GoldenSounds passes apparently ABX test for DACs (NOT Really)

[sonitus mirus]

So either he has really amazing listening capabilities or he is fooling us big time.

Yep

 

[Axo1989]

The test I proposed is to see if one can hear the uploaded music samples (he used music not test tones) and bring the disputed part of the frequency range into the range where even old farts can detect differences and then ABX to see if one can detect differences with that recording.

It is not to debunk anything. Just a method that could be used.

I'm not fussed myself. I can't hear it anyway (I am DAC-deaf)

Good idea actually. I expect I'm also DAC-deaf though.

I was hoping (as we all were no doubt) that the listening tests which are the theme of this thread were going to provide an aural window into the hypnotically beautiful world of audio plankton. I suppose they could be caught up in the sieves of different filters for analysis, but I want to see (or hear, or hear about) them cavorting delightfully and swimming free.

 

[restorer-john]

I am not suggesting that the YouTuber is using a similarly simple trick to deceive his viewers, but everything should be done to prevent such doubts from arising in the first place.
So, for example, show the test procedure in the video (perform the ABX-Test before the camera) and give a description of how others can also replicate the result, i.e. what to look out for in the music sample and which is the best place in the music sample to differentiate.

It's been patently obvious to me that any ABX 'test' even performed in a one-take video can be gamed.

step one: identify the measureable difference.
step two: run a live ABX with headphones, speakers whatever in real time with a spectrogram/fft running live out of camera shot. With something this obvious, you could have a bandpass filter set to a threshold trigger point to make it easier.

The Foobar ABX results are only useful to the person that took them- not anyone else. With people on ASR displaying them like an "audiophile golden ear badge" or "my ears are better than yours", I don't participate in public "post your ABX results" threads anymore.

 

[MacClintock]

It is starting to become more suspicous still, our all well known ex-member Sharur claims to have passed the ABX test as well with 10/10:

 

[Coverpage]

Are they ever compared in a way that there is actual evidence of this sound difference? I haven't really seen that. All I've seen is people making unsupported claims.

I agree, my whole post is agreeing with you that we should test that.

There’s no science that tells us that two different measuring equipment will have to sound the same. If you’re suggesting it hasn’t reach the threshold for audibility, that in itself is an unsupported claim.

 

[MacClintock]

I agree, my whole post is agreeing with you that we should test that.

There’s no science that tells us that two different measuring equipment will have to sound the same. If you’re suggesting it hasn’t reach the threshold for audibility, that in itself is an unsupported claim.

There are well established and thoroughly investigated facts about human hearing, for example that a range of 115 dB is far than enough to cover everything possible.

 

[Coverpage]

There are well established and thoroughly investigated facts about human hearing, for example that a range of 115 dB is far than enough to cover everything possible.

Yup and that’s not what is being discussed.

 

[Sokel]

I had a look at Holo May hoping to find any limitations about hi-res playback often tied to R2R designs so to explain any difficulty or glitch but no.
Goes up to PCM 1536Khz and DSD1024,the 176.4Khz task is a piece of cake.
I then had a look at it's filters and also nothing really strange:



It's THD+N Ratio is also beyond doubt even measured with 90Khz bandwidth by Wolf so no added garbage up high:


(link)

I don't know where else to look.

 

[VientoB]

I agree, I think. The assertion that someone of his age can discern subtle differences around 21kHz is quite extraordinary.

People seem to be obsessing about his age, but what evidence do we have that this is so unusual?

 

[Geert]

People seem to be obsessing about his age, but what evidence do we have that this is so unusual?

A study of 352 human subjects between 10 and 65 yr old having clinically normal-hearing thresholds. The graph includes error bars which represent the spread for 95% of the participants. You can see for yourself how exceptional it is to have a meaningful level of hearing above 20kHz.

https://www.researchgate.net/publication/221903004_Behavioral_Hearing_Thresholds_Between_0125_and_20_kHz_Using_Depth-Compensated_Ear_Simulator_Calibration

 

[VientoB]

A study of 352 human subjects between 10 and 65 yr old having clinically normal-hearing thresholds. The graph includes error bars which represent the spread for 95% of the participants. You can see for yourself how exceptional it is to have a meaningful level of hearing above 20kHz.

