Well, it seems happening quite randomly… it does not happen always and on the same file (if I remember good). It’s quite hard to reproduce it because it happens once maybe or twice, then it works fine for hours and then it can happen again…
OK obviously something going a miss here. Some questions:
What kind of audio file - wave, mp3 or flac?
When it goes mute you say it continues playing but do wondering if the level meters are still lighting up on the media player slot?
Do VSTi’s continue to sound out? eg: is it just the media player or everything?
What’s required to get it working again? If you stop it and then hit play again does the sound resume? What if you switch files in the media player and/or stop/restart the engine. (not suggesting as a solution - just after clues where the problem might be).
I’ll do a code review on the audio player code base and perhaps setup a load test for it.
OK thanks. And to be clear: nothing else of significance relating to the media player was happening when playback stopped? Also, the file wasn’t on removable media was it?
If it happens again before I get this sorted, please check point 2 above. Very interested to know about this.
This used to happen to me. The meters would still be responding but the wav files would stop playing half way through a song. It happened randomly but some times more than others.
I turned on the media file display to try to see if it always happened at the same point in the file and it never happened again.
The meters do keep running! Jeez I didn’t expect that.
Well that eliminates a whole bunch of places where this could be failing so that’s good I guess. I’ve been testing here but haven’t been able to reproduce it.
I’ll setup another test now and run it over night and see what happens.
I’ve had your audio file on loop for over 48-hours now in a debug build of Cantabile and it’s still playing fine. I’m going to switch over to the release build now and test again.
A couple more points:
If this happens again for you could you please also check the media player’s audio output (send) route and see if it’s level meters are lighting up
Also check that neither the media player nor its routes gain settings have changed
Also try adjust the gain setting on both the media player and it’s sending audio route to see if that wakes it up again (even if they haven’t moved, move them anyway and see if it makes a difference)
If you can, grab me a screen shot of the media player, its output routes and the master output levels.
(None of this is intended as a solution - this is for diagnosis to try and figure out what’s going on here)
No problems here running it over night again. I’m a little confused over exactly when this problem happens - sounds like it just happened randomly in the middle of playing which is why I set the long loop mode test in the hopes of reproducing it.
The fact that it plays past the end when in the mode and the fact that the level meters are still lighting up doesn’t make much sense to me - perhaps there’s two issues here but I’ve been unable to reproduce either.
Do you have anything else going on in that song file that might be impacting it? Eg: bindings to the media player, bindings to the media players routes etc…?
Hi @brad, I understand. Well, in the last 4-5 days this issue is not happening. However, I’ve some bindings in the background rack to change program and control the expression pedal. Nothing related to the media player.
Update: I received another report of this yesterday but this time it was reported that the audio player’s level meters stopped lighting up (which seems contrary to what was described here).
So I’ve been looking at this again from this new angle and after running some more experiments and I think I’ve found a way to simulate the problem by deliberately stalling the audio reader thread in a debug build. The symptoms seem the same as what’s been described:
Audio goes silent
Level meters stop lighting up
Play position continues to move
Play doesn’t stop at the end of the file
For this to happen in a release build the audio reader thread would need to be stalled for quite a while (but obviously that’s what’s happening).
I need to think about how to best fix this but will try to get something out asap…