Originally Posted by hifisponge
From all of the posts I've read of your's, you seem like a knowledgable guy with a no nonsense approach. And since you are an amp guy, I've always wondered something that I hope you have some insight on. If a solid state amp has flat frequency response (+/- .3 dB) and low distortion (less than 1%), and it is not being driven to clipping, can it still have a sound of it's own? Are there other things that are measurable or unmeasurable that can affect the sound quality of an amp?
If the amp is genuinely accurate, it will not have a "sound of its own." And it will in fact sound just like any other highly accurate amp (allowing for differences in power). Any two amps that put out identical signals will sound the same through the same loudspeakers in the same listening environment; other than the signal there is no other conveyer of sound quality to the loudspeakers--no side channel or metadata telling the loudspeaker to act refined because it's hooked up to a cryogenically-treated and magnetically-aligned Celestial Sound Megamoolah 12000, or squirrelly because the amp is a PA piece bought used. If it's not in the signal, it's not there.
But there is much more than just frequency response and distortion, although those are two huge defining factors in sound quality. At QSC and probably many other manufacturers, there are hundreds of tests done on amplifer designs from development through qualification. Many of them are directly related to sonic performance, like THD, frequency response, noise, et al, and many of them are tangentially about performance but more directly about reliability and usability. For example, does the amp circuitry generate any artifacts that don't show up in a distortion test? How hot does it get (excessive temperature is a major killer of electronics, which is why I never advocate disabling the fan)? Will it run on a 15A outlet? How does the amp behave when it clips?
Excessive hum and noise can squash sound quality, primarily during quiet passages, in between tracks, etc. Instabilities like Chu Gai mentioned can cause audible problems, too. Different types of distortion can have varying effects on audibility, too. For example, we're all familiar with the distortion that arises from clipping, due to non-linearity as the output voltage reaches the maximum limits of its swing. But the distortion from light and even moderate clipping can be inaudible or maybe barely discernible, even if it measures rather high. Crossover distortion, OTOH, might measure extremely low on a THD test using a sine wave at or near the amp's full output, but it could have a devastating on sound quality because its biggest effect is on the smaller signal levels.
If any artifact or oddity in the amp's performance is significant enough in the signal to be heard, it can be identified one way or another. If it can be identified, it can be measured. If it can be measured, it can be understood. If it can be understood, it can be corrected.