 David Piuva
|
ba162b1f68
Found a bug in clamping of mipLevel in texture_sample_bilinear, but it is much faster to just clamp it as a single scalar in texture_getMipLevelIndex, so MIP_INSIDE is currently mandatory until I find an efficient way of limiting mip levels on Intel processors.
|
6 months ago |
|
David Piuva
|
e69f53aaaf
Moved outrendering logic from modelAPI.h and replaced raw pointers with reference counting.
|
7 months ago |
 David Piuva
|
339b50e00c
Allowing the mip level to be uniform to increase performance in texture sampling.
|
10 months ago |
|
David Piuva
|
8ba2e9c9de
Moved lots of code into an implementation folder.
|
10 months ago |
