Goldensound and his golden ears...

[JJB70]

He seems a genuine enough guy to me, he measures for those who prioritize measurement and offers subjective opinions for those that want that type of review, offering something to the widest audience seems sensible if you are a YouTube channel. I tend to be in the middle myself and see no inherent conflict between valuing both measurement and subjective impressions.
That said, I would love to see a lot more level matched double blind tests as to me it is the most interesting and informative type of test. Not only would it challenge subjective reviewers, it would also challenge those chasing measured metrics.

 

[Hark01]

I was also surprised at first about his detailed subjective opinions, but now I think it's ok, we all have our impressions. His measurements are first class and a great source of information.

 

[garbulky]

Btw, we had a nice discussion on Telegram and I just want to reinforce that if we still disagree, he is a cool fellow

But what if you agree?

 

[Lupin]

I actually think he is a cool guy and I know he posts here from time to time. Buuuuut the amount of differences he claims to be able to hear from dac to dac is bullshit.

So he listens to dacs generating 99,9995% equal signal and yet he perceives differences in soundstage, bass, "tembah", instrument separation and so on. Those speakers of his must have some 0,0001% thd to be able to catch such details and sound so different from dac to dac.

Most likely the unpopular opinion here but... what exactly is the purpose of this thread?

What you say is true for all subjective reviewers. Replace Goldensound with any other random subjective reviewer on YouTube and you can copy that to a new thread.. Is this just to bash the subjective side and rack up some likes at the same time? This has been done many times before, I don't see why this need a new thread.

Goldensound is perhaps not even the worst subjective reviewer as he also does objective measurements and as for as I know the subjective reviews are not changed based on the measurement results. So at least you get both subjective and objective point of view.

And of course every subjective YouTube reviewer can hear the difference between DACs as that is literally their income.
The reviewer will loose his/her viewer base the moment the reviewer admits he/she can't hear the difference between DAC x and DAC y.

 

[antennaguru]

Maybe he won't get caught up in the trap of ranking amplifiers by SINAD, which gives the appearance that the "best" ranking amplifiers will sound better than some other "median" ranking amplifier in the list, when IMO there is really no difference other than hanging your hat on the convenience of a "single figure of merit" - which unfortunately doesn't really exist in the real world of what's audible. In actuality, and IMO, "best" really has to do with the quality of the source material, and NOT 120 dB SINAD versus 80 dB SINAD. BTW, if you're determined to sell to SINAD "purists" all you have to do is jack up the negative feedback to play that "numbers game".

To the comment about tinkering with small car changes in SCCA racing I say sure it's "clean fun" that doesn't matter much in where you finish in a race. After a number of years of racing on the SCCA circuit I feel I can pretty fairly rank where results improvements are generated: 92% driving skill and course knowledge; 6% the right tires; 2% tweaking engine and other stuff. I eventually replaced my need for speed with Masters ski racing - where results are even more about "the driver" than the equipment. In one ski race a couple of years ago I had a great first run on the ski race course and one of my competitors commented that it was because of my really expensive skis. I immediately offered him to swap skis for the second run since we both had the same boot sole length. He took me up on my offer and I beat him by an even wider margin on the second run. Both runs added to our final times, which pushed him even further back in the results. He hasn't whined about equipment since!

 

[Human Bass]

I actually think it's OK to hear some very subtle difference between high-end dacs. Emphasis on very subtle. Now claiming major sonic changes that rustles my jimmies.

Also the clichés of "delta Sigma detailed but cold, r2r musical" and such seem to be very convenient. And never mentioning the limitations of human hearing and perception. Say whatever to want about Zeos, but he is not afraid to say "i can't hear a difference".

 

[Jimbob54]

Now , I can't say that I have been following all his posts. Has he yet been faced with "poor" measurements on a device he subjectively approves of?

Lets say something that measures akin to some of the "classic" Audio GD stuff.

 

[GaryH]

That's just 2 weeks ago, didn't notice is. So things might become interesting. Nice to see he puts some effort into it.

Looks like @Resolve has one too:

Good to see subjectivist-leaning reviewers getting into ABX testing. Now we need the objectivist reviewers to do the same and (fail to) demonstrate an audible difference between an average and high-SINAD DAC/amp.

 

[Jimbob54]

He did buy an ABX board so he probably will experiment with that. At least he's got a sense of humor.

Once he has trained himself he will find out he can't tell the perceived differences blind.
let's wait and see. He seems to be serious about audio measurements.
No idea why he keeps posting linearity plots without the essential filter though and finds his subjective findings valid.

Do those things help you measure/ match levels too?

 

[solderdude]

I have no idea. It looks like a bunch of relays and some stuff to generate random switching.
It can be a very educational tool when used correctly.

