<html dir="ltr"><head></head><body style="text-align:left; direction:ltr;"><blockquote type="cite" style="margin:0 0 0 .8ex; border-left:2px #729fcf solid;padding-left:1ex"><pre>On Sat, Nov 16, 2019 at 02:47:26PM -0700, Ethan Funk wrote:</pre><pre><br></pre><blockquote type="cite" style="margin:0 0 0 .8ex; border-left:2px #729fcf solid;padding-left:1ex"><pre>1. With the low-latency kernel, jack dropouts/underruns are a problem</pre><pre>when jack is configured to bridge to pulseaudio. Without a pulseaudio</pre><pre>bridge, the dropout are nonexistent.</pre></blockquote><pre><br></pre><pre>I'd suggest to get rid of pulseaudio. It doesn't do anything useful</pre><pre>for an application like yours.</pre></blockquote><div><br></div><div>Getting rid of pulseaudio is probably the right thing to do. Thanks for pushing me in that direction.</div><div><br></div><blockquote type="cite" style="margin:0 0 0 .8ex; border-left:2px #729fcf solid;padding-left:1ex"><blockquote type="cite" style="margin:0 0 0 .8ex; border-left:2px #729fcf solid;padding-left:1ex"><pre>2. I have a similar issue with zita-a2j and zita-j2a with lost of</pre><pre>dropouts on the ALAS side, even though the jack side shows no overruns.</pre></blockquote><pre><br></pre><pre>How does this show up (i.e. how do you know it is the ALSA side) ?</pre></blockquote><div><br></div><div>It shows up as broken audio to/from the device, with no under-runs logged by jack. All the rest of the jack audio flow is unaffected.</div><div><br></div><div><blockquote type="cite" style="margin:0 0 0 .8ex; border-left:2px #729fcf solid;padding-left:1ex"><pre>What do you get with the -v option ?</pre></blockquote><div><pre><br></pre><div>I don't know because I am using Ubuntu Studio Control to manage the create of zita-j2a/a2j instances on my test system. I wanted an "easy" way for potential users to set up jack and Ubuntu Studio Control is nice and easy. zita bridges are working fine on my development machine, where I manually set them up, so there must be something in how Ubuntu Studio Control is spawning the zita bridges. That's why I am also looking for the source code for Ubuntu Studio Control. Ubuntu Studio Control make jack setup supper simple, and almost a pleasure for the average user. </div></div></div><div><br></div><blockquote type="cite" style="margin:0 0 0 .8ex; border-left:2px #729fcf solid;padding-left:1ex"><blockquote type="cite" style="margin:0 0 0 .8ex; border-left:2px #729fcf solid;padding-left:1ex"><pre>Can extra buffering be applied to the zita programs? Maybe that is</pre><pre>what the -n option is for?</pre></blockquote><pre><br></pre><pre>Using more periods could hide the problem, but it's never a real</pre><pre>solution.</pre><pre><br></pre><pre>Are you sure things are running with real-time scheduling ? </pre></blockquote><div><br></div><div>Yes.</div><div>ps -eLfc shows FF for for the scheduling policy of at least one of the threads in the processes of interest.</div><div><br></div><blockquote type="cite" style="margin:0 0 0 .8ex; border-left:2px #729fcf solid;padding-left:1ex"><pre>This requires some configuration in /etc/security/limits.conf,</pre><pre>as well as for Jack itself.</pre></blockquote><pre><br></pre><div>I had gone down that road originally on the machine I am using for development. On my test system, I used the Ubuntu Studio Installer, which does a very nice, friendly job of setting up a stock Ubuntu machine for use with jack.</div><pre><br></pre><blockquote type="cite" style="margin:0 0 0 .8ex; border-left:2px #729fcf solid;padding-left:1ex"><pre>Why do you need zita-a2j/j2a anyway ? Using a single multichannel</pre><pre>card is usually the better solution.</pre></blockquote><pre><br></pre><div>I have one multichannel audio interface for everything important: program out, studio monitors, headphones, guest mic, host mic, etc. But it sure is nice to be able to use the built-in audio for a cue channel and talkback mic, where latency is not important. Also handy for USB turntables, and other random devices that are occasionally used in a radio show without latency being important.</div><div><br></div><div>Thanks,</div><div>Ethan...</div><div><br></div></body></html>