Save-to-webm: use timestamp from samplebuilder #130
Conversation
This file contains 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
|
If/when this gets merged I can do the same for the Twitch example. |
tmatth
force-pushed
the
bugfix/pop-with-timestamp
branch
from
3dc7e3a
to
4f95b29
Compare
tmatth
force-pushed
the
bugfix/pop-with-timestamp
branch
from
4f95b29
to
83158f2
Compare
|
Tweaked to use |
This avoids desyncing by accumulating drift.
tmatth
force-pushed
the
bugfix/pop-with-timestamp
branch
from
83158f2
to
5bab82e
Compare
at-wat
reviewed
Nov 11, 2022
tmatth
force-pushed
the
bugfix/pop-with-timestamp
branch
from
ac7d2b9
to
051d328
Compare
In the event that the packet timestamp wraps past uint32max, cast to int64 and add uint32_max so that the delta (and thus the webm timestamp) stays positive.
tmatth
force-pushed
the
bugfix/pop-with-timestamp
branch
from
051d328
to
aca9a01
Compare
Hopefully something like aca9a01 is what you had in mind, I would squash before merging I just wanted the change to accommodate overflow to be clear. |
tmatth
commented
Nov 17, 2022
Instead we use the sample.Duration (as before) for the first packet only, subsequent packets will be computed as a delta.
tmatth
force-pushed
the
bugfix/pop-with-timestamp
branch
from
523f40f
to
c827071
Compare
at-wat
reviewed
Nov 22, 2022
Comment on lines
+77
to
+82
| timestampSinceStart := int64(sample.PacketTimestamp) - int64(s.startAudioOffset) | ||
| // handle range where PacketTimestamp has wrapped past uint32 by operating in int64 range until the timestamp has caught up to the offset | ||
| if timestampSinceStart < 0 { | ||
| timestampSinceStart = int64(sample.PacketTimestamp+math.MaxUint32) - int64(s.startAudioOffset) | ||
| } | ||
| ts = timestampSinceStart / 48 // convert from RTPTime to sample time |
There was a problem hiding this comment.
Does it work after the second overflow of packet timestamp?
|
Sorry for the radio silence, I finally circled back to this and I'm tempted to drop it as in my own application, it is not fixing the problem it should be. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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.
This avoids desyncing by accumulating drift.
Description
In my own application I've been able to force audio and video to get badly out of sync by interfering with the source (causing dips in FPS etc., I don't fully grok the desync, but I can reproduce it reliably), which would cause the mkv written to disk to be badly out of sync. This change uses the sambuilder packet timestamp directly - the starting offset instead of accumulating durations.
Some caveats: