X here everyone:
So I’ve been working the last few weeks on version 2 of my rig rebuild project as I’ve been migrating from GP3 on Windows to GP4 on Mac as well as very significant hardware changes in my rig. Sometimes, you get your mind stuck on an idea and you just keep chasing down that path until one day you realize… “I think I took the wrong road”.
My original plan was to put an instance of Kontakt into the global rackspace, and then route midi from local rackspaces to the global rackspace by implementing virtual midi ports on the Mac via the IAC driver. Now, I can tell you that this part of the plan worked. Once I figured it out, it worked really well! I was so pleased with myself and I created a number of other constructs around this to further the agenda.
- Since I only use this global instance of Kontakt in approximately 45% of my rackspaces, I decided to add a widget button to the global rackspace to disable the Kontakt instance, and then created a widget in my local rackspaces to send the ‘disable’ or ‘enable’ as appropriate to the global widget.
- Loaded up about 10 instruments (each on its own channel) to basically stay ‘memory resident’ through an entire gig as ‘ready to go’ as necessary
- Did all sorts of weird routing through a mixer in the global rackspace to (to local rackspace) and then sent all of those channels through to the local rackspace for final mixing
This increased the general coupling between local rackspaces and the global rackspace significantly. The other side effect I could see immediately was that the amount of channels I was typically routing between local and global was extremely high. It just looks ugly in the wiring view in general.
So as all of these design ‘smells’ were stewing, I read the following post which just added to my discomfort regarding my design choice.
Send “All Notes Off” On First Note-Off After Rackspace Change
So after reading multiple comments, I realized that the original author was trying to essentially do the same thing I was. Then I read so many comments that made me realize the folly of my approach. I will be running back to the drawing board again, this time using a predictive loading approach to solve the problems I was trying to solve with loading 10 or so instruments into a globally available instance of Kontakt.
X