I was using Audacity 2.0.3 for some time with no problems. More than a year ago, when an update was available, I upgraded to a newer version, but I noticed static and audio destruction happened when I recorded in the new version. I downgraded back to Audacity 2.0.3 and the problem went away.
This past month, I upgraded to 2.1.0. It took a while before I started to realize that occasional and unpredictable static (in the form of little pops like what you might hear on a vinyl record) had been developing here and there in my recordings. Also, occasionally I’d get audio corruption (for example, a few words would be destroyed by a high-pitch sound, and the waveform would reflect destruction).
Today, I upgraded to 2.1.1 to see if that would fix the static and audio corruption. After a few tests with a different microphone, I got the same issues.
So, I just downgraded back to 2.0.3. After some tests, the problem went away and I wasn’t able to trigger static or the audio corruption.
I am using Mac OSX 10.10.5, with 2.2 GHz Intel Core i7, and 4 GB 1333 MHz DDR3.
I searched the forums, and I wasn’t able to locate any advice on the issue. I’m generally happy with 2.0.3, though I long a little bit to have a more recent update for Audacity. Any input or thoughts on my issue are much appreciated.
In Audacity 2.1.1, reducing the “Audio to buffer” setting in “Audacity menu > Preferences > Recording”. The default setting is 100 ms. A smaller setting may solve the “static” problem, so try decreasing it a little at a time and test for a “sweet spot” setting (too small and the problem will probably be worse).
I’m a bit surprised that it works set to 0, but if it is, then no need to change it. The default setting is 100. Modern Mac computers seem to like values less than 100. If the problem reoccurs, try tweaking that setting.
These cases are often reported to record on 0 milliseconds, probably because it is relative to a misreported PortAudio input latency that is higher than it really is. Some users cannot set it higher than 10 milliseconds, so my suggestion is to set it to 0 and work up from there.