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

[totti1965]

David Mellor from audio masterclass and Alain Pauls are YouTubers and Audio engineers.
They wanted to find out, if Differences between DAC are audible.

So both of them started polls in which they asked the audience if it is possible to hear, how often they switched between an original audio file and it looped signal with an extra conversion.

They also provided the original wave files, because it’s known that the YouTube soundqualtiy is compressed.

The results were telling: No audible differences at all!


David Mellor has the results now: 505 Participants!
And I have left a comment, as Davis suggested: Beginn your comment with Your test is invalid because……
So I wrote: „Your test is invalid because you have crushed all my audible Illusions!“

Check also out Alain Pauls (AP Masterings) video (same test, same result, 300 Participants)

 

[DanielT]

It was good. Well done with music files to download. Check out the first video regarding the test:

Screenshot above from this video:

 

[ahofer]

Should cross post in the DAC difference garbage thread, which I can't locate right now.

 

[DanielT]

For those of you who don't want to watch the video, here is the file. How many times does it play with a mediocre DAC?

loopback-test.wav

drive.google.com

 

[totti1965]

For those of you who don't want to watch the video, here is the file. How many times does it play with a mediocre DAC?

loopback-test.wav

drive.google.com

53.50 Great - Britain- Pounds (ADC included) !!!!!

 

[dominikz]

Thanks for sharing!

After more than 500 attempts in the listening test I prepared a few years ago (link to test and link to latest test results) the conclusion was unsurprisingly similar.

Even the fact that the cheap DAC in my test introduced significant high-frequency response deviation didn't seem to help much with audibility... going into it I was half-expecting the test to be too easy because of this, but that was not the case at all, it turned out.

So we're talking about a SINAD difference of roughly 30dB between the two DACs, with the cheaper DAC also having non-flat frequency response, and still the vast majority of test attempts indicated no difference was perceived.

Of course many people will point out limitations of my test (some IMHO valid, and some not so much), which was also extensively discussed in the thread containing it.

 

[DanielT]

Then you can wonder if that is the most optimal setup for a blind test? He should have the guitarist Mick Hutchings record a new track and then have two versions of the same track. One that has passed his mediocre DAC and a "clean" version. Then blind test them AB. The result would most likely be the same but still.
Or there is an advantage with the quick changes that occur with the test track as it is. The auditory memory factor can then be ruled out.

With that said, good job with the test David Mellor (Audio Masterclass). A good test recording. Clean guitar sound that David Mellor or rather Mick Hutchings created.

 

[DanielT]

Thanks for sharing!

After more than 500 attempts in the listening test I prepared a few years ago (link to test and link to latest test results) the conclusion was unsurprisingly similar.

Even the fact that the cheap DAC in my test introduced significant high-frequency response deviation didn't seem to help much with audibility... going into it I was half-expecting the test to be too easy because of this, but that was not the case at all, it turned out.

So we're talking about a SINAD difference of roughly 30dB between the two DACs, with the cheaper DAC also having non-flat frequency response, and still the vast majority of test attempts indicated no difference was perceived.

Of course many people will point out limitations of my test (some IMHO valid, and some not so much), which was also extensively discussed in the thread containing it.

Good job. Thanks for the information you just provided. That was exactly what I was thinking about so now I, and other curious people, got those answers.

 

[DSJR]

What the hell is that 'music' track used? I mean, no bass, no top, no reverb tails to concentrate on, just a VERY close-miked guitar strumming of the type many gear-loving audiophiles tend to listen to whole albums of

Even on more demanding full range well produced music, I'd say that dac differences, unless they've been heavily 'tuned' to have a 'sound,' don't really exist these days, but there are many gear-enthusiasts way back on the journey I completed years ago, that have serious trouble with this concept and will fight it to the end (forgive me if I sound patronising there, I really don't mean to, but my 'sonic epiphanies' came many, many years ago now)

 

[ahofer]

What the hell is that 'music' track used? I mean, no bass, no top, no reverb tails to concentrate on, just a VERY close-miked guitar strumming of the type many gear-loving audiophiles tend to listen to whole albums of

I know. Needs a good symphony. Make it the Mahler 6 hammer strike, or Sibelius 5 horns off-stage, for good measure.

 

[DSJR]

I know. Needs a good symphony. Make it the Mahler 6 hammer strike, or Sibelius 5 horns off-stage, for good measure.

There's some good rock music too - of the varied type rather than full-on 'metal.'

