Godot Engine – Multi-platform 2D and 3D game engine https://godotengine.org https://git.neuromancer.sk/godot.git/atom/scene/3d/voxel_light_baker.cpp?h=3.0.1-stable 2018-01-18T21:01:42Z 2018-01-18T21:01:42Z Rémi Verschelde 2018-01-18T20:37:17Z urn:sha1:9f479f096cb50fa8d1215e68c262f110116114ef Using v1.11.0 from https://github.com/lucasdemarchi/codespell 2018-01-11T23:47:04Z Juan Linietsky 2018-01-11T23:47:04Z urn:sha1:c48aab2f05d931fbab6bf0702e81e45b5cd88aa1 2018-01-11T23:23:41Z Juan Linietsky 2018-01-11T23:23:19Z urn:sha1:eb16f9939a1ccb1e44d230f7ed2811ea55c25c34 2018-01-01T13:40:47Z Rémi Verschelde 2018-01-01T13:40:08Z urn:sha1:b50a9114b105dafafdda8248a38653bca314a6f3 Happy new year to the wonderful Godot community! 2017-12-26T15:45:56Z Juan Linietsky 2017-12-26T15:44:58Z urn:sha1:c1153f5041f896536957b851329b5adaa1de6dc4 2017-12-24T12:32:12Z Juan Linietsky 2017-12-24T12:31:17Z urn:sha1:021f3c924be29cafe9d8d50bf00ecc6f13675e87 -Disabled Opus, implementation is wrong. 2017-12-20T13:37:00Z Hein-Pieter van Braam 2017-12-20T13:30:50Z urn:sha1:0db512354807318646ac9884f3702733a56b3bb1 The previous commit corrected the RNG behavior for the lightbaker but also made it significantly slower on high core count systems. Due to the vector of states being physically close together in RAM we force a cache synchronization across all cores whenever we call for the next random number to be generated. This will create a temporary local copy of the RNG state before entering the loop and then saving it back to the global state when done. This will preserve the per-thread RNG state (and random number quality) while significantly improving performance. On my 16 thread box it saves 3 minutes baking the Sponza scene, bringing performance back in line to before the various RNG fixes were introduced, being slightly faster than the first implementation. 2017-12-20T11:09:09Z Hein-Pieter van Braam 2017-12-20T11:04:52Z urn:sha1:ccbb5923ac12f38b9bc47448a45340dd0534eabc In our previous attempts to fix the lightmapper we may have inadvertently introduced the same issue we were trying to fix. It appears that rand() will on some platforms introduce a mutex making it slower and on others may have a per-thread state that would need to be initialized with srand() on each thread. This slows down the lightbaking further. This sets up a separate rng state for each OpenMP thread by calling rand() only in the single-threaded part of the code. We then keep a vector of states. I believe this solves our problems. 2017-12-20T00:04:30Z Hein-Pieter van Braam 2017-12-20T00:04:30Z urn:sha1:7a10d3ab78975a95f55b1bf97705ff8cc300da08 Sorry for the churn 2017-12-20T00:02:03Z Hein-Pieter van Braam 2017-12-20T00:02:03Z urn:sha1:9fa5a588f0ad42c58b07b0c30bf56a29d59378bf