Master Thread: Are measurements Everything or Nothing?

[HarmonicTHD]

I'm sorry but I have to ask: why is there a child photo on that page ? What's the point?

I think you need to ask Paul that.

 

[OnLyTNT]

I think you need to ask Paul that.

I didn't want to jump on conclusions and thought I must missing something but I see no correlation and by stating that I already jumped on conclusions I guess.

 

[antcollinet]

I didn't want to jump on conclusions and thought I must missing something but I see no correlation and by stating that I already jumped on conclusions I guess.

It is stated at the end of the article. I don't think we need to risk concluding anything other than that.

 

[Newman]

It is stated at the end of the article. I don't think we need to risk concluding anything other than that.

I saw that and thought it gave no explanation as to why a little girl.

 

[antcollinet]

I saw that and thought it gave no explanation as to why a little girl.

Even so, speculating any other motivation is neither necessary nor desirable.

 

[Newman]

I guess we’ll never know.

 

[Ken1951]

Good luck finding a hifi store that would actually set up a valid test. It isn't in their best interest to show the emperor has no clothes after all.

So far, no one on the planet has provided actual evidence of what you are claiming.



If you can provide evidence that you can consistently identify something that doesn't show up in measurements (by using your ears, not including your eyes or through other clues, i.e. controlled testing), you would be the first ever.



Speakers and transducers in general are much harder to fully characterise, and good luck finding a cheap speaker that can do a flat 20-20kHz at 120dB at <1%THD. High SPL and low extension are where money typically goes with speakers.



It may be that you've had bad luck buying crap or improperly chosen gear, but it's more likely that your comparisons were done in a way that basically guarantees you'd feel that way.





How well do you understand the measurements done? This is the fallacy the high end world counts on people literally buying into.

It was about time for another posting about music, not measurements. $$$$$$ are always better equipment, etc, etc, etc... Frickin' Whack-A-Mole.

 

[Crosstalk]

^^^ Ya still think he's not a troll?

Jim

How did Jim's profile get deleted?

 

[BDWoody]

How did Jim's profile get deleted?

That was his request, and he is welcome back anytime.

 

[Crosstalk]

I guess we’ll never know.

@Bozon left the forum I guess, so we never know.

 

[welwynnick]

Actual DBT tests of ClassD vs ClassA & ClassAB showed them to be identical...
A "good" amp, driven within its design envelope (ie: well matched to the speaker requirements) - will sound the same as any other decent amp doing the same job.

More and more people on ASR seem to be coming to the opinion that all amps sound the same unless broken or clipping.

If there are any superficially scientific tests that show this, then we need better tests instead of a race to the bottom.

A negative result from a scientific test is just a piece of information, it's not actual proof. Many here can't see the difference. I'll give you an example......

 

[welwynnick]

Here's a well-known published paper by Kaoru Ashihara , Shogo Kiryu, Nobuo Koizumi, Akira Nishimura, Juro Ohga, Masaki Sawaguchi and Shokichiro Yoshikawa at various reputable Japanese institutions.

https://www.jstage.jst.go.jp/article/ast/26/1/26_1_50/_pdf

Abstract:
Detection threshold for distortions due to time jitter was measured in a 2 alternative forced choice paradigm with switching sounds.
Music signals with random jitter were simulated on the digital domain.
The size of jitter was arbitrary controlled so that the detection threshold could be estimated.
Professional audio engineers, sound engineers, audio critics and semi-professional musicians participated as listeners.
The listeners were allowed to use their own listening environments and their favourite sound materials.
It was shown that the detection threshold for random jitter was several hundreds ns for well-trained listeners under their preferable listening conditions.

On the surface this looks like a genuine piece of scientific research to those that don't understand science, with a surprising but credible conclusion.
This was gleefully accepted by those on a race to the bottom as proof that large amounts of jitter are inaudible. Not just a piece of information, but actual proof.
There are a few problems though:

1. This paper was popular because it was readily available without having to pay the AES, while proper papers cost proper money.
2. The testers didn't actually add jitter to the test signals, they manipulated the digital data to simulate jitter.
3. The testers did nothing to quantify or minimise the jitter in their baseline, so it's no surprise that their well-trained listeners couldn't distinguish between 250ns of simulated jitter and 250ns of real jitter. It's analogous to testing a Purifi amplifier using a sine wave with 1% THD, which will conclude that the amplifier has 1% THD.

However, because this paper looked and sounded and smelled like science, it was taken to be good science, with a conclusion that could be relied on.
Instead, it was simply a piece of information that in these particular tests, their well-trained listeners could not detect small amounts of jitter.
To get a proper conclusion you have to look at other tests that have found different detectability thresholds - you need sources with low jitter, you need sine waves with low distortion - then you start to get the right conclusions. And the right conclusions come from the best detectability threshold, not from the lowest.
There are lots of instances of this sort of thing. Meyer and Moran is a good example, and set the world of HiFi back several years when they concluded that no-one could hear the difference between 1) CD quality audio, and 2) CD quality audio.

