bttv msp3400 2.6.15.6 - > 2.6.16 stereo stop working

[Roland Scheidegger <rscheidegger_lists (at) hispeed.ch>]

Hans Verkuil wrote:
> Please apply the attached patch to msp3400-kthreads.c and let me know if 
> this fixes the problems.
I've tried that patch and it doesn't help. I think the chip gets 
misprogrammed somewhere, the stereo detect reg always hovers around 0 
even after a long time (nowhere near that 4096 it needs to be for stereo).

>> 3) when pres enter(xawtv)  sound mute but when increase  volume or
>> switch channel some glitches are present.
>
> It's not likely that that will be solved. It's not present with msp3400d
> and newer versions. It may be a limitation of the msp3400c.
Sorry, but this is a lame excuse at best :-) (Not that I'd accuse anyone 
of not fixing it, I understand it may not exactly be the most important 
thing...)
One reason (there may be more) is that the chip simply is never muted 
when it does its am/fm scan (which takes a while). xawtv (and probably 
other progs) will mute sound, then switch channel, then unmute. Now, 
what happens is that xawtv mutes (which causes a restart scan but it 
doesn't really matter) and audio is muted. Then, channel is switched, 
which will cause the next restart scan (and audio is still muted). Then, 
xawtv unmutes, so audio gets unmuted, even though the msp3400 is still 
trying to find the carriers (and I think this causes another restart 
scan). I'm not exactly sure what happens next, but I see another 
restart_scan which mutes again, until finally audio gets unmuted for the 
last time. No wonder there's a lot of popping noise...
Other things are wrong too. For instance that muting audio causes a full 
restart scan is just evil (and no wonder the popping noises are back 
with that).
There is a fast_mute value too which could be used instead of setting 
volume to 0, which might prevent some popping noises, though I don't 
think it is needed (it certainly won't help with the fact that the chip 
simply is not muted while searching for the carriers). The 3410d may be 
way faster with detection so it doesn't have the problem that sound gets 
unmuted when it's still busy with detection, or it may just mute its 
output anyway when doing autodetection, so I'm not surprised there is no 
problem there.

Roland

--
video4linux-list mailing list
Unsubscribe mailto:video4linux-list-request (at) redhat.com?subject=unsubscribe
https://www.redhat.com/mailman/listinfo/video4linux-list