Serious Question: How can DAC's have a SOUND SIGNATURE if they measure as transparent? Are that many confused?

[SIY]

Feels good knowing that Apple makes the best sounding DAC in the world.

They make a DAC that sounds identical to all other competently engineered DACs.

 

[Darkweb]

They make a DAC that sounds identical to all other competently engineered DACs.

Yes, by the transitive property they make the best sounding DAC in the world. Pretty neat to do that for $9.

 

[SIY]

Pretty neat to do that for $9.

Indeed, it's impressive. They have very smart engineers and manufacturing people. We truly live in the golden age of audio.

 

[LTig]

Isn’t it true, that any listening room has its own sound signature and terribly distorts the sound, 10-100x more than any DAC? What’s the point of looking for transparency if it’s impossible to achieve outside anechoic chamber?

Don't mix linear (FR in room) with nonlinear (THD, IMD, ..., in equipment) distortion. The former can be fixed with room EQ and other means.

 

[PenguinMusic]

Hi,
Do I get you correctly when I understand that you are saying that no matter what DAC I take, they will all sound equal, no matter the made, chip or power supply ?
If so, does it mean that when Amir does measure the output of DACs, if there are differences in those measurements, they are not at all audible in any way ?

If so, I must say that I am a little confused...
I thought that if there are differences in measurements, it COULD BE that those differences would have be reflected in the sound produced...

But mayber I get all this totally wrong :-(

 

[Frank Dernie]

Hi,
Do I get you correctly when I understand that you are saying that no matter what DAC I take, they will all sound equal, no matter the made, chip or power supply ?
If so, does it mean that when Amir does measure the output of DACs, if there are differences in those measurements, they are not at all audible in any way ?

If so, I must say that I am a little confused...
I thought that if there are differences in measurements, it COULD BE that those differences would have be reflected in the sound produced...

But mayber I get all this totally wrong :-(

Well it does depend.
Certainly an electronic device can produce audible levels of noise and distortion.
Personally I mainly listen to classical music and have made quite a lot of recordings, originally to reel-to-reel tape, later to DAT and other digital recorders.
Having a noise level lower than the background noise of the audience and auditorium noise in a live music recording is fairly easy. Tape recorder hiss is generally audible between songs, if the audience don't cough too much, but I have never heard any noise whatever from a digital device at listening volume.
Theoretically a recording of sound effects for a film, made in a special super quiet studio, may use all the dynamic range of readily available inexpensive DACs (though personally I doubt it).
I don't seem to be able to hear distortion of 0.1% on music (that is -60dB) but would want the noise to be 90dB down to be 100% sure I would never notice it.
Based on this it is very unlikely that I could hear the difference between the huge majority of DACs Amir has tested.
Personally I want distortion less than 0.1% for music listening (I don't watch many films) so for me, certainly, there is no indication that the huge majority of DACs tested here would sound different to each other listening to music.
That is just me though.

 

[PenguinMusic]

Well it does depend.
Certainly an electronic device can produce audible levels of noise and distortion.
Personally I mainly listen to classical music and have made quite a lot of recordings, originally to reel-to-reel tape, later to DAT and other digital recorders.
Having a noise level lower than the background noise of the audience and auditorium noise in a live music recording is fairly easy. Tape recorder hiss is generally audible between songs, if the audience don't cough too much, but I have never heard any noise whatever from a digital device at listening volume.
Theoretically a recording of sound effects for a film, made in a special super quiet studio, may use all the dynamic range of readily available inexpensive DACs (though personally I doubt it).
I don't seem to be able to hear distortion of 0.1% on music (that is -60dB) but would want the noise to be 90dB down to be 100% sure I would never notice it.
Based on this it is very unlikely that I could hear the difference between the huge majority of DACs Amir has tested.
Personally I want distortion less than 0.1% for music listening (I don't watch many films) so for me, certainly, there is no indication that the huge majority of DACs tested here would sound different to each other listening to music.
That is just me though.

Hi,

Thanks for your answer.
To sum it up : difference may be mesurable but, unless it is reall y huge, will probably not be audible ?
I can take that... Like people saying "my smartphone's camera has 108 dXO mark when your's only have 106"... Not sure I'd be able to tell which one produces better photos :-(

But thinking about the OP, I think that what is meant is that, if we assume that a device has everthing common and we change simply the DAC, then there would be no difference in sound ?

