Commit Graph
5 Commits
Author SHA1 Message Date
Miguel Ángel 3d45d1b037 fix(engine): audio encodes on FFmpeg 4.4, which has no setts duration option (#6219)
* fix(engine): audio encodes on FFmpeg 4.4, which has no setts duration option

The segmented AAC encode ended its stream with setts=duration, added in FFmpeg 5.0, so every render
with audio failed on Ubuntu 22.04's FFmpeg 4.4. An FFmpeg without the option now encodes the audio in
one pass, which ends the stream at the duration itself.

* fix(engine): keep the setts capability check private to the AAC encoder
2026-10-24 16:10:44 +00:00
Miguel Ángel 9da107aef5 fix(audio): fit PCM to video frames before AAC delivery (#6089)
* fix(audio): fit PCM to video frames before AAC delivery

* fix(audio): migrate the CLI diagnostic stage fixture

* fix(test): measure the whole unscaled audio residual

* test(audio): refresh the speed-ramp delivery golden
2026-10-22 23:12:56 +00:00
Miguel Ángel a8a80323f6 perf(engine): cloud renders cut a video's audio from its URL instead of downloading the file (#5972)
* perf(engine): cloud renders cut a video's audio from its URL instead of downloading the file

* refactor(engine): keep the audio mixer's comment share at the base

* fix(engine): a cloud-held file's audio probe retries once, then fails instead of mixing unprobed

* fix(engine): a cancelled probe is not retried, and the encode test noise is seeded so it repeats
2026-10-20 20:52:10 +00:00
Miguel Ángel 87c6e50f5e fix(engine): a sharp sound in a film's first moments no longer grows louder with each audio pass (#5956)
* fix(engine): a sharp sound in a film's first moments no longer grows louder with each audio pass

Every AAC encode now starts in digital silence that is dropped afterwards. Opening cold on quiet sound, the native encoder coded the first loud frame three scalefactor steps hot (4.5 dB), and each re-encode (pad, true-peak correction) did it again.

* fix(engine): AAC encodes stay in step with audio that is not 48 kHz before the encoder

* test(producer): regression goldens for audio that now opens without the encoder's cold start
2026-10-20 16:32:03 +00:00
Miguel Ángel 8e4ce74118 perf(engine): long renders encode their audio in parallel segments (#5788)
* perf(engine): long renders encode their audio in parallel segments

* fix(engine): parallel audio encode counts packets from whole samples

The expected packet count came from a floating-point duration, so frame-aligned lengths such as 147.52 s (one ulp above 7080960 samples) expected one packet more than the encoder writes, the last segment check threw, and the render lost its audio. Round to whole samples first; a 147.52 s test fails without the fix.
2026-10-17 22:51:25 +00:00