GoldenSounds passes apparently ABX test for DACs (NOT Really)

[withoutsuit]

I am more interested in his ABX device than DACs.
Do we have info about it?

haha me too - I already asked him in the comment section some hours ago

 

[goat76]

I'd suggest that was Archimago.


JSmith

Archimago is mostly known by other technically knowledgeable people, Golden Sound made the video on MQA that went viral so that even many of the less technically skilled audiophiles started to understand what MQA really was.

 

[CedarX]

I see he's using Deltawave also (best app on the known Universe) and I also saw something like the AP's meters.
Does he uses AP5xx?

Cause if he does that could explain his shift into more objective views.

He does have (or have access to) an APx555-B, see this review as an example:

Ferrum WANDLA GoldenSound Edition Measurements & Discussion

[TableOfContents] DISCLAIMER - THIS IS NOT A REVIEW Design and I/O Internals GoldenSound Edition Changes and Features Measurements [End] [Product:6934312910919] DISCLAIMER - THIS IS NOT A REVIEW This post is NOT a review, and is intended to provide some transparency regarding the objective...

headphones.com

 

[stumat]

Archimago is mostly known by other technically knowledgeable people, Golden Sound made the video on MQA that went viral so that even many of the less technically skilled audiophiles started to understand what MQA really was.

A very skilled audiophile-Amir-showed how his test was misconceived. I suggest you re-read the 18-page thread

 

[solderdude]

haha me too - I already asked him in the comment section some hours ago

He made a video about the ABX tester about 2 years ago:

 

[BDWoody]

Leaving aside the technical details of the test, it seems to me that the title of the video leads to confusion and deception...

It is a GoldenSounds production after all...

 

[OK1]

Spoiler: Apparently he does it by hearing above 20kHz.

MOD EDIT: GS only tests two different filters created with a software program. At no time he tests different DACs let alone passing such a difference.

This video, which I had to repeat sections of, to fully appreciate the conclusions, made me want to post a new thread here. But you beat me to it. Thanks.

For a decade, I've put so much faith in audio interface/DAC specs. Same with microphone preamplifiers, cos my critical application of all this knowledge, is in audio engineering and music production, in my home studio, and in my work as a volunteer in audio departments of churches, I attend.

Based on reviews and published measurements, I bought two dongle devices, the Apple USB-C dongle (bought directly from Apple.com in the United States, and shipped to the UK), and a Samsung USB-C dongle also from Amazon.co.uk. While various reviews, measured them as probably indistinguishable from the main measured criteria - SINAD, and frequency response, I'm both sad and happy to report, these dongles do NOT sound anywhere similar. I did not do any blind testing, as I certainly do not have the resources to achieve that with these dongles. But extensive listening back and forth, over a week, made it clear to me, the Apple dongle had a restricted frequency range and sounded restricted in the stereo field, with restricted bass. So I proceeded to continue to use the Samsung USB-C dongle, which I was quite happy with, UNTIL.

The original reason for investing in these dongles, was as an alternative to my on-board headphone DAC, on my DELL laptop, which had a huge variance between the sound of the ASIO driver (which was shockingly terrible and worse) and the output of the WASAPI driver. I'm on Windows 10. Not too long ago, I refreshed some drivers, on the laptop as advised by Dell's support web site, and one of these was the driver for the Realtek audio device. So I tested the onboard audio, and lo and behold, in my opinion, the WASAPI driver sounded better than the Samsung USB-C dongle. And I switched to that. Only to discover that the ASIO driver, to my ears, further to the device driver update, actually was the most balanced, surpassing the quality of the Realtek WASAPI driver (which sounded shrill and upper frequencies boosted - relative to the ASIO driver).

I have no data to publish, only my own perceptions from listening. And when I saw this video, and a few others, it aligned so closely with my own personal experience, where I had been investing in DAC' s based on measurements, but clearly from my own listening, the measurements do not tell the whole story.

One aspect of the published measurements which were spot on, was that the Samsung dongle, had a higher output than the Apple dongle. I did my best to adjust the level correspondingly, as best as I could, during my listening, using gain plugins in my DAW, as I switched from one device to the other. Used the same headphones in all of the tests.

