[mythtv-users] Really bad sound quality from /dev/dsp1

Jim Carter jimc at math.ucla.edu
Fri Dec 19 19:38:04 UTC 2008


On Tue, 25 Nov 2008, jb17bsome wrote:

> For 1, Did you see the ticket for alsa sound capture?
> http://svn.mythtv.org/trac/ticket/3405

I thought I'd close out this thread with a semifinal report.

Following the suggestion of jb17bsome I checked out the future v0.22
from SVN and applied the patch for ALSA sound capture from
http://svn.mythtv.org/trac/ticket/3405 
This was successful in using ALSA to capture the sound, but the audible
result was the same: terrible sound.  As that particular SVN state had
GUI issues which detracted from WAF, I reverted to v0.21, unfortunately. 

(Note to the developers: ALSA is the way of the future, OSS is in the
past...)

While arecord was able to record from the ALSA source, GStreamer and
Xine (the GUI player) were unable to set sound parameters and refused
to use the device.  My best guess on the scenario is like this: when
the Myth backend uses either ALSA or OSS to capture, it sets parameters
such as the sample rate but does not abort when this fails.  Then the
sound is trashed, for example the card sends at 48000Hz stereo but the 
OSS API thinks it is receiving at 8000Hz mono, which would pretty well
match what I'm hearing.

Other people say on the Wiki that they deal with the pcHD-5500 by 
connecting the analog sound jack over to line-in of the sound card.  It's 
tricky (and implementation dependent) to get the sound shut off in the 
output mixer, so it doesn't play all the time (out of sync with live TV 
images), but turned on in the capture mixer, but eventually I figured out 
what the hardware was trying to accomplish.  I'm now able to record good 
quality sound by the roundabout route.

Comments in the sources for the cx88_alsa driver indicate that in kernel
2.6.22 the sound support exists but is far from complete.  I plan over 
the holiday to upgrade to a recent kernel and to see how much better the
new version might be.

Maybe I'll be able to record direct from the capture card if I use the
latest driver version.  Thanks to all who responded to my posting.


James F. Carter          Voice 310 825 2897    FAX 310 206 6673
UCLA-Mathnet;  6115 MSA; 405 Hilgard Ave.; Los Angeles, CA, USA 90095-1555
Email: jimc at math.ucla.edu  http://www.math.ucla.edu/~jimc (q.v. for PGP key)


More information about the mythtv-users mailing list