At Wed, 9 Aug 2006 12:08:00 +0200, Markus Rechberger wrote: > > Hi, > > On 8/8/06, Takashi Iwai <tiwai@xxxxxxx> wrote: > > At Tue, 8 Aug 2006 03:09:43 +0200, > > Markus Rechberger wrote: > > > > > > Ok I managed to get rid of that garbled background noise now I get > > > underrun messages... > > > > > > $ aplay -D hw:2,1 5_5_2006__07_00_59.wav > > > Playing WAVE '5_5_2006__07_00_59.wav' : Signed 16 bit Little Endian, > > > Rate 9600 Hz, Mono > > > underrun!!! (at least 0,271 ms long) > > > underrun!!! (at least 0,248 ms long) > > > [...] > > > > > > I can hear the sound clearly, but looks like these 200ms cause a choppy > > sound... > > > I also tried to add support for this device to snd-usb-audio, but > > > since this device has no synchronisation packets the driver fails... > > > Does anyone know what causes such a buffer underrun? > > > > This is fixed now I set the buffer size 3 times as high as the period > size.. though I still have the same choppy effect. Seems like a > problem with the filled USB packets.. I wonder why the supported > packet size is 128 bytes but the windows driver fills max. 38 bytes of > every packet... My driver currently fills only 20 bytes since this is > the average of what the windriver does too. Hm, this sounds like a 1ms alignment problem. What happens if you play a sample with n * 1000Hz ? > > For example, the calculation of the current pointer doesn't match with > > the actual position. > > > > there might be something wrong in line 350/351 but it has a lower > priority at the moment for me... > > > > I think it still has something to do with the snd_pcm_hardware_t values... > > > > My guess is that the problem is the implementation of pointer > > callback. > > > > > maybe the time will come when I'm not the only one on this ML who > > > answers my questions.... > > > > Maybe you'll get more response if you show your codes more openly... > > Otherwise it's just a wild guess. > > > > there we go: > http://linuxtv.org/hg/~mrechberger/v4l-dvb?mf=598d5b40d82a;path=/v4l_experimental/usbaudio/;style=gitweb > > the projectwebsite: > http://alsa.opensrc.org/au600 > > the driver was always available uppon request... thought it would be > nice to host it in an alsa development tree or so but I haven't got > any answer regarding this a while ago.. I put it into my video4linux > development tree now.. We can put it to alsa-driver tree and later push it to upstream when the code is well brushed up. So, could you post the code, or as a patch, to alsa-devel ML? The post is easier for reviewing and correcting some issues. But, please fix the following for the next version: - Don't use snd_xxx_t but use struct snd_xxx. - Follow the coding style described in Documentation/CodingStyle. For example, the opening brace of a function should be in the next line. Also, take care about spaces. - We have an easier procfs helper functions. - printk() should be either a debug macro (that can be disabled easily) or persistent (for proper information or error messages). Takashi ------------------------------------------------------------------------- Using Tomcat but need to do more? Need to support web services, security? Get stuff done quickly with pre-integrated technology to make your job easier Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642 _______________________________________________ Alsa-devel mailing list Alsa-devel@xxxxxxxxxxxxxxxxxxxxx https://lists.sourceforge.net/lists/listinfo/alsa-devel