 

[westsail42]

question is, what is this "cheap dac" that not only does D/A but also A/D? Is there any gain stage in between? nothing? he does not say what the device is.

 

[DanielT]

What the hell is that 'music' track used? I mean, no bass, no top, no reverb tails to concentrate on, just a VERY close-miked guitar strumming of the type many gear-loving audiophiles tend to listen to whole albums of

Even on more demanding full range well produced music, I'd say that dac differences, unless they've been heavily 'tuned' to have a 'sound,' don't really exist these days, but there are many gear-enthusiasts way back on the journey I completed years ago, that have serious trouble with this concept and will fight it to the end (forgive me if I sound patronising there, I really don't mean to, but my 'sonic epiphanies' came many, many years ago now)

I know. Needs a good symphony. Make it the Mahler 6 hammer strike, or Sibelius 5 horns off-stage, for good measure.

Maybe you both have a point but it sounds more like you are suggesting music with lots of dynamics, lots of instruments. That in itself might make sense if you want to push the pedal to the metal with your speakers and or amplifier to see what they can achieve before they say no more. But for the purpose of a "mediocre DAC" that "pollute" vs the clean signal, it should work with a man who plays a little electric guitar.
By should work I mean functioning as a basis for carrying out the test itself.

 

[ahofer]

But for the purpose of a "mediocre DAC" that "pollute" vs the clean signal, it should work with a man who plays a little electric guitar.
By should work I mean functioning as a basis for carrying out the test itself.

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.

 

[MAB]

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.

Is it this Sean Olive chart?

SoundStage! Access | SoundStageAccess.com (GoodSound.com) - Dr. Sean Olive on 35 Years of “Fast Car”

Of the female pop-rock, Tracy Chapman's 'Fast Car' was the song studied with the highest score for evaluating reproduction.

People often misinterpret this to mean it's because other genre require different reproduction qualities... Not true, other genre have lower performance of the listener to hear actual differences in reproduction. The best test signal for determining reproduction performance is pink noise.

 

[ahofer]

The best test signal for determining reproduction performance is pink noise.

Thanks for retrieving that!

Pink noise has other performance problems:Nobody wants to spend time listening to it.

 

[kemmler3D]

Personally if I want to evaluate gear, I listen to pink noise and sweeps first. Whatever I'm listening for is easiest to find that way. If I hear anything in particular, I can evaluate the subjective effect on music once I establish it's not a piece of junk.

I used to go to trade shows and need to check dozens of speakers or headphones in a day, so we could find factories worth following up with. Test signals save a lot of time in that scenario.

"In real life", testing gear with music is mostly a waste of time. For hobby purposes or for writing flowery prose it's useful, but not if you just need to identify flaws quickly. Also, listening to the first 15 seconds of the same song 40 times a day isn't much more fun than listening to pink noise.

The reviewers use case is evaluating one piece of gear thoroughly, but for the buyer checking many systems against each other, pink noise makes it much easier. Not sure if they would let people play pink noise at a HiFi show, but they should.

 

[ahofer]

Personally if I want to evaluate gear, I listen to pink noise and sweeps first. Whatever I'm listening for is easiest to find that way. If I hear anything in particular, I can evaluate the subjective effect on music once I establish it's not a piece of junk.

I used to go to trade shows and need to check dozens of speakers or headphones in a day, so we could find factories worth following up with. Test signals save a lot of time in that scenario.

"In real life", testing gear with music is mostly a waste of time. For hobby purposes or for writing flowery prose it's useful, but not if you just need to identify flaws quickly. Also, listening to the first 15 seconds of the same song 40 times a day isn't much more fun than listening to pink noise.

The reviewers use case is evaluating one piece of gear thoroughly, but for the buyer checking many systems against each other, pink noise makes it much easier. Not sure if they would let people play pink noise at a HiFi show, but they should.

Some day I'm going to go around to the remaining hifi boutiques in the city and blast pink noise over the Wilsons, Audio Notes, and B&Ws.

 

[Vuki]

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?

 

[MAB]

Thanks for retrieving that!

Pink noise has other performance problems: Nobody wants to spend time listening to it.

Agree, only so much of that we can take. However, many years ago I used to use a beat up Velodyne sub and a tone generator to chase down rattling noises in customer's rooms, mostly ducting and windows. I have used pink noise and a laptop to evaluate a pair of JBL XPL at a garage sale. I ended up not buying them despite they were fine sounding...