• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

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

Killingbeans

Major Contributor
Joined
Oct 23, 2018
Messages
4,089
Likes
7,547
Location
Bjerringbro, Denmark.
So I feel some of these things do have a sonic signature which maybe only become "visible" while dealing with real complex sets of tones and transients (music).
Do any such tests exist? If not what would they look like?

Even the most "complex" piece of music in the world is still just a sum of sine waves. Intermodulation distortion is also measurable.

An ideal transient requires infinite bandwidth, meaning that an impulse response doesn't tell you anything that a frequency response won't.

If you want to compare two DACs reliably, make sure they have the same frequency response within the audible band, do careful level matching and design the test to exclude as much cognitive bias as possible.
 

Sokel

Master Contributor
Joined
Sep 8, 2021
Messages
5,844
Likes
5,784
Just a disclaimer.
Level matching accurate settings must include the type of medium too.
Specially in DSD,level matching sometimes also includes double conversion (if done by DAC so VC can be applied) in some dacs or can be applied the analog way.
Whether the difference is audible or not...way above my pay grade.
 

Ported

Member
Joined
Jan 8, 2022
Messages
60
Likes
72
So far you have seemed to miss the point .. I would like to see a scientific measurement that compares one with another using music (so ears and arbitrary tones do not get involved) .. seemes a bit closed shop to assume we have all the measurements we need as just a bunch of tones.

Also if a difference is obvious at pretty much any level why would you need level matching? So simpler test to illustrate.. my TV has line out from its DAC .. it sounds pretty *hit and is easy picked out at any level agains the rme... They don't need to be level matched to assess that.
 

Soria Moria

Senior Member
Joined
Mar 17, 2022
Messages
395
Likes
810
Location
Norway
So far you have seemed to miss the point .. I would like to see a scientific measurement that compares one with another using music (so ears and arbitrary tones do not get involved) .. seemes a bit closed shop to assume we have all the measurements we need as just a bunch of tones.

Also if a difference is obvious at pretty much any level why would you need level matching? So simpler test to illustrate.. my TV has line out from its DAC .. it sounds pretty *hit and is easy picked out at any level agains the rme... They don't need to be level matched to assess that.
Music as far as I know is also just 'arbitrary' tones. Your gear doesn't know how to differentiate between music and test tones.
 

Sokel

Master Contributor
Joined
Sep 8, 2021
Messages
5,844
Likes
5,784
So far you have seemed to miss the point .. I would like to see a scientific measurement that compares one with another using music (so ears and arbitrary tones do not get involved) .. seemes a bit closed shop to assume we have all the measurements we need as just a bunch of tones.

Also if a difference is obvious at pretty much any level why would you need level matching? So simpler test to illustrate.. my TV has line out from its DAC .. it sounds pretty *hit and is easy picked out at any level agains the rme... They don't need to be level matched to assess that.
For me there are three measurements that show me what I get:
SINAD vs freq.
SINAD vs level.
Multitone.
Covers all the scenarios.
 

Blumlein 88

Grand Contributor
Forum Donor
Joined
Feb 23, 2016
Messages
20,524
Likes
37,057
So far you have seemed to miss the point .. I would like to see a scientific measurement that compares one with another using music (so ears and arbitrary tones do not get involved) .. seemes a bit closed shop to assume we have all the measurements we need as just a bunch of tones.

Also if a difference is obvious at pretty much any level why would you need level matching? So simpler test to illustrate.. my TV has line out from its DAC .. it sounds pretty *hit and is easy picked out at any level agains the rme... They don't need to be level matched to assess that.
You can try this bit of software. One method is to record the output of two devices and let this compare them to each other or record one device and compare to the original source. It will match levels for you even if you simply get them kind of close. It even lets you hear only what the difference is.


As for the idea differences are so obvious you don't need level matching, I could go thru all the reasons why. I suppose the short version is as obvious as you think it is level matching would leave it still obvious. A very, very simple variable you can eliminate so as not to cloud the results. What you'll likely find is suddenly some of those differences disappear. Even if you don't you'll be one step closer to convincing other people. There is good research showing even tiny level differences make people hear sound quality differences that aren't there.
 

antcollinet

Master Contributor
Joined
Sep 4, 2021
Messages
7,409
Likes
12,294
Location
UK/Cheshire
So far you have seemed to miss the point .. I would like to see a scientific measurement that compares one with another using music (so ears and arbitrary tones do not get involved) .. seemes a bit closed shop to assume we have all the measurements we need as just a bunch of tones.

That is only ever requested by people who don't understand how the waveform of real music is made up of a set of sinewaves, AND / OR the mechanisms by which distortion/noise are created in audio equipment. You don't need to test a device with real music in order to understand how real music is changed by that device.

I keep having to say this - but there is nothing mysterious about audio electronics. It is pretty much the simplest application of electronics - so much so that it was the first application of electronics over 150 years ago. We know, by now, how to do it, how it works, and how to test it.

Also if a difference is obvious at pretty much any level why would you need level matching? So simpler test to illustrate.. my TV has line out from its DAC .. it sounds pretty *hit and is easy picked out at any level agains the rme... They don't need to be level matched to assess that.
A difference can *seem* obvious if your sources are NOT level matched. IE you hear the different level of one and that:

a - allows you to identify which you are listening to, so not blinded .. and..
b - might make one sound better than the other.

If you don't level match (as well as the other necessary controls) you haven't blind tested.
 

Ported

Member
Joined
Jan 8, 2022
Messages
60
Likes
72
Happy for all those who think the current set of measurements is full and complete.. rest easy.

