Originally Posted by kbarnes701
Here's a graph of the Emotiva XPA-2 from Audioholics. It looks real flat to me from 20Hz to 20kHz... you saying this one is no use either?
And here's a couple from Stereophile:
The one thing they all seem to have in common, to me, is that they show a remarkably flat FR from 20Hz-20kHz. Am I missing something?
BTW, if you were referring to me as someone with a sales/marketing background, you have the wrong guy. I have worked all my life in Advertising and never, even briefly, in either sales or marketing.
In all of my use of an Audio Precision over the last 17 years, I can't think of a time I ever had the scale bigger than -2 to +2 dB. At that range, the curves are, well, curves, ie, not flat.
And I was the one who referred to your sales/marketing background. OK, it's advertising. To me, that's all the same. Just like to you, "flat" means, well, whatever you want it to mean.
The issue here, as Greg Lee seems to be getting at, is that plots not zoomed in to a range of +/1dB, or +/-2 dB, are worthless.
Furthermore, plots that show the freq resp into a simple 8 Ohm or 6 Ohm load are worthless.
Give me plots with a real speaker load, and then we're back where I started: At what point are the freq resp variations "neglibible" or "inaudible?"
It might be a non-issue to you, Keith, but then again you don't design mass quantity consumer A/V electronics, do you? You don't have to review CPK data taken from thousands of units manufactured in a factory in China, do you? You don't get to see that in worst case scenarios, some of those amps, when measured with a more "real" load, might just have variations that approach audibility.
It appears to me you've taken a religious position on the "all amps sound the same" idea, and while it may surprise you to know that I *generally* fall in that same camp, I'm aware of enough issues to know that the "all" part of that statement needs some qualifiers...