A mix of very thorough and educated measuring/examination and some guessing I would say.Coconutty wrote: Mon Feb 27, 2023 2:10 pm Hi OxyOTT, out of curiosity, how did you go about matching the compressor algorithms for each band to the original. I’m guessing neither you, nor Xfer, nor Slate got a chance to peak at the code. I also noticed that your plugin doesn’t null with the original (although it does sound very very close to me - if not identical). You put 2 presets in, the second one imitating the Xfer version, so I’m guessing that the Xfer one isn’t identical either (all these years I just assumed that it was). So, any insights you have here on the development, or, the challenge of cloning a sound like this would be interesting! Thanks!
There are only so many common compressor designs, it's rare for people to completely reinvent the wheel here.
I think the not nulling part could be related to slight differences in the crossover filters or slightly different timings, nothing you would really hear in a side by side comparison of course.
I believe the actual dynamic processing algorithms to be quite accurate though, in my measurements the dynamic transfer curve visually aligned 1:1 with Ableton with every setting I've tried.
There are much more components in an effect like this where a small difference would make it not null than for example a waveshaper.
