The Merits of 'DBT'

[twich54]

Lets try and keep this civil.........

Gordon asked me in the Oppo thread about my phrase 'legitimate DBT' so here goes......

by legitimate I just meant that DBT (double - blind - testing) be done / executed in a controlled, sceintific and measureable enviroment. I'm not sure how others feel but I see nothing wrong with this and as it relates to the Oppo thread, well if a company produces a 'upgraded model' at twice the price I for one would expect to 'hear' a sonic improvement.

I know there are those who feel DBT's are a bunch of nonesense but I think there's merit.

 

[amey01]

Interesting. DBT certainly has merit, but the major problem I see with it is that when dealing with sound (and music as a sound in particular), we're not only dealing with the scientific aspect of sound, but all the emotional aspects that go with it.

I have no doubt that these also can be scientifically measured, however for present day science it is either:
* too complex and involved - there are too many factors and too many variables for us to get a thorough grip
* one or more variables of which we are unaware

Therefore, when listening/evaluating/auditioning we talk about things that we feel when we listen. Such terms as timing, melodic, convey emotion, etc.

For these, there is no known measurement. Well - as I said above, there may be, but at this stage either we don't know what we're looking for, or what we're looking for is far too involved to be practical to measure.

Add to this - there is no way to record what we hear or feel. Our auditory memory is very short.

I feel that these factors combine to make DBT very hard to execute successfully.

I have been involved with DBT on various occasions and while sometimes I have found little difference, it is only when you spend time with a particular product that its particular strengths and/or vices become apparent.

Case in point - play a single piano middle "C" (261Hz) through a high end system, then through a cheap Sony system with 128K MP3 source in a DBT. Most of us would be hard pressed to tell which was what in a single DBT. It would make the high end system look like an extraordinary waste of money. But after time (when you allow yourself to see the "forest" instead of the "trees") things begin to become apparent.

This scenario extrapolates when engaging in DBT - the emotional aspects get pushed aside for the technical aspects. And we all know that the technical aspects of any modern day sound system are pretty damn good.

 

[tsv_1]

Our auditory memory is very short.

Yes... I was going to mention this as well. I'll try to find the article I read recently, but I think it mentioned something like full auditory memory fades rapidly after just 2 seconds. But I don't remember exactly because I read the article (out loud so I could hear it) more than 2 seconds ago.

 

[timm]

I just look at it this way. If one thing sounds 'better' to you then why should you be swayed as to what it actually is? Isn't that the point of DBT? Why are people afraid to let their ears do the talking - so to speak.

 

[twich54]

I just look at it this way. If one thing sounds 'better' to you then why should you be swayed as to what it actually is? Isn't that the point of DBT? Why are people afraid to let their ears do the talking - so to speak.

Tim that's exactly it, some prople ARE afraid to admit that their ears may mislead them..... 'psychoacoustics'..... in the simplistic way of looking at DBT it proves or disproves that differences do exist between the objects being tested.

For the record I firmly believe in psychoacoustics.

 

[amey01]

Yes... I was going to mention this as well. I'll try to find the article I read recently, but I think it mentioned something like full auditory memory fades rapidly after just 2 seconds. But I don't remember exactly because I read the article (out loud so I could hear it) more than 2 seconds ago.

It would be very interesting if you can find it!

I really don't think I'd dispute that two seconds.

Of course, we remember what things sound like for longer than that, but it is always the emotional things - things like "this had better timing" or "this sounded warmer".

But for the technical aspects of sound - yeah - two seconds is probably about right. This is where DBT has a hard time!

 

[amey01]

For the record I firmly believe in psychoacoustics.

I do too. But then it becomes like the placebo pill. Who cares if it is placebo - if it works, it works!

 

[tsv_1]

I just look at it this way. If one thing sounds 'better' to you then why should you be swayed as to what it actually is? Isn't that the point of DBT? Why are people afraid to let their ears do the talking - so to speak.

But then where is the absolute qualification? Is that not the ultimate goal of any test - to determine the "right answer"?

If we're saying DBT is only useful for the individual doing the listening, then OK, what's right for him is right for him and he's done. But if we are to use his opinion and label it "right" then how does that benefit the masses any more than a typically subjective TAS or Stereophile component review?

It doesn't surprise me that DBT has it's supporters and it's detractors.

Now, where did I put that blindfold?... oh, there it is over by the latex unitard (don't ask)

 

[Hocky]

I love the idea of the DBT - personally, I want to buy the right stuff, not necessarily the stuff that is "cool."

 

[timm]

I guess I don't get it. I thought audio was supposed to be an 'ear' thing. Why do the eyes need to get involved. I would think that adding another one of the senses would create more conflict for the brain to decyphor. I bet if someone is baking cookies - the system sounds better. I like things to look good too. I just think this debate is out there so the so-called experts don't get embarrassed when it is found that their 'golden ears' can't tell a difference

 

[Gordon Gray]

Dave,

That's why I asked what makes a legitimate DBT. I assume you are talking about the methodology. How often do you switch, how long, etc. I'd love to hear what methods make one DTB more legitimate than another.

My issue with this method is that, by nature, DBT is an instant in time. You can say that voice sounds better, the piano fuller, the cymbals less splashy if that is the focus (material being played) at that moment in time.

I listen to alot of classical and with all the various instruments involved, a DBT may reveal that the violins sound thin whereas, if you were listening to brass, you might say that they sound sharp and "blatty", which is the way they sound in a live concert.

Two instruments, two different conclusions.

Interesting topic.

