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

[antcollinet]

If people can not tell apart a dac with -70dB THD+N VS a dac with -120dB THD+N, folks who think they can tell them apart are myth, wha

Test yourself here: https://www.klippel.de/listeningtest/

The vast majority of people who take that test can’t hear below around -45dB distortion with real music. Probably most who get close to that are using headphones or IEMs Personally with my speakers I struggle to get to -25dB.

What do you get? (Don’t forget to post your chart)

 

[oleg87]

If people can not tell apart a dac with -70dB THD+N VS a dac with -120dB THD+N, folks who think they can tell them apart are myth, what is the point of even doing reviews and measurements on DACs here. Why not just put a banner at the top of the DAC section. "Look elsewhere, all dacs sound the same regardless"

I don't really think we need "reviews" per se, but how would you know their measured performance if you don't measure their performance?

 

[Blumlein 88]

I don't really think we need "reviews" per se, but how would you know their measured performance if you don't measure their performance?

Maybe an AI, listening over artificial ears (e.g. microphones) and then generate flowery prose indicating a go/no go condition. Artificially subjectively measured? Then we'll need a new website labeled ASM instead of ASR.

 

[solderdude]

I am not disputing well measured dacs should sound different.

Why should well measuring DACs sound different ? Because of the price tag ?

Lets also put aside differences in R2R, delta sigma dacs for a moment.

Both produce an analog output signal that is dictated by the incoming digital signal, just using different methods.

If people can not tell apart a dac with -70dB THD+N VS a dac with -120dB THD+N, folks who think they can tell them apart are myth,

Nope, they are not and it depends on the frequency response, type of distortion and noise levels as well as the used recordings where they can be told apart in a proper controlled statistically valid 'blind' test (regardless of the protocol)

what is the point of even doing reviews and measurements on DACs here.

The point is to check manufacturer specs, operational issues, output levels and technical performance.

Why not just put a banner at the top of the DAC section. "Look elsewhere, all dacs sound the same regardless"

Because that would be really silly and not make any sense.

I previously read reviews here with headless panthers on the PS audio dac and a few other dacs with SINAD in the 60s db, maybe it don't matter much too.

One could ask themselves why they got a headless panther. Functionality, technical performance, certain issues, price/performance ?

Maybe a touch off topic, but does it mean that as long as say a tube preamp has decent linearity, even measured at -70dB THD+N would sound no different than say a Benchmark LA4 close to -120dB THD+N? I am confused

You may be confused, 'we' are not. When the word 'sound' comes into view it should be determined by a well performed, statistically valid and controlled blind test IF the goal is to check for audible differences.
When one knows what one is listening to and has a certain 'belief' then I am sure you will hear differences. You might not when tested as described above, which could also depend on the used recording, transducers and 'training'.

 

[Tinker]

Someone remind me where the term "placebo effect" comes from again?

A poem for you...

In labs where logic reigns supreme,
Where facts unfold, precise and clean,
The test tubes hum, the data sings,
And science spreads its measured wings.

Equations etched in rigid form,
Dispel the myth, outlast the storm.
It seeks the truth, the purest light,
To break the dark, and show what's right.

But in the shadows, quiet, sly,
Placebo waits, a trick of mind,
With whispered hope and phantom grace,
It changes hearts, though leaves no trace.

"Believe," it says, "and feel the glow,
Though proof eludes the charts you show."
A sugar pill, a harmless sham,
Yet somehow pain dissolves like sand.

And so they clash, these ancient foes—
One grounded firm, one ebb and flow.
Is it the pill or the belief,
That truly brings the sought relief?

Science scoffs, demands its due,
"Results must stand, be clear, be true!"
But placebo smiles and gently sighs,
"For sometimes faith is just as wise."

In this great dance of mind and nerve,
The lines grow thin, the truths may curve.
For in the space where both collide,
Might something deeper yet reside?

