Romain Beauxis points out a possible link with this bit of code:
> See for instance output/ao_out.ml, line 86:
>   self#log#f 4 "Reader not ready!"

This bit of code works as follows: when liquidsoap wants to send data to AO,
if there is already too much data that AO didn't play (read) yet, liquidsoap
simply drops some samples.

We use a similar device for jack and other outputs. I don't think that this
problem happens much for me when I use the AO output, because the buffer
is big enough.

The question concerning the initial bug report is: do you see any such
message in your logs when you hear those pops?

We are planning to build a more modular buffering framework that should
allow more control on which external interface (jack/alsa/ao..) is buffered,
which forces the clock, etc. That should help with this kind of issue.
--
David

------------------------------------------------------------------------------
This SF.net email is sponsored by:
High Quality Requirements in a Collaborative Environment.
Download a free trial of Rational Requirements Composer Now!
http://p.sf.net/sfu/www-ibm-com
_______________________________________________
Savonet-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/savonet-users

Reply via email to