Use music you don't particularly like for the initial evaluation - Page 2

Dusty,

Qvortrup isn't just thinking out loud. He has used this comparison by contrast method for years. Audio Note systems are extremely musical to my ears and his company is very successful. He must do something right during his listening sessions.
Auditioning with different pieces each time, that's what Qvortrup suggests, yes. It's probably not feasible for mere mortals. Once again, Dusty: his point is that he doesn't want to "assess the equipment", he wants to assess the recordings first and foremost. He wants to listen to the music. The conclusion which component is the more neutral and transparent one is a byproduct of this process.
Dusty, why the L would you want to assess "soundstage"? Why is this the first hi-fi criterion that comes to your mind? Do you listen for "soundstage" in a concert? Or do you listen for the music? Our methodology obviously influences our buying decisions. And if you listen very analytically for soundstage aspects, you'll probably succeed in building a system that creates a soundstage like you want it. It is possible to compensate for the lack of visual clues in audio reproduction, but how important can it be to "see" the instruments? Is it really helpful? Doesn't it mean applying flawed methods if all you want, in the end, is to "listen" to the music?

Tom,
I am afraid you have flunked "Comparison by Contrast 101". Tom, you're supposed to listen to your three unknown CDs using source A, then you are supposed to listen to the same CDs using source B. First, you listen for differences between the three CDs using source A. And then you listen for differences between the three CDs using source B. You can even compare the soundstages on the different recordings, if you like. Qvortrup's theory is that the better and more transparent source will show greater differences bewteen the recorded soundstages of the three CDs. And, more importantly, the better source will facilitate a better understanding and easier access to the musical meaning of the three CDs you haven't known before. Taking three unknown recordings and just listening with source A without performing the second part of the comparison with source B isn't really a comparison of sources, is it?

Dusty,

I am still not sure whether you really understood the problem that Qvortrup tries to address with his method. Dusty, you left out the second part of the paragraph you quoted, and the second pargraph is crucial: A byproduct, Peter. In the end, it is about comparing components, but Qvortrup tries to address what he has identified as a heuristic problem by performing this comparison in an indirect way.

He believes that the problem is this: In any direct AB comparison with a single recording, we judge the respective sonic attributes of component A and of component B. But how do we do this? Do we really know how the recording would be supposed to sound if reproduced by the perfectly transparent and neutral component? Does anyone know? Actually, not even the guys who have been present in the recording studio can know this, because it can't be said with any degree of certainty what influence the recording equipment has had. Sure, the engineer has been monitoring the recording, but are his monitors perfect? Or the amps he has used for driving them? The recording has no absolute reality against which the sonic accuracy of the music reproduction chain (and of our components) can be judged. We simply don't know how the recording would sound through the perfectly neutral and accurate equipment because such equipment doesn't exist and any recording is flawed - or at the very least, unique.

Now for the problem: how do we determine that one component is more accurate than another when the only method we use is that of the AB comparison? Since perfection doesn't exist, every component will impart a sonic character of its own on any music signal that passes through it. We may like the sonic character of component A more than that of component B, but how do we judge accuarcy through this AB process? How do we know which sonic character is truer to the reality of the recording? Is the soundstage supposed to be as wide as component A likes us to believe? Is there as much instrument seperation captured on the recording as component B suggests? How the hell do we decide this? We may have developed an idea what we would like a certain well known recording to sound like, or what we expect it to sound like, but actually, we have no factual basis for this expectation.

Qvortrup's solution: don't AB when comparing components. Instead, compare different recordings (especially ones you're not familiar with) with component A, then compare the same recordings with component B. The more neutral and accurate component will obscure the differences between the recordings to a lesser degree, it will show more contrast, and it will provide more musical information, and it will get us closer to the truth of the recording and thus closer to musical truth.

The AB method is always in danger of focusing on the sound of the equipment, and losing sight of the objective of music reproduction. Qvortrup's comparison by contrast method tries to prevent this by focusing on the music instead, and on the uniqueness of the recording and the interpretation. And this works best when you have no preconceptions about the music you listen to. As I said, the conclusion which component is better is reached indirectly, but it is reached nonetheless and probably with greater certainty than through ABing of our favourite reference recordings. Dusty, I simply don't see the flaw in his reasoning.

Dusty,

You're right, of course. I guess we have reached a point where we just start repeating ourselves. So, Dusty and Tom, let's agree to disagree.

Hooray!

You have saved my day, shivohum.