NetBSD-Bugs archive
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Old Index]
Re: kern/54230: mpv drops video frames with `--ao=oss' (probably after isaki-audio2 merge)
The following reply was made to PR kern/54230; it has been noted by GNATS.
From: Tetsuya Isaki <isaki%pastel-flower.jp@localhost>
To: gnats-bugs%netbsd.org@localhost
Cc: kern-bug-people%netbsd.org@localhost,
gnats-admin%netbsd.org@localhost,
netbsd-bugs%netbsd.org@localhost
Subject: Re: kern/54230: mpv drops video frames with `--ao=oss' (probably after isaki-audio2 merge)
Date: Fri, 31 May 2019 15:18:42 +0900
> Probably since isaki-audio2 merge mpv started to drop several frames
> making video playback hard to follow.
Thank you for your detailed report!
It is not a new problem.
Please specify --autosync=30 for mpv. It's suggested by mpv manpage.
Our audio driver + libossaudio layer have not implemented "a perfect
audio delay measurement" mechanism that mpv says(or requires).
On 8.99.37 (before merge) with hw.hdafg0.latency=128 (default),
mpv -> Dropped: 740
mpv --autosync=30 -> Dropped: 3..4
On 8.99.37 with hw.hdafg0.latency=16,
mpv -> Dropped: 0 [1]
mpv --autosync=30 -> Dropped: 0
On 8.99.41 (after merge) with hw.audio0.blk_ms=40 (default),
mpv -> Dropped: 1906 [2]
mpv --autosync=30 -> Dropped: 0..1
On 8.99.41 with hw.audio0.blk_ms=10,
mpv -> Dropped: 0
mpv --autosync=30 -> Dropped: 0
Your report compared case [1] and [2]. But this table shows
the different facts.
(And there is one more note. The previous audio driver and audio2
have different relationship between blocksize and latency. So you
don't compare them simply.)
Thanks,
---
Tetsuya Isaki <isaki%pastel-flower.jp@localhost / isaki%NetBSD.org@localhost>
Home |
Main Index |
Thread Index |
Old Index