Double Blind Testing of HT Components - Necessary for you? (1 Viewer)

[John Morris]

If accuracy is not a major reason for buying a preamp, that the above is a rather moot exercise.

No, what I hear and what I think I hear is the ONLY reason, other than price, to buy one preamp over another. But, if that reason is satisfied to the BUYERS satisfaction during careful A/B listening, why isn't that enough...?

 

[John Morris]

George: I think we may agree yet, let's consider this:

Take speakers. Alot of people love Paradigms, and they have a huge following. The people who like Paradigms say they are detailed, and revealing (a good quality). However (not to knock Paradigm), I find them bright and fatiguing (a bad thing). So two different people can "hear" the same thing, and come to different value judgements, which is why a DBT is not appropriate for answering those types of questions. They are good only for determining whether there is a difference.

Well, what if I A/B tested those same speakers in a system which used tubed amps. Or used a Ref 30 which I found to slightly dis-emphasize those top octaves? As I think Martice mentioned previously, IMO, each new component has to be evaluated within its' new home and not in some sterile sound lab... or does it?

 

[chung]

This is exactly where I DID NOT want this thread to go, but, WHICH straight wire(used in the ABX box) are you talking about? Is it made of the exact same cable that I use in my, or your, system? Does it use pure silver contacts, traces and such? SEE how convoluted THAT can get?

The straight wire may be a misnomer (as in straight-wire-with-gain). It is a short pair of cables that connect between the CD player and the ABX box's inputs. If you have doubts about what cables to use, then use the same cables as those connecting the preamp under test. You can also use DBT to establish that there is no difference betwteen cables. I don't think that is convoluted at all.

If a buyer is happy to make a decision based on A/B testing, I don't see any problem at all. But do realize that if a preamp is not accurate, it may not sound good after extended listening, which you cannot really do well at a store.

 

[John Morris]

If a buyer is happy to make a decision based on A/B testing, I don't see any problem at all. But do realize that if a preamp is not accurate, it may not sound good after extended listening, which you cannot really do well at a store.

Hi Chung! Thank you for re-affirming my previous purchases... Second, accuracy is relative. And, if one preamp is convincingly more ear accurate than the other one... then isn't that as accurate as each buyer needs to be to make a buying decision?

 

[John Morris]

If a buyer is happy to make a decision based on A/B testing, I don't see any problem at all. But do realize that if a preamp is not accurate, it may not sound good after extended listening, which you cannot really do well at a store.

Hi Chung: I don't know if you meant to do this... but you may have given a sign that DBT and ABX is in fact not valid due to extended listening factors. Those factors can either be due to listener acustomization(sp) or due to gear break in? Did you mean to make this statement in a preamp testing thread?

 

[chung]

John:

Hi Chung: I don't know if you meant to do this... but you may have given a sign that DBT and ABX is in fact not valid due to extended listening factors. Those factors can either be due to listener acustomization(sp) or due to gear break in? Did you mean to make this statement in a preamp testing thread?

No, I don't think I gave such a sign. You can run ABX with gear that has been thoroughly broken in (if you believe that makes the sound different). It is definitely recommended that the gear under test is warmed up. Remember you have as much time to spend with the gear as you want. You are trying to detect a difference. If you believe that you are not capable to detecting difference unless you need extended time, then spend as much time as necessary to zero in onto the "sonic signatures".

What I said about someone not being happy with an inaccurate preamp is this: In a store, you have a limited number of sources you listen to. A slightly colored preamp can make certain kind of music sound "better" (particularly when mated with certain speakers, for instance). In the long run, it is best to choose a preamp that does not add any color, unless you are counting on that coloration to compensate for something else in your system. Or if you really like the coloration, like what tube amps give.

It is really personal how one selects equipment. Like I said, you don't need to use ABX to buy stuff, and there are many factors going to the buying decision besides accuracy. However, if one is contemplating spending $5K on interconnects, some kind of DBT may be wise. Just my opinion.

 

[Chris Tsutsui]

#2

Actually I'd pick the Rotel because I've used their products before. The difference between them is probably too subtle for me to do any double blind testing.

I usually get embarrassed in blind tests anyways (when I choose the cheaper product).

 

[John Morris]

Actually I'd pick the Rotel because I've used their products before. The difference between them is probably too subtle for me to do any double blind testing.

Chris: I guess that you haven't had any problems with your preamp? Then I guess that you didn't buy one of the recent... aw nevermind. Dang! Once I decide that I don't really need to do this severe testing, along you come... Anyone that buys the same brand due to previous performance, in spite of current performance, is just not testing at all, IMO. Dang, how much does this ABX box cost?