So to get the information we really want - we need to find the best positive test result, not the worst negative result.

 

[depo196]

Hi all. Using measurements, you can quite accurately describe mathematically the operation of an acoustic system. But that is not all. You can also visualize the process of operation of the transducers - this will help you understand not only why, but also how sound is formed. Let me show you the behavior of a dome tweeter on a Klippel laser vibrometer. The video is in Russian, but even without words everything is clear. And this is just the top of the iceberg...
Dome tweeter

 

[welwynnick]

So - what examples are there of good, credible, positive test results. Amir has obliged:

Discrete OpAmp Review: Sonic Imagery vs Sparkos

Recently I reviewed the Nord Three SE 1ET400A Dual Mono Stereo Amplifier. That unit came with the Sonic Imagery's 990ENH discrete operational amplifier ("opamp") in the buffer stage (pre-amplifier). The owner then ordered a set of Sparkos Pro SS2590 discrete opamps. This gave me a chance to...

www.audiosciencereview.com

He performed multiple ABX DBT tests using Foobar200 and was able to reliably distinguish between 24-96 and 16-44 audio. Not only that, he did it with a very high level of confidence - not just 95% confident.

So that's a piece of robust, scientifically derived information that says CD quality audio is not transparent, and the bar needs to be set higher than 20kHz / 93dB.

That's about the level achieved by the AMP10, so that's why I had some doubts about it.

 

[Newman]

So - what examples are there of good, credible, positive test results. Amir has obliged:

Discrete OpAmp Review: Sonic Imagery vs Sparkos

Recently I reviewed the Nord Three SE 1ET400A Dual Mono Stereo Amplifier. That unit came with the Sonic Imagery's 990ENH discrete operational amplifier ("opamp") in the buffer stage (pre-amplifier). The owner then ordered a set of Sparkos Pro SS2590 discrete opamps. This gave me a chance to...

www.audiosciencereview.com

He performed multiple ABX DBT tests using Foobar200 and was able to reliably distinguish between 24-96 and 16-44 audio. Not only that, he did it with a very high level of confidence - not just 95% confident.

So that's a piece of robust, scientifically derived information that says CD quality audio is not transparent, and the bar needs to be set higher than 20kHz / 93dB.

That's about the level achieved by the AMP10, so that's why I had some doubts about it.

If you check the way he aced that in his audiblity video (approx 18m50s timestamp), you will see that it's not relevant to listening to music. This has been mentioned previously. It doesn't 'say' what you think it does about 'the bar'.

cheers

 

[welwynnick]

If you check the way he aced that in his audiblity video (approx 18m50s timestamp), you will see that it's not relevant to listening to music. This has been mentioned previously. It doesn't 'say' what you think it does about 'the bar'.

I'd encourage everyone to watch the whole of that video and read the whole of the ensuing thread. Amir rejects your scepticism :

It doesn't get more real than the way I passed the test. Every listener has a volume control for their headphone. If you know what to look for, have proper training, you can pass the test showing the format is not transparent. No trickery was used as you were suggesting with software manipulation.
Remember, there are people with better hearing than me. So if I can pass the test, others may be able to pass it with even less preconditions.

No, your point doesn't stand. You modified the file if using software which is not allowed in controlled testing. And if you used analog amplification, you had to demonstrate that you could pass such listening test as I did. No pontification about 150 dB, noise-free amplification. I suggest you watch the video again as you have lost the plot here.

And just to re-iterate that he did multiple ABX tests , not just the one in question, and he did it just using a laptop and in-ear buds, so hardly state of the art gear.

One of the reasons I'm sceptical about CD being transparent is because 7kHz sine and square waves are audibly different, which theory says should be impossible.

I can't explain that, except that just because a 20kHz tone is inaudible doesn't mean a 20kHz filter is inaudible - which is the assumption used for the CD spec.

 

[antcollinet]

Hi all. Using measurements, you can quite accurately describe mathematically the operation of an acoustic system. But that is not all. You can also visualize the process of operation of the transducers - this will help you understand not only why, but also how sound is formed. Let me show you the behavior of a dome tweeter on a Klippel laser vibrometer. The video is in Russian, but even without words everything is clear. And this is just the top of the iceberg...
Dome tweeter

Hi - welcome to the forum.

Your link appears to be a paste of the text of your post.

 

[AdamG]

@Bozon left the forum I guess, so we never know.

@Bozon is still an active member and can post anytime he wants. I see no restrictions or removal of privileges.

 

[depo196]

Your link appears to be a paste of the text of your post

Oh I'm sorry, it's destroed. see the link o video please

 

[antcollinet]

Oh I'm sorry, it's destroed. see the link o video please

Well that was fascinating. Never seen anything like it. Is this what we can expect from all tweeters, or only dome tweeters?

What about metal domes - do they behave similarly?

I was particularly intersted in the animation of a "pressed in dome" after it had been pulled out again. I've sometimes seen it said that this won't make any difference - your video clearly showed that it does.

Thanks for posting that.