If I open the project from the UI, and use Nodes | Compute All I have the option to increase the process count and this leads to a almost linear speedup (project does repair, delete small components, remove spikes, decimate, render a few cameras, save as drc, a lot of which is understandably single threaded)
Memory wise I can run 4 frames in parallel which is great, however when I run the project using WrapCmd this option is not available to me and everything is serialised.
I have a lot of scans to process, and I spent a lot of time automating generating wrap projects so we could add Wrap to our scan processing pipeline. We’ve also bought more floating licenses to use for this. Now I have to chose between manually running hundreds of projects around the clock or explain to my boss it will take four times as long :-\
I can see the following strings embedded in the executable which are exactly what I would like to use (but obviously it doesn’t work yet):
-p/–processCount/Number of parallel processes. If greater than one, each frame is computed in a parallel process. Default is 1
Are there any plans to add/enable this functionality?