David Michael Barr

We have started documenting encoding performance in release announcements. See for example: https://github.com/xiph/rav1e/releases/tag/v0.4.0

I consent to re-license my contributions to build.rs.

This has been partly superseded by the changes in #2682. In particular, it was made explicit that the activity scale is implied by `ssim_boost` as a function of variance. The...

We might need to supplement with https://github.com/est31/serde-big-array to cover `--features=serialize`.

To clarify, the **Contributing** chapter has moved to [CONTRIBUTING.md](https://github.com/xiph/rav1e/blob/master/CONTRIBUTING.md). This section should be added to that document.

This might need to be manually reapplied to the new file but it would be great to finally merge this.

The ffmpeg wrapper currently maps `matrix`, `primaries` and `transfer` from ffmpeg-native configuration: `colorspace`, `color_primaries` and `color_trc`. Mastering and content-light appear not to be mapped in ffmpeg master.

@lu-zero Are `mastering-display` and `content-light` available through the C API yet?

A first step toward this was landed in #1866.

See also the `dav1d` tracking issues for [AVX2 SIMD](https://code.videolan.org/videolan/dav1d/issues/78), [SSSE2 SIMD](https://code.videolan.org/videolan/dav1d/issues/216) and [NEON SIMD](https://code.videolan.org/videolan/dav1d/issues/215). Decoder kernels are best contributed first to [dav1d](https://code.videolan.org/videolan/dav1d) as there is rigorous testing and expert review...