He's a hero among coders for particularly elegant programming that pushes the limits of hardware. (bolding by me)
Carmack may be known for many things, but elegant code isn't one of them.
That's actually untrue, or at least misleading. In addition to the x86 processor, the XBox includes a GPU (NV25) dedicated to graphics processing. This doesn't just draw triangles either, it processes shader programs and handles transforms and lighting. This is by far the bulk of the console's work.
The type of processor is pretty much irrelevant, which is presumably why they went for a bog standard, cheap x86.
Well, there are pixel shaders and there are vertex shaders.
Pixel shaders are small programs which allow you to specify per-pixel operations applied to surfaces, such as how to combine the textures, and how to calculate lighting equations. This gives a lot more flexibility than standard OGL calls allow, and you do do some very complex things, like shadows, bump mapping, per-pixel lighting and environment mapping.
As for vertex shaders, I'm no expert because I've never used them, but I understand they're a similar sort of deal except dealing with sets of vertices.
All of this is a good thing if you want more flexibility in your graphics algorithms, and don't want to be constrained by the API.
Until recently, these programs have been written in a low level language similar to ASM, and were only accessable to OpenGL through hardware specific extensions. The main reason OpenGL 2 is so great is that it makes all this hardware independent, and it uses a high level language (similar to C).
There's plenty of info on nVIDIA and ATi's sites about this stuff, including pretty pictures.
Slashdot Mirror
He's a hero among coders for particularly elegant programming that pushes the limits of hardware.
(bolding by me) Carmack may be known for many things, but elegant code isn't one of them.
I almost fell off my chair and woke up half the house, you bastard.
Where did I say clockspeed matters exactly?
That's actually untrue, or at least misleading. In addition to the x86 processor, the XBox includes a GPU (NV25) dedicated to graphics processing. This doesn't just draw triangles either, it processes shader programs and handles transforms and lighting. This is by far the bulk of the console's work.
The type of processor is pretty much irrelevant, which is presumably why they went for a bog standard, cheap x86.
Well, there are pixel shaders and there are vertex shaders. Pixel shaders are small programs which allow you to specify per-pixel operations applied to surfaces, such as how to combine the textures, and how to calculate lighting equations. This gives a lot more flexibility than standard OGL calls allow, and you do do some very complex things, like shadows, bump mapping, per-pixel lighting and environment mapping. As for vertex shaders, I'm no expert because I've never used them, but I understand they're a similar sort of deal except dealing with sets of vertices. All of this is a good thing if you want more flexibility in your graphics algorithms, and don't want to be constrained by the API. Until recently, these programs have been written in a low level language similar to ASM, and were only accessable to OpenGL through hardware specific extensions. The main reason OpenGL 2 is so great is that it makes all this hardware independent, and it uses a high level language (similar to C). There's plenty of info on nVIDIA and ATi's sites about this stuff, including pretty pictures.