Regrettably the only way to learn this is by investing and doing one's own listening. Shocked that an on-board DAC, on a laptop from 2017 or earlier, sounds better (with revised audio drivers) than two of the current dongles from major consumer electronic products makers (Apple and Samsung). If anyone had told me this, I would have disbelieved them.

Moral of all this. Those publishing tests, well meaning as they are, cannot test everything. You have to do your own testing. By yourself, I mean listening tests.

This also makes me question, with all due respect, a recent video by Amir, summarising his paper to the AES, which concluded that many DACs used in professional audio, do not measure as well as consumer DAC's. It begs the question, is there something not being measured by Amir's tests which explains why professionals in audio are not switching over to consumer DACs (in spite of the higher measured performance of consumer DACs).

The only way to find out is to compare a professional audio device, to a consumer one, and do a head to head listen. I've a couple of professional and prosumer audio interfaces from Echo and EMU, which measured pretty well for devices of their time, and I'll try to, when I have the time, compare with one of the more recent higher end dongle or desktop DAC's that also measure well. My interest is not in the measurements, but deciding for myself, which one sounds most accurate, to me, which I can rely on the most when working in my home studio, or listening to music/audio recreationally or professionally.

Definitely this video has made me think again. Measurements can be only one aspect of the purchasing criteria, for any audio device., cos the reviewers may not be measuring everything that is material, to what I like to hear.

 

[JSmith]

the measurements do not tell the whole story

Uh huh.


JSmith

 

[PBSequence]

Moral of all this. Those publishing tests, well meaning as they are, cannot test everything. You have to do your own testing. By yourself, I mean listening tests.

Level-matched, blind listening tests, that is. No matter how good you think you’ve done channel matching by hand, minuscule differences will make you prefer one DAC over the other.
Even the video above (leaving all skepticism aside) shows very very tiny differences, far smaller than what level-matching by hand could achieve.

 

[Koeitje]

So he proved that filters can sound different (we knew that) and didn't test any DACs with similar filters. So what does this video actually try to achieve?

 

[FINFET]

What's weird to me about this video is that, Goldensound in many of his previous DAC review videos mentioned how some DACs just sound better with many adjectives but suddenly there are no descriptions of differences about music in this video, such as sound stage, depth, layering, density, musicality, or more detailed resolution and resolving capability, etc. Instead, it becomes very subtle changes in the highest frequency response from reconstruction filters, which you literally need superpowers to perceive, or quote "most people will not be able to hear it, and those who can will find it's too small a difference to care about." That's not what I know about people talking about DACs, night and day differences disappeared, suddenly wife in the kitchen can't hear any differences.

 

[Blockader]

What's weird to me about this video is that, Goldensound in many of his previous DAC review videos mentioned how some DACs just sound better with many adjectives but suddenly there are no descriptions of differences about music in this video, such as sound stage, depth, layering, density, musicality, or more detailed resolution and resolving capability, etc. Instead, it becomes very subtle changes in the highest frequency response from reconstruction filters, which you literally need superpowers to perceive, or quote "most people will not be able to hear it, and those who can will find it's too small a difference to care about." That's not what I know about people talking about DACs, night and day differences disappeared, suddenly wife in the kitchen can't hear any differences.

Isn’t it great that people are changing and becoming more rational in how they approach audio?

 

[Tell]

What's weird to me about this video is that, Goldensound in many of his previous DAC review videos mentioned how some DACs just sound better with many adjectives but suddenly there are no descriptions of differences about music in this video, such as sound stage, depth, layering, density, musicality, or more detailed resolution and resolving capability, etc. Instead, it becomes very subtle changes in the highest frequency response from reconstruction filters, which you literally need superpowers to perceive, or quote "most people will not be able to hear it, and those who can will find it's too small a difference to care about." That's not what I know about people talking about DACs, night and day differences disappeared, suddenly wife in the kitchen can't hear any differences.

I remember him having a video about a streamer if I recall correctly where he talks quite lengthy about jitter, yet there's not any proof that jitter has been an audible problem for many years now, and especially even with that particular streamer (of course I can't remember which one it was). So while he is objective many times he also have some weird side tracks in his videos, just like in this file ABX test video where he talks about DACs and his ABX hardware, yet he doesn't even use that to test DACs with, so I don't even understand why he brought it up.

