Suggestion: ban all cable debates

[Jud]

2 hours ago, SoundAndMotion said:

There is no "Blind Testing Police". Temptations to use an inappropriate method does not force it, and anyone doing any test, blind or sighted, should have an idea about what they are doing. That said, take any sighted test you want and add only "blinding", and I'm prepared to argue that although perhaps inconvenient, even prohibitively so, the results will be closer to the truth.

 

Had to smile at the thought of "Blind Testing Police." 

 

Regarding temptation, I thought something GUTB said was interesting:

Quote

while I was doing the ABX, at first I tried to consciously identify the "marker" I had heard during casual listening, and my results showed that I got most of those wrong.

 

I have had similar experiences when taking blinded tests and not knowing anything about what I was doing.  I tried to remember how something sounded in the middle of a musical passage that I'd listened to only 30 seconds or a minute before, and it seemed the harder I tried the more it frustratingly slipped away, like trying to make a fist to hold onto water. 

 

Very many of the blind tests that people want others to take in audio contexts aren't double blinded (the test subject knows the thinking of the person who set up the test), and often neither the person who set up the test nor the subject understand that (1) our sensory memories of sound aren't going to last more than a handful of seconds (loudness differences seem to be an exception; any others?), and (2) a sequence of sounds will confound our sensory memories of any particular sound in the beginning or middle of the sequence, as research by Diana Deutsch and others shows.

 

So yes, I very much agree people doing these tests (and of course people conducting them) *should* know what they are doing.  Often, in what passes for blind testing in the audio world, they do not.

 

With regard to taking a sighted test and adding only blinding: Perhaps, but in the context I've been discussing, where neither the people conducting the tests nor those taking them really know what they are doing, how close to the truth is anyone getting, sighted or blind?  To really get at the truth, we will certainly want to control for any sighted bias; but of course we're not nearly done at that point.

 

I wanted to mention two other points:

 

- Nearly all of these tests require a conscious response.  We may be aware of differences subconsciously ("instinctively") long before we are able to articulate a conscious preference.  (E.g.. the "Iowa gambling task" presents subjects with four decks of cards.  The object is to maximize the gambling return.  Some decks are better than others.  Research has shown elevated galvanic skin response - subjects with the bad decks are perspiring - long before conscious realization that the deck is bad.)  I'm aware of very few tests in audio that seek to eliminate this potential problem, except for the Oohashi ultrasonics experiment that has been criticized by many on various grounds.

 

- I conducted a little test on the forum here (that would never pass scientific muster), that eliminated echoic memory as a factor and attempted to see to what degree the sort of quick switching done in many A/B or A/B/X tests itself may hamper a common audio task, distinguishing one instrument from another similar one.  Have a look; I had someone run some statistical testing on the results, and I'll try to dig that up in the next day or two.

 

https://www.computeraudiophile.com/forums/topic/22709-heres-a-little-test-you-can-all-join-in/#comment-390666

 

[Jud]

1 minute ago, Speedskater said:

In fact a well done ABX test is just the opposite! In that it's way too sensitive. First they have to remove all the uncontrolled variables and then skilled listeners can hear differences of a small fraction of a dB. These differences can be too small to hear in real music listening experiences and way too small to  have a preference.

 

I think SoundAndMotion has said it very well in this thread:

 

Quote

ABX in certain implementations can be great for identifying certain minor differences. So stating it is "useless" is not true. But certain implementations can be rather weak in other situations....

 

[Jud]

17 minutes ago, STC said:

For an example, if you are going to say you could tell the difference between 16/44 and 24/96, then why not  do a ABX with those two samples and another one with 16/44 vs 16/44. 

 

The point is - you will always hear a difference even when listening to  identical samples if you believe there must be a difference. 

 

If you play a sample at 16/44 and one at 24/96:

 

- If there is any difference, can untrained people tell what it is within, say, 5 seconds?

 

- You would want to use music, right?  But then what happens when the notes of the 24/96 sample crowd out all valid sensory memories of what the 16/44 sample actually sounded like (research by Deutsch and others)?

 

So you may easily get a negative response, but you don't know whether it's because humans can't hear a difference or due to your test setup.

