[quote user="Eddy Current"]I don't know, but who said it was already perfect?[/quote]
Well, in one of you responses to me you said :
[quote user="Eddy Current"] There are no "harmonics which break through the rectification and smoothing stages" to eliminate. Get a 'scope and look at the ripple on the low signal supply lines in your amp if you doubt my word. You can try it with or without your variac, because it won't change a thing.[/quote]
Now my understanding of a power supply ( in a hifi context) is simply to provide a stable voltage regardless of whats happening on the main supply. Your statement (to me at least) suggests that a" perfect power supply" condition as been reached in that there is supposedly no measurable ripple/noise harmonics present.
I am not in a position to actually check physically the power rails (in that I need an oscilloscope and a spectrum analyser ). All I can do is listen (blind tested when possible).On every occasion I have tested a variac,I have heard a difference.It is also significant that so many hifi manufactures (who don't manufacture or sell conditioners or have a vested interest in conditioners) where using variacs.
I don't know your electronics background is but when people like the hifi guru :Tim de Paravicini (who's electronics credentials are impeccable in hifi and more importantly the professional sector) state things like "Regulation is only a device used by conventional circuits because they conventionally are too sensitive to their power supplies." or "I try to design the beast to be relatively immune to the supply.",the implication is that power supplies are less then perfect and can compromise the overall performance of the"conventional circuits".
I am still amazed that you won't even try a conditioner or a variac just out of interest and scientific curiosity.
I am wondering whether you fit "reality to fit theory" instead of "theory to fit reality "??? If it was the latter then (despite your gut feeling or what theory currently says) you would have no qualms in doing some blind tested listening tests.