 

[FINFET]

Isn’t it great that people are changing and becoming more rational in how they approach audio?

Sure, the direction of change is good, but I thought Goldensound knew these things many years ago. If he will changes his subjective system for evaluating DACs, then some previous review videos were either not be genuine or not knowledgeable enough, thus misleading the audience who actually believed in these pink bubble descriptions of DACs and spent their savings based on them.

 

[VientoB]

This video, which I had to repeat sections of, to fully appreciate the conclusions, made me want to post a new thread here. But you beat me to it. Thanks.

For a decade, I've put so much faith in audio interface/DAC specs. Same with microphone preamplifiers, cos my critical application of all this knowledge, is in audio engineering and music production, in my home studio, and in my work as a volunteer in audio departments of churches, I attend.

Based on reviews and published measurements, I bought two dongle devices, the Apple USB-C dongle (bought directly from Apple.com in the United States, and shipped to the UK), and a Samsung USB-C dongle also from Amazon.co.uk. While various reviews, measured them as probably indistinguishable from the main measured criteria - SINAD, and frequency response, I'm both sad and happy to report, these dongles do NOT sound anywhere similar. I did not do any blind testing, as I certainly do not have the resources to achieve that with these dongles. But extensive listening back and forth, over a week, made it clear to me, the Apple dongle had a restricted frequency range and sounded restricted in the stereo field, with restricted bass. So I proceeded to continue to use the Samsung USB-C dongle, which I was quite happy with, UNTIL.

The original reason for investing in these dongles, was as an alternative to my on-board headphone DAC, on my DELL laptop, which had a huge variance between the sound of the ASIO driver (which was shockingly terrible and worse) and the output of the WASAPI driver. I'm on Windows 10. Not too long ago, I refreshed some drivers, on the laptop as advised by Dell's support web site, and one of these was the driver for the Realtek audio device. So I tested the onboard audio, and lo and behold, in my opinion, the WASAPI driver sounded better than the Samsung USB-C dongle. And I switched to that. Only to discover that the ASIO driver, to my ears, further to the device driver update, actually was the most balanced, surpassing the quality of the Realtek WASAPI driver (which sounded shrill and upper frequencies boosted - relative to the ASIO driver).

I have no data to publish, only my own perceptions from listening. And when I saw this video, and a few others, it aligned so closely with my own personal experience, where I had been investing in DAC' s based on measurements, but clearly from my own listening, the measurements do not tell the whole story.

One aspect of the published measurements which were spot on, was that the Samsung dongle, had a higher output than the Apple dongle. I did my best to adjust the level correspondingly, as best as I could, during my listening, using gain plugins in my DAW, as I switched from one device to the other. Used the same headphones in all of the tests.

Regrettably the only way to learn this is by investing and doing one's own listening. Shocked that an on-board DAC, on a laptop from 2017 or earlier, sounds better (with revised audio drivers) than two of the current dongles from major consumer electronic products makers (Apple and Samsung). If anyone had told me this, I would have disbelieved them.

Moral of all this. Those publishing tests, well meaning as they are, cannot test everything. You have to do your own testing. By yourself, I mean listening tests.

This also makes me question, with all due respect, a recent video by Amir, summarising his paper to the AES, which concluded that many DACs used in professional audio, do not measure as well as consumer DAC's. It begs the question, is there something not being measured by Amir's tests which explains why professionals in audio are not switching over to consumer DACs (in spite of the higher measured performance of consumer DACs).

The only way to find out is to compare a professional audio device, to a consumer one, and do a head to head listen. I've a couple of professional and prosumer audio interfaces from Echo and EMU, which measured pretty well for devices of their time, and I'll try to, when I have the time, compare with one of the more recent higher end dongle or desktop DAC's that also measure well. My interest is not in the measurements, but deciding for myself, which one sounds most accurate, to me, which I can rely on the most when working in my home studio, or listening to music/audio recreationally or professionally.

Definitely this video has made me think again. Measurements can be only one aspect of the purchasing criteria, for any audio device., cos the reviewers may not be measuring everything that is material, to what I like to hear.

