20 amps times 120 volts is 2400 watts at 100% efficiency. Take that time maybe 85% efficiency per pass through device (power conditioners and power supply) and you get a more realistic 2400 *.85 *.85 = 1734 watts (1300 watts with a 15 amp circuit).
That's for everything sharing that outlet. Most systems can live with that just fine.
Short of CLASS D amps, that have real issues with RF noise (and a bunch of other bad stuff in wide-band use)if not handled properly, good capacitor filtered power supplies are pretty quiet.
I still haven't seen any results on Vcc DC bias variation on "sound" and fundamentally this is really what all that upstream stuff is supposed to improve. OK, people "say" when the Vcc bias changes, imaging and dynamics suffers.
That's one of the general "improvement" of class A amps, the engine is at red-line all the time so it's ready to go NOW! But at an efficiency cost. A/B amps have to accelerate up to speed, slowing transient responses. High bias A/B amps cheat a little and hold the throttle in class A for 15-30 watts. Other improvement is distortion which is inversely related to bias current and of course, no PNP to NPN notch distortion to manage.
Some of the "power" improvements I'll try to hear with a PLINIUS SA-103 amplifier verses my ODYSSEY KISMET MONO low bias A/B amplifiers. This, with the units AC line noise defended by the on board power supply of each amplifier.
To say a power supply is inadequate is somewhat intriging. The caps recharge at the RC time constant of the circuit, which is pretty fast and with full voltage charge from ZERO in five or six time constants. If a big drum beat pulls down the supply, then the only drum beat that will sound good is the first one you hear! The time even between drum hits is PLENTY of time for the caps to recharge. There is a measurable supply sag that can be designed for.
As far as caps needing to sit for "days" to charge fully? This sounds like a fairy tale as again, music is happening faster than that, and the power supply of a good amps seems to manage without waiting days between bass transients or high intermittent current draw. A more continuous draw? Well, I guess your amps don't sound good anymore until you leave and come back in a day? Amps are tested at a steady state 3/4 load cycle and then swept to full power to make sure long duration operation is decent.
The data says power supplies work on a much FASTER charge cycle than "days" or even seconds. The Vcc sag is an indicator of this attribute as that's where the rubber hits the transistors. No supply can never sag, so it's the amount (dips) and character (spikes) that is in question to fidelity.
