The famous AB test between amplifiers

[MAB]

A lot of people get offended when I say that listening to two twin children for the first time would be almost impossible for them to tell apart and for their mother the task is easy but we think it would be a bit different if we listened to two amplifiers in an A/B test for the first time.
This is just my analogy for this test.
In this case we are not dealing with another language, take it as if they speak their native language and see if you can find differences the first time you hear it.

Geert isn't saying he is offended. He is saying your analogy is incorrect. On top of that, I don't struggle with identification they way you describe.

Back you your OP. Please demonstrate (unsighted, level matched, rapid switching or otherwise) that you can actually hear differences not previously heard in the studies you mention. Otherwise your thought experiment and arguments by analogy aren't useful.

 

[Blumlein 88]

Never really sold anything to a moth, perhaps if we stayed open later.
Keith

Moths really like the glow of tube gear. However, it can be dangerous to them. Plus they usually lack the disposable income.

 

[Blumlein 88]

False equivalence is false. News at 11.

 

[END WAY]

I'm not talking about things she can hear - or at least not about the things that are defined by identical genetics. There are a million other verbal and non verbal/auditory clues she will be picking up on subconsciously also.

It is a meaningless analogy. It simply doesn't apply to a blind test where ONLY the tonality is available.

I'll change the analogy since you say that in this test there is only one variable, which is the tonality, although there are several other things such as the dynamics or the damping factor (let's suppose that one has a factor of 5 vs another of 100) etc...
But let's see, touch 5 people when one of them has 39 degrees Celsius vs all the others who have 37, you can almost certainly identify those two degrees of difference (a single variable as you mention) we live with that temperature range all the time and our brain managed to adapt to be able to differentiate those tiny changes.
Now do the exercise and touch other things that are at 65 degrees Celsius and others at 67 to see if you can identify them quickly.

 

[mhardy6647]

I don't care at all about your sarcasm or MAB's
A moth or an elephant could hear some frequencies better than a human being, but obviously they do so because they have evolved in their environment for survival. I know that some people are not able to understand the concept and that is understandable.

Actually we are in complete agreement. You mistake facetiousness for sarcasm.
We have auditory systems well evolved to do what we need(ed) them to, as do other critters. I just find it endlessly amusing that many folks here check their analytical thinking at the door when they consider UFOs -- but not esoteric cables or harmonic room purifiers... or whatever.

 

[Geert]

Now do the exercise...

This discussion leads to nothing. You have the assumption fast switching is no good, and you believe you have good reasons to think so. All fine. Next step is to do an experiment to see if your assumption is valid and not to try convincing people of your beliefs. That's the scientific approach this forum supports.

So it's up to you to do an exercise. Show us evidence of how another method than fast switching improved identification of differences in sound. To make it easier, you can use devices or tracks which measure differently and go from there.

 

[NTK]

Actually we are in complete agreement. You mistake facetiousness for sarcasm.
We have auditory systems well evolved to do what we need(ed) them to, as do other critters. I just find it endlessly amusing that many folks here check their analytical thinking at the door when they consider UFOs -- but not esoteric cables or harmonic room purifiers... or whatever.

Do you know of an explanation (in terms of evolution) as to why the human auditory system also enables us to enjoy music?

 

[Geert]

Do you know of an explanation (in terms of evolution) as to why music the human auditory system also enables us to enjoy music the human auditory system ?

 

[antcollinet]

I'll change the analogy since you say that in this test there is only one variable, which is the tonality, although there are several other things such as the dynamics or the damping factor (let's suppose that one has a factor of 5 vs another of 100) etc...
But let's see, touch 5 people when one of them has 39 degrees Celsius vs all the others who have 37, you can almost certainly identify those two degrees of difference (a single variable as you mention) we live with that temperature range all the time and our brain managed to adapt to be able to differentiate those tiny changes.
Now do the exercise and touch other things that are at 65 degrees Celsius and others at 67 to see if you can identify them quickly.

Output impedance high enough to impact the audio - would be one of those things that makes the amp poorly designed so doesn't fall into the category of being transparent.

And your new analogy is equally poor - you are talking about differentiating things that are within the capability of the senses. Blind testing is about finding out if what you are detecting is real or if the differences (if they exist) are smaller than your senses are able to detect.

Now - you can carry on coming up with nonsense arguments and analogies, or you can prove we are wrong by taking two amps that measure as transparent, and tellling them apart in a properly controlled blind test. You can listen to them for as long as you like prior to - or during - the test. Until you do though, I'm done here.

 

[voodooless]

But let's see, touch 5 people when one of them has 39 degrees Celsius vs all the others who have 37, you can almost certainly identify those two degrees of difference (a single variable as you mention) we live with that temperature range all the time and our brain managed to adapt to be able to differentiate those tiny changes.

