To see micro-optimizations and how we iterated, you may look at the history of the repository: https://github.com/ctrl-alt-test/mouton/commits/main/ https://git

To see micro-optimizations and how we iterated, you may look at the history of the repository: https://github.com/ctrl-alt-test/mouton/commits/main/ https://github.com/ctrl-alt-test/mouton/commits/main/ There's a commit (https://github.com/ctrl-alt-test/mouton/commit/79d2d1eab7a222b434b7041d8fc9bce4e60524f8 https://github.com/ctrl-alt-test/mouton/commit/79d2d1eab7a22...) where we save many bytes by removing a performance optimization. We originally wanted to keep it, but we realized we were short on bytes and that optimization was not required on recent-ish GPUs.