View attachment 367520

https://www.researchgate.net/publication/221903004_Behavioral_Hearing_Thresholds_Between_0125_and_20_kHz_Using_Depth-Compensated_Ear_Simulator_Calibration

What is “n”? I am not very familiar with statistics and standard deviations. I find it difficult to interpret that graph so that it tells me the percentage of people that could hear above 20kHz.

 

[voodooless]

What is “n”? I am not very familiar with statistics and standard deviations.

The number of participants, most likely.

 

[VientoB]

The number of participants, most likely.

Ah, probably. What I find interesting is the threshold drops from 19 to 20 kHz.

I could certainly hear 21kHz when I was 18 as I’ve already said in this thread.

 

[Ken Tajalli]

A study of 352 human subjects between 10 and 65 yr old having clinically normal-hearing thresholds. The graph includes error bars which represent the spread for 95% of the participants. You can see for yourself how exceptional it is to have a meaningful level of hearing above 20kHz.

View attachment 367520

https://www.researchgate.net/publication/221903004_Behavioral_Hearing_Thresholds_Between_0125_and_20_kHz_Using_Depth-Compensated_Ear_Simulator_Calibration

i.e. his ability to hear 21kHz at 100dB or more, has nothing to do with him being able to distinguish the tracks which contain music with 20kHz contents at less than 40dB below midrange level (approx).

 

[VientoB]

i.e. his ability to hear 21kHz at 100dB or more, has nothing to do with him being able to distinguish the tracks which contain music with 20kHz contents at less than 40dB below midrange level (approx).

We don’t know what volume he was testing his hearing at. When I did the hearing test when I was 18 it certainly didn’t seem anywhere near 100db at more normal frequencies and is volume not kept constant for that test?

 

[raest]

If it is being suggested that the whole thing has been faked:
- then let's just move on, it is all lies!
- Or, we don't like the results, so we use some excuse to accuse him?
I like to believe ir is not fake. Or we are all wasting our time here.

WTF

no. the issue is that the test has (nearly) NOTHING to do with the claim. it's pretentious clickbait trying to 'prove' subjectivists had it right all along, when in fact to anyone with any kind of audio knowledge and at least 2 working neurons, this test proves the contrary

 

[Ken Tajalli]

WTF

no. the issue is that the test has (nearly) NOTHING to do with the claim. it's pretentious clickbait trying to 'prove' subjectivists had it right all along, when in fact to anyone with any kind of audio knowledge and at least 2 working neurons, this test proves the contrary

. . . those bloody subjectivists!
If only you could get your hands on 'em . . .

 

[Geert]

i.e. his ability to hear 21kHz at 100dB or more, has nothing to do with him being able to distinguish the tracks which contain music with 20kHz contents at less than 40dB below midrange level (approx).

If 100dB is the hearing threshold for 20kHz and the tracks midrange is more than 40dB higher, then he would need to listen at a volume where that midrange is at least 140dB to hear 20kHz frequencies. Can his equipment handle that? If so, how good do you think his hearing still is if this is how he treats it? And if these are the levels that are needed to hear a difference (at one particular spot in a track), then how relevant is this test for the average audiophile.

 

[Ken Tajalli]

If 100dB is the hearing threshold for 20kHz and the tracks midrange is more than 40dB higher, then he would need to listen at a volume where that midrange is at least 140dB to hear 20kHz frequencies. Can his equipment handle that? If so, how good do you think his hearing still is if this is how he treats it? And if these are the levels that are needed to hear a difference (at one particular spot in a track), then how relevant is this test for the average audiophile.

absolutely!
hence, it could not be the 21kHz audibility.
Look for another reason.

 

[ads_cft222]

For a 20khz fundamental there are harmonics at lower frequencies

For instance, a cymbal crash might have a strong element at 20kHz, but it will also have harmonics at 10kHz, 5kHz, and so on, which contribute to the bright, sharp sound we perceive.