 

[krabapple]

This guy's videos are infuriating but I seem to watch anyway, it's like watching a car crash- I just can't look away. The comment section is even more disappointing. He supposedly also has a degree in neuroscience but can't appreciate a blind test...?

I took some neuro in grad school. Got to handle and slice up some brains!

neuroscience != cognitive psychology

Neuroscience is mostly the study of the physical apparatus and operation of the nervous system.

If there were studies of what brain centers lit up during 'sighted' bias, that would be neuroscience.

But the studies that 'discovered' sighted bias in the first place and explored what beliefs and externalities influence it -- that's psychology.

He could easily be quite clueless about psychoacoustics and audio-related cognitive bias, beyond a knowledge of the auditory pathway.

 

[Mart68]

When I was at university they voted to ban the Professor of Neuroscience from the student union bar because it was rumoured he was experimenting on monkeys.

The Neuroscience building was a large bunker bristling with security cameras. I used to imagine that inside it he had a thousand monkeys working at a thousand typewriters.

 

[aandres_gm]

When I was at university they voted to ban the Professor of Neuroscience from the student union bar because it was rumoured he was experimenting on monkeys.

The Neuroscience building was a large bunker bristling with security cameras. I used to imagine that inside it he had a thousand monkeys working at a thousand typewriters.

Typing posts like this very one? Creepy.

 

[amirm]

Maybe he won't get caught up in the trap of ranking amplifiers by SINAD, which gives the appearance that the "best" ranking amplifiers will sound better than some other "median" ranking amplifier in the list, when IMO there is really no difference other than hanging your hat on the convenience of a "single figure of merit" - which unfortunately doesn't really exist in the real world of what's audible. In actuality, and IMO, "best" really has to do with the quality of the source material, and NOT 120 dB SINAD versus 80 dB SINAD. BTW, if you're determined to sell to SINAD "purists" all you have to do is jack up the negative feedback to play that "numbers game".

There is no "trap" there. The reason you have any idea about SINAD in this manner is because you saw them collated and organized so you can make inferences. Without them, after testing 350 DACs, where on earth would you be? I will answer: completely lost. You also have some misunderstanding of the audibility of SINAD which I won't go into.

My job here is to tear apart a piece of audio product and tell you if it is good, bad or in the middle. Ranking SINAD into four broad categories does this very nicely, thank you very much.

 

[amirm]

He is always publishing the complete automated measurement results pdf. from its AP software.
(Some other reviewers with AP don't... i wonder why?)

I tell you why. You just said you like fast food better than high end restaurant.

Audio Precision software has two modes of operation: Sequence and Bench. The former is what produces the pdf and its most common use is in end of line testing of assembled audio products. You create a sequence of tests, tell it the low and high limits and it goes to town and spits out a report at the end. You may have seen this from the pdfs that Schiit publishes now for their products.

The bench mode on the other hand is interactive and doesn't generate any reports. This is what I use for most of the test (for a few, it is not possible and I am forced to use the sequence mode). I have individual templates for each test that is run in this mode. There is no automation here and no pdf generation facility. Worse yet, AP software is dog slow in bringing up a new template so I waste a lot of time using this mode of operation. Why do I do it? Because it is a far superior system if you want to properly understand what makes the device tick.

I run one test and look at the results. Let's say I run the dashboard and see SINAD is 90 dB whereas the company advertises 100 dB. I stop there and work to find the cause. I change bandwidth, filtering, grounding, etc. If I can improve the results using those changes, then I gain insight into what the problem is. Or if it is not, then I report as such. The process then repeats for the next test.

Unless I make a mistake, or there is some device limitation, same set of tests are run on every product. I have numbered the templates as a matter of fact and walk though them one at a time. Again, at every stage there is analysis going on which you don't see in the review. They are all diagnostic probes intended to see the product in a different light. This is why you see specific commentary on each graph.

As to your comment regarding "completeness," I am completely against spitting out dozens of graphs for numerous reasons. Having a graph may impression the novice as to the quantity of data presented, but works to lose the interest of many to the point where they will ignore it all and just go by the text. There is a measurement guy on another forum that does this and that is precisely what their membership does. Ignore the 30 graphs and go by his one liners which are routinely different than his measurements. For this reason, I have a very high bar for adding measurements to what I already do. If the measurement can't be defended to be useful and provide additional information, I will simply not run it. Damn whoever thinks more is the better.

Finally, the PDF export in AP sucks from formatting point of view. It is hard to get it to export graphs with the right resolution for our platform here. Hard to go and ad markers, etc. Again, it is not meant for this use (publishing and educating).

Bottom line, I am not the quality control arm of an audio company. It is not my job to produce those reports. They should and provide them to you all.