Except in audio, we’re talking about differences of less than .1 deg. Do you think you’d still confidently spot the different temperatures?

 

[antcollinet]

Do you know of an explanation (in terms of evolution) as to why the human auditory system also enables us to enjoy music?

No-one knows but there are hypotheses:

Evolutionary musicology - Wikipedia

en.wikipedia.org

My pet one is that all human art is about knowledge capture.
Storytelling, Song, Dance, Painting, etc - all arose from the need to encapsulate group knowledge in an easy to remember fashion for passing on down the generations prior to the written word.

 

[END WAY]

Output impedance high enough to impact the audio - would be one of those things that makes the amp poorly designed so doesn't fall into the category of being transparent.

And your new analogy is equally poor - you are talking about differentiating things that are within the capability of the senses. Blind testing is about finding out if what you are detecting is real or if the differences (if they exist) are smaller than your senses are able to detect.

Now - you can carry on coming up with nonsense arguments and analogies, or you can prove we are wrong by taking two amps that measure as transparent, and tellling them apart in a properly controlled blind test. You can listen to them for as long as you like prior to - or during - the test. Until you do though, I'm done here.

A/B testing is a matter of perception combined with science, we will never get anywhere. I can tell you that I have done it and you will challenge me on what methodology I have used, what spl, what microphone, how I equalize gains, in short, you will question everything I do, even so, in case you want to know, I have made the comparison between 2 amplifiers. I have had a HK990 for more than 10 years
and recently I have changed to two MC611, leveling the spl with a pink noise and measuring with the multitester supported by the measurement of the microphone used with the Fisher audio software, I measured the frequency curve both equal or at least without visible differences that indicate a deviation
Used speakers, Dynaudio, Elac, Magnepan, Morel and my favorites which are some made by me that use Satori driver, with all of these there is a constant
Do you know what the conclusion is for me? They sound almost the same but the difference is in the mids, which with the MC611 sound a little softer and less cold than with the HK990, this is a little noticeable in the roughness of the voice.

 

[krabapple]

We are underestimating the processing of our brain through our ears, we are the most advanced that exists on the planet

Not in terms of hearing, we aren't.

Your analogy to Chinese speakers is inane -- as if the sonic differences between amps were on the order of different languages -- not to mention borderline racist : I have no doubt plenty of nonspeakers could STILL tell two native Chinese speakers apart by their voices.

You are very confused about 'fast switching', which refers to the speed of the actual switching -- 'slow', if you disconnect wires and reconnect them, 'fast' if you use a switchbox to which both devices are connected. Fast is better because audio memory for detail is short. Short *samples* (aka short audition time) is also better for the same reason. That's why both are standard in psychoacoustic research. As, of course, is double-blind methodology. The last factor for an amp comparison would be careful level matching as well.

But hey, feel free to set up a 'slow-switching' (by which you mean, long sample/audition time) double blind, carefully level matched comparison between two non-broken solid state amps operating within their design limits, and get back to us ASAP with your results. You probably think no one's done it before, which would be another thing you're wrong about.



(I dedicate this post to the spirit of Tom Nousaine)

 

[krabapple]

maybe it is just me -- but I have a sense that the signal to noise ratio of this thread is rather low.

Well it started out with a blast of noise, followed by another later. And here's another:

A lot of people get offended when I say that listening to two twin children for the first time would be almost impossible for them to tell apart and for their mother the task is easy but we think it would be a bit different if we listened to two amplifiers in an A/B test for the first time.

Record their voices saying the same thing. Set up a blind A/B/X, with short audition times (and fast switching -- which are *two different things*). I'd bet you good money your average person could easily tell A from B.

Along with many things, you seem unaware that an an ABX test (or other blind A/B) the listener is free to switch between sample A and B as often as thely like and listen to them as long as conditions permit, before marking "A' or "B' as their answer.

This is just my analogy for this test.

It's lousy.

In this case we are not dealing with another language, take it as if they speak their native language and see if you can find differences the first time you hear it.

Again, it's not like you only get 'one shot' in a blind A/B. It's not a contest.

 

[END WAY]

Not in terms of hearing, we aren't.

Your analogy to Chinese speakers is inane -- as if the sonic differences between amps were on the order of different languages -- not to mention borderline racist : I have no doubt plenty of nonspeakers could STILL tell two native Chinese speakers apart by their voices.

You are very confused about 'fast switching', which refers to the speed of the actual switching -- 'slow', if you disconnect wires and reconnect them, 'fast' if you use a switchbox to which both devices are connected. Fast is better because audio memory for detail is short. Short *samples* (aka short audition time) is also better for the same reason. That's why both are standard in psychoacoustic research. As, of course, is double-blind methodology. The last factor for an amp comparison would be careful level matching as well.

