Measurable aspects of sound perception

[trl]

The most important measurement system for the quality of audio-equipment, food, wine, video-equipment etc. is our individual brain

It has not been invented yet a measuring system and the right equipment to measure how food or wine smells or tastes, so they're still using TDS-meters, reflectometers and stuff like that, but human subjects are a must in this area for sure.

Although, for video we have calibration spiders.

For audio, we do have enough good mics and REW or, why not, a good ADC and DeltaW. Not saying that measurements are always perfect and tell the entire story about an audio gear, but if there are audible differences that more human subjects can actually hear, then I'm sure this can be measured with audio equipment as well. After all, similar audio equipment was used in the recording studio, so if it was good for recording it must be good enough to record the output of a speaker and then compare the files on a PC.

 

[Wes]

the brain is not a measurement system - I see what you are getting at tho

What listening tests have you done to compare different gear and.... how did you control for biases?

 

[LTig]

no thx for inviting

First thing before posting again please learn to cite correctly, especially do not replace words within my postings! This changes the meaning to something I have neither written nor intended to say. Therefore I ask you to remove those words you've changed as soon as possible (use the Edit function). Thank you for your cooperation.

 

[audiophile]

Paul from PS Audio talks about Ethan Winer’s null test, difference between wires, snake oil, science, quantum mechanics, and how engineers at PS Audio use repeatable level-matched blind A/B comparisons to evaluate the immeasurable.

From 8:37 to 15:55:

 

[Blumlein 88]

Paul from PS Audio talks about Ethan Winer’s null test, difference between wires, snake oil, science, quantum mechanics, and how engineers at PS Audio use repeatable level-matched blind A/B comparisons to evaluate the immeasurable.

From 8:37 to 15:55:

Sometimes when you appeal to authority, it doesn't help your case. Like this time here. Complete with a bit about quantum mechanics.

 

[audiophile]

Sometimes when you appeal to authority, it doesn't help your case. Like this time here.

That’s a universal answer, I’ll definitely use it in the future

Folks here on ASR constantly ask subjectivists to present an evidence like this and when it is presented it is always not the right test, or not the right authority or just a “lucky coin”. Is Paul lying when claiming that he and his engineers are able to consistently pick the differences in USB cables in blind level-matched tests? Btw he invites everyone to visit their offices in Boulder and listen for themselves. Any ASR members living in the area?

 

[majingotan]

Is Paul lying when claiming that he and his engineers are able to consistently pick the differences in USB cables in blind level-matched tests?

He might be perceiving it since his DAC's distortions are creeping into the audible territory. Then again, without a 19/20 evidence that he can tell the differences between USB cables on a DBT test with a blind/unbiased 3rd party umpire then all his claims are just his own opinion and cannot be considered scientifically credible

 

[audiophile]

He might be perceiving it since his DAC's distortions are creeping into the audible territory.

Ok, maybe, this is at least an attempt to explain what’s going on.

Then again, without a 19/20 evidence

Why not 99/100? Will be more credible then

 

[majingotan]

Why not 99/100? Will be more credible then

Would you put yourself up to DBT A/Bing for 100 times? I'd be over exhausted to the point that I won't be able to focus on the task before I even reach 50

 

[audiophile]

Would you put yourself up to DBT A/Bing for 100 times? I'd be over exhausted to the point that I won't be able to focus on the task before I even reach 50

Absolutely not, I’m getting tired after first 10 minutes. I saw a paper some time ago claiming that 40 min for such tests is the maximum, after that listener fatigue starts to distort the results.

 

[Blumlein 88]

That’s a universal answer, I’ll definitely use it in the future

Folks here on ASR constantly ask subjectivists to present an evidence like this and when it is presented it is always not the right test, or not the right authority or just a “lucky coin”. Is Paul lying when claiming that he and his engineers are able to consistently pick the differences in USB cables in blind level-matched tests? Btw he invites everyone to visit their offices in Boulder and listen for themselves. Any ASR members living in the area?

No we aren't asking for "evidence like this", there is no evidence there. Some questionable claims. And yes, I am saying Paul is not able to pick out USB cables blind. He of course has said he won't accept Ethan's challenge. So he isn't going to let you test him on it. Again where is the evidence?

 

[Blumlein 88]

Absolutely not, I’m getting tired after first 10 minutes. I saw a paper some time ago claiming that 40 min for such tests is the maximum, after that listener fatigue starts to distort the results.

Hey 19 of 20 is good. You can do it on two consecutive days. 10 each day if you wish. If he already does blind testing as part of his design there, he'd be used to it anyway. If they do that with him and his employees they must already have a setup for that purpose. And I'd like to know how many trials they normally do for themselves. I'll not hold my breath for those answers however.

