Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
Use radsort for Transparent2d PhaseItem sorting (#9882)
# Objective Fix a performance regression in the "[bevy vs pixi](https://github.com/SUPERCILEX/bevy-vs-pixi)" benchmark. This benchmark seems to have a slightly pathological distribution of `z` values -- Sprites are spawned with a random `z` value with a child sprite at `f32::EPSILON` relative to the parent. See discussion here: #8100 (comment) ## Solution Use `radsort` for sorting `Transparent2d` `PhaseItem`s. Use random `z` values in bevymark to stress the phase sort. Add an `--ordered-z` option to `bevymark` that uses the old behavior. ## Benchmarks mac m1 max | benchmark | fps before | fps after | diff | | - | - | - | - | | bevymark --waves 120 --per-wave 1000 --random-z | 42.16 | 47.06 | 🟩 +11.6% | | bevymark --waves 120 --per-wave 1000 | 52.50 | 52.29 | 🟥 -0.4% | | bevymark --waves 120 --per-wave 1000 --mode mesh2d --random-z | 9.64 | 10.24 | 🟩 +6.2% | | bevymark --waves 120 --per-wave 1000 --mode mesh2d | 15.83 | 15.59 | 🟥 -1.5% | | bevy-vs-pixi | 39.71 | 59.88 | 🟩 +50.1% | ## Discussion It's possible that `TransparentUi` should also change. We could probably use `slice::sort_unstable_by_key` with the current sort key though, as its items are always sorted and unique. I'd prefer to follow up later to look into that. Here's a survey of sorts used by other `PhaseItem`s #### slice::sort_by_key `Transparent2d`, `TransparentUi` #### radsort `Opaque3d`, `AlphaMask3d`, `Transparent3d`, `Opaque3dPrepass`, `AlphaMask3dPrepass`, `Shadow` I also tried `slice::sort_unstable_by_key` with a compound sort key including `Entity`, but it didn't seem as promising and I didn't test it as thoroughly. --------- Co-authored-by: Alice Cecile <[email protected]> Co-authored-by: Robert Swain <[email protected]>
- Loading branch information