Set messaging has bad math somewhere #61
Labels
Comments
|
Ok, tested it all with 1/2/4 and the same problem is there (the overdub/display problem). (I also adjusted the code example above to make it easier to test with 1/2/4 channel versions) |
|
a rather confusing collection of bugs, fixed in commit 10e4b62. |
|
Man, fantastic catch there! (this is in the current help patch too, which I guess would never really manifest given how things are setup) |
|
oops! |
|
Actually yeah, or rather, that was how it functioned up to this point. I guess there could be more granular outputs when the list outlet becomes modular, but knowing the lowest and highest point in the buffer being used is handy for the purposes of [waveform~]. |
|
|
ultimately, in karma~ v2, we'll have modular outputs for the list, and this will include the option of outputting loop size directly, rather than having to make it in max. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Summary:
When you use a
setmessage, it appears to be either sending the wrong start/end points out the list outlet, or the pointer is wrong, or both.Steps to reproduce:
Send
recordto create a long (>2000ms) silent loop. Sendstopto definite the initial loop. Sendset whatever 500. 1000.to define just a subset of that loop. Now start audio playback into [karma~], sendplayto [karma~] (so it starts playing back the silent loop), then sendrecordto enter an overdub state.Actual results:
You see that the playhead (from the list outlet) and what is being written into the buffer phase away from each other and that what has been written into the buffer doesn't fill the entire area being shown in [waveform~] (as defined by the start/end points in the list outlet).
Expected results:
The playhead should match the visible record head, and the material that is recorded should cover the whole [waveform~] display.
Comments:
Playback itself doesn't have a silent gap in it (like [waveform~] would lead one to believe) so this must mean that the list outlet start/end points are wrong.
This may have to do with the issue I mention in #57, or be like #57 in general.
(I thought there might have been a different issue going on with jumping as well, but it appears it's actually the same problem. It may be worth double checking in the code that the boundaries defined by
setmessages clamp thejumpmessages.)Example code:
The text was updated successfully, but these errors were encountered: