Conclusive "Proof" that higher resolution audio sounds different

[arnyk]

Right, then the other ABX test could be flawed or the devil is in the details of what's being tested?

Cutting to the chase, the devil is in the details of what is being tested.

At least you are open to some different possibilities that the evidence suggests.

Well, one of the possibilities is if you actually read it, totally open-ended. Therefore the meaning of the following statement is hard to understand:

But there are more possibilities than you have listed. I too am interested in how all this pans out & what the conclusions are - if conclusions are reached.

I agree that we've been seeing a major rush to judgement.

I've got about two years experience conferencing with the relevant parties and I'm not the least bit surprised.

Are you talking about me damning ABX or Tim (Phelonious) damning it as being of no consequence for ordinary listening?

Let's leave personalities out of this.

My opinion is that ABX has it's place as do measurements as does long-term listening - they are each of particular sensitivity in revealing articular aspects of the playback chain

My opinion that listening tests work best the fewer constraints you put on the listeners. A certain number of basic controls such as controlling relevant variables and trying to minimize bias are necessary. But beyond that, give people the chance to live with the test over a period of time in their own context and try their own thing.

At this point the myth that short selections are necessarily bad has again been spiked. The myth that small differences can't be heard in an ABX test has been spiked. Again.

 

[jkeny]

I would change that slightly, Arny:

Didn't the people who successfully ABXed the keys-jangling files have to listen to them for a long term until they started conciously hearing the difference?

 

[arnyk]

I already reported, I think, one guy relating how he couldn't hear any differences in Winer's 20 pass generational file & it was only when other forum members started to report they could that he actually made a more concerted effort & found out he could. He didn't know how but the ABX results proved again & again that he intuitively picked the correct file 100% of the time. He moved down to 1 pass & this also was 100% correct. He learned from this that his negative bias was such a strong bias that it was preventing him from hearing 20 pass differences.

Can you provide a URL to the posts in the conference where this occurred?

 

[arnyk]

I would change that slightly, Arny:

Didn't the people who successfully ABXed the keys-jangling files have to listen to them for a long term until they started consciously hearing the difference?

I think that having some insights into people's unconscious thoughts may be possible, but discerning them reliably at at that level of detail has proven to be very difficult.

 

[jkeny]

Can you provide a URL to the posts in the conference where this occurred?

Conference? Are you looking for submitted papers, Arny?

 

[Phelonious Ponk]

Tim,
Back up the wagon a bit, there.
You were the one who started the conversation about ordinary (long-term) listening by saying that this test was of such difficulty that you doubted it had any significance for ordinary listening. You are now trying to say I made the leap in my assumptions based on the evidence here, when, in fact, it was you who made the false conclusions that I & Orb have been trying to correct ever since.

I didn't begin the dialogue about long-term listening, John, but I do believe that long-term listening to music will deliver a completely different result than Amir's clinical analysis has. I believe what Amir hears under close scrutiny will be inaudible to average, untrained audiophiles during casual listening, or even during normal critical listening, because while you seem to be saying that Amir's methods are creating the difficulty in differentiating the files, I believe the difficulty of differentiating the files is necessitating Amir's training and methodology.

Do you agree with how I've characterized your position (see bold)? Do you accept that I am allowed a different point of view? Then I believe we're done here.

Tim

 

[jkeny]

I think that having some insights into people's unconscious thoughts may be possible, but discerning them reliably at at that level of detail has proven to be very difficult.

You could be right?

 

[arnyk]

Conference? Are you looking for submitted papers, Arny?

Conference as in online forum. What was it, Gearslutz or something like it?

 

[arnyk]

I didn't begin the dialogue about long-term listening, John, but I do believe that long-term listening to music will deliver a completely different result than Amir's clinical analysis has. I believe what Amir hears under close scrutiny will be inaudible to average, untrained audiophiles during casual listening, or even during normal critical listening, because while you seem to be saying that Amir's methods are creating the difficulty in differentiating the files, I believe the difficulty of differentiating the files is necessitating Amir's training and methodology.

Do you agree with how I've characterized your position (see bold)? Do you accept that I am allowed a different point of view? Then I believe we're done here.

Tim

The outcome that we've seen before is more like:

"OK, I can hear the difference under highly idealized conditions but I'd never even look across the street to see a difference this small".

 

[jkeny]

I didn't begin the dialogue about long-term listening, John, but I do believe that long-term listening to music will deliver a completely different result than Amir's clinical analysis has.

Sorry to differ, Tim but you did begin the dialogue by saying that ordinary listening (which is long term listening) would make these results meaningless as they required training, etc to hear

I believe what Amir hears under close scrutiny will be inaudible to average, untrained audiophiles during casual listening, or even during normal critical listening, because while you seem to be saying that Amir's methods are creating the difficulty in differentiating the files, I believe the difficulty of differentiating the files is necessitating Amir's training and methodology.

I'm saying the forced choice test itself creates the difficulty & Amir's method of overcoming that difficulty demonstrates this. Arny made a good point - the section to focus on in these tests had first to be identified by long-term listening, right?
What I think you are saying is that casual, ordinary listening will not pick up these differences & what I'm saying is that over the long term it actually will - maybe just as a dissatisfaction with the sound or an unconscious uneasiness.

Do you agree with how I've characterized your position (see bold)? Do you accept that I am allowed a different point of view? Then I believe we're done here.

Tim

I've expressed & explained my position, Tim.

 

[jkeny]

Conference as in online forum. What was it, Gearslutz or something like it?

Oh, OK - PinkFishMedia - I'll try & find the posts but it is such a wayward, meandering thread (much like the AVS one) that I may not be successful.
My memory is that a poster could unconsciously reliably identify the differences (between Winer's loopback files) but was not able to state what that difference was.

 

[jkeny]

I found the post in the other forum & thought I posted it here
Was it removed?I'm quite happy to not confuse this thread as it is Foobar ABX test for Winer's Loopback test
So I'll start another thread if you want to discuss it there

 

[arnyk]

I found the post in the other forum & thought I posted it here
Was it removed?

I think I found it here:

http://www.pinkfishmedia.net/forum/showpost.php?p=2368120&postcount=606

The test fails the requirement that the number of trials be planned before the results are analyzed.

There were 3 different tests and after the fact, and after each one was analyzed, they were added together.

The person making this post should be invited to redo his work as one ABX test with a number of trials that he specifies in advance of the test.

 

[jkeny]

I think I found it here:

http://www.pinkfishmedia.net/forum/showpost.php?p=2368120&postcount=606

The test fails the requirement that the number of trials be planned before the results are analyzed.

There were 3 different tests and after the fact, and after each one was analyzed, they were added together.

The person making this post should be invited to redo his work as one ABX test with a number of trials that he specifies in advance of the test.

I've started another thread so as not to pollute this one, Arny

 

[amirm]

Didn't the people who successfully ABXed the keys-jangling files have to listen to them for a long term until they started hearing the difference?

Not in my case. For some tests, I could instantly tell from the first second or two of the clip. For others, I would randomly pick a segment and analyzed it.

 

[treitz3]

Tom, tubes will last that long, I agree with you BUT what I'm talking about here is that your only going to get about 800 hours of *GOOD SOUND* out of those 10K hours...

Hello, Audio_Karma. We will have to agree to disagree on that one. All of my tubes are well beyond the 800 hour mark and they all still sound superb. If this is what you are being told, I would personally question the source in which it came. If this came from personal experience I couldn't begin to explain as to why (bad/knockoff/cheap tubes perhaps?).

Tom

 

[esldude]

I'm saying the forced choice test itself creates the difficulty & Amir's method of overcoming that difficulty demonstrates this. Arny made a good point - the section to focus on in these tests had first to be identified by long-term listening, right?
What I think you are saying is that casual, ordinary listening will not pick up these differences & what I'm saying is that over the long term it actually will - maybe just as a dissatisfaction with the sound or an unconscious uneasiness.

Rather curious thinking here in my opinion. A test can be easy or more difficult. Not surprising that with increasing difficulty where only portions are on the edge of being perceptible one gets results by narrowing the scope, and the length of what is being listened to for comparison. Versus the easier test where differences are larger and easier to hear where paying even a little attention the difference is readily apparent. You are taking just that situation and saying the fact it was a hard enough a test to require such narrowing of scope and focus means the test methodology itself was creating the difficulty. You are saying since it was close enough to not being perceptible that proves the methodology created the difficulty. That involves a leap of logic (or faith) that appears without support.

