You couldn’t hear the difference - Proof! Beautiful video of David Mellor (Audio Masterclass) about sound differences between DAC

[MAB]

What's the reference for the pink noise reproduction? When one, listening to pink noise, says this loudspeaker is better than that one.
Based/compared to what?

It's ability to resolve differences in reproduction. Not just speakers.
The question is:

• Are unit a and unit b different?

Not better.
Better is a population study of statistical preference. Be careful to not confuse differences with preferences.

To determine preference, first need to be able to identify differences.

 

[Vuki]

It's ability to resolve differences in reproduction. Not just speakers.
The question is:

• Are unit a and unit b different?

Not better.
Better is a population study of statistical preference. Be careful to not confuse differences with preferences.

To determine preference, first need to be able to identify differences.

Ok, then what about this Tracy Chapman track? Is it also used just to identify differences?

 

[ahofer]

What's the reference for the pink noise reproduction? When one, listening to pink noise, says this loudspeaker is better than that one.
Based/compared to what?

guess my memory wasn't so great on "girl with guitar".

 

[DanielT]

I'm fine with the test, but using more complicated music isn't just about "pedal to the metal", it's about exploring a much broader palette of dynamics and frequency where defects might be revealed.

I can't come up with the right search string, but I do believe @amirm has posted some studies showing how people hear differences with different kinds of music. What I remember is the summary: the girl-with-guitar stuff that everyone uses in hifi shows is not very demanding. Can't be much better when you subtract the girl.

Ok, I was thinking more about not stressing out test equipment with "demanding" high dynamic music so that instead it becomes speaker/headphone/amp capacity that is tested instead of in this case looped through a "mediocre" DAC or not.

If it would play any role in practice if you try to detect different DACs or different OP amps whether the amp is driven into clipping a little now and then or not, whether the speaker shows twice as much distortion or not? In these different scenarios, or let's say in a scenario where the amp is toughing it out smoothly and nicely and the speakers are not played at such a high volume, you will still not hear any differences (other than perhaps in exceptional cases, extreme cases).

 

[Vuki]

So, it actually isn't a track to judge sound quality with - just the differences, else one falls into "circle of confusion" trap?

 

[kemmler3D]

What's the reference for the pink noise reproduction? When one, listening to pink noise, says this loudspeaker is better than that one.
Based/compared to what?

I *think* the graph earlier in the thread is about listeners ability to tell stuff apart, which is definitely easier with PN.

PN makes obvious flaws really obvious and subtle flaws noticeable more quickly, but you don't get much of a subjective experience from it, so coming up with a quality rating based on listening to one speaker with PN only might be tough.

 

[krabapple]

Over the years on audio forums I've repeatedly seen common 'intuitions' about what music is best for detecting difference, and what music is hardest to lossy encode....many seem to think 'symphonic' is most revealing or 'most difficult', perhaps on the presumption that 'more' is better/harder. It's not true.

 

[kemmler3D]

Over the years on audio forums I've repeatedly seen common 'intuitions' about what music is best for detecting difference, and what music is hardest to lossy encode....many seem to think 'symphonic' is most revealing or 'most difficult', perhaps on the presumption that 'more' is better/harder. It's not true.

It's often said that hard / alt rock is best because of the dense spectrum... I think this is more or less correct for the same reason that pink noise works well.

Symphonic stuff can be useful for the same reason, but it's less common in that genre to have the density that makes it easy to hear problems with FR.

 

[Honken]

While the use case of 'Fast Car' is similiar to that of pink noise, at least you can tell with your own intuition if things sound completely wrong with the former. If a simple guitar + vocal track doesn't sound decent in whatever system you are listening to at the moment, something's off. But yes, apart from that, naturally you can notice differences between gear quite easily with this track as well.

When listening to two speakers blind with pink noise, with one of the speakers being borderline broken - it's still kinda hard to tell which one is broken. Even if telling the speakers apart is easy. With Fast Car, it's quite easy to tell which one is 'broken' (in pathological cases).

Edit: Personally, I like using a track by Sharon van Etten called 'I told you everything'. I find it less forgiving than Fast Car in terms of bass reproduction ('Fast Car' has almost no bass), and the lower treble is audibly muddy with shoddy gear that reproduce 'Fast Car' just fine (a good test for IEMs, and their adherence to the Harman curve). But as a track on it's own, it still is mostly female vocals + a backing instrument, so it's "clear" in that regard.

 

[Palladium]

Over the years on audio forums I've repeatedly seen common 'intuitions' about what music is best for detecting difference, and what music is hardest to lossy encode....many seem to think 'symphonic' is most revealing or 'most difficult', perhaps on the presumption that 'more' is better/harder. It's not true.

A/B testing to test the limits of lossy codecs is mostly using electronica now.

 

[FrantzM]

While I can understand the need for a more complex (or "better") signal for this experiment, does anyone believe that the mega dollars DAC would sound different? My money is that the results would be the same ...

Peace.

 

[westsail42]

Ok, maybe I am way off base here...

