[Occam]

I'm looking forward to the release of more information and details.
Its potentially a great tool for product design and evaluation.

......
There are absolutely no active components in the ABX box, nothing to "mask" or "ruin" the system sound.  What you have set up is exactly what you are listening too.  With identical components in the test chains, you can use the ABX box to listen to interconnects, speaker cables, or even high end fuses too.  Even more fun.....

Its and interesting and telling use of those parentheticals around "mask" and "ruin".

[sts9fan]

From mgalusha Welcome to the lab. See link http://www.audiocircle.com/index.php?topic=71571.0
 If this ABX Tester is offered for sale by AVA then we have a potential problem with what amounts to a promotional post in the Lab Circle by a vendor that might be better placed within the Audio by Van Alstine Circle in the form of a sticky.
Scotty

He ain't the facilitator

[*Scotty*]

Is DBT discussion in the LAB now allowed?
Scotty

[sts9fan]

I see no reason to limit discussion in a hidden , barely traveled circle.  That said it has to stay civil.
I am shocked at how quickly the anti abx folks attacked. If you don't like it don't use it.

[*Scotty*]

It works for me, let the games begin!
Olympics Fan.
Scotty

[Freo-1]

I see no reason to limit discussion in a hidden , barely traveled circle.  That said it has to stay civil.
I am shocked at how quickly the anti abx folks attacked. If you don't like it don't use it.

I can see that side of the argument, but by the same token, if one thinks the science is flawed in one or more respects, the reasoning for that ought to be able to be voiced as well.  After all, how does one get past the arguments if they are not hashed out (in a civil manner?).
 
In that spirit:
 

https://passlabs.com/articles/audio-distortion-and-feedback
 

[planet10]

DBT may be based on science, but, it is clearly an incomplete science at best.

In this case, talking about ABX, the test is not imcomplete, it is just that as the test is designed it is a statistically weak test and easy to confound. People who use ABX to say there is no difference are using the science incorrectly. That is not within the capability of an ABX test.

A null result only says that in that specific test that no one heard any differences, that is all.

An ABX that produces a statistical difference is a strong indicator that the differences exist (because the test is weak)

I found the part of Kunchur's papers that covered ABX testing very interesting.

http://www.physics.sc.edu/~kunchur/Acoustics-papers.htm

dave

[JohnR]

Enter the test mode and all the indicator lamps are turned off, the starting point is randomized the dispaly lights up saying "Test 1".  Listen away, make notes, and when you want to, press the test button again.  This will then switch to Test 2 and the power amp under test will switch to the other power amp, or --  maybe not.  This depends upon the random number generator making life difficult for you.  Listen again, make notes again, and press test again and switch to Test 3. Repeat for the entire 8 test program session.

Frank, to be accurate, this is not ABX.

Here is an explanation of ABX: http://en.wikipedia.org/wiki/ABX_test

Here is a product that does ABX (no longer available, I believe): http://www.ackthud.net/shawnfogg/pics/temp/ABX_manual.pdf

[MaxCast]

Forget about DBT. To me, the real value of the ABX test box is not to prove anything to anyone except for the owner or renter of the test box. Imagine having an ABX test box and you want to see how your tube amp compares to your friend's SS amp because you're considering purchasing the same amp. You can use the ABX test box to determine if you prefer the sound of one amp over the other. And being able to change amps on the fly would be wonderful. The test box would also be very useful for borrowing items from your local high end audio store to test the potential significance of an "upgrade" before you buy it. I've been looking for a device like this for 15 years. Price notwithstanding, if you can think of a better way to make such a comparison, I'd like to hear it.

This is the best post in this thread!
I love auditioning components in my system, I hate getting behind the rack turning everything off, switching cables then turn everything back on again...multiple times.


On a side note.  If you are not interested in the device and have said as much you no longer have to post in this thread.  Get it?

[sts9fan]

Exactly! I doubt Frank is using it blind. For me I would love to use it to listen to compare a few amps. I would not use it blindly.
That said I don't understand why people are calling DBT statistically weak.
You may say the data is weak but don't think that makes data statistically weak.

[jtwrace]

Exactly! I doubt Frank is using it blind. For me I would love to use it to listen to compare a few amps. I would not use it blindly.
That said I don't understand why people are calling DBT statistically weak.
You may say the data is weak but don't think that makes data statistically weak.

Yes!   

[JohnR]

Exactly! I doubt Frank is using it blind.

I think you have to use it "blind" - it doesn't tell you which one you're listening to. You can't tell (unless there are meters on the amps or something like that).

[avahifi]

The main purpose for our ABX box is to help us understand what, if any, sonic improvements result from new design ideas.  Used in my sound lab with the same speakers in the same acoustic space over and over we try and eliminate all the variables possible to get more insight into our design ideas.

Note that being able to switch speaker systems is also very helpful.  The results of electronics testing will vary depending upon the speaker loads being used.  It is really easy to tell the difference between speakers, but which electronic design shines with which speakers provides interesting and sometimes unexpected results.

Finally it will make life a lot easier at consumer audio shows as we will be able to easily be able to hook up speakers C to amplifier A or whatever with a push of a button.

I never said that the ABX box is the living end.  It is just a useful tool for us, and maybe for others too.

To deem it to be simply a piece of crap is rather disingenuous.  Perhaps some are actually afraid of it? It might prove that the earth is not the center of the universe?  It is a useful tool, nothing more, nothing less.  If you are afraid of it, then avoid it.

Regards,

Frank Van Alstine

[Devil Doc]

Frank hit the nail on the head. People get very uncomfortable when their belief systems are challenged.

Doc

[JohnR]

It's not an ABX box.

(Man it's hard to even get the tiniest fact through in these discussions.)

[avahifi]

Hi John,

Please tell me why our box is not an ABX box?  Is simply misnamed?  If so what should it be called?

It does allow double blind testing.  What am I missing?

Thanks in advance for your response.

Frank

[*Scotty*]

I think what JohnR is referring to is that the ABX Company designed and sold the ABX Double Blind Comparator System. AVA builds a device which allows ABX testing to be performed. Link to ABX home page http://home.provide.net/~djcarlst/abx.htm
Scotty

[JohnR]

Hi John,

Please tell me why our box is not an ABX box?  Is simply misnamed?  If so what should it be called?

It does allow double blind testing.  What am I missing?

Thanks in advance for your response.

Frank

Frank, the ABX protocol involved listening first to A, then to B, and only then to "X" which is randomly chosen as either A or B. The listener is then to tell whether X=A or X=B.

From your description, your box randomly selects either A or B each time.

[*Scotty*]

If it randomly choses A or B every time and then switches to X, that could get damn confusing and lead to a null or no difference result in every trial. 
Scotty
 
werd, its stoke, not stoak.
From  Dictionary.com
Stoak\, v. t. [Cf. G. stocken.] (Naut.) To stop; to choke.
Webster's Revised Unabridged Dictionary, © 1996, 1998 MICRA, Inc.

[JohnR]

Well you wouldn't do it like that - the protocol and the analysis would be different.

I suppose that with the AVA box you would leave the indicators on for as many trial runs as you like to get familiar with the sound of each component (randomly selected, but you can always skip). Then turn the indicators off and do another series, noting your judgment of A or B each time.

PS. I thought a stoak was a small animal. Like a ferret.