Latency Issue in Protools With Multiple Instances of VEPro/MIR

[strytten]

I am working with a template of roughly 25 instances of VIPro and Kontakt, all loaded into one instance of VEPro/MIR Server (decoupled). This has run consistently in Protools 10.3.5 without any latency issues.

Today I decided to split 6 instances of VIPro into a separate instance of VEPro to put these 6 instruments into a different MIR room, so I am now running two instances (both decoupled). Everything went smoothly and sounds fine, except that I now have a latency issue. The original template is still in correct time with the audio in Protools, but the new instance of VEPro that holds the 6 instruments is ahead (maybe 1/16th note ahead, tempo about 108).

Protools 10.3.5 is set to maximum latency compensation. I am on 12 core Mac with lots of memory and Activity Manager doesn't show strain on either CPU or memory. VEPro has plug-in latency compensation turned on.

First question: Any one else experincing this? Should I be checking any other settings?

Second question: I've always been fuzzy as to exactly what a project or a metaframe is. When I created the separate instance of VEPro, I went to the smaller window that shows the name of the one instance that is open (bottom right corner has "preferences" and "exit"). I clicked the box to "add" an instance, and a new mixing board came up. Does that mean I've opened a second "project"? Or is the whole thing a "project"? And what is a "metaframe"? Can I save this whole thing in one command - for example, if I go to that small window and save a new metaframe file have I saved everything? Or do I need to go into each of the two instances and save them separately as projects? I will be most grateful to be straightened out on these points.

Thanks!

Steve