News Nvidia's RTX 5090, 5080 reportedly have the same L1 cache size per SM compared to RTX 4090, 4080

[bit_user]

OMM is needed to remove the work of calculating areas for which calculation is not required.

Nvidia says: "Micro-Mesh increases real-time performance and memory compression for complex geometry such as scanned artifacts, creatures, rocks, and trees."

I think there's your answer. Most games probably aren't using geometry detail that high. If they reduce the mesh complexity, as you'd want for supporting lower-end hardware and lower LODs, then OMM isn't needed.

This work only for engine and game developer, not for driver

I read that thread. It doesn't refute what I said, in any way. The answer by the Nvidia person doesn't distinguish between game vs. engine developer.