 

[raif71]

Maybe he won't get caught up in the trap of ranking amplifiers by SINAD, which gives the appearance that the "best" ranking amplifiers will sound better than some other "median" ranking amplifier in the list, when IMO there is really no difference other than hanging your hat on the convenience of a "single figure of merit" - which unfortunately doesn't really exist in the real world of what's audible. In actuality, and IMO, "best" really has to do with the quality of the source material, and NOT 120 dB SINAD versus 80 dB SINAD. BTW, if you're determined to sell to SINAD "purists" all you have to do is jack up the negative feedback to play that "numbers game".

To the comment about tinkering with small car changes in SCCA racing I say sure it's "clean fun" that doesn't matter much in where you finish in a race. After a number of years of racing on the SCCA circuit I feel I can pretty fairly rank where results improvements are generated: 92% driving skill and course knowledge; 6% the right tires; 2% tweaking engine and other stuff. I eventually replaced my need for speed with Masters ski racing - where results are even more about "the driver" than the equipment. In one ski race a couple of years ago I had a great first run on the ski race course and one of my competitors commented that it was because of my really expensive skis. I immediately offered him to swap skis for the second run since we both had the same boot sole length. He took me up on my offer and I beat him by an even wider margin on the second run. Both runs added to our final times, which pushed him even further back in the results. He hasn't whined about equipment since!

Interesting ski race example there, put that into audio perspective, gears like DACs don't matter much but skill like hearing, eye and brain coordination can make all DAC shine?

 

[khark]

You guys are just jealous of his British accent

 

[Lambda]

filtering, grounding, etc. If I can improve the results using those changes, then I gain insight into what the problem is. Or if it is not, then I report as such. The process then repeats for the next test.

This Is all good and very appropriated!

But once Ideal or Operating conditions for the DAC have been Evaluated i it seams to me its easy and fast to run the automated Sequence test.
I do not say to put Everything and Every graph in the Review!

But if there would be a download link at the and the view users that are curious. can go and look at what ever they are interested in.

I have a very high bar for adding measurements to what I already do. If the measurement can't be defended to be useful and provide additional information, I will simply not run it. Damn whoever thinks more is the better.

But for some how want to see "what makes the device tick." at home its interesting to have the data.
For example Impulse response can be interesting to understand what makes the device filter tick but i don’t want you to shove this "unnecessary" data into everyone's face.

Edit: other things might be interesting but are not need to be included in the review:
FFT spectrum with output at -infdB

The difference in distortion between Generator Level: -3dB and 0dB
The difference in distortion between 1khz test signal and 50Hz signal

The niche thing about the Automated test is they are alle very reproducible and every test is labeled with what signal was exactly used.

for example if i look at this verry usfull an highly appreciated graph:

https://www.audiosciencereview.com/forum/index.php?attachments/smsl-su-9n-measurements-thd-vs-level-balanced-xlr-usb-stereo-dac-png.158934/

i have to assume its done with a 1khz tone?! in the automated test it is all labeled.

in the automated test we have also IMD Level Sweep
Maybe the IMD+N sweep showed something different compared to THD+N sweep.

Maybe the Cross talk Sweep shows something unusual?

Maybe Wideband view FFT to 1Mhz showed interesting HF noise.


99% won’t care and don't look at the data but im sure ther always someone who will and double check.

 

[Zensō]

I actually think he is a cool guy and I know he posts here from time to time. Buuuuut the amount of differences he claims to be able to hear from dac to dac is bullshit.

So he listens to dacs generating 99,9995% equal signal and yet he perceives differences in soundstage, bass, "tembah", instrument separation and so on. Those speakers of his must have some 0,0001% thd to be able to catch such details and sound so different from dac to dac.

Goldensound is no different than Guttenberg, Darko, and the rest of the YouTube "miracle ears" gang who want you to believe they have superhuman hearing abilities.

 

[PierreV]

I think we can all agree that there is more value in active testing done by a competent tester, and even more value in the explanations of what is wrong and why. However, that doesn't mean there is no value in the batch measurements he provides.

Additionally, he seems to get his hands on interesting (or at least, hyped) hardware and doesn't hesitate to criticize expensive stuff now and then.

And, when Darko spends parts of his profits on more lights and flowers for the IKEA-styled studio where he delivers his simplistic monologues, Goldensound did actually go as far as procuring an AP. Subjective reviewers are usually more into getting free stuff, long-term loaners or, steep discounts.

That being said, the constant differences he claims to hear are... a tiny bit fuzzy. I usually skip that part of the monologue. He does, at least, spare us the lyric, utterly deluded flowery prose that has become the trademark of other reviewers.