Gordon

 

[Rich]

I believe that many of the differences people swear they hear between components disappear under DBT. I think that psychoacoustics play a much larger role in this hobby than anyone is willing to admit. I believe it is very difficult to accurately hear differences between similar quality products.

As for myself, I conducted DBT between my ARC Ref 3 and my Sanders Sound preamp, using a switch provided by Sanders himself for the purpose. One of these components is tube and one is solid state. Both are incredibly neutral. I was able (just barely) to detect the slightest sibilance in the highs in my main system with the Sanders vs. the ARC. I was able to pick this out accurately and repeatably enough under DBT that I decided to keep the ARC in that system. But the sound between the two was so close and so good, that I decided to keep the Sanders in my secondary system hooked up to my Pass Labs X-350.5 and my Ascents. There is no hint of sibilance with it connected in this system and the sound is superb. This experience proved to me once again the importance of synergy between components.

 

[Speedskater]

I think that the meaning of "DBT" and "ABX" is getting blurred in this thread. There are countless protocols and situations for "DBT"s and "ABX" is only one of them. "ABX"s can only test for small differences. While "DBT"s can test for differences or preferences.

 

[robertawillisjr]

Listening to assess the quality of equipment should not be short term. We need time to understand the nuances of what we are hearing. Music is very complex and it is difficult to determine if a component is conveying that complexity with any degree of accuracy.

While we can determine the gross factors of a component (shrillness, unbalance, etc) in a few moments, it can take some time to review a unit or system. INMHO, DBT is not conducive to a thorough understanding of how the component of system is performing.

It can be a good first step in the process, but if I were going to spend $23k on a pair of CLX's, I wouldn't depend on DBT alone.

 

[tsv_1]

if I were going to spend $23k on a pair of CLX's, I wouldn't depend on DBT alone.

If you were going to spend $23K on a pair of CLX's, we would have to first hook you up with a le$$ demanding retailer

 

[Rich]

I think that the meaning of "DBT" and "ABX" is getting blurred in this thread. There are countless protocols and situations for "DBT"s and "ABX" is only one of them. "ABX"s can only test for small differences. While "DBT"s can test for differences or preferences.

That's a good point. To be clear, my testing of preamplifiers was ABX testing.

Listening to assess the quality of equipment should not be short term. We need time to understand the nuances of what we are hearing. Music is very complex and it is difficult to determine if a component is conveying that complexity with any degree of accuracy.

While we can determine the gross factors of a component (shrillness, unbalance, etc) in a few moments, it can take some time to review a unit or system. INMHO, DBT is not conducive to a thorough understanding of how the component of system is performing.

Not sure why this has to be a problem. With DBT, you can set up protocols that take this into account. When I was doing ABX testing, I would listen to a passage for a long time, then switch and listen again for a long time, then switch, etc. Then I might switch back and forth a little quicker to really hear the immediate differences.

But with a little help, you could easily set up a protocol where you listened to one component for days or even weeks, then switched to another, and then either switched back or not, and listened for weeks again, all without knowing which component you were listening too at any given time and all the while taking extensive notes about what you were hearing.

It can be a good first step in the process, but if I were going to spend $23k on a pair of CLX's, I wouldn't depend on DBT alone.

Of course, you should never rely on just one thing alone when determining what you want to buy. Lots of other factors play into that decision. But I think DBT/ABX testing will quickly make someone realize that a lot of the differences they think they are hearing turn out to be non-existent or minimal. People that can wax poetic for paragraphs about the huge differences they hear between different components suddenly can't tell one component from the other in a reliable manner. I think that speaks volumes about the reality of psychoacoustics.

 

[Bernard]

In an article I read by Bob Stuart (of Meridian) on DBT, he made the point that you could switch from one component to a better component and hear stuff that you missed before, then if you switched back to the first component you would hear the stuff you missed before, because you would now be listening for it. I have found this to be true. Then perhaps you would conclude that there was no difference between the two, when in fact there was.

 

[Rich]

Then perhaps you would conclude that there was no difference between the two, when in fact there was.

Either the information is there or it isn't. If it's there, then there is no difference between the components (even if you didn't notice it on the first listen). If it isn't, then there is an obvious difference. If the information is presented poorly by one of the components, for whatever reason, you should be able to pick out those differences on close listening to both components. If one piece is more articulate in its imaging, for instance, I would expect that would jump out at you as you go back and forth between them.

I can't imagine you would not notice a trumpet in the background on one component, then notice the trumpet very clearly with the next component, then switch back and hear the trumpet with the original component, and then not be able to tell a difference in the way that trumpet sound is presented as you switch back and forth between the components.

 

[twich54]

I think DBT/ABX testing will quickly make someone realize that a lot of the differences they think they are hearing turn out to be non-existent or minimal. People that can wax poetic for paragraphs about the huge differences they hear between different components suddenly can't tell one component from the other in a reliable manner. I think that speaks volumes about the reality of psychoacoustics.

Exactly Rich, I totally concur ! and it's the Audio press that are some of the worst offenders which in turn has transcended into all the various forums.....this one included.

 

[Gordon Gray]

Had an interesting experience at the Linn Room at 09 RMAF. Demo between a CDP and a computer based system.

Presenter put on 15 seconds or so of music. Paused it for five seconds and replayed the same 15 seconds.

My ear told me I liked the latter (post pause) and asked the presenter which media he switched to during the 5 second pause.

He didn't change a thing. My ear was readjusting to the volume level, which appeared to me to be louder than the pre pause level.