There is a famous quote - often attributed to Einstein (probably incorrectly).
It goes: “Everything should be as simple as it can be, but not simpler”
The simplest way to test amplifiers is by connecting them all to a nice simple resistor load...
It's very standard, big load resistors aren't very expensive, and it's a convenient way to compare things equally....
And, for all those reasons, measuring them with a resistive load is in fact the most common way to measure amplifiers.
The problem is that, while purely resistive test loads are the most common type, virtually
ALL real world speakers are nothing like that.
Almost all speakers are the exact opposite - in fact, most represent very reactive loads, with very complex impedance characteristics.
And, yes, different amplifiers do act
VERY differently when connected to reactive loads.
In fact, being asked to drive a speaker that is almost purely resistive, like a Magnepan,
IS "the wild and wacky exception" for an amplifier.
You might as well compare the handling of several cars - by testing them on a straight road, with a gentle but steady downhill angle, and a perfectly smooth dry surface...
And then claim to be comparing how they handle and how fast they go under "real world driving conditions".
In fact, a large part of designing a useful test is to choose test conditions, and test samples,
representative of real world usage conditions.
You can drive a VW Beetle, and a Formula 1 racer, straight down the street in front of your house at 20 mPH...
And then gleefully claim: "Wow... we tested a VW Beetle and a Formula 1 racer - and they handled just the same."
Your statement won't be a lie - but it won't contain any useful information either.
You aren't going to be able to hear the differences in distortion between two even really poor quality amplifiers while playing fuzz guitar music...
And you aren't going to be able to hear differences in frequency response by playing vocals that contain a range of frequencies that any telephone could easily handle...
And you aren't going to find out how they sound with real typical consumer speakers... unless you test them with a variety of real typical consumer speakers.
In that Stereo Review test, they almost certainly chose Magneplanars because they are highly favored by certain audiophiles...
However, they are not at all representative of "the load offered by most typical loudspeakers"...
And neither are they "the loudspeaker most likely to reveal a wide variety of flaws or limitations if they're present"...
- Magneplanars have a very resistive and "benign" impedance characteristic
- Magneplanars have poor low bass response (which is obvious if you look at their specifications)
- Magneplanars are also dipoles (so they tend to highlight certain types of errors very well - while totally obscuring others)
- Also, because of their low mass, and weakly coupled drive mechanism, Magneplanars are extremely INSENSITIVE to variations in amplifier damping
(while virtually all other speakers in common use are the exact opposite)
In short, other than being well liked by a certain group of audiophiles, Magneplanars are a very POOR approximation of "a typical speaker".
They are in fact a very ATYPICAL speaker.
Therefore, unless you happen to own or plan to own Magnepan speakers, the results of their test are quite likely to be useless to you.
There is no need for evidence to "prove the test wrong".
Based on the methodology of the test itself it is simply inadequate to support the convlusions you're looking for.
If you look at the products they chose...
And actually read the verbiage about how they chose what to test...
It becomes obvious that they chose products to satisfy certain requirements.
They chose products that were currently popular among audiophiles.
They also SPECIFICALLY chose products that many audiophiles would be likely to EXPECT to sound very different.
They compared a low cost, solid state, consumer receiver to an expensive tube amplifier of rather unique design.
They didn't even pick a typical tube amplifier - they instead chose a unique and atypical "output transformer-less" one.
Rather than chose typical products - they chose products that would "appear to be as different as possible".
And, as such, they succeeded quite well.
sorry. not buying it. They were testing to see if any difference could be noted between two amps. no need to test them under a wild array of conditions. here's two amps. nothing else is changing in the test except the amp. do you hear a difference? Simple. And again, if you have a problem with the test, go ahead and present the evidence you have that proves it wrong...