We already know sticking with an ABX methodlogy, that little focus is needed to have it work for large obvious differences. That smaller differences require more focus. That even smaller differences get detected with shortened focused attention though still using ABX methodology. Also that small though detectable differences can get missed by overly long and casual listening during ABX testing. All of which directly support the idea that the reason these files are difficult to detect as different is due to them being barely perceptible.

To jump in at this point and say this testing methodology caused the difficulty goes against most evidence. You need evidence that the long term listening not only allows these differences to be heard, but heard more easily, and more reliably. The anecdotal reports aren't sufficient for this.

 

[Orb]

.....
Either people are hearing this difference, or they are being too tolerant during the IM test or there's something completely new and different going on here.

Just to clarify it has been proved Arny IM is NOT an issue (context this thread) in several posts here including measurements by JA using your IM test tones, unless one pushes the amp into clipping/strong distortion which as I keep saying many of your tests are designed to do.
Your really breaking test comparison guidelines to hear IM, and using poorly designed-engineered equipment should be avoided anyway by listeners as they have non-linearity in many ways/unpredictable behaviour.

No need to debate this because the facts (with measurements) have been laid out earlier showing why IM is not causing the issues you state for most listeners, and those posts align fully with Dr David Griesinger, what Amir shown checking for IM with his gear, and a few others also did a check for IM who passed the ABX using the check you provided (those who failed IM hardware tone check I noticed mentioned turning the volume very loud hence stress testing the amp).

Thanks
Orb

 

[jkeny]

Rather curious thinking here in my opinion. A test can be easy or more difficult. Not surprising that with increasing difficulty where only portions are on the edge of being perceptible one gets results by narrowing the scope, and the length of what is being listened to for comparison. Versus the easier test where differences are larger and easier to hear where paying even a little attention the difference is readily apparent. You are taking just that situation and saying the fact it was a hard enough a test to require such narrowing of scope and focus means the test itself was creating the difficulty. You are saying since it was close enough to not being perceptible that proves the methodology created the difficulty. That involves a leap of logic (or faith) that appears without support.

No, I'm saying that the test is usually passed by focussing on specific parts of the audio file. The finding of that specific part is difficult to start with & requires perhaps a bit of training & exposure to other audible differences in the past. This is the normal approach & it's not easy for the casual listener to listen in this new, more focussed manner & to be able to decompose what he's hearing into specifics that can be cross-compared with the other audio file. This is all the training needed before doing the test itself - Amir, has practise & doesn't need much of this training

I'm also contending, although there is no evidence for this, that long-term listening will be more likely to allow differences to be heard. I'm talking about weeks of listening here, not minutes. In other words living with the sound & then living with the sound from another file - this, to be honest, is usually done with devices & not files - I really can't imagine doing it with files.

What I'm saying is that there is more audible differences in sound than A/B testing (usually quick A/B) will reveal & what will be revealed in A/B testing will (just as easily?) be revealed in long-term listening

 

[Phelonious Ponk]

You're wearing me out, John...

Sorry to differ, Tim but you did begin the dialogue by saying that ordinary listening (which is long term listening) would make these results meaningless as they required training, etc to hear

From page 4. Long before I made the remark above, From Ron Party:

Hi Orb.
Hi Orb.

I suspect for many the whole point of the exercise is to ascertain whether one can reliably and repeatedly differentiate with real music. I mean, isn't that what Audiophiles claim, that they can reliably and repeatedly do so? That Amir could do so with jingling/jangling keys is great but we need to know the effect of something like masking. We must not a priori conclude music is not suitable.

As to the use of small segments in music, great. Use whatever one believes will create the greatest likelihood of passing the test. Strike of a cymbal? Great. Decay of a bass note? Great.

Yes, longer listening might be useful in identifying preferences, tolerances and/or threshholds. Of course, that presumes there is an audible difference in the first place.

I would like to see properly designed tests -- plural intentionally used -- using long term listening to music, not just slices thereof.

A good point, by the way. Back to your post:

I'm saying the forced choice test itself creates the difficulty & Amir's method of overcoming that difficulty demonstrates this.

OK. It's not the method of listening Amir uses to differentiate that makes it difficult, it's the testing methodology that makes it difficult. I disagree, but OK.

We disagree. Good enough for you?

Tim