From 1.5 to 1.6 the biggest changes are related to the addition of the plugin store. There’s very little we implemented that justifies a big change in performance.
There are however a few scenarios where I noticed some UI performance issues:
- When you turn on ‘beta plugins’ in the Plugin Store the UI at this points gets a bit overloaded with “too many objects” to handle. I consider that a corner case scenario for most users and beta plugins are off by default.
- When loading the mod-ui there are a few extra requests being issued to retrieve commercial plugin data. Depending on your internet speed or wifi signal it’s possible to notice some delays in loading content. Nothing major, but noticeable.
I have tested recently some of the things you mentioned (dragging plugins, loading PB, etc.) and in my unit I see no problems whatsoever. UI feels very snappy to me (I do have a powerful desktop running the browser, which certainly helps). The UI takes about 7 seconds to load in my setup for example.
So the bad news is: we don’t have any bug (with proper steps to reproduce) waiting to be fixed
The good news is: I’m here to help
The best way for us to get to the bottom of this is really to isolate the issue. If you can make one particular problem reproducible (by me), then I can fix it.
For all of you a few tips:
- Try going back to a clean PB and checking the CPU/memory level. Do you see a performance issue at this point? Does it still take forever to load?
- Try dropping a single plugin, what happens if you address it and change parameters? Can you also check the load on your computer?
- Is the browser giving you crazy CPU / memory spikes?
- Try changing the browser and tell me how it affects the issue.
For those of you a bit more technical I’ll ask you to use chrome devtool (F12) and capture the network requests statistics and send it to me (if you right click in the network tab you can save it as a HAR file).
I appreciate all the help you can give us to isolate those issues listed.
The “increase in xruns” hasn’t been really confirmed to be an issue with 1.6. There are so many factors involved, it’s hard to make this correlation (not that I don’t believe you, we just need to keep our minds open about the source of the problem). In order to confirm it’s a 1.6, you’d have to go back to 1.5 and do whatever steps you do to produce xruns and conclusively show that in 1.6 they happen at least 20-30% more frequently.