Random Xruns and dropouts when exporting

Hi. So this was happening on April (version 6.6):

I’m getting Xruns everytime I export a file. Not during the process itself (I think), since when I do it in RT doesn’t show anything until the process is done. Then I usually hear a glitch (that doesn’t appear on the exported track) and the xrun counter goes up. Until now it hasn’t done any damage to any exported file that I’m aware of, but it still worries me. Is it just a result of an overworked PC or something?

Now I moved on to the 6.7 version, and it’s giving me the following message:

error

Notice the timespan going 8 out of one. This happens now every time I export more than one range. So I tried exporting not on realtime, and it goes randomly. Sometimes it works, sometimes it shows the same message about dropouts and now, contrary to what happened in April, it is showing on the actual audio files, with random glitches and cracks and stuff like that.

So what do I do to solve that?

It’s a bug that can happen when exporting multiple timespans. There can also be some false positives when switching from/to faster-than-realtime processing. It should be fixed since 6.7.39 which you can get from: https://nightly.ardour.org/

Cool. What does it mean that these versions have no support though?

They’re built from whatever code happens to be in the git repository at that specific time of the night so there’s no guarantee that they’ll even work at all.

They’re probably just as good, or even better, than the official release but if they don’t work you’ll just have to try the next nightly build and see if that’s better.

2 Likes

Ok, so I installed the latest nightly build, and everything was still the same. Then I turned off the Buffered I/O option, and both xruns and dropouts went away (even those random xruns that generated whenever I finished exporting a file, so great). I guess that was screwing with a VSTi or something.

The timespan bug going 10 out of 2 is still there, although it only appears for when I export multiple timespans and doesn’t seem to affect the final result (no glitches at all), so I guess the problem is fixed for now.

This topic was automatically closed 91 days after the last reply. New replies are no longer allowed.