So I got Amir totally wrong when he writes that the ESS Sabre DAC's have the same "flaw" or that the Burr-Brown/TI are now old models that are outperformed by newer ESS or AKM DAC's...
I thought that meant that the AKM or ESS would sound better (point 2) and that the ESS Sabre have a kind of "signature" (point 1).
But maybe I am totally wrong :-(

 

[Julf]

To sum it up : difference may be mesurable but, unless it is reall y huge, will probably not be audible ?
I can take that... Like people saying "my smartphone's camera has 108 dXO mark when your's only have 106"... Not sure I'd be able to tell which one produces better photos :-(

Exactly.

So I got Amir totally wrong when he writes that the ESS Sabre DAC's have the same "flaw" or that the Burr-Brown/TI are now old models that are outperformed by newer ESS or AKM DAC's...

Not wrong. Amir is really looking at very small measurable differences - but at this point, they are way beyond audible.

 

[PenguinMusic]

Hi guys,

thanks a lot for your answers... Helps me to get my way through this DAC Jungle

So, ESS or AKM or Wolfson (if they still exist) should basically sound about the same.

But can I take it that if a device has better power supply, better capacitors (or whatever the names...) and better "circuitry (I am a noob :-(...) it may sound better, but that the better sound is not because of the DAC chip, but to all the surroundings of that DAC Chip ?

Or does all that stuff not matter at all ?

 

[Julf]

But can I take it that if a device has better power supply, better capacitors (or whatever the names...) and better "circuitry (I am a noob :-(...) it may sound better, but that the better sound is not because of the DAC chip, but to all the surroundings of that DAC Chip ?

Yes, especially the analog buffer stage after the DAC. Of course some DACs provide optional sub-optimal filter algorithms to please audiophiles...

 

[majingotan]

But can I take it that if a device has better power supply, better capacitors (or whatever the names...) and better "circuitry (I am a noob :-(...) it may sound better, but that the better sound is not because of the DAC chip, but to all the surroundings of that DAC Chip ?

Better as long as it is objectively better. You can have all that branded capacitors and whatnot but if it’s objectively inferior then it’s not better

 

[PenguinMusic]

Better as long as it is objectively better. You can have all that branded capacitors and whatnot but if it’s objectively inferior then it’s not better

Hi,
Can a sound be "objectively" good or bad, better or worse ?
Most probably. That is what measurements are for.

Can music sound objectively good or bad, better or worse ?
I doubt that. That is what our ears are for.

Not wanting to start a debate... Purely personal opinion.

 

[Frank Dernie]

Hi guys,

thanks a lot for your answers... Helps me to get my way through this DAC Jungle

So, ESS or AKM or Wolfson (if they still exist) should basically sound about the same.

But can I take it that if a device has better power supply, better capacitors (or whatever the names...) and better "circuitry (I am a noob :-(...) it may sound better, but that the better sound is not because of the DAC chip, but to all the surroundings of that DAC Chip ?

Or does all that stuff not matter at all ?

It is not the chip you are listening to but the whole device.
This means shielding, board layout and the analogue stage probably make more difference to the output of the box than which DAC chip is being used.
As far as boutique expensive components and power supplies are concerned they may make a difference, better or worse, but since the only thing connecting the box to the outside world are the output connectors any change to the amplitude, frequency and phase here are measurable, there is nothing else.
If there is no change to the output the fancy components can not be changing the actual sound, though if they are reassuringly expensive one may well imagine an improvement.

 

[PenguinMusic]

It is not the chip you are listening to but the whole device.
This means shielding, board layout and the analogue stage probably make more difference to the output of the box than which DAC chip is being used.
As far as boutique expensive components and power supplies are concerned they may make a difference, better or worse, but since the only thing connecting the box to the outside world are the output connectors any change to the amplitude, frequency and phase here are measurable, there is nothing else.
If there is no change to the output the fancy components can not be changing the actual sound, though if they are reassuringly expensive one may well imagine an improvement.

