I use Audacity 3.7.7 daily. Today, track operations are very slow and it didn’t happen gradually. It started today.
Suddenly, MUTING or UNMUTING a track takes 5 seconds. Moving tracks around is impossible. It takes 5 seconds to drag one track above another track. Duplicating a track takes 5 seconds. Deleting a track takes 5 seconds. Selecting a track takes 5 seconds but extending that selection to multiple tracks is instantaneous.
Playback is instantaneous. Recording is instantaneous. Applying effects is also as fast as it was yesterday.
I rebooted, defragmented, emptied recycle bin, stopped all other processes, checked disk health, ran windows defender.
I know its a Windows problem but what the heck do I try next?
1 Like
Those are all things that involve writing data to the project file.
Is the project file on a local hard drive? Is it getting full? Could it be starting to fail? Fragmented?
1 Like
I checked all that. It’s local, less then half full, and not fragmented .
I did just now try an idea to rule out the drive. I copied the aup3 to my local C SSD and I encountered the same issue. I then tried another aup3 and … it’s fast. So, the problem is just this aup3.
I looked in my Audacity directory and the au3p is WAY bigger than it needs to be and WAY bigger then other aup3 files that have WAY more tracks. The “slow” aup3 has only 5 tracks while most of my other projects a have a dozen.
I did a “save as” to give it another name then I closed the aup3. The disk space now occupied by the aup3 shrunk by 75%. I opened the newly saved aup3 and … it’s slow.
So, it must be one of the tracks.
Next, I created a new aup3 project. I copied tracks one-by-one from the “slow” aup3 to the new aup3. I was able to isolate this down to two of the six problematic tracks.
So, I did an export to wav for each problematic track and subsequently imported them to the new aup3.
Problem solved.
If this happens again, we can try to isolate root cause.
1 Like
Based on what you wrote I would guess it was a problem with the blob data in SqlLite. There may be ways to optimize SqlLite that wouldn’t involve extracting the data and reloading it, but in the end, would be about as complicated.