 

[Wombat]

I find the uncalibrated fingers test a hoot.

Pauls hearing test..

 

[audiophile]

No we aren't asking for "evidence like this", there is no evidence there.

Every time someone mentions that he can hear differences between components people ask to do a blind level-matched comparison, ask a friend to switch components without telling what’s playing. Well, Paul did just that, and been practicing it with his team for couple decades as a part of their design process. Blind or non-blind - they still hear the differences. He mentions that the results of these tests are good enough for their team to make design decisions, but they have no time and motivation to do a full-blown research trying to figure out which measurements can explain what they hear. He hopes that some open-minded scientist will do such research some day.

Hey 19 of 20 is good. You can do it on two consecutive days. 10 each day if you wish.

For me short A/B tests work when the differences are huge, or when I know exactly what to listen for. In case of subtle differences I personally prefer much longer tests. I spend at least a week with a single component just to have a chance to understand how it affects or doesn’t affect the sound. The science that talks about limited bandwidth of consciousness perfectly explains to me why this approach is better.

 

[Blumlein 88]

Every time someone mentions that he can hear differences between components people ask to do a blind level-matched comparison, ask a friend to switch components without telling what’s playing. Well, Paul did just that, and been practicing it with his team for couple decades as a part of their design process. Blind or non-blind - they still hear the differences. He mentions that the results of these tests are good enough for their team to make design decisions, but they have no time and motivation to do a full-blown research trying to figure out which measurements can explain what they hear. He hopes that some open-minded scientist will do such research some day.


For me short A/B tests work when the differences are huge, or when I know exactly what to listen for. In case of subtle differences I personally prefer much longer tests. I spend at least a week with a single component just to have a chance to understand how it affects or doesn’t affect the sound. The science that talks about limited bandwidth of consciousness perfectly explains to me why this approach is better.

Yes, sorry, the limited bandwidth does not at all fit as an explanation for what you experience long term. I've read your prior posting on it, and it does not fit.

I don't know what Paul did, I know what Paul said. I'd like to know Paul's procedure for his design decisions, and assuming it isn't obviously faulty, let him use what he is used to and show he can hear what he claims he can hear. Paul's not about to do that. His claims are nothing, but claims. Some of his claims are very dubious given how things work. Burden of proof is on him.

 

[solderdude]

Paul McClown of BS Audio is a master at mixing truths with 'theories' he has and has things to sell.
Often you need electronics knowledge and experience to tell what are theories and is 'real'.

He mentions that the results of these tests are good enough for their team to make design decisions, but they have no time and motivation to do a full-blown research trying to figure out which measurements can explain what they hear. He hopes that some open-minded scientist will do such research some day.

This is a joke. If he were really on to something there would be vids with proof and there would be conclusive evidence. Afterall design decisions are made on this so there must be logs and documentation of it.
He would also gladly take up Ethan's challenge or let some of his co-workers with special hearing abilities prove Ethan is wrong and make a nice video of it showing the events. There is none of that. There is only truths mixed with theories.

How does Paul explain people also hear (obvious and clear) differences even when there are none ?

 

[audiophile]

Yes, sorry, the limited bandwidth does not at all fit as an explanation for what you experience long term.

Why not? Let’s say there is a certain distortion in the sound that a trained listener can spot in a quick test. But someone, who is not trained what to listen for, may spend many days, weeks or even months until he can figure it out, because his brain will be wandering almost randomly consciously focusing on different aspects of the sound at any particular moment. I am not aware of any better explanation of this phenomena than limited bandwidth of conscious mind.

 

[audiophile]

How does Paul explain people also hear (obvious and clear) differences even when there are none ?

I’d guess he would accept that this happens sometimes. Anyone can experience it: just listen to a 10-second fragment of a song and focus on the vocals during the first listen and on bass during the second one - you will hear new stuff, even when nothing was changed.

 

[Wombat]

I prefer "it can be proved" to "let's say".

 

[solderdude]

I’d guess he would accept that this happens sometimes. Anyone can experience it: just listen to a 10-second fragment of a song and focus on the vocals during the first listen and on bass during the second one - you will hear new stuff, even when nothing was changed.

exactly... now extend this to the individuals that have changed their cable or whatever controversial tweak and here 'more' and then decide it is definitely better and quickly run to Reddit and post their experiences. How credible are their findings ? Should we give those people the benefit of doubt and assume they actually heard an improvement and those that do not expect it and don't hear any improvement should also be victim to bias ?

When one wants to really know and confirm findings more rigid testing is needed. Rarely to people do that and the fact they heard a change is enough for them. After all they read it everywhere on the internet searching for the next upgrade.