For those who believe they have detected a sound signature where those measurements suggest that would be impossible then I would like to see a digital process that can compare the outputs using music .. eliminate the similarities see if anything is left ... Thereby possibly forever shutting up anyone who thinks there may be any significant differences between all sorts of audio offerings these days.
 

voodooless

Grand Contributor
Forum Donor
Joined
Jun 16, 2020
Messages
10,227
Likes
17,807
Location
Netherlands
Thereby possibly forever shutting up anyone who thinks there may be any significant differences between all sorts of audio offerings these days.
It's very unlikely that people will shut up about it. They will just find a new audiophile of the gaps argument.
 

Julf

Major Contributor
Forum Donor
Joined
Mar 1, 2016
Messages
3,004
Likes
3,998
Location
Amsterdam, The Netherlands
For those who believe they have detected a sound signature where those measurements suggest that would be impossible then I would like to see a digital process that can compare the outputs using music .. eliminate the similarities see if anything is left ... Thereby possibly forever shutting up anyone who thinks there may be any significant differences between all sorts of audio offerings these days.
That is exactly what the null comparator (link a few messages back in the thread) does. The results show the absolute difference, but that says nothing about if the differences are audible or not.
 

antcollinet

Master Contributor
Joined
Sep 4, 2021
Messages
7,409
Likes
12,294
Location
UK/Cheshire
Happy for all those who think the current set of measurements is full and complete.. rest easy.

For those who believe they have detected a sound signature where those measurements suggest that would be impossible then I would like to see a digital process that can compare the outputs using music .. eliminate the similarities see if anything is left ... Thereby possibly forever shutting up anyone who thinks there may be any significant differences between all sorts of audio offerings these days.

When those people have clearly demonstrated that they actually can hear those differences, on devices which otherwise measure as transparent, with a properly controlled statistically valid blind test: Then I'm sure someone will be happy to help you further do that measurement with the devices tested.

You can start with the info you've been given above in post 3898, if you want to do it prior to properly validating the listening tests.
 

BDWoody

Chief Cat Herder
Moderator
Forum Donor
Joined
Jan 9, 2019
Messages
6,948
Likes
22,625
Location
Mid-Atlantic, USA. (Maryland)
For those who believe they have detected a sound signature where those measurements suggest that would be impossible then I would like to see a digital process that can compare the outputs using music .. eliminate the similarities see if anything is left ... Thereby possibly forever shutting up anyone who thinks there may be any significant differences between all sorts of audio offerings these days.

Why not actually demonstrate that you can determine the difference under the described conditions? Sounds like it should be easy, so why not put us all in our place and show what you claim?

From this end, the question would be why go through contortions when a difference can't be determined in the first place? Once someone can do that part...find that they can reliably (done with enough trials to be statistically valid) differentiate between two devices where measurements suggest that shouldn't be possible, then there is more testing to find. Until then, it's just more claims based on uncontrolled anecdote.

All that's missing is the evidence that more tests are needed.
 

Blumlein 88

Grand Contributor
Forum Donor
Joined
Feb 23, 2016
Messages
20,524
Likes
37,057
Happy for all those who think the current set of measurements is full and complete.. rest easy.

For those who believe they have detected a sound signature where those measurements suggest that would be impossible then I would like to see a digital process that can compare the outputs using music .. eliminate the similarities see if anything is left ... Thereby possibly forever shutting up anyone who thinks there may be any significant differences between all sorts of audio offerings these days.
The link in post #3898 is exactly what you are looking to find. Designed for this purpose and more.
 

Killingbeans

Major Contributor
Joined
Oct 23, 2018
Messages
4,089
Likes
7,547
Location
Bjerringbro, Denmark.
So far you have seemed to miss the point .. I would like to see a scientific measurement that compares one with another using music (so ears and arbitrary tones do not get involved) .. seemes a bit closed shop to assume we have all the measurements we need as just a bunch of tones.

We get the point. But our point is that those tests aren't needed. Music is "just a bunch of tones", when you get to the meat of it.

It's not a very romantic thought, but that's just the way it is. And it's a good thing. None of the technology we enjoy today would be around if fourier transforms didn't work.
 

fpitas

Master Contributor
Forum Donor
Joined
Jul 7, 2022
Messages
9,885
Likes
14,191
Location
Northern Virginia, USA
Why do audiophiles hate the Fourier transform so much? Were they promised there would be no math, or what?
 

Julf

Major Contributor
Forum Donor
Joined
Mar 1, 2016
Messages
3,004
Likes
3,998
Location
Amsterdam, The Netherlands
Why do audiophiles hate the Fourier transform so much? Were they promised there would be no math, or what?
"Music is art, not mathematics - but the mud-ear measurement fans seem to think mathematics is more important than the music". :)
 

Blumlein 88

Grand Contributor
Forum Donor
Joined
Feb 23, 2016
Messages
20,524
Likes
37,057
Why do audiophiles hate the Fourier transform so much? Were they promised there would be no math, or what?
Even if you don't grok the math, it can be explained in basic principle with simple pictures so one can at least get the concept. Which should make it easier to accept the accuracy of the idea. Not to mention all the signal analysis that can be done getting correct results using that principle.
 

fpitas

Master Contributor
Forum Donor
Joined
Jul 7, 2022
Messages
9,885
Likes
14,191
Location
Northern Virginia, USA
"Music is art, not mathematics - but the mud-ear measurement fans seem to think mathematics is more important than the music". :)
Must be how it is. What an ignorant position.
 

fpitas

Master Contributor
Forum Donor
Joined
Jul 7, 2022
Messages
9,885
Likes
14,191
Location
Northern Virginia, USA
Even if you don't grok the math, it can be explained in basic principle with simple pictures so one can at least get the concept. Which should make it easier to accept the accuracy of the idea. Not to mention all the signal analysis that can be done getting correct results using that principle.
I agree, but it's a common refrain, that all we use are tones for testing, but music is so much more! I mean, WTF.
 
Top Bottom