Since upgrading, the day of a gig  to both MainStage 3.5 and Catalina, I find that when I use MS's native Record function, I now have many audio drop-outs in the recording of complex Concerts where previously I had none. Same Concerts, same computer, same lack of external CPU load. The interesting thing is that if instead of using the native Record function I use AudioHijack to grab and save the output, it has no dropouts whatsoever.
To me this points the finger squarely at MainStage. If it were an I/O issue it would affect Audio Hijack too. If it were a CPU issue ditto. It seems prima facie to me that MainStage has gotten less efficient at writing to disk for no apparent reason. Is anyone else seeing this or anything similar?
iMac (Retina 5K, 27-inch, 2017), 48 GB RAM, i5, 1TB SSD MainStage 3.5 Max CPU 160% (100% == 1 core, and the i5 has 4) Apple and Arturia AUs.
Woah yes! I was just coming to this forum to post this exact same issue in 3.5... Small recording drop outs in the internal recording which are not present in Audio Hijack recordings. My setup is quite complex, but I run at about 40% CPU and haven't had this problem before.
The gaps can been seen in the attached waveform.
I submitted a bug report to Apple
You do not have the required permissions to view the files attached to this post.
I believe i saved my patch that was created in 3.4 in 3.5, then was able to open it in 3.4. But not 100%. Sampler doesn’t exist in 3.4 that’s for sure. Although all my esx24 instruments that opened as Sampler instruments in 3.5 still worked back as esx24 instruments in 3.4.