I haven't ever tried to sample a compressor with NAT, and so I don't know how the process goes, but I had a couple of thoughts about it. I doubt I am offering any new ideas here but I'm just kind of curious about it. I know that when sampling preamps, the sweep tone is repeated whatever number of times at fixed intervals of db levels. So maybe each one is 1db lower, to capture the dynamic behavior of the hardware. I was wondering though, wouldn't it be better to have a higher density of samples at the 'knee' (in the case of the compressor) where the change starts to occur? Since there will be more of a rapid change there?
So in other words, if I had a compressor with a threshold setting of -14db, and a ratio of 6:1, and a hard knee, then maybe 1 or even 2db steps would be fine from -30db up to -16 or so db (since there would be little change anyway), up to when the gain response starts curving. But then maybe have steps more tightly bunched, like every .5db, to more accurately represent the curve?
I don't know, maybe thats how it already works or maybe it wouldn't be good to do it that way. I guess it would obviously require NAT to be more complicated, and Nebula for its reading of the programs. I guess you would have to run an extra tone out to NAT with a sweep in volume gain to map the curve first, before the sampling at different levels, to see where to put more of the samples gain-wise. Then you'd have a program where each different threshold setting would have a different non-linear pattern of dynamic samples.
Really I'm just kind of thinking out loud because I think its interesting and I know there are difficulties with compressors, and I was just wondering if it already works like that.
Tips & tricks, working results, technical support
2 posts • Page 1 of 1