• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required as is 20 years of participation in forums (not all true). There are daily reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

A More Useful Way of Measuring the Performance of DACs?

Status
Not open for further replies.

tonycollinet

Major Contributor
Joined
Sep 4, 2021
Messages
3,609
Likes
5,591
Location
UK/Cheshire
With the Prism, the dCS now measures the best in dBA.

Am I misunderstanding the result that the PK software gives?

In both your real music measurements, the dCS has a higher level of difference than the other two - eg Diana Kral -77.57 versus the 3+db lower -80.96 for the RME


Loving the measurement attempts in general though.
 

Blumlein 88

Grand Contributor
Forum Donor
Joined
Feb 23, 2016
Messages
17,127
Likes
29,824
Am I misunderstanding the result that the PK software gives?

In both your real music measurements, the dCS has a higher level of difference than the other two - eg Diana Kral -77.57 versus the 3+db lower -80.96 for the RME


Loving the measurement attempts in general though.
The PK Metric is like null results or SINAD. Negative db and the lower the number the better. Yes -77.57 db is a less good PK result than -80.96 db.
 
  • Like
Reactions: EdW
OP
manisandher

manisandher

Addicted to Fun and Learning
Joined
Nov 6, 2016
Messages
523
Likes
429
Location
Royal Leamington Spa, UK
Am I misunderstanding the result that the PK software gives?

In both your real music measurements, the dCS has a higher level of difference than the other two - eg Diana Kral -77.57 versus the 3+db lower -80.96 for the RME

I think I understand the confusion here. There were 3 DACs tested, shown in the column 'DAC (2V output). Meanwhile, there were 2 ADCs used: RME and Prism. So, the RME was used both as a DAC and as an ADC.

The results for the two music tracks once again:

DeltaWave - Original2.JPG

DeltaWave - Temptation.JPG


The PK Metric measurements suggest that the dCS has the best audible and perceptual accuracy, but not the best objective accuracy. And yet the latter is all that is tested for here and elsewhere.

Hope this helps.
 
OP
manisandher

manisandher

Addicted to Fun and Learning
Joined
Nov 6, 2016
Messages
523
Likes
429
Location
Royal Leamington Spa, UK
Can you scientifically define music?

I can't define love... but I know it when I feel it. I can't define music, but I know it when I hear it.

To my ears, single or multiple sine tones are not music, no matter what anyone else here says.
 

Sokel

Major Contributor
Joined
Sep 8, 2021
Messages
2,241
Likes
1,755
To my ears, single or multiple sine tones are not music, no matter what anyone else here says.
Taste is always respected,for example I don't think there's real music for the last 100 or so years (with some exceptions) and that everything are covers of classical.
But that's me,alone.
People love different kind of music each one for it's own reasons.
An example:
If you ever listen to Pan Sonic you will find out that they're using all kinds of test tones some of them combined to each other or even with "real" music.
A complete torture to gear (and my ears) but some people love it!
Any way,I like the reasoning behind your test,I'm all for real world measurements that come close to practice.

Edit: If anyone go ahead and listen to them don't get scared about some parts that sound like 50 or 60Hz passing right though the mains like when an amp fails :facepalm:
 
Last edited:

voodooless

Master Contributor
Forum Donor
Joined
Jun 16, 2020
Messages
6,123
Likes
10,151
Location
Netherlands
I can't define love... but I know it when I feel it. I can't define music, but I know it when I hear it.
Then scientifically it very hard to make this useful.
To my ears, single or multiple sine tones are not music, no matter what anyone else here says.
It being subjective makes it even more problematic.
 

solderdude

Grand Contributor
Joined
Jul 21, 2018
Messages
13,033
Likes
29,356
Location
The Neitherlands
PK Metric with real music suggests clear differences

I would argue that there are no 'clear' differences (in PK) between DcS and RME. With 'clear' I don't mean the numbers but the audibility of it.
There are small differences and these differences are less than -80dB.

Now about perception.
Listen to the music at the levels you did.
Now attenuate that music 80dB and tell me what you hear.

Alternatively, listen to the music followed by the null (without any amplification). This is the perception part.
 
Last edited:

ahofer

Major Contributor
Forum Donor
Joined
Jun 3, 2019
Messages
2,984
Likes
5,242
Location
New York City
Interesting measurements. I’m still a bit hung up on the idea that what you hear and describe may not be the difference measured here - that the source of the PK score differences may not be audible.
 
  • Like
Reactions: MAB

PierreV

Major Contributor
Forum Donor
Joined
Nov 6, 2018
Messages
1,390
Likes
4,510
I’m 100% certain I’d pass an ABX with the dCS against either of these modern DACs.

First things first: congratulations on actually doing the extensive testing. It is an interesting post.

Just one comment about the prior certainty of passing an ABX test. I've been there, multiple times. I know that, if I had been challenged in my home by friends, my answer would have been "come on, pal- this is obvious: hear this, hear that, and so on!" I confess I may then have rationalized the differences on technical grounds explaining why the "this" and "that" were logical consequences of the design, the material, the whatever...

But, in fact, when I attempted a controlled ABX, I failed resoundingly in all cases but one...
 

tuga

Major Contributor
Forum Donor
Joined
Feb 5, 2020
Messages
3,599
Likes
3,699
Location
Oxford, England
It has to be anti-imaging filter response, as long as the power spectra match completely, or perhaps different frequency response, but that generally goes right back to anti-imaging filter.

That is, assuming other things in the test were done properly. Any shift in frequency response could have that effect, as could different bass responses due to cheap DC blocking capacitor in the DAC circuit, come to think of it.
But exactly what is annoyingly hard to determine. Maybe do an impulse response measurement of all 3?
What about noise-shaping, could it also be playing a part here? (maybe also affecting the frequency response)
I think I can hear differences between some of HQPlayer's noise-shaping algorithms, but I can't be bothered to blind-test them...
 

tuga

Major Contributor
Forum Donor
Joined
Feb 5, 2020
Messages
3,599
Likes
3,699
Location
Oxford, England
Perception is a totally different field of expertise than measuring electronics and acoustic things though.
Maybe a new website is needed APS (Audio Perception Science) but the abreviation is already taken.:)

In my view the two should not be split. Otherwise you'll have one side reporting listening observation and the other saying that it can't be because measurements bla, bla.
 

solderdude

Grand Contributor
Joined
Jul 21, 2018
Messages
13,033
Likes
29,356
Location
The Neitherlands
And this is exactly what is happening in this thread so yes, measurements and perception are different things.
Unfortunately, in the end, all that remains is perception.
Measurements, the correct ones, and these have to been done properly, can only tell something about technical performance/signal fidelity.
In audio, unfortunately there, is an overlap.

When you meant the part about a second website... this was a joke...
 
Last edited:
OP
manisandher

manisandher

Addicted to Fun and Learning
Joined
Nov 6, 2016
Messages
523
Likes
429
Location
Royal Leamington Spa, UK
My heartfelt thanks to everyone who's contributed to this thread to date, even if I happen to have disagreed with you.

I'll continue to do more null testing with DeltaWave. I intend to switch the RME (as ADC) to M/S-Proc mode to eke out the last vestiges of performance from it. I have a number of other DACs here (including the SMSL DO200 I bought for my 10-year-old) and will create my own 'PK Metric table'. If it proves useful, who knows, perhaps Amir will adopt a similar approach as a standard test for DACs?

Mani.