I've certainly run the reverb version down to a DSPBUFFER size of 128 samples, but the CPU overhead I measured was double that at 512, so I stuck with that so I could run twice as many instances of Nebula
This is something i changed in xml , but is it really changing the value? , sounded like it was having a effect but of course it does not show on nebula when starting up. If you could remove this limitation it would be great .
Could you make a mod for me even down to 32 samples?
I just want to know why minimizing latency is so important.... Do you really need to track through the Nebula instance? I find it more productive to deal with tracking first (at the lower latency), and use Nebula in the mixing process, when I raise my buffers anyway. I am not questioning the logic of anyone's work-flow, I am merely trying to learn more and gain a better understanding. Is it really that critical to be able to use Nebula at the tracking phase of production?
I understand the function of this, I just don't understand why not use a CPU lighter effect at low latency, and replace it with the Nebula instance at mixing and at higher buffer settings. It is so easy and allows a person to keep the momentum and inspiration. But of course, this is just my opinion and what works very well for me. Other people have their own needs, I am merely offering a different solution that requires no new hardware, no new settings, and no "optimizations".