"https://www.temporalcoherence.nl/cms/images/docs/FourierConditions.pdf
«The effects in time domain of non-linear behaviour in combination with memory effects could explain why e.g. amplifiers with similar properties regarding frequency response and distortion
levels, sound different. It is to be expected that ten (10) different designs will produce ten different responses to music signals and thus receive a different perceptual qualification.»
This physicist seems to know better than Amir ... 😊"
I read through it. There is little there to comment about. Your audio gear does NOT have memory in it. He creates a simple circuit that does and shows trivial scope simulations of it. Enough to fool a layman into thinking there is some measuring going on.
He also makes other dubious comments:
". Another well-known example is the upper frequency hearing limit: as humans cannot hear above 20 kHz, the reasoning is that there is no use in reproducing higher frequencies, as these will not contribute to the signal, reaching the brain. This argument has often been brought to the table to disqualify high-resolution audio. However, many high-end enthusiasts claim they can clearly hear the difference and even seniors, with an upper frequency upper limit of 10 kHz (like the author) can distinguish the difference."
First, this has nothing to do with fourier transform. Second, I post ABX tests of high res that I did pass. But I am confident none of you can, including the author or said paper. You never throw such random claims in a proper paper. If what he says is true, you need to show it in controlled listening tests which he has none.
Really, none of what you post has anything to do with ASR and value of what we are doing.