Ok, full interview from anonymous third party about Wii GPU.

If you can find the new MP3 pics laying around the web I think the debate that devs are half hearting effort on Wii can be put to rest.

Here's a quote from the interview on MP3 that play online did.

"Play: Okay, gotcha! So moving on to the visual side of things, I was wondering if you could talk a bit about how developing for the Wii is different from developing for the GameCube.

MP: Well, our production pipeline was similar, so it was a lot faster for us to get up and running on the game hardware, which was great. But we were also able to make a great deal of enhancements to the game world. There’s more polygons, higher-res textures, bloom lighting—a lot of graphical enhancements that we were able to make on the Wii that we couldn’t do on the GameCube.

MK: We really looked at the Wii from a philosophy of “wow, look at all the stuff that we can do now,†as opposed to “well, it doesn’t do this, and that’s kind of a bummer.†So for us, we were really able to accomplish a lot more things than we could do on the GameCube. And in my opinon [laughs], it’s the best-looking Wii game out there. I won’t say it was seamless, but it certainly wasn’t like developing on a brand new piece of hardware with a six-month learning curve. The development environment is a cakewalk to work with, and we feel we took advantage of what the hardware has to offer. I’m partial to the bloom lighting, it’s a really nice effect, and we could never have done that on GameCube."

Source
 
Last edited by a moderator:
you can not do normal map on the cpu.That is a per pixel instruction.
On a console it's pretty doable. Wii/GC/PS2 can do regular per channel multiplication for sure if not a dot product. All you have to do is a full screen pass with the CPU where you add R+G+B and write it into A. With only 256K pixels, that's a piece of cake on a CPU, and they can access video memory very efficiently. Of course, you have to draw everything again to use the value, and there's no normalization.

Still, it's pretty likely that Wii/GC has a dot-product operation on the GPU. They said Geforce2 class, right?
 
Mintmaster said:
All you have to do is a full screen pass with the CPU where you add R+G+B and write it into A. With only 256K pixels, that's a piece of cake on a CPU, and they can access video memory very efficiently.
You can do better then that - if you run deferred shading numbers on PS2, @30fps you get something like 32instructions/pixel(single VU), which allows for a lot more then just dot product.
Basically you get you SM3.0+, with dynamic branching @ 1pixel granularity, and no texture lookups. Unfortunately there are other issues that make this less then practical, but cpu setup is up to it.

Still, it's pretty likely that Wii/GC has a dot-product operation on the GPU.
No dot-product, dependant lookup with a hard-wired 3x2 matrix transform.
 
Last edited by a moderator:
Once again I'd like to remind some posters that this is not the Console Games forum but the Console Technology forum. A few off-topic posts have been moved.
 
You can do better then that - if you run deferred shading numbers on PS2, @30fps you get something like 32instructions/pixel(single VU), which allows for a lot more then just dot product.
Basically you get you SM3.0+, with dynamic branching @ 1pixel granularity, and no texture lookups. Unfortunately there are other issues that make this less then practical, but cpu setup is up to it.
Oh I know (about doing better than just a dot3), I just wanted to point out to bomlat that it was doable.

I'm interested in what the other issues are, but I guess they're covered under NDA, right?
No dot-product, dependant lookup with a hard-wired 3x2 matrix transform.
What exactly can you put into that matrix? A vertex attribute? Only a constant? Can the vector being transformed come from a texture?
 
Back
Top