 

[Sal1950]

If people can not tell apart a dac with -70dB THD+N VS a dac with -120dB THD+N, folks who think they can tell them apart are myth, what is the point of even doing reviews and measurements on DACs here. Why not just put a banner at the top of the DAC section. "Look elsewhere, all dacs sound the same regardless"

I previously read reviews here with headless panthers on the PS audio dac and a few other dacs with SINAD in the 60s db, maybe it don't matter much too.

I don't really think we need "reviews" per se, but how would you know their measured performance if you don't measure their performance?

Someones got to hold those manufacturers feet to the fire.
This passion is already full of snake-oil peddlers. As consumers we deserve the right to know who's building quality products and who's selling junk.
A DAC doesn't necessarily need to have blatantly audible problems to be considered a rip off. Specially when the MSRP is 10x that of units that measure 10x better.
"They say" CD's with their edges painted with CD Stoplight magic markers sound better too. Snake-oil and BS is what it is and needs to be exposed.
It's a damn shame the circus that the High Fidelity passion has turned into.

 

[dlovesmusic]

Why should well measuring DACs sound different ? Because of the price tag ?

Both produce an analog output signal that is dictated by the incoming digital signal, just using different methods.


Nope, they are not and it depends on the frequency response, type of distortion and noise levels as well as the used recordings where they can be told apart in a proper controlled statistically valid 'blind' test (regardless of the protocol)


The point is to check manufacturer specs, operational issues, output levels and technical performance.


Because that would be really silly and not make any sense.


One could ask themselves why they got a headless panther. Functionality, technical performance, certain issues, price/performance ?


You may be confused, 'we' are not. When the word 'sound' comes into view it should be determined by a well performed, statistically valid and controlled blind test IF the goal is to check for audible differences.
When one knows what one is listening to and has a certain 'belief' then I am sure you will hear differences. You might not when tested as described above, which could also depend on the used recording, transducers and 'training'.

Yes I am definitely confused. I know you are not confused and many here sounded like they are not.
So you are saying if level matched, no one will be able to tell apart say a dac with tube output with heavy distortion over a SOTA dac with minimal distortion.

I own a Schiit Freya tube preamp and a Benchmark LA4, if I level matched them, I will not be able tell it apart?

 

[Blumlein 88]

Yes I am definitely confused. I know you are not confused and many here sounded like they are not.
So you are saying if level matched, no one will be able to tell apart say a dac with tube output with heavy distortion over a SOTA dac with minimal distortion

No that is not what is being said.

 

[Demmers]

Test yourself here: https://www.klippel.de/listeningtest/

The vast majority of people who take that test can’t hear below around -45dB distortion with real music. Probably most who get close to that are using headphones or IEMs Personally with my speakers I struggle to get to -25dB.

What do you get? (Don’t forget to post your chart)

That was interesting! And apparently so am I?! This is through headphones: Topping D10s DAC > mk1 JDS Labs Atom amp > Beyerdynamic DT 770 Pro X

 

[Blumlein 88]

That was interesting! And apparently so am I?!

Yep that is a much better than average result. Congratulations.

 

[dlovesmusic]

My question arise from a previous comment that the person claimed he had heard many different dacs, from the worst dac with THND < -70db to SOTA and would not be able to tell a difference. That was my confusion. So are we suggesting something like the PS Audio measured to have transformer noise and THND < -70db will sound the same as say a Topping D90 with much better THND measurements?

 

[solderdude]

You are obviously confused.
I did not say: if level matched, no one will be able to tell apart say a dac with tube output with heavy distortion over a SOTA dac with minimal distortion.

I stated that: a tube preamp has decent linearity, even measured at -70dB THD+N would sound no different than say a Benchmark LA4 amp close to -120dB THD+N in a well performed blind test. In the assumption that 'decent linearity' for you meant no audible roll-off in the audible range (with linearity one usually means distortion).

