Hi Dietz,
Here is an update on my system's performance with MIR Pro. I ran tests at various latency combinations. and here is what I feel is the best my system can offer :
I am able to get a maximum of 16 instruments (VI-Pro2) into the MIR Pro stage, basically each instrument's stereo output is going to Cubase 7, where I have a MIR Pro inserted into each of the 16 audio channels. Cubase 7 is set to a latency buffer of 256 samples, MIR Pro is set to a latency of 1024 samples. more channels of MIR Pro would cause pops/clicks to appear due to CPU overload. I could incerase Cubase 7 (I'm using RME UFX ASIO audio interface) latency to 512 samples, but playing in real time does not feel very comfortable with latency set at 512.
I'm guessing that I can work with a max. of 16 MIR channels, since I might be able to group some instruments into the same channel. (i.e. all Hi-Brass can be assigned to the same MIR Pro output channel, All Low-Brass assigned to another channel, ...etc. etc. for other groups of instruments).
I can most likely have more MIR Pro channels if I increase both MIR Pro and Cubase 7 Buffers, but playing in real time at these high-buffers is not feasible. Which means I could use these high-settings to use MIR Pro in a Post-Production scenario. Where I would assign, and edit the instruments on the stage after having recorded the performances into my DAW. But, I'm not sure I need to go this far. Since I feel 16 audio channels of MIR Pro could be sufficient to get a good spatial placement of the various groups of instruments, I can then apply an algorithmic reverb, or VSL's Hybrid Reverb to polish the overall mix. (By the way, I love the VSL Suite's EQ, it is one of my favorite EQs).
Do you think I will be able to load more instruments if I was using MIRx , I will be running each MIRx plug-in on each of the two slave-PCs and not in Cubase 7 (given that MIRx works withing VI-Pro2) as a plug-in, I would guess that the two PCs running MIRx would be able to handle this task more efficiently, since I will be distributing the processing power required by on the two Slave-PC systems ? What are your thoughts on this detail ?
Please let me know if there is anything I'm missing here, or if you have any additional tips to further improve the results I posted.
Thanks & All The Best,
Muziksculp