v0.3.95

Documentation

• Settings → Encoding help text on the Intel encoders now flags the gotchas users hit when copying libx265 / NVENC tuning intuition over: (1) QSV's preset cost curve is nearly flat (slower is ~10-20% slower than medium, not 5×), so look-ahead is the meaningful quality knob for QSV, not preset; (2) rough cross-encoder CQ↔CRF conversion to make CQ values comparable across NVENC / QSV / libx265; (3) VAAPI is most useful on AMD — on Intel hardware where both QSV and VAAPI work, QSV produces better quality per bit at similar throughput.
• README "Intel/AMD GPU support" section now reflects that one external tester has confirmed both encoders end-to-end on a modern Intel iGPU (Ubuntu 24.04, ~145 fps QSV / ~190 fps VAAPI on 1080p HEVC) as of v0.3.92. Still flagged experimental until we have data from a broader hardware spread (older Coffee Lake, AMD, Arc / Battlemage).

Docker images

Pinned to this release:

docker pull ghcr.io/i-ial9000/shrinkerr:0.3.95         # portable (amd64 + arm64, CPU)
docker pull ghcr.io/i-ial9000/shrinkerr:0.3.95-nvenc   # x86_64 + NVIDIA GPU (NVENC)
docker pull ghcr.io/i-ial9000/shrinkerr:0.3.95-edge    # as above but ffmpeg master
docker pull ghcr.io/i-ial9000/shrinkerr:0.3.95-edge-nvenc

Floating tags (:latest, :edge, :nvenc, :edge-nvenc) track the most recent release of each lineage.