Your listening test findings don’t seem to correspond with Goldensounds tests. You’re mentioning changes in the bass and treble which do not match. GoldenSound went to great lengths at the start of his video to emphasise that non-blind listening tests are flawed.

 

[solderdude]

GoldenSound went to great lengths at the start of his video to emphasize that non-blind listening tests are flawed.

Ironically this video kind of invalidates most of his reviews that he did earlier as his reviews are all sighted.

 

[MacClintock]

I actually agree with (almost) everything he says... except for the hasty conclusion and 'bridge' towards the incorrect title of the vid.
Just because he was able to distinguish 2 measurable different files does not mean that competent measuring DACs thus still may sound different.
The guy seems to be perfectly on 'our side' about 95% of the time in that video.

Of course he could be kind of right because a DAC that measures perfectly well, with certain filter settings, may not measure that well and be audibly distinguishable when a not so good filter is used.
That filter may well even be a default filter for some manufacturers that want to distinguish themselves or have weird notions based on whatever theory they might have.

Filters can make a difference and some of them could be audible different for young folks.
The conclusion that DACs do sound different is a long stretch and the wrong conclusions as is the clickbait title he gave the video.
That video is going to get a lot of hits, 'fame' and possibly even some small extra income ?

At least he seems to have grown in knowledge after the 'incident' at ASR. Which by the way was a spin-of from the MQA test he snuck in a official file for MQA rendering. I liked that a lot.

The one filter he is using, the extremely sharp one, is apparently the one, or at least very similar to the one in the Chord M-Scaler. So supposedly repeating his ABX test, he might score the same, statistically relevant, result when comparing the Chord Dave with the M-scaler against any other good measuring DAC with a very common "lazy" filter. But still, the main point remains, this difference, while discernable for him, is so itsy bitsy tiny and has nothing to do with other claimed differences in sound, like soundstage or thin mids or whatever, that it wouldn´t be worth mentioning or investing in.

 

[MacClintock]

I see he's using Deltawave also (best app on the known Universe) and I also saw something like the AP's meters.
Does he uses AP5xx?

Cause if he does that could explain his shift into more objective views.

Yes, he does have a AP5xx, and, being the person that he is, made a video about how measurements can be misleading......

 

[Keith_W]

Based on reviews and published measurements, I bought two dongle devices, the Apple USB-C dongle (bought directly from Apple.com in the United States, and shipped to the UK), and a Samsung USB-C dongle also from Amazon.co.uk. While various reviews, measured them as probably indistinguishable from the main measured criteria - SINAD, and frequency response, I'm both sad and happy to report, these dongles do NOT sound anywhere similar. I did not do any blind testing, as I certainly do not have the resources to achieve that with these dongles. But extensive listening back and forth, over a week, made it clear to me, the Apple dongle had a restricted frequency range and sounded restricted in the stereo field, with restricted bass. So I proceeded to continue to use the Samsung USB-C dongle, which I was quite happy with, UNTIL.

Happy to believe you if you perform a level matched blind test and publish measurements showing a difference between the two dongles you mentioned. Otherwise, your anecdote is indistinguishable from all other anecdotes and can be dismissed.

It begs the question, is there something not being measured by Amir's tests which explains why professionals in audio are not switching over to consumer DACs (in spite of the higher measured performance of consumer DACs).

The reason why pro's stick with their DAC's and their pro equipment in general is because of features. You would be shocked at the performance of some of those pro audio DAC's and amps. Some of them even have fans built into the chassis. But it does not matter, when you are providing sound for a concert hall, or an airport, or a stadium. Nobody is going to hear a SINAD of -60dB or noisy fans in your amps. What matters is that they can gang together dozens or even hundreds of DAC channels, manipulate each one, amplify it to earsplitting volumes, and do it with bulletproof reliability.

Just last night I was watching a documentary on the new Las Vegas dome. The resolution of the screen is 3 ppi, which is laughably low compared to your phone which is >200 ppi. So why aren't they switching to giant banks of consumer phone screens? Because that low resolution does not matter when you are sitting that far away from the screen.

 

[ObjectiveSubjectivist]

Ironically this video kind of invalidates most of his reviews that he did earlier as his reviews are all sighted.

Yeah, thats a good one.