Audacity Forum

Gale Andrews wrote:

otey wrote:In all likelihood I will submit the new imminent release (I have not been idle but am still reluctant to wrap up). I will take this deadline seriously regardless.

Just to keep things up to date, Michael has done a hack fix to potentially fix Bug 172 - see r10820. Not tested yet by anyone except Michael.

Clayton, Michael's changes in r10820 aren't working for me at all - please see Bug 172. If you still can't commit, can you suggest a viable quick fix for the pitch change reversal and the problems at > 44100 Hz until you are ready to commit?



Gale

I was working on this a couple weeks ago with the intent of finalizing libsbsms-2.0.0 -> audacity, but was abruptly interrupted when my father passed away ~Dec 23. Sorry for not notifying earlier. Should I follow mchinen and modify his audacity patch to better address the bug, or continue with the pre-christmas plan and finalize libsbsms-2.0.0 -> audacity as I've been planning for months? The bugfix would be a few lines of code and ~1 day of (probably redundant) testing. The finalization would involve ~1000 lines of code mods which I've done over the past year or so and ~2-3 days of parameter mods/testing/committing.

Clayton,

I am sorry to hear about your father - please accept my condolences.

Please see the current position in bug 172. Michael made another fix and added the patch for that to the bug. The problem with entering pitch as semitones not changing the pitch is indeed fixed but there are still two problems:

1) Tempo change is reversed (entering a negative speed change now speeds up and vice versa)

2) If you enter the pitch change in a % box the value that appears in the corresponding semitones box has the wrong sign

I assume Michael will try another fix so it's whoever gets there first and removes the bugs without introducing others. If Michael commits another fix and attaches the patch before you can commit, and his fix seems OK, that will close the bug. Then when you are ready to commit you would need to specifically test that none of the problems mentioned at the various stages of testing the bug exist in your patch, then revert all the attached bug patches in sequence before applying your own patch and committing.

I think I may have mentioned it before but the volume changes that TimeScale applies are a bit of a drawback (it modifies the volume even if you run it at zero values). This isn't just with tones - you can demonstrate it with any music by looking at the values in Effect > Amplify before and after TimeScale. Is there a longer term fix for this do you think?


Thanks


Gale

Just to note that Michael seems to have fixed Bug 172 so that bug and the related Bug 170 have been "resolved-fixed". I attached the final patch that fixed 172 to the bug report so you can download and revert those patches when you are ready to commit.

As you can see from the patch comments Michael thinks some fixes for bug 172 are hacks for sbsms bugs (such as "the pitch part of sbsms is backwards") so I assume your eventual commit will address those.


Thanks.



Gale

Thanks again for addressing this, guys. I still feel bad about the interruption to my planned commit, and it is currently a bit buried in my stack of things to do. I will reprioritize, though, such that I can get to it this weekend.

I just yesterday committed (11214) more permanent changes to sbsms time scaling
"Major update to TimeScale effect, incorporating libsbsms-2.0.0, fixes to bug 172, changes to ui"
I also committed (11215) a spectrogram narrow/broaden view menu item (ctrl-[ & ctrl-]) which borrows the mixed-radix fft code from libsbsms-2.0.0

r11214 fails to build on Ubuntu.
See here: http://forum.audacityteam.org/viewtopic ... 19&t=58380
and here: https://launchpad.net/~audacity-team/+archive/daily
and here: http://audacity.238276.n2.nabble.com/au ... 47285.html

Reverted so that Audacity builds again. This will give us time to look at Clayton's improvements in less of a rush and see if they are safe to put in 2.0 or if they have to be held back till after 2.0.



Gale