 

Then you play a sample twice at 16/44:

 

- Did you instruct the subject to tell you if there were any differences between samples?  Just the word "samples" may be enough of a suggestion to cause a subject to think there is a difference.  What does the subject know about your attitude on the matter?  What is the subject's attitude on the matter?  But let's say everything is nicely double blinded, the subjects have been instructed that the second presentation may be identical to or different from the first, etc.  Then you have determined that people's answers on the tests you ran are unreliable.  Congratulations! 

[Jud]

45 minutes ago, wgscott said:

 

For the same reason I take antibiotics when I get a bacterial infection, rather than sacrifice chickens and pray for recovery.

 

But chickens are better eating.

[Jud]

4 hours ago, wgscott said:

 

More often than not, no:

http://www.ahjonline.com/article/S0002-8703(05)00649-6/abstract

 

 

 

"Get this man to a hospital!"

 

"But he's a Christian Scientist!"

 

"Take him to a Reading Room!"

[Jud]

I think esldude's post is quite fair.  My only small quibbles are to say that I think a lot of the blinded listening tests haven't been more probative than the sighted tests, and a lot of the measurements haven't considered cables "in situ."  But yes, the burden would certainly be on those who think there are objectively verifiable differences to show it, or to simply say they are happy with their subjective choices.  (I'm in the latter category, though I haven't spent a king's ransom on cables either.)

[Jud]

13 hours ago, esldude said:

Sadly the confluence of anti-biotics and chicken consumption is about to take a big hit thanks to advocacy groups pressuring KFC. 

 

https://arstechnica.com/science/2017/04/kfc-phasing-antibiotics-out-of-its-chicken-after-intense-pressure/

 

Depends - Will chicken farmers change the horribly overcrowded conditions that created the necessity for antibiotics in the first place?  They may have to or risk losing flocks to disease.  And there are many successful chicken farmers who provide better conditions already.  So I guess we'll see.  Expect prices to rise, of course.

[Jud]

3 minutes ago, plissken said:

Also the whole SuperDad / Uptone debacle at WBF where Alex said he would do a bias controlled test and then up and disappeared immediately after. 

 

Not this again....

 

Might've had something to do with the whole Amir thing, you think?

[Jud]

Just now, plissken said:

 

It's pertinent. It's certainly a data point.

 

Cue "I can pull an Ethernet cable..." in 3, 2, 1.... 

[Jud]

1 minute ago, Ralf11 said:

 

but medical problems will go down as we will not be ruining the efficacy of our antibiotics

 

There are already bugs that are resistant to everything we now have.  But it'll be nice if they don't immediately gain resistance to the next ones.  Now if we could do something about prescribing them for colds....

[Jud]

1 hour ago, esldude said:

 I needed to edit the above from my prior post.  The tear down was of a Transparent audio speaker cable not MIT.  The time for editing that post had passed before I realized that I had confused the two companies.  Of course the three principles of Transparent worked for a few years with MIT before starting their own company that also had the networking boxes like MIT did.

 

I also think they sound similar, but....

[Jud]

16 hours ago, esldude said:

Unfortunately colds are a virus.  Having previously had a Dr. who always gave anti-biotics the placebo effect on my father's illness was substantial. 

 

Solution: Placebo "antibiotics!"  (Can't be done of course, not ethical.)

[Jud]

16 minutes ago, Danaudio said:

 

 

In the Audio forums the best advice on Equipment is to hear the equipment before buying because everyone hears differently and have different abilities. I guess you are the exception where you already know how a product sounds or doesn't sound by just reading about it on forums and taking sides. 

Don't really understand how you can have an opinion on something when you didn't even try it. Did you buy all your other equipment by just reading reviews or reading about it in forums?

 

He doesn't think cables make a difference.  Not hurting your listening experience.  Don't bug thy neighbor.

[Jud]

1 hour ago, GUTB said:

 

Objectvists need to talk about it, its a necessary component of that belief system.

 

But it was a subjectivist I quoted.

[Jud]

4 hours ago, Danaudio said:

He is taking my time away from listening to music reading his posts.

 

...and it is necessary to defend the honor of your way of listening and that of others like you, I get it.

 

But really, the world's plenty big enough for both, and for both to be discussed with understanding, humor, and, dare I say, perhaps a little humility.  Each of us doesn't absolutely have to be right always.  We can have great conversations about why we think we're enjoying ourselves more doing it the way we're doing it, 'cause that's really the point, isn't it?

 

And if you find you just can't keep yourself from a heated response, that's why the ignore list exists.