For my sessions and compositions, I used SWAM Instruments to create a large horn section in Digital Performer, which is externally controlled by Divisimate. I utilized a DP template provided by Divisimate, and it works perfectly on my MacBook Pro M2 Max with 64 GB of memory. Control is assigned in Divisimate using two MIDI controllers and a TEC Breath Controller, and Digital Performer handles it without any issues.
Encouraged by this success, I created a single rackspace in a new GIG file. In this rackspace, I duplicated the MIDI inputs as set up in Digital Performer and added thirteen instances of SWAM Instruments, outputting without any plug-in effects into a 16-channel GIG Performer. While it performs the Divisimate MIDI assignments perfectly, the CPU spikes up to 90% immediately, resulting in sound crackling and dropouts. I cannot use the Auto Sampler patch as different horn configurations and panning are controlled by Divisimate. Examples, Trumpet and Trombones with mutes on and off, instruments on and off, solo Sax etc.
Do you have any suggestions on how to match DPs’ performance with Gig Performer?
Using multiple instances of Gig Performer, you can thereby distribute SWAM sections over multiple cores of your CPU. For instance
Instance 1 - Saxophones
Instance 2 - Trumpets and Trombones.
Keep in mind that Digital Performer in DAW mode will use multiple cores because it doesn’t have to process real-time audio. When you switch to Live Performance mode, I guarantee you would see the same kind of CPU spikes as you experienced in GP. They even note this in their manual, to watch the real-time meter in the Audio Performance window when this mode is enabled.
Multi-instances can be easily created, primarily because I used this horn section extensively. Thanks for the tip. To do a test, I switched my DP to Live Performance mode, which is what I would have used before using Gig Performer, to see if there was a CPU spike, which there was; however, there was no audio loss, dropouts etc. Thanks all for speedy responses!
You could also try Bluecat Patchwork. There’s a demo version that should let you pretty easily test out how that solution stacks up to your other options.
Patchwork is a plugin container. You could put, for example, 8 instances of SWAM in parallel inside of Patchwork and it will distribute them across different processor cores.