How much difference in sound quality will you hear between qutest and tone board ?

[Calexico]

Can you explain it? I chek the link ak240 did you find the etc result good?
I compared the etc from the ak240 with etc from nos dac and it's totally different looking. Wich is better?
https://reference-audio-analyzer.pro/en/report/dac/muse-audio-4x-tda1543-nos-dsc.php#rw31

For curiosity i compared the etc with different filters on the project box s2
Optimal transiant looks like nos dac
https://reference-audio-analyzer.pro/en/report/dac/pro-ject-pre-box-s2-digital-transient.php#rw31
And fast roll of looks like classic ds dacs
https://reference-audio-analyzer.pro/en/report/dac/pro-ject-pre-box-s2-digital-fast.php#rw31

 

[solderdude]

Look at the whole suite so also waveform, aliasing, multitone, frequency response.
Those will tel you more about the sound than the ETC.
It does have its place in a measurement suite.

 

[amirm]

2. It could show other differences between dacs. Like sinad show differences although almost all the time unhearable.

This is shown in my measurements of THD+N versus frequency:

 

[amirm]

3. It could tell us on anither parameter about the quality of engineering.

It takes time to make measurements and documented them. And more measurements serve to confuse and frustrate readers who don't want an encyclopedia of graphs to go through. I want to post the minimum that characterizes the performance of a product, not the maximum.

The current set of measurements have evolved through the above constraints. If you can clearly demonstrate the value of such a measurement, then I would consider it. But "it could tell us" is not it.

 

[amirm]

4. And there s some debate and some scientist have done tests that leads to conclusion that unhearble high frequency made a difference in the brain activity that was associated with enjoyment.

None of that is remotely reliable data. Regardless, I run tests like multi-tone which have 32 frequencies and use 192 kHz sampling to encompass a much wider bandwidth than our hearing.

 

[Calexico]

This is shown in my measurements of THD+N versus frequency:

A low thd 105db at 19khz means you have more than 16 bits of precision at 19khz ???
You say that only by measuring the thd you can get more than 65000 different levels at 19khz ??
I was thinking it was not correlated i'm quite surprised by your answer.

 

[Calexico]

It takes time to make measurements and documented them. And more measurements serve to confuse and frustrate readers who don't want an encyclopedia of graphs to go through. I want to post the minimum that characterizes the performance of a product, not the maximum.

The current set of measurements have evolved through the above constraints. If you can clearly demonstrate the value of such a measurement, then I would consider it. But "it could tell us" is not it.

How do you demonstrate the value of your measurements as a lot of them most of the time doesn't make differences to the hear?

 

[Calexico]

None of that is remotely reliable data. Regardless, I run tests like multi-tone which have 32 frequencies and use 192 kHz sampling to encompass a much wider bandwidth than our hearing.

Good point that you try new tests

 

[Xulonn]

This thread on a science-based website is totally ridiculous, because the antagonist never demonstrated that he eliminated the sight bias with proper double blind testing. He may "believe" that he hears differences, but he has not "demonstrated" it.

 

[amirm]

How do you demonstrate the value of your measurements as a lot of them most of the time doesn't make differences to the hear?

And you are saying your 19 kHz test is different? That tone is far less relevant to hearing than 1 kHz is which I and the industry in general use. Harmonics of 1 kHz fall in the most sensitive part of our hearing:

Harmonics of 1 kHz will be at 2, 3, 4, and 5 kHz, all of which are the part of the threshold curve that is the lowest (almost -10 dBSP).

See how I can tie my measurements to our ability to hear distortions? You need to do the same with your suggestions. Your 19 kHz tone harmonics will be at 38, 57 kHz, etc. which all fall outside of our hearing range.

 

[amirm]

A low thd 105db at 19khz means you have more than 16 bits of precision at 19khz ???
You say that only by measuring the thd you can get more than 65000 different levels at 19khz ??
I was thinking it was not correlated i'm quite surprised by your answer.

You can use SFDR to compute equiv. number of bits if you wanted. See https://www.audiosciencereview.com/...ital-audio-converters-dacs-fundamentals.1927/

A much more relaxed version is whether the distortion products are no higher than the noise level which would be what you state.

