GoldenSounds passes apparently ABX test for DACs (NOT Really)

[Geert]

Looking at the DAC I don't expect any weirdness at 176kHz sample rate. I would suspect the Susvara (in combination with his ears) a bit more than some minutia coming from an amplifier or DAC or re-sampler.
Transducers usually have the biggest impact and distortion.

I wonder which headphones Sharur used. He didn't mentioned it in his video. If he used the same ones because they're the only ones they're aware of that allow them to detect a difference, than it would have been fair to mention it.

 

[Thomas_A]

Looking at the DAC I don't expect any weirdness at 176kHz sample rate. I would suspect the Susvara (in combination with his ears) a bit more than some minutia coming from an amplifier or DAC or re-sampler.
Transducers usually have the biggest impact and distortion.
Note that both Sharur and Cameron had to find a 'tell' in the song and needed to concentrate on that in order to tell the files apart.
I am sure when just listening to file A and then listening to file B and would have say which was what they might not be able do any better than chance.
It is not that one of them sounded 'obviously' different.

Both used Susvara?

 

[solderdude]

I think you got it there! let's not complicate it.
If he can do the same with say a Chord TT2 and the top Topping DAC.

When he just records the output of both DACs that should be enough input for DeltaWave.

Doing an actual test IRL with those 2 DACs real time is really problematic. Maybe it should have short 'mutes' and somehow the timing differences should be made small enough. Level matching (for both channels) should be < 0.1dB.
Not an easy task to ABX. Also replaying a part of a song could have different 'tells' that may not be present when just listening.

 

[solderdude]

Both used Susvara?

probably not but Sharur most likely used a headphone/earphone the point is that in both cases transducers are used which may or may not play 'nice' around 21-22kHz ...

 

[Thomas_A]

probably not but Sharur most likely used a headphone/earphone the point is that in both cases transducers are used which may or may not play 'nice' around 21-22kHz ...

Well see tomorrow if I find something. Most likely not since my previous attempts with HF IMD did not give anything. At least what I remember. Have a bit better equipment now however. To work ear canal must be loaded with 21 kHz, combined with 14 kHz to give an audible 7 kHz buzzz.

 

[MacClintock]

There will always be outliers, for various reasons though. I wonder if it is only with Susvara HP it is heard? Pressuring high enough HF into the ear canal might introduce non-linearities and perhaps some ear IMD. Not cheating but possible explanation besides hearing 21 kHz.

Sharur, if his test results are valid, did not use the Susvara. He has the LCD-5 and the Truthear Nova

 

[MacClintock]

Sharur is a well known troll/attention seeker.

That is only half of the truth. He has proven excellent listening skills (Harman certified listening test) and knows a lot about audio. Sure, sometimes his judgements are a bit extreme, but he is definitely more than a troll.
Once he ABXed high-res vs red book. I then prepared two files where the original was high-res and the other upsampled from red book and he could not tell them apart anymore, admitting that he was using originally some artefact as a tell.

 

[MacClintock]

Imaging (precise placement of a specific element) in the soundstage or a tiny imbalance is usually the most overlooked but easy "tell" there is. But, all of that relies on the two reference tracks (A+B) and X and the ability to switch really rapidly between them.

Let us assume for a moment that it is not any content above 20kHz, but the difference in imaging. This wouldn't show up in a difference file, right? Although the common lore here is/was that soundstage and imaging is just determined by the transducers, this sound to me like a more probable explanation.

 

[MaxwellsEq]

This wouldn't show up in a difference file, right?

If you doing a difference analysis on the correct thing, yes you will measure it. An image shift change implies a relative difference in level and/or phase between left and right. Measuring systems have sufficient resolution for this. Whether anyone has checked for this, I don't know.

 

[MacClintock]

If you doing a difference analysis on the correct thing, yes you will measure it. An image shift change implies a relative difference in level and/or phase between left and right. Measuring systems have sufficient resolution for this. Whether anyone has checked for this, I don't know.

So one should look for that, phase included, not only on the difference file of SPL.

 

[Blockader]

I wonder which headphones Sharur used. He didn't mentioned it in his video. If he used the same ones because they're the only ones they're aware of that allow them to detect a difference, than it would have been fair to mention it.

LCD-5

 

[Sokel]

What was that test (actual measurements,just not the ones we do here) in a thread comparing an E1DA dac with an SMSL (I think but I'm probably wrong about the later) which showed differences otherwise not evident in the usual tests and no one really cared about?
(not debunked either as far as I remember)

I can't find it,maybe it contains some more data to consider.

 

[dasdoing]

an image shift would suggest a phase difference since the filters resulting in SPL difference between the channels is not likely?!

 

[MacClintock]

an image shift would suggest a phase difference since the filters resulting in SPL difference between the channels is not likely?!

Yes, that was hinted on before, so could anyone using Deltawave (I only have a Mac) create this phase difference file?

 

[Sokel]

Yes, that was hinted on before, so could anyone using Deltawave (I only have a Mac) create this phase difference file?

You mean between the test files?
Here:



Left




Right

 

[MacClintock]

You mean between the test files?
Here:

View attachment 367770

Left


View attachment 367771

Right

Well, they are slightly different well below 20 kHz. Isn't this the right direction of investigation?

 

[Sokel]

Well, they are slightly different well below 20 kHz. Isn't this the right direction of investigation?

Very slightly though,look at the scale.

 

[dasdoing]

You mean between the test files?
Here:

View attachment 367772

Left


View attachment 367771

Right

well, it will average out over the song. what about that small passage where he observed the difference?

 

[Blumlein 88]

I was about to post the phase info when Sokel did. It is about .1 degree barely, and mostly less.

I've been quiet about it, but having already looked in Deltawave I don't see any reason one could hear these different. If he does, he does and that makes this very interesting. So if one wished to narrow down what is being heard there are things that could be done involving him doing more tests. One is to try just one channel instead of stereo or to mono the stereo channels. One could come up with others, but if it is an image shift claimed audible, then that seems a place to start. I have my doubts about an image shift as our acuity for that isn't really very deep in terms of db of difference.

 

[Blumlein 88]

well, it will average out over the song. what about that small passage where he observed the difference?

Here is the difference waveform. Peaks do reach almost to -40 db for short periods of time. The highest of those is not where others have said he was narrowing down his listening.

Here is a zoom in on the largest of those, notice how short the time interval. Bottom of the chart in seconds.