Fast seek in bug516323.indexed.ogv doesn't work as expected

NEW
Unassigned

Status

()

Core
Audio/Video: Playback
P5
normal
2 years ago
10 months ago

People

(Reporter: jwwang, Unassigned)

Tracking

Firefox Tracking Flags

(Not tracked)

Details

(Reporter)

Description

2 years ago
The video key frames of bug516323.indexed.ogv are:
[0,41666]
[583333,625000]
[666666,708333]
[3333333,3375000]

If we call fastSeek(1.76), the 1st decoded video frame after seek is [0,41666] which seems wrong to me. For 0.666666 < 1.76 < 3.333333, we should get either [666666,708333] or [3333333,3375000] after seek.

I found this problem when trying to fix bug 1193124.
(Reporter)

Updated

2 years ago
Flags: needinfo?(jyavenard)
(Reporter)

Updated

2 years ago
Blocks: 1193124
I don't know anything about ogg, and it's not using the new media architecture.

Chris knows all about it though :)
Flags: needinfo?(jyavenard) → needinfo?(cpearce)
(In reply to Jean-Yves Avenard [:jya] from comment #1)
> I don't know anything about ogg, and it's not using the new media
> architecture.

Bug 1168674 claims to have working patches to move it to the new media architecture. Maybe it would be better to get them landed than try to fix the old one?
(In reply to Timothy B. Terriberry (:derf) from comment #2)
> Bug 1168674 claims to have working patches to move it to the new media
> architecture. Maybe it would be better to get them landed than try to fix
> the old one?

We shouldn't be making investments in the old OggReader for sure. I'd like to see the new Ogg MFR demuxer landed. Is Byron still working on this?
Flags: needinfo?(cpearce)
Brion posted working patches in his Github (linked from the bug) shortly before Christmas.
New test file for you, Brion.
Flags: needinfo?(brion)
Priority: -- → P5
Clearing needinfo. The old OggReader has been replaced, but I don't believe anyone is investigating this.
Flags: needinfo?(brion)
You need to log in before you can comment on or make changes to this bug.