I really do need help. I reported before that I had some crackling and
other problems. I though I had it all cracked, but one of my systems is
still playing up.
I am using a Meridian Explorer USB DAC. On my main (ubuntu) desktop
this now works fine, and I can listen using headphones or via optical
output to a mid-fi system getting really excellent results.
However, I also have a 4 * Intel-Atom Vortexbox (fedora), which runs
LMS. I'd prefer to use squeezelite on this directly so that it doesn't
matter if my main PC is up and running or not.
I have squeezelite 1.1 installed, and it runs without error according to
the debug log. However, it crackles and pops at me sometimes every few
minutes, sometimes with long breaks. I am using a new USB cable, and
have tried different USB ports to no avail. When I plug the DAC in it
reports it as a USB 2.0 (not 1.1) high-speed USB device.
Looking at CPU loads, squeezelite is < 1%, LMS is < 1% and flac is ~1%.
Total CPU averages < 10% on each CPU even when I am monitoring with
webmin, which takes up more CPU than the processes I am monitoring.
Given no diagnostic from 'output debug', what can I do? I know the DAC
is OK since it is perfect on another system. It's not the cable or
port. I have tried various -a options but I don't understand what they
mean and without a diagnostic showing an error it's hard to know whether
what I'm doing has any effect or not without listening again, sometimes
for ages, sometimes for minutes.
The output from the current test of several tracks, with several short
crackly periods is below. I had one glitch getting it running which was
the cause of the one START failure.
Code:
--------------------
[20:49:07.830156] output_thread:583 open output device: hw:CARD=DAC
[20:49:07.830760] alsa_open:269 opened device hw:CARD=DAC using format:
S32_LE sample rate: 44100 mmap: 1
[20:49:07.830875] alsa_open:317 buffer time: 20045 period count: 4 buffer
size: 884 period size: 221
[20:49:08.282078] ALSA snd_pcm_hw_start:587 SNDRV_PCM_IOCTL_START failed (-32)
[20:49:14.394675] output_flush:1411 flush output buffer
[20:49:14.687671] output_thread:773 track start sample rate: 44100
replay_gain: 0
[20:51:18.625773] output_thread:773 track start sample rate: 44100
replay_gain: 0
[20:55:26.944251] output_thread:773 track start sample rate: 44100
replay_gain: 0
[20:56:56.014382] output_thread:773 track start sample rate: 44100
replay_gain: 0
[21:01:10.547396] output_thread:773 track start sample rate: 44100
replay_gain: 0
[21:05:14.898673] output_thread:773 track start sample rate: 44100
replay_gain: 0
[21:07:02.100659] output_thread:773 track start sample rate: 44100
replay_gain: 0
[21:09:18.400950] output_thread:773 track start sample rate: 44100
replay_gain: 0
--------------------
I have played with a number of -a settings and priorities, to no
noticeable effects either way. Is there any reason to try the buffer
settings (-b) and if so where should I start? or -a options 24 bit, mmap
0 , or....?
Is there a description or explanation of how all this works anywhere at
all? I have got a copy of squeezelite.c, but I don't have enough
knowledge to work out what it all means and how to formulate a
methodical approach to solving this.
------------------------------------------------------------------------
PasTim's Profile: http://forums.slimdevices.com/member.php?userid=41642
View this thread: http://forums.slimdevices.com/showthread.php?t=97046
_______________________________________________
unix mailing list
[email protected]
http://lists.slimdevices.com/mailman/listinfo/unix
