Originally Posted by arnyk
Test fixture for what test?
This one Arny:
Originally Posted by arnyk
If the parties to these forums were all in the same metro area, it would be feasible to set up an ABX test and cut to the chase.
You said it was feasible to set up an ABX test. I am asking for an explanation of the set up for that test so that people know what to build.
Amir, you are in my opinion most definitely not an appropriate or relevant spokesperson for the pro-DBT viewpoint. The AVS record shows little but both categorical and personal criticism and also no hands-on involvement with DBTs, ever. The leading stumbling block to setting up DBTs is access to equipment related to the issue and also access to audio systems and listeners of such a nature that they would be acceptable to the other side. If a person who claims intimate involvement with a high end audio dealer/installer is not such a person, I don't know who would be.
I am sorry my posts come across to you that way Arny. I am a strong supporter of blind testing when it comes to audio. Let there be no doubt about that.
As to Madrona Digital, I will venture to say there is not one other system integrator's web site that talks about double blind testing on their web site as much as we do. Here are some examples: http://www.**************.com/Showroom/HomeTheater.html"You may have heard of room optimization logic that exists in traditional mass market products. Alas, while they can improve the audio at times, their capability is highly limited due to cost restrictions. Blind testing done by Harman, the parent company of JBL, shows that some of these systems are actually worse than doing nothing at all!"
Here is another example: http://www.**************.com/Showroom/Revel%20Performa3%20Speakers.html"When we put listeners in (blind) controlled testing, we discover that there are key traits that lead to the listener voting one speaker more realistic than another."
And http://www.**************.com/Products/architectural.html"As with other Revel products, they are fully engineered in house and go through rigorous blind testing to make sure they outperform all competing products."
But wait, there is more
: http://www.**************.com/Products/audio/audio.html"Revel speakers are at the pinnacle of sound reproduction. Their extremely low distortion provides a superbly enjoyable musical experience. Designed using objective double-blind studies..."
I hope this eases your concern regarding my stance on blind testing. If I say the above to our potential customers, you better believe that it is core to my audio beliefs. I have written all the pages on our web site by the way.
This statement: " I worry however that we lose credibility as a camp when it is used hoping the other side doesn't question it, showing that we have not used the very tool we advocate. If you don't care about us losing objectivity that way, I understand. " IMO starts out with a misplaced we and goes downhill from there.
Appreciate your concern Arny but I think you are equating me questioning something you said as being synonymous with questioning validity of double blind testing. I don't know how you arrive at that. At the risk of stating the obvious, one is a methodology and the other a person
. People should be free to question us without the implication being that we throw them out of the camp if they do. This is the nature of this forum and this subforum specifically. We question each other and discuss the topic.
I live by a line I learned when I was a teenager: "If you can't criticize something, you don't know it well enough!"
Years ago, I made the mistake of accepting the offer to write a book
once about Unix. In there, I point out flaw after flaw in its instrumentations for determining the computer performance. To make sure no one takes that criticism as me not liking my beloved Unix operating system (grand daddy of Linux), I wrote that very line in the preface.
It is my strong belief that we need to be fully transparent about any flaws that exist in our camp. The limitation of any test needs to be front and center before results are accepted. In this argument it appeared to me that we were try to imply there were ABX test results when in reality it seems there were none. And asking how one would create such a test has turned into me having to spit about the above
. Let's be upfront about these things. If we have never created an ABX for passive bi-amping, let's say it. Implying that it is easy for folks to get together and run it when we have not comes across as non-sincere. I apologize for using that word
. But that is how I read it and I worry that others will too.
But to address this new made-up issue, I see no evidence that anybody who is pro DBT who hopes that they won't actually be used. I would love to see DBTs be used more frequently and I have encouraged their usage in any way that I could reasonably do so.
I too have the same wish. The problem is that we mention the words ABX about a million times more than actually showing their results. Current discussion is a good example. A bunch of back and forth on ABX but no test results related to the topic at hand. Do you believe that is the right way to discuss science?