GX: a few technical questions

[tom34]

Hello,

Here are my questions about GX :

- Using 2 TEV stages on a polygon seems to take as much fillrate as drawing that polygon twice. Is it normal? Is it the same with any number of TEV stages?

- Is there any way to read the bounding box coordinates we can clear with GX_ClearBoundingBox? I could use them for some optimization.

- I'd like to render a scene into a texture which is smaller than the screen size, but i don't want it to be too pixelated.
Can i do that using GX_SetCopyFilter to add some anti-aliasing? How do i setup the anti-aliasing? I have to render my scene in RGBA6_Z24.
The only other way i found was rendering a full size texture and downsizing it using bilinear filtering, but it's a waste.

BTW thank you so much for your amazing tool chain. It's an incredible work you've done, congratulations.

[Samson]

- Using 2 TEV stages on a polygon seems to take as much fillrate as drawing that polygon twice. Is it normal? Is it the same with any number of TEV stages?

Do both the TEV stages require a texture fetch? And is the vertex cost low? Then I assume it must be nearly the same.

[tom34]

Yes they do. Thanks for the answer.

I hope to get an answer for the 2 other questions too

[tom34]

After some research :
- GXReadBoundingBox seems to exist in the official library, but it seems to be missing in libogc.
- Antialiasing is only supported with RGB565_Z16 pixel format.