But don't most "DACs" have some sort of analog gain stage? Doesn't that have (more of a?) chance of affecting sonic differences than the actual D/A conversion?

His tests say nothing about the D/A -> A/D converter (make model etc.)

If the goal is to compare "DAC chips" thats one thing. But most DACs have a gain stage which also affects sonic quality depending on how it is designed.

 

[raif71]

Better to loop 5 or more different DACs with various dac chips, couple of r2r and throw in some fpga implementation in the mix too.

 

[Purité Audio]

Ok, maybe I am way off base here...

But don't most "DACs" have some sort of analog gain stage? Doesn't that have (more of a?) chance of affecting sonic differences than the actual D/A conversion?

His tests say nothing about the D/A -> A/D converter (make model etc.)

If the goal is to compare "DAC chips" thats one thing. But most DACs have a gain stage which also affects sonic quality depending on how it is designed.

Dacs are measured at their output which includes of course the output stage,
Keith

 

[DanielT]

Better to loop 5 or more different DACs with various dac chips, couple of r2r and throw in some fpga implementation in the mix too.

One wonders how many times it would take before it became audible to most people?

Incidentally, a problem we young teenagers struggled with when we copied cassettes and VHS with each other. Especially VHS, damn how bad the quality could be. But we had to put up with that in order to to watch the latest James Bond or now that I think about it; the most copied (of course copies of copies of ...and so on) in our youth circles: Raiders of the Lost Ark
Tell that to today's young people and they wouldn't believe it!

You've probably already seen this sketch with Monty Python, but have you seen this one:

Edit:
Here are the test results from David Mellor (Audio Masterclass), the Youtube video in #1. By the way, if you haven't tested it yourself, these results won't "help" you in the slightest in arriving at the right result:

 

[mcochris]

If we can't hear a difference between DACs, then respectfully, why does ASR measure and rate them? Does a DAC with a high SINAD actually sound better than one with a low SINAD?

 

[diablo]

I saw the original video and couldn't spot any changes in the test when listening on my pretty decent stereo system. Didn't think it would help if I downloaded the file. I didn't think the music played was suitable really.

I quite like the pink noise test that Erin uses when comparing speakers, the differences are usually very obvious. Not quite sure how he prepares the tests though - but maybe someone could repeat the DAC test with pink noise?

 

[Mart68]

Edit: Personally, I like using a track by Sharon van Etten called 'I told you everything'. I find it less forgiving than Fast Car in terms of bass reproduction ('Fast Car' has almost no bass), and the lower treble is audibly muddy with shoddy gear that reproduce 'Fast Car' just fine (a good test for IEMs, and their adherence to the Harman curve). But as a track on it's own, it still is mostly female vocals + a backing instrument, so it's "clear" in that regard.

But the quote in post 23 says 'Fast Car' - 'Has energy from 20Hz to 20Khz' - which it can't have and have no bass. It certainly has bass on my speakers.

I actually only bought that album recently. My test has always been 'Thin Lizzy - Live And Dangerous'. If a speaker can make sense of that it can play anything.

 

[DanielT]

If we can't hear a difference between DACs, then respectfully, why does ASR measure and rate them? Does a DAC with a high SINAD actually sound better than one with a low SINAD?

For my part, I don't bother studying SINAD on the DAC tests Amir performs. Or rather, I quickly check to make sure they are ok, that is, below audibility levels.
As far as I've seen, all DACs (modern) that Amir tests meet that criterion. I don't even know if he has tested any vintage DACs?

What I look at are functions, how does the build quality seem, what about warranties, user experience about reliability, does the remote control work well, do users have sad experiences with it picking up noise, if it has Bluetooth (even though I don't use it often) does the LDAC codec work well and so on and so forth.
Nowadays, I read Amir's tests AND the users' practical experiences of the same DAC. I see it as a whole in the evaluation. Or I would actively read the users' comments if I were interested, in need of buying X DAC.

Dear topic, which is shown here. A bit unbelievable: 545 pages with 10877 posts:

How can DAC's have a SOUND SIGNATURE if they measure as transparent?

I know there are people out there that think cables affect sound, which is much worse, but there really is no response to something like that, but just to smile and nod. But what about people who talk about DACS as if they were headphone drivers or speakers, and talk about the SOUNDSTAGE...

audiosciencereview.com

This thread, one among many, on roughly the same DAC theme:

Benefits of using expensive DACs

Vorrei segnalare il seguente articolo: https://www.headphonesty.com/2024/10/expensive-dacs-duping-wasting-money/ Invito i nostri amici del blog a leggere e valutare attentamente quanto scritto. Sono totalmente d'accordo con l'autore. Translated by moderator. Please post in English going...

audiosciencereview.com

 

[Honken]

But the quote in post 23 says 'Fast Car' - 'Has energy from 20Hz to 20Khz' - which it can't have and have no bass. It certainly has bass on my speakers.

Technically, both "Has bass" and "have almost no bass" can be true at the same time.

I personally consider this to be quite lean. Maybe it is just my copy.