Quote:
Originally Posted by
Currawong 
Running sweeps through an amp to measure the output, which is what test programs do, wont give the same results as playing music through it. That's what I meant. All the extra parts in TOTL amps aren't there for nothing.
So what you can also do (and what people do in practice), is measure the output of an amp given an input signal of music. However, it's been found that amps that do very well with a broad range of traditional audio benchmarks, usually or pretty much always also do well with any kind of audio signal including various types of music. That's pretty much the point of benchmarks in any field, though quite arguably, the standard ones in audio exist for historical reasons and may not be the most revealing. That said, I'm of course not satisfied with only one source of performance figures. I'd still like to see more people run different amps through an AP, dScope, or other sophisticated audio analyzer. It would be better to see the O2 results confirmed on another system, and I'd like to see audio differencing tests too. Nobody seems to have stepped up though. I certainly don't own an audio analyzer or even a decent enough sound card to make a reasonable line-in recording of an amp, or I would have done it myself, at least for the O2.
As for extra parts in other amps or more sophisticated designs, there are a number of potential reasons:
1. Designing simple barebones cookbook amps can be boring.
2. For pride, a sense of accomplishment, or just because you can, you want to outperform the competition.
3. You need something more complicated to handle very high power outputs well.
4. You're trying to implement a different kind of sound (less accurate).
5. The consumers want to buy something more fancy, with or without certain key attributes, whether or not that actually makes sense from a performance point of view. Many people don't buy toys based on what they need, but what they want.
and so on.
Edited by mikeaj - 4/7/12 at 9:15am