It is highly likely that a DAC with a tube output (where FR and 'heavy distortion' was not defined) sounds different from a SOTA DAC so I have no idea how you could come to that conclusion based on your amplifier thesis.
-70dB THD+N is actually quite good and you did not even specify what distortion you meant nor what the noise component is.
If it is mostly noise then yes... you could certainly hear it.

 

[oleg87]

My question arise from a previous comment that the person claimed he had heard many different dacs, from the worst dac with THND < -70db to SOTA and would not be able to tell a difference. That was my confusion. So are we suggesting something like the PS Audio measured to have transformer noise and THND < -70db will sound the same as say a Topping D90 with much better THND measurements?

That is not a sufficiently thorough metric to make such a claim. For example, you can find DACs with significant high-frequency rolloff, which 1khz THD+N measurement won't capture. And of course, distortion and noise are different things.

You would need to look through a reasonably complete set of measurements and see if any of its defects are likely to reach audibility in a realistic use case.

 

[antcollinet]

That was interesting! And apparently so am I?! This is through headphones: Topping D10s DAC > mk1 JDS Labs Atom amp > Beyerdynamic DT 770 Pro X

Pretty good result.

But still only half the level of the 45 db I mentioned. And still 100 times larger distortion than Thd of (say) -91 db.

 

[Blumlein 88]

My question arise from a previous comment that the person claimed he had heard many different dacs, from the worst dac with THND < -70db to SOTA and would not be able to tell a difference. That was my confusion. So are we suggesting something like the PS Audio measured to have transformer noise and THND < -70db will sound the same as say a Topping D90 with much better THND measurements?

Okay, so THD+N or SINAD could be -120 db noise and -70 db THD. You wouldn't hear that versus one that had THD+N of -120 db if both have flat frequency response. OR -70 db SINAD could be -70 db noise and -120 db THD. You'll hear that. Especially with efficient speakers or headphones. And sometimes with tubes it might be the noise limiting performance (though good tube circuits won't have that problem).

 

[dlovesmusic]

You are obviously confused.
I did not say: if level matched, no one will be able to tell apart say a dac with tube output with heavy distortion over a SOTA dac with minimal distortion.

I stated that: a tube preamp has decent linearity, even measured at -70dB THD+N would sound no different than say a Benchmark LA4 amp close to -120dB THD+N in a well performed blind test. In the assumption that 'decent linearity' for you meant no audible roll-off in the audible range (with linearity one usually means distortion).

It is highly likely that a DAC with a tube output (where FR and 'heavy distortion' was not defined) sounds different from a SOTA DAC so I have no idea how you could come to that conclusion based on your amplifier thesis.
-70dB THD+N is actually quite good and you did not even specify what distortion you meant nor what the noise component is.
If it is mostly noise then yes... you could certainly hear it.

Thanks for the clarification.

 

[Tinker]

"They say" CD's with their edges painted with CD Stoplight magic markers sound better too. Snake-oil and BS is what it is and needs to be exposed.

I had a friend who loved his green pen. Why would you go out of your way to ruin his experience? Actively trying to dispel something he's perfectly happy with and ruin his enjoyment is somehow a noble mission because YOU don't like it? I just let him be. Ignorance is bliss. No harm done.

 

[ahofer]

All he said was "the dots" aren't real. When in fact the white dots are real. Just saying.

If you want to be pedantic, they aren’t real, they are pixels on a screen. Just sayin’

 

[ahofer]

I had a friend who loved his green pen. Why would you go out of your way to ruin his experience? Actively trying to dispel something he's perfectly happy with and ruin his enjoyment is somehow a noble mission because YOU don't like it? I just let him be. Ignorance is bliss. No harm done.

Did I miss Sal going into your friend’s house and berating him?

 

[Tinker]

Did I miss Sal going into your friend’s house and berating him?

It "needs to be exposed". Why? If people can't see for themselves that it's a bunch of BS then why ruin it for them?. Why should Sal or you or anyone care that he uses a stupid green pen?