Is there an explanation somewhere about the decision to use RED (R,0,0,1) instead of INTENSITY (I,I,I,I) behavior for single-channel textures in GPU programming? It seems like the universally wrong method to standardize, but maybe I'm missing something?
Eg., the shader is compiled to take the texel fetch and use all four channels. The texture unit is going to return _something_ in those four channels, and the question is, does it replicate or does it have hard-coded 0's and 1's in places.
-
-
If you were going to standardize on something there, it doesn't make sense to me why you'd pick (0, 0, 0, 1) as the default for missing channels instead of replicate.
-
It doesn't have to make sense, it's legacy :-)
End of conversation
New conversation -
-
-
The texture unit doesn't have to return anything. Perfectly normal to have the texture unit only return 2 registers of data if it's given a 2-channel texture format. Returning more just wastes power.
-
Remember all this .rgba and .xyzw stuff is just a compiler hallucination. All the hardware I know of is "scalar"
- Show replies
New conversation -
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.