You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
https://github.com/cschied/q2vkpt/blob/master/src/refresh/vkpt/shader/asvgf.glsl#L24 defines it as 3, but for 0-2 (0-GRD_DWN-1) only 2 bits are needed. This normally wouldn't matter, but with bigger resolutions (e.g. 8kx8k), this would be hit. This is because for the 1D location in previous frame it uses a 32 bit int and packs 7 bits for 1 (dirty), 2x3 (x and y GRD_DWN), leaving only 25 bits for texture location. This would not be hit by regular 8k though, as it just about fits, but I'd not recommend it for future proofing.
The text was updated successfully, but these errors were encountered:
https://github.com/cschied/q2vkpt/blob/master/src/refresh/vkpt/shader/asvgf.glsl#L24 defines it as 3, but for 0-2 (0-GRD_DWN-1) only 2 bits are needed. This normally wouldn't matter, but with bigger resolutions (e.g. 8kx8k), this would be hit. This is because for the 1D location in previous frame it uses a 32 bit int and packs 7 bits for 1 (dirty), 2x3 (x and y GRD_DWN), leaving only 25 bits for texture location. This would not be hit by regular 8k though, as it just about fits, but I'd not recommend it for future proofing.
The text was updated successfully, but these errors were encountered: