Forums
Hi Rui, if you got no time for https://github.com/rncbc/qtractor/issues/510 then never mind.
I just want to make sure that this issue doesn't go unnoticed.
Hi Rui, if you got no time for https://github.com/rncbc/qtractor/issues/510 then never mind.
I just want to make sure that this issue doesn't go unnoticed.
re. Crash with WSOLA...
confirmed. somewhat historically it only got tested on stereo at most (long ago soundtouch legacy you know?)
good catch! anyway now I'm on summer holidays and that might be just the right time to deal with it ;)
cheers
UPDATE: Hopefully fixed in qtractor >= 1.5.7.2git.36b58c, please test && tell.
Summer holidays? Then I have 2 more issues for you :)
Missing transients with rubberband transposing and an off by one matrix routing error for plugins.
I hope it's not too difficult to repair.
re. too difficult to repair...
of course it is
even more so if no detail or suggestion to reproduce are given ;)
cheers
See GitHub issues
Sorry, I forgot to mention that there are two (hopefully) proper issues on GitHub with sample sessions to reproduce them: 513 and 514
#514
I had to close #514 because today I couldn't reproduce the underlying problem with CALF Vocoder and CALF Sidechain Compressor anymore.
But the rubberband issue #513 is still reproducable.
Another Save/Reload thing
#514: Creating a 4ch bus and some 2ch buses and making AUX sends from the 2ch buses to the 4ch bus doesn't always honor the matrix when the routing is different from 1->1, 2->2. Saving and reloading the session corrects it.
Maybe it needs more time to stabilize?
Try compiling with a time of a quarter of a second (250 ms): stabilize(250)
https://github.com/rncbc/qtractor/commit/aef4bd515
From what I've read in other threads, we're working with computers that are over ten years old, and Rui's on a really powerful machine. Maybe our machines need more time.
If that fixes it, a quarter of a second is a negligible amount of time for a UI interaction.
It should simply be increased in the master version of Qtractor.
Will it fix it?
bad luck
Setting
static void stabilize(int msecs = 200);
in src/qtractorSession.h didn't solve the problem but led to some strange queuing of actions. So I use 20 again.
It was attempted :)
I meant to add it to the stabilize that Rui created to fix the previous error.
https://github.com/rncbc/qtractor/commit/aef4bd515
When this happens, are the modified channels displayed correctly in the connections window?
My guess is that it might be the same problem, or equivalent.
Although if changing the global stabilize value didn't fix it, changing the local value most likely won't either.
Confirmed, it's a different error
In my tests:
Newly created auxiliary sends ignore the matrix. They send to the destination by default. 1>1, 2>2, etc.
The matrix doesn't work correctly until the session is restarted.
This only happens with newly created auxiliary sends. Existing ones work correctly when opening a saved session.
Did this happen before? I don't think so. We would have noticed in tests. Since when? It's a complete mystery.
re. different error...
maybe fixed in v1.5.7.3git.cde637
check it out!
thanks
it's fixed
Thanks Rui. Issue can be closed.
Bus creation
I will do some more tests because IIRC then it's the creation of the new bus that has to be saved/restored, not the new AUX send.
Bus and AUX Send creation
Creating the bus and save/restore is not enough. Also the AUX sends have to be created before save/restore.
Is the matrix
Yes, in this case the error seems to be in the matrix, not the buses.
It's as if Qtractor doesn't recognize the new matrix's existence until the session is restarted.
I've confirmed that when the estabilize bug was fixed, this error was already present.
I only tested the channel modification because I assumed the rest were still working correctly.
The error wasn't there originally. It must have been some modification to correct the display of multichannel audio files, or perhaps a failed merge... who knows.
Add new comment