Hi,
sorry for the (probably too) late response.
On Wed, 6 Dec 2006, Paul Knowles wrote:
>>> first thing that comes to my mind is a glitch in JACK
>>> transport support. There're some messages about seeking in
>>> the middle of recording, and this shouldn't be
>>> happening. Could you retry the tests with adding the option
>>> "-G:jack,eca,notransport"...?
>>
>> This seems to have done the trick. At least I have now 12
>> hours of uninterrupted recording. The analysis for phase
>> jumps (lost samples) is currently underway.
>
> Nope, spoke too soon. After analysis, the data loss is
> still there, It seems to be only one sample at most, not a
> whole buffer. With the notransport option, ecasound just
> doesn't tell me about it.
You could try recording directly with ecasound+ALSA, i.e. without JACK. It
shouldn't make any difference, but you'd have one less moving part in the
puzzle. Missing a single sample does suggest a hardware/driver level
problem though.
> I did isolate a problem with the word clock: while in use,
> occassionally the sound cards would drop the PLL coherence
> momentarily. I could not correlate that to the loss of
> data, but when the PLL problem was made worse, there was no
> increase in (supposedly lost) sample corruption, so I don't
> think that the word clock is the problem there.
>
> Any hints on how to make ecasound exit when this seek event
> occurs?
I'm afraid this is a similar hw/driver level issue that is not
escalated/signaled to application layer (i.e. to apps using
alsa-lib/libasound). Recording directly from ALSA, and with '-z:xruns'
ecasound option, might be worth a try though.
-- links, my public keys, etc at http://eca.cx/kv ------------------------------------------------------------------------- Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT & business topics through brief surveys-and earn cash http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV _______________________________________________ Ecasound-list mailing list Ecasound-list@email-addr-hidden https://lists.sourceforge.net/lists/listinfo/ecasound-listReceived on Mon Mar 5 00:15:03 2007
This archive was generated by hypermail 2.1.8 : Mon Mar 05 2007 - 00:15:04 EET