We are OT on this.
It is all a question of differentiating between linear and non-linear distortions of DUT. Just let me say that through a "DRC" filter for both curve and envelope it is possible to perfectly match two devices, like the one I mentioned above (we have done it with Emm Labs too), excepts for their non-linear component.
Well I'm 100% open to anything that can be preformed on a PC to make the hardware end as simple and pure as possible. Now that I've decided to go 100% chipless DSD, I want to learn all I can about filter algorithms that can be handled on software the make the best out of it. Jussi and HQplayer has really opened my mind to the possibilities
. Perhaps you should start a thread we can discuss software based filter algorithms in detail?