a)well, yes 2 versions of the same patch can exist like you suggest. but your idea of having a button switch between them would be problematic because a patch saves the state of all of the controls too, so your settings would be erased, just like when you load a patch.RJHollins wrote:
No one has outright said that 2 versions of a particular
patch can exist [simply by renaming the patch name] so
that a WORKING copy, and a FINAL RENDER copy could
be selected. [the CPU hit seems a major deterrent].
2nd ... The mod being referenced to a 'clean' kernel
ONLY is something that raises the eyebrows
If only a single, clean kernel is being used ... does
this not fly totally against the efforts of NEBULA's
initial strength ... multi-level convolution ???
b)the clean kernel is going to be the most important by far. it's the main body of what you hear. but, YES, the harmonics are also part of what makes nebula what it is. it's just one of those diminishing returns kind of things. switching clean kernel would make a much more noticeable difference compared to switching the others. still, you can switch them if you want, just keep in mind what people have said about artifacts, and realize you're taxing your cpu much more for what might not even be noticeable at that point. especially with a high end console program, for example, where those kernels are so low, due to those more expensive consoles being so expensive partly because of their low distortion levels. If it's debatable how much of an improvement switching to timed makes with the clean harmonic, think about what difference it makes with h2-h10 if they are 50+db below clean harmonic and masked by it.
well, for most things besides compressors, 50ms for clean kernel is the default that NAT 'renders' stuff to. i'm sure giancarlo/acustica did some research/testing to come to that figure. it probably had a little to do with balancing the program rate (getting it as fast as possible), but also having a fair length to the kernel to impart the tone of the impulses onto the signal. longer kernels mean slower prog rate, so the two would have to be weighed out for a figure to be decided on. i noticed one thing myself not too long ago when looking at some impulses, a 'low frequency' cyclic part of the impulse lasted almost exactly 50ms. it was really prominent then right after it came back down to 0dbfs, it flattened out, at 50ms. it made me wonder if that had anything to do with the 50ms figure. could have been pure coincidence though.david1103 wrote:
interesting! if a program, like doc fear has a clean kernal length of 90ms, would reducing it to 50ms not effect quality? some eq's have 180ms clean kernal. has anyone tested the effect of reducing kernal length? why would developers have such long ones?
anyway, i've always thought 50ms seems to be the upper limit for having meaningful information (significant stuff still above the noise floor) still occurring in the impulses (for preamp type dynamic programs). so i don't see any reason why you'd go over 50ms. BUT maybe those developers who made those programs actually did listening tests and thought that it made an improvement to have those longer lengths. or maybe it was somehow accidental. there are a lot of little things to keep track of when developing programs, and it's REALLY easy to forget to re-adjust something, etc. you'd have to ask them. i would say that reducing a 180ms kernel for an eq down to 50 would be barely noticeable, if. but again that's my opinion.