most developers were too lazy to take the time to figure out how to properly utilize the TEV unit.