 

[Kevin. W]

Never down a DBT in my whole time in HT. When I first got in at the DPL level with my Sony SAV-10 setup it was due to price(store was closing). Got it hooked up and it sounded awesome, compared to my TV speakers. During my move to DD/DTS everything was done on recommendations of this forum. Again my prsent setup sounds awesome. If I honestly knew what to listen for I'd do a DBT.

Kevin

 

[JaleelK]

No, you had it right the first time. The key thing is, X is either A or B, but you don't know which one it is. A machine has chosen it for you. So, you sit with a remote control - when you hit button 'A', the machine playes the Rotel. When you hit 'B', it plays the Outlaw. When you hit 'X', it plays either the Rotel or the Outlaw, but you don't know which one it plays.

I'm so glad you have come in to explain DBT, you are doing a great job!!

 

[JaleelK]

Its easy to show that you can score perfectly on an ABX test if you have a reliable answer to just one of these 4 questions. Such a deal ? in ABX testing you only need to answer 25% of the "questions" reliably to score

Also, some of you interested in training your ears to become a critical listener and to start performing your own ABX test should www.PCABX.com and download the program. It's quite interesting and even challenging once you start ABX-ing different audio sources.

 

[Saurav]

I'm so glad you have come in to explain DBT, you are doing a great job!!

I assume you're being sarcastic here?

 

[JaleelK]

I assume you're being sarcastic here

No sarcasm coming from me.

 

[Saurav]

OK, just checking

 

[Larry B]

Ian:

What reading audio boards has throroughly convinced me of, is that many audiophiles, who spend tens of thousands of dollars on audio equipment, are spending _well_beyond the point of diminishing returns. Many are spending at the point where a 100% increase in price will buy you only 1% in perceivable quality.

I know that I am in a very small minority but for the life of me, I do not know what a 1% increase in aduio quality is. To use the common analogy of cars, a Corvette is about $45K, and a Ferrari 360 Modena is about $175K. Is the Ferrari 4 times as good a car as the 'Vette? I can't answer that, as I have no idea what it means. Put another way, I don't know how to quantify automobile overall perfomance (even though things such as 0-60mph and lateral G force as easily quantifiable). Similarly, I coundn't tell you whether a particular dish in a restaurant (or at home, fo that matter) is x% better than another dish, as I do not concider gustation to be quantifiable.
A more meaningful way to look at it (in my opinion) is to ask the question, are you willing to pay the $130K premium for the Ferrari (assuming one is lucky enough to be able to )? Based on your terminology, the Ferrari is 4 times as good to the individual who is willing to pay the premium, but is not to the individual who isn't.
If this is the method you are applying to audio equipment, then it simply boils down to your unwillingness to pay the going price for that level of performance, while others may be willing to do so (for any of a variety of reasons).
Make sense?
Larry

 

[Larry B]

Merc:

DBT, if conducted (1) in an appropriate environment, and (2) for a sufficient length of time, would almost certainly be preferable. As I am unaware as to how this could practically be accomplished, I am forced to rely on unblinded comparisons. (Similar to the way I choose, for example, restaurants.)

Larry

 

[Larry B]

Shawn:

There are lots of things out there that can bias/alter your opinion of the sound. The 'X' is used as a check so you know if you are only using your ears to listen to the products in question. IOW if what you think you are hearing is real.

If an A/B test is properly blinded, what other than your ears could you be using?

Merc and others:

I believe one relevant point has be omitted from discussion of A/B vs A/B/X tests; namely, what it is you are trying to accomplish. If your aim is to determine which of 2 components you sonically prefer, a double-blind A/B test is sufficient since it removes prejudices such as appearance, price, manufacturer, etc.

On the other hand, if you are trying to prove (to yourself, or others, or both) that you can distinguish between 2 products, then the A/B/X test is required.

Larry

 

[Saurav]

I don't see how you can decide on a sonic preference without first establishing that you can hear a difference. If you prefer A over B, doesn't it automatically imply that you can hear a difference? And if it does, it seems to me that even if you're trying to find the one you prefer, first you need to ensure that they sound different. In other words, the first step is always to establish that an audible difference exists, regardless of whether you want to stop there or go on to establishing which one you prefer.

 

[Larry B]

Saurav:

If no audible differences exist (to your ear), then you would be unable to choose a favorite. I have experienced this very phenomenon when first listening to 2 pieces of equipment. (Though differences became apparent after extended listening.) Again, is the issue to pick your favorite, or prove that you can distinguish?

Larry

 

[Larry B]

Saurav (continued):

Let's say you were having trouble deciding which you preferred. An A/B/X test would show that (1) you had no statistically significant preference, or (2) that you had a prefernce, but it was very small (and thus revealed only after an extensive number of trials). In either case, you shoul logically go for the less expensive unit (unless there were mitigating factors, such as looks, size, etc.)

Larry