But hey, feel free to set up a 'slow-switching' (by which you mean, long sample/audition time) double blind, carefully level matched comparison between two non-broken solid state amps operating within their design limits, and get back to us ASAP with your results. You probably think no one's done it before, which would be another thing you're wrong about.



(I dedicate this post to the spirit of Tom Nousaine)

My analogy with the Chinese is not absurd, you are distorting it and have even introduced racism, I just want to make it clear that listening to Chinese, Russians, Italians, etc., at first glance it seems that they have the same tone of voice, then you differentiate as you pay attention.

The issue with the rapid switching between amplifiers is what I question with the famous A/B test, with these tests you do not give the brain the opportunity to process and memorize the differences between amplifiers if there are any.

 

[krabapple]

A/B testing is a matter of perception combined with science, we will never get anywhere. I can tell you that I have done it and you will challenge me on what methodology I have used, what spl, what microphone, how I equalize gains, in short, you will question everything I do,

Yes, of course we will. Extraordinary claims demand extraordinary evidence. And also, I would say, in your case, independent verification.

What is your extraordinary claim? You say you 'level matched' with 'pink noise' and did comparative frequency sweeps "measuring with the multitester supported by the measurement of the microphone used with the Fisher audio software", huh, is that how measuring amps is done? , and saw no 'visible' difference in the result; yet you say one amp sounded 'warmer' in the 'midrange' than the other.

You don't say anything about double blinding. Huh.

 

[robh36062]

The Carver Challenge

Is it possible to make a $700 "mainstream-audio" power amplifier sound exactly like a high-priced perfectionist amplifier? Bob Carver, of Carver Corporation, seemed to think he could, so we challenged him to prove it.

www.stereophile.com

Rob

 

[Blumlein 88]

A/B testing is a matter of perception combined with science, we will never get anywhere. I can tell you that I have done it and you will challenge me on what methodology I have used, what spl, what microphone, how I equalize gains, in short, you will question everything I do, even so, in case you want to know, I have made the comparison between 2 amplifiers. I have had a HK990 for more than 10 years
and recently I have changed to two MC611, leveling the spl with a pink noise and measuring with the multitester supported by the measurement of the microphone used with the Fisher audio software, I measured the frequency curve both equal or at least without visible differences that indicate a deviation
Used speakers, Dynaudio, Elac, Magnepan, Morel and my favorites which are some made by me that use Satori driver, with all of these there is a constant
Do you know what the conclusion is for me? They sound almost the same but the difference is in the mids, which with the MC611 sound a little softer and less cold than with the HK990, this is a little noticeable in the roughness of the voice.

Yes bad matching methodology. Mics are not a good way to match levels. Tones with a voltmeter at the speaker terminals is. As pink noise is dominated by low frequencies, one amp rolling off a bit early in the very low end could cause a mismatch that would explain your results.

You also don't mention any blinding. If no blinding, nothing you have to say is credible on relative sound quality.

 

[krabapple]

My analogy with the Chinese is not absurd, you are distorting it and have even introduced racism, I just want to make it clear that listening to Chinese, Russians, Italians, etc., at first glance it seems that they have the same tone of voice, then you differentiate as you pay attention.

And who says proper amp comparisons don't involve paying attention, sir?

The issue with the rapid switching between amplifiers is what I question with the famous A/B test, with these tests you do not give the brain the opportunity to process and memorize the differences between amplifiers if there are any.

Again, you are confusing two aspects of best practice test methodology: the switching interval, and the listening interval. You want the first to be as instanteanous as possible; and research shows that short listening intervals are more revealing than long ones for detecting difference --- both for the same reason: because audio memory for detail is short.*

No any says you only get to listen to A and B once before deciding. So you are proposing a strawman version of an A/B test.

(What's also farcical is how often golden ears claim to immediately hear a difference between amp A and amp B...and attribute that to the amps, without basic method controls. You should really go and challenge them.)



*If the difference is real and so unsubtle that memory for detail isn't an issue, then neither is there much need for 'getting used to' either amp.

 

[Blumlein 88]

My analogy with the Chinese is not absurd, you are distorting it and have even introduced racism, I just want to make it clear that listening to Chinese, Russians, Italians, etc., at first glance it seems that they have the same tone of voice, then you differentiate as you pay attention.

The issue with the rapid switching between amplifiers is what I question with the famous A/B test, with these tests you do not give the brain the opportunity to process and memorize the differences between amplifiers if there are any.

Yet as alluded to by others, the test has been done without the quick switching. The results indicate quick switching is far more discriminating vs longer auditioning periods. Despite how it may seem, unless you have new evidence it contradicts what has already been done many times to test whether quick switching over short auditioning periods is better than longer periods. Short wins.