Regardless, as I explained you don't need the full "resolution" at 19 kHz. Your ears are not acute there as they are in mid frequencies.

 

[Calexico]

You can use SFDR to compute equiv. number of bits if you wanted. See https://www.audiosciencereview.com/...ital-audio-converters-dacs-fundamentals.1927/

A much more relaxed version is whether the distortion products are no higher than the noise level which would be what you state.

Regardless, as I explained you don't need the full "resolution" at 19 kHz. Your ears are not acute there as they are in mid frequencies.

Thank you. By the way i'm still curious if there are differences between dac chipsets or dac output stages at those frequencies.

 

[Blumlein 88]

Thnak

Thank you. By the way i'm still curious if there are differences between dac chipsets or dac output stages at those frequencies.

The answer is yes there are differences at 19 khz that are measurable. But why the focus on 19 khz?

As explained, it is a frequency you likely can't hear at all. Or if young enough you can hear it, but only at very high sound levels. Distortion from that is even higher. I like an intermodulation test at 18 khz and 19 khz though my reason is the likelihood of a good result there means as good or even better results lower in frequency.

 

[solderdude]

19-20kHz is more 'detected' as a weird 'pressure' than a high 'tone'.
That is my experience from about 30 years ago when I could just detect such frequencies.
Now my hearing tops out at around 14kHz, yet I don't think I hear any less 'hifi' and hear less 'treble extension' then I did then.

Most audiophiles I know that whine about 44/16 not being enough are older folks that most likely do not detect any sounds above 15kHz anyway.

Even if 19kHz distorted heavily the 2nd harmonic is at 38kHz ... yes... that will be very audible indeed.

 

[JJB70]

To answer the OP, if listening blind then none. If you did a sighted test you may think that you could hear a difference.

I'm sort of perplexed by the heavy focus on DACs these days and the proliferation of expensive DACs as of all the parts of the audio chain the DAC may be the link for which audible transparency can be achieved for the lowest price. Unless you need the functionality offered by some professional audio DACs or want a luxury product offering premium materials, build and tactile feel (and regardless of my views on the audible transparency of DACs I do understand the appeal of products from RME, Benchmark etc) then any competently designed DAC will be fine. In most cases I don't even think you need a separate DAC as the DACs in most devices seem to be audibly transparent.

 

[Calexico]

The answer is yes there are differences at 19 khz that are measurable. But why the focus on 19 khz?

As explained, it is a frequency you likely can't hear at all. Or if young enough you can hear it, but only at very high sound levels. Distortion from that is even higher. I like an intermodulation test at 18 khz and 19 khz though my reason is the likelihood of a good result there means as good or even better results lower in frequency.

Most tests dones show unhearable things.
If dac perform very good at 19khz i guess that it prooves at least that it is very good engeneered.
And does 18khz sine produce harmonics at 9khz for exemple ?
By the way if this test reveals big differences i think it's a interesting one (as most tests don't show big differences)

 

[Nango]

Yes, properly conducted blind tests are supposed to be level matched but I am quite sure some folks test 'blind' but do not level match. They just let someone 'switch' between 2 inputs and call it a blind test

... and are probably aware of which two or three items they are testing, they know their specs and their colors, etc.

 

[Nango]

Yes but by how much ? Should a dac convert data not change it.

By the way, who or what does convert the spaces between the bits (supposedly "1"s and "0"s) ... or is it nothing even no vacuum between them?

 

[andreasmaaan]

And does 18khz sine produce harmonics at 9khz for exemple ?

No, all harmonics are higher in frequency than the fundamental.

But when two or more tones are mixed together, this produces intermodulation products that can be lower in frequency. For example, if you ask a device to reproduce an 18kHz tone and a 19kHz tone at the same time, it will produce an intermodulation product at (19-18=) 1kHz.

 

[JanRSmit]

By the way, who or what does convert the spaces between the bits (supposedly "1"s and "0"s) ... or is it nothing even no vacuum between them?

The digitization is in fact points on a continuous sound. The pitch being the sampling rate. So the sound part in between is gone forever. In the analog reconstruct these in between sound parts are synthesized, and here the difference starts. So no it is not true that alle dacs sound the same. Simple frequency tests are not really close to a sound like music.