Noors, you probably dont want to apply gamma correction in a shader for lightmaps. I think that would make banding and compression artifacts more visible. Instead if you need gamma correction render your lightmaps and save them as 16 bit images and do the corrections in photoshop (loading them into the engine or into max as a standard 8 bit per channel image), or render them directly with the gamma correction you want.
Thanks for answering !
My goal is to have the same result in my vray renders and my engine, so i will know by doing a quick render, what exact result i (and my customer) will have in the end.
With linear workflow, Vray render all the passes, do all the operations between them in gamma 1, then apply a gamma 2.2
If if mix an already corrected gamma lightmap and gamma corrected textures (that's what i do actually), it doesn't give the same result. It's not that far, but the white are very burnt, forcing me to do adjustements in photoshop, and still it can't mathematically be the same, i think.
Anyway, how all the tone mapping that we see in UDK or mirror's edge could be done, if the lightmaps were 8 bits ? There would be horrible banding aswell, don't ya think ?
Replies
My goal is to have the same result in my vray renders and my engine, so i will know by doing a quick render, what exact result i (and my customer) will have in the end.
With linear workflow, Vray render all the passes, do all the operations between them in gamma 1, then apply a gamma 2.2
If if mix an already corrected gamma lightmap and gamma corrected textures (that's what i do actually), it doesn't give the same result. It's not that far, but the white are very burnt, forcing me to do adjustements in photoshop, and still it can't mathematically be the same, i think.
Anyway, how all the tone mapping that we see in UDK or mirror's edge could be done, if the lightmaps were 8 bits ? There would be horrible banding aswell, don't ya think ?