Replace gifsicle GIF pipeline with gifski#1282
Open
artisticmedic wants to merge 1 commit into
Open
Conversation
Kap's GIF export ran ffmpeg palettegen -> paletteuse -> gifsicle. On Apple Silicon the bundled gifsicle is x86_64-only, so it runs under Rosetta and triggers the macOS Intel deprecation warning; it is also the slowest export path. Replace it with gifski, a multithreaded GIF encoder that ships a universal (x86_64 + arm64) binary. gifski's binary statically links FFmpeg, so it reads the recording directly: convertToGif now runs ffmpeg only to trim the selected range (gifski can't trim), then hands the clip to gifski for fps resampling, scaling, and encoding -- no intermediate frames, which also removes a per-frame argument list that overflowed ARG_MAX on long recordings. Benchmarks (native arm64): ~1.2-1.5x faster and ~15% smaller than the old pipeline, plus no Rosetta. - The npm gifski package is frozen at 1.7.1; scripts/download-gifski.js fetches the official universal 1.34.0 release on postinstall, verified against a pinned SHA-256, into vendor/gifski (gitignored, packaged via asarUnpack). - GIF frame delays use 1/100s units, so fps is capped at 50. - The "Lossy GIF compression" setting now maps to gifski --quality. - gifski merges duplicate frames, so the GIF tests assert playback duration rather than a constant nominal frame rate. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
Kap's GIF export ran
ffmpeg palettegen → paletteuse → gifsicle. On Apple Silicon the bundledgifsicleis x86_64-only, so it runs under Rosetta and triggers the macOS Intel deprecation warning — and it's the slowest part of the export.This PR replaces that pipeline with gifski, a multithreaded GIF encoder that ships a universal (x86_64 + arm64) binary.
How it works
gifski's binary statically links FFmpeg, so it reads the recording directly.
convertToGifnow runs ffmpeg only to trim the selected range (gifski can't trim), then hands the clip to gifski for fps resampling, scaling, and encoding. No intermediate frames — which also removes the per-frame argument list that overflowedARG_MAXon long recordings.Benchmarks
Native arm64: ~1.2–1.5× faster and ~15% smaller output than the old pipeline, plus no Rosetta.
Notes for reviewers
gifskipackage is frozen at 1.7.1 (unmaintained since 2022).scripts/download-gifski.jsfetches the official universal 1.34.0 release onpostinstall, verified against a pinned SHA-256, intovendor/gifski(gitignored, packaged viaasarUnpack). macOS-only, usescurl+tarthat ship with macOS.--quality.gifsicledependency removed (large drop inyarn.lock).🤖 Generated with Claude Code