On Thu, 15 Nov 2007, gnittor@xxxxxxxxxxx wrote: > > hello, > What is the best way to use this card, for single application at a time operation, without resampling? I am interested in best quality as I want to only use this card for music. software mixing is not important (i plan to use another soundcard as the default). Sampling rate should depend on the audio source. > Use any card at the highest sampling rate it natively supports. Do NOT use resampling. While there are certainly resampling techniques which work well, (see my analysis of the ones used by sox at http://www.theory.physics.ubc.ca/soundcard/resample.html I would not trust the sound card manufacturers to have gotten it right, especially as they have to do it really really fast, and thus probably use linear interpolation-- a terrible technique. I do not know that particular sound card the MAudion card I have-- the cheap Transit, really works very well at its native rates. But it has variable hardware rates, and I would expenct the Revolution to have that as well. On the other hand if you are going to be using it for CD work, sample at either 44100 or 88200 Do not bother trying to go to 96000 and then software resampling. > MAudio Revolution 7.1 > linux 2.6.23-rt1 #1 PREEMPT RT > alsa 1.0.14 > debian unstable > > what i've done so far: > > - using it as default seems to work ok in every app (sw mixing ok) but plays at 48khz, while 44.1khz audio sounds slower than original > cat /proc/asound/card0/pcm0p/sub0/hw_params > access: MMAP_INTERLEAVED > format: S32_LE > subformat: STD > channels: 2 > rate: 48000 (48000/1) > period_size: 1024 > buffer_size: 16384 > tick_time: 1000 > > > - using hw:0,0 does not seem to work in any application. i get device busy. (also for hw:0,1 and hw:0,2) > > > - using plughw:Revolution71 in audacious works fine, but the value of period_size looks odd > cat /proc/asound/card0/pcm0p/sub0/hw_params > access: MMAP_INTERLEAVED > format: S32_LE > subformat: STD > channels: 2 > rate: 44100 (44100/1) > period_size: 4412 > buffer_size: 22052 > tick_time: 1000 > > > - using plughw:Revolution71 in xine produces crackles (stops every 1-2 sec for a little) Sounds like your system is not keeping up and you are getting buffer overflows. > cat /proc/asound/card0/pcm0p/sub0/hw_params > access: MMAP_INTERLEAVED > format: S32_LE > subformat: STD > channels: 2 > rate: 44100 (44100/1) > period_size: 4096 > buffer_size: 32768 > tick_time: 1000 > > > - using xine & jack (at 44100) produces crackles (no matter what settings for frames/period & periods/buffer i've tried) > cat /proc/asound/card0/pcm0p/sub0/hw_params > access: MMAP_INTERLEAVED > format: S32_LE > subformat: STD > channels: 8 > rate: 44100 (44100/1) > period_size: 16 > buffer_size: 96 > tick_time: 1000 > > > - using xine & jack (at 48khz) no crackles but sounds slow (using either audacious or xine) You are recording at 48 but playing back at 44100 Yes, that will "sound slow" and about a tone flat. > cat /proc/asound/card0/pcm0p/sub0/hw_params > access: MMAP_INTERLEAVED > format: S32_LE > subformat: STD > channels: 8 > rate: 44100 (44100/1) > period_size: 16 > buffer_size: 96 > tick_time: 1000 > > > > The problem is with 44.1 audio and xine. It seems that it is played without crackles only with the card in 48khz but sounds slower. > You are playing back at the wrong speed. tell xine to play back at 48K. > > I have tried setting period_size and buffer_size in asoundrc (as this is the only difference I could find between xine and audacious playng using plughw:Revolution71) but with no success. > Also the only way I could get a virtual device to work in asoundrc was using plughw as such: > > pcm.rev { > type plug > slave { > pcm "plughw:Revolution71" > format S32_LE > } > } > > > Any help appreciated > > > ------------------------------------------------------------------------- This SF.net email is sponsored by: Splunk Inc. Still grepping through log files to find problems? Stop. Now Search log events and configuration files using AJAX and a browser. Download your FREE copy of Splunk now >> http://get.splunk.com/ _______________________________________________ Alsa-user mailing list Alsa-user@xxxxxxxxxxxxxxxxxxxxx https://lists.sourceforge.net/lists/listinfo/alsa-user
