Es schrieb hermann pitton: > > Es schrieb Tim Connors: > > [snip] > > > In 0.2.8, the timing seems worse, so that muting upon channel changing > > happens after the channel has already changed, and the sounds has alreday > > gone through its stabalisation period. Rather a waste muting it after, > > because then I just have to put up with the ~1 second wait before I get > > sound again. > > > > Seems that is the small price we pay for now to have get language > switching :) > With the above snapshot on my card (md9717) this is well adjusted again. > > Incidentally, a couple of days ago, I found out about the oss=1 dsp_nr=1 > > option, but haven't gotten this to work. I get the video volume slider in > > mixers, but it does nothing, and if I unplug the loopback cable, I get > > nothing - it is all still coming though line-in only. > > > > It looks like the other mixers "steel" the capture flag from the saa7134 > mixer. > Closing all mixers and unloading all sound modules and putting oss=1 > dsp_nr=0 mixer_nr=0 oos_debug=1 to the saa7134 will give a first proof, > that this is not a phantom in userland. (Taking also care of > autoreloaded modules.) > > With audacity at 32000 Hz (48000 is not ready for our chip), setting the > volume with v4lctl to 90% after "volume mute off", I did nice recordings > with definitly no audio cables connected, but I'm not finally through, > how to set this for daily use, trying on the 2.6.0-test1-ac2 with the > kraxelpatch for test1 ;) > The saa7134 oss dsp and mixer now peacefully coexists with my two other sound devices and mixers. With nvrec (divx4rec) I directly record from the saa7134 adjusting the values something like "divx4rec -D -vb 1800 -ab 128 -d /dev/dsp2 -r 32000 -s -o pcisound-out.avi". The file can be played back during recording with xine (timeshift 1 second, or what I have free on disk delay) on a audio device of my choice and all over the net with nfs mounts. So that's now about daily use :) Hermann -- p.s. The saa7134 mixer doesn't show up in kmix, but in gnome-volume-control and "xmixer -device /dev/mixerX" on RH-9 with 2.4.20-v4l2.
