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

[antcollinet]

This is my first post and, as the name implies, I don't know much. But I like to learn.

So please forgive my naive question: Do different DAC chip brands sound differently? Is there an ESS "sound" v AKM?

I ask because DAC companies often boast what DAC chip they're using. Is this relevant or marketing? If the latter, how did it become a differentiator versus SINAD etc?

Confusion is my business,
FG

I'll just get this in before this thread gets merged into that other one.


A DAC has one, and only one job, and that is to accurately convert the digital representation of music from the source into an analog representation of that music. Well measuring DACS do that with inaudible levels of noise and distortion, and with flat frequency response in the audible band. In other words the analogue output is (audibly) a perfect representation of the digitally encoded music. Any defects are below the level any human ear can detect.

If two DACS both achieve this (and well measuring DACS do, and nearly all modern DACs are well measuring) then the analog signal from both must be identical within audible limits. By definition, they must sound the same.

And that is the case regardless of the brand, or model of DAC chip used.

 

[Timcognito]

Asked and answered. @antcollinet got it right, not much to add by me or anybody else?

 

[krabapple]

The need for a READ THIS POST BEFORE YOUR FIRST POSTING requirement on ASR screams out yet again.

 

[bravomail]

I'll just get this in before this thread gets merged into that other one.


A DAC has one, and only one job, and that is to accurately convert the digital representation of music from the source into an analog representation of that music. Well measuring DACS do that with inaudible levels of noise and distortion, and with flat frequency response in the audible band. In other words the analogue output is (audibly) a perfect representation of the digitally encoded music. Any defects are below the level any human ear can detect.

If two DACS both achieve this (and well measuring DACS do, and nearly all modern DACs are well measuring) then the analog signal from both must be identical within audible limits. By definition, they must sound the same.

And that is the case regardless of the brand, or model of DAC chip used.

We all agree with that in theory, myself, having a degree in electronics.
But I still have many anecdotes to share on how different dacs sound.
Here are some:
AKM has velvet sound, CS is ear-piercing, ESS don't have proper stereo imaging etc etc.
Human ear and brain are easily fooled.

 

[krabapple]

By 'have many anecdotes' do you mean you've read these, or that they are yours?

 

[Saponetto]

Ran away.

 

[Jimbob54]

The need for a READ THIS POST BEFORE YOUR FIRST POSTING requirement on ASR screams out yet again.

You know what the problem with that is though ....

 

[ISchmidt]

Ran away.

is not too late...

 

[JayGilb]

Ran away.

Run Forrest, run !

 

[Saponetto]

So, honestly, I understand the need of getting tidy the threads but I also remember that one of my first posts was about abnormal behavior of rca cables.
I have to be solidal with anybody asks naive questions, because I'm still one of them.

 

[Jimbob54]

Predictably a similar new thread moved in .

 

[antcollinet]

So, honestly, I understand the need of getting tidy the threads but I also remember that one of my first posts was about abnormal behavior of rca cables.
I have to be solidal with anybody asks naive questions, because I'm still one of them.

Nothing wrong with naive questions. But if there is already a collection thread of the kind "all posts on this topic go here" then related posts will get moved to it.

 

[Timcognito]

Change the name of the thread to something like: Properly Functioning DACs Have Undetectable Audible Differences: No Need to Explore.

 

[solderdude]

Some DACs (purposely designed or not) DO have a different 'sound signature'.
Most (competently designed and built) DACs measure different but are audibly the same under equal conditions.
Most of the competently designed DACs that measure differently may not be perceived as sounding the same when the conditions aren't the same (filter settings, output levels for instance).

So ... not all DACs 'sound' the same to everyone and may do so for different reasons.

 

[krabapple]

Nothing wrong with naive questions. But if there is already a collection thread of the kind "all posts on this topic go here" then related posts will get moved to it.

There's no reason ASR couldn't have a Stack Overflow-like function that tells the poster, before their post gets published, that their question may have been asked before in <shows threads>.

Better still would be a FAQ that actually answers those questions that are asked with mind-numbing frequency, like, 'do DACs sound different|all the same?' The function could point them to that.

It would require some work up front -- maybe crowdsourced to ASR veterans -- but after that, less work for mods!

 

[Jim Creek]

All DACs have extremely low distortion and are extremely accurate. If you can hear the microscopic differences in the sound of different DACs then you have ears like a bat. Maybe you do have bat-like ears or maybe you are just fooling yourself.

 

[voodooless]

There's no reason ASR couldn't have a Stack Overflow-like function that tells the poster, before their post gets published, that their question may have been asked before in <shows threads>.

In fact, it already does that!

 

[StewartThomson]

I caught the DAC bug about 12 months ago. I thought I had found an acceptable "bang per buck" solution with the Gustard R26, a well received R2R box. Indeed it swept the floor with a Topping DX5 and Sabage A20d that used ESS chips and both of those sounded pretty good albeit for around £400. The Gustard is four times that. I'd also add that I am using a Sugden class A amp which brings instrument separation and soundstage width and depth goodies as well.

However serious audio as we know is a demanding mistress. I have invested in a second hand Chord Hugo TT2 and M Scaler and have been listening and comparing with the R2R for about two months. The Chord combo is unambiguously superior in every respect and a big surprise was the improved depth and tonality (colour not colouration) to the bass part of the spectrum. I can live with the frankly odd and obscurest layout and controls.

I am not going to exercise myself or any reader with any cost justification but just listen to kit before you comment. I do believe that the common measurement suites employed are good but only to a point and are currently not comprehensive enough to confirm or dispute with subjective impressions. For example "timing" appears to be a bigger subject than simply measuring jitter values.

 

[threni]

I caught the DAC bug about 12 months ago. I thought I had found an acceptable "bang per buck" solution with the Gustard R26, a well received R2R box. Indeed it swept the floor with a Topping DX5 and Sabage A20d that used ESS chips and both of those sounded pretty good albeit for around £400. The Gustard is four times that. I'd also add that I am using a Sugden class A amp which brings instrument separation and soundstage width and depth goodies as well.

However serious audio as we know is a demanding mistress. I have invested in a second hand Chord Hugo TT2 and M Scaler and have been listening and comparing with the R2R for about two months. The Chord combo is unambiguously superior in every respect and a big surprise was the improved depth and tonality (colour not colouration) to the bass part of the spectrum. I can live with the frankly odd and obscurest layout and controls.

I am not going to exercise myself or any reader with any cost justification but just listen to kit before you comment. I do believe that the common measurement suites employed are good but only to a point and are currently not comprehensive enough to confirm or dispute with subjective impressions. For example "timing" appears to be a bigger subject than simply measuring jitter values.

It's not just cost you shouldn't feel the need to justify; it's also your belief you're getting good separation or soundstage from an amp, or that the Chord sounds better than far cheaper DACs etc. And let's not even mention the M Scaler which is an (expensive) piece of crap which actually makes the signal worse!

Getting back to the science, though - it's great that you've discovered this site, so a big hello, and we look forward to your suggestions for improvements to the test suites so that we can finally pin down exactly why we're failing to see any benefit in expensive audiofool jewelry.

 

[oleg87]

I do believe that the common measurement suites employed are good but only to a point and are currently not comprehensive enough to confirm or dispute with subjective impressions. For example "timing" appears to be a bigger subject than simply measuring jitter values.

"I don't understand any of it, but I'm sure the people who do are wrong"
No evidence of proper controls, no validity, nothing to discuss.