Here are some initial readings from my Poniie PN2000 meter, purchased on Amazon. No idea how accurate it is but the reviews compared favorably to Killawatt. My AC currently measures in the 124 to 125V RMS range @ 60Hz.
Standby: power draw read as zero on the meter, well within the < 0.5W specification.
Soft start: The amp starts at around 100W, and very gently ramps up. Note however this wasn’t a completely cold start - the amp had been powered down maybe 10 minutes after idling in low bias mode, and was slightly warm still.
Warmup in high bias: After half an hour, it was drawing around 220W and still rising, indicating bias and temperature still hadn’t settled. So I would not expect it to sound its best yet.
After 1 hour: things have stabilized around 277W, give or take a watt. This is considerably less than the rated 350W, and probably on par with a Pass XA30.8. Again, this wasn’t a fully cold start, but you can expect the amp to be fully warmed up at this point. Power factor is 66% (421VA), indicative of a pretty inductive load as expected (the transformer is a huge inductor after all). I haven’t measured PF before and will compare to some smaller amps, but it may be worthwhile investigating a conditioner with power factor correction (e.g. Furman Elite) to improve performance of other gear in your home, e.g. a home theater.
Green bias: switching to low bias immediately drops draw under 110W. After 3 minutes it is around 105W and seems stable there despite heat sink temperature taking a while to come down. So the bulk of power drop is immediate and while device temperature is a factor, I would expect the sound to largely reflect within a few seconds of switching.
Switching back to high bias: If done while the amp is still warm from prior high bias use, it quickly ramps back up to 275W within 20 seconds. I would have to perform this test after an extended cool down to see how much heat sink temperature factors.
So to summarize:
1. Give it at least half an hour warmup for high bias
2. It uses a lot of power in Class A, but not as bad as specified
3. Bias changes take effect pretty quickly, temperature stabilization not withstanding
4. Low bias cuts power by more than 60% to 105W, and heat is quite manageable. While still not a green practice, I would be ok leaving it on all day in this setting.
Depending on how bored I get, I may plot this over time more accurately and from a cold start... I’ll probably have to be really, really bored though...