More power for better sound at low volumes?

I would think your amp's input sensitivity and your speakers' sensitivity are the main factors. Also I think the noise floor and distortion figures are a contributing factor in that the lower they are, the higher the volume you can comfortably tolerate, therefore perceived loudness will be less. So i find some of these responses a bit skewed. Ssglx's point is well taken. The smaller the amp, the more relevant the parameters I just mentioned will be. This has been my experience.

I would expect those Martin Logan's and and any good electrostat to sound pretty good on the grand scale compared to other speaker types at low volume. Same with Maggies/planars. Traditional cone designs can sound as good at low volume or even better (Example the Triangles I own) but are more hit or miss in general.

Not true, electrostatics do not usually sound good a low levels, quite the contrary. Low efficient speakers with high output amps are what you are seeking.

Rustler 06-19-12:
I'm wondering if a more powerful amp will provide better sound at lower volumes, all other things being equal.

If everything else is indeed equal, including the input sensitivity of the power amps (or the power amp sections of integrated amps), noise levels will be worse with a more powerful amp. If input sensitivities (the input voltages required to drive the amps to full power) are equal, it would mean that the more powerful amp will have higher gain, which will amplify any noise that is introduced or generated at circuit points downstream of the volume control by a greater amount.

I suspect that in those cases where it has been found that a higher powered amp provided better results at lower volumes than a lower powered amp (with the same speakers), the reason was not the difference between the power ratings of the amps in itself, but other differences between the designs of the two amps.

Have you tried listening from different distances, compared to the distance you listen from at normal volumes? Perhaps you can find a listening distance at which differences in room acoustics will partially compensate for the reduced sensitivity to low and high frequencies that our hearing mechanisms have at low volumes (the Fletcher-Munson Effect, that was alluded to by Wolfie and Mapman).

Also, along the lines of Newbee's comment, it would seem to make sense that an amp that operates in Class A, at least when it is putting out low power levels, will generally tend to have lower distortion at those levels than a Class AB design.

Regards,
-- Al

My old Quad ESL57's were champs at low volume, so I have to disagree with Tmsorosk.

Anything you can do to lower the noise floor should help. Do you use any power conditioning?