Hi,

Ok, so there may be a differnece between DAC (device) but that won't be the DAC (chip) itself.
I can understand that.

Just to clarify : I wasn't talking about the price. I know some well designed devices can go cheap... And that is exactly what I am looking for

 

[majingotan]

Ok, so there may be a differnece between DAC (device) but that won't be the DAC (chip) itself.
I can understand that.

Sonic difference or in your terms music sound difference? Highly doubt that unless you're able to prove it with 95% consistency using DBT volume matched test. IMO, even those DACs with tube analog stage or NOS R2R would most likely sound indifferent under those controlled conditions for those average audiophiles and even seasoned trained listeners. Unless there's some broken implementation in the analog stage or even in the digital stage (wonky DSP or horrendous oversampling filters) then maybe there's a music sound difference and those differences are obviously measurable.

When listened sighted without volume match, our biases just make up unicorn music sound improvements unless that said DAC has THD+N of 5% or higher.

 

[Milesian]

What I find hard to deal with is 24 pages of responses devoted to one person’s inability to accept that some folks hear things he says they ought not to hear.

 

[majingotan]

What I find hard to deal with is 24 pages of responses devoted to one person’s inability to accept that some folks hear things he says they ought not to hear.

I would accept what that person's claim of hearing differences between DACs IFF there's an absolute objective evidence of that person's claims, e.g. 19/20 guessed correctly on which DAC is playing. Else, that person's opinion is nothing more of a delusional belief and should not be taken into factual context. I myself have undergone quite a few volume matched A/B (mostly sighted and one DBT) and in both cases, there is absolutely no difference in music sonic reproduction once levels are tightly matched to 0.1 dB or less. It does not matter if that said DAC is R2R (NOS or oversampling) or Chord FPGA oversampling DS DAC, regular off-the-shelf DAC chips, discrete analog stage and so on and so forth

 

[PenguinMusic]

Hi majin and Milesian,

Not sure what both of you mean with theior respectives posts.

majingotan : if I get you right, you are writing that every device that is used as a DAC will give more or les the same output, with no consideration of the rest of the components besides the DAC chip itself. Is that right ?

Milesian : if I get you right, you're saying (probably as I do) that some people are "subjectively" believing that they hear differences (and if they believe they do, they are indeed hearing them) and that people saying "it cannot be measured so it cannot be" should leave them alone and let them hearing/beliving they are hearing differences ?

If that is correct, I must say that I am more on Milesian's side.

 

[majingotan]

f I get you right, you are writing that every device that is used as a DAC will give more or les the same output, with no consideration of the rest of the components besides the DAC chip itself. Is that right ?

I'm stating that a well engineered DAC, regardless of DAC chips used, digital filters used or NOS (unless the filters are deliberately horrible but is measurable), regardless of analog stage will more or less sound the same under DBT and volume matched to < 0.1 dB. I repeat well engineered. There's also R2R NOS DACs that well engineered such as the Holo Audio May DAC with unprecedented measurements for a discrete R2R DAC. Even the ones that are mediocre (e.g. 90 dB SINAD) (but not abysmally horrible i.e. < 60 dB THD+N) are inaudible even for trained listeners under that same conditions. Obviously, the amps where these DACs are connected are also audibly transparent without wonky input impedance (at least 30 K ohm and at least have over 100 dB SINAD) as eliminate all other variables that are in the system.

The key is when performing these controlled experiments, every component must be audibly transparent and do not have issues regarding damping factor and other variables that can affect the frequency response to ensure that the DBT volume matched test is valid and not borked.

 

[solderdude]

It has been established that NOT all DACs sound the same.
It also has been established that all competently designed DAC's with a proper reconstruction filter can not be told apart in level matched blind listening tests.
It has certainly been established that all DACs measure differently. That does not mean they THUS sound different.
It has since long been established that people can hear differences even when there are none.
And it has also been established that substantial measured differences don't always translate to audible differences.
Ears are not reliable analyzers. Ananlyzers can be used but need to measure all relevant aspects and need to be interpreted as a whole.
It has also been known for a long time that people who do not understand measurements choose to say that measurements don't matter.

The above can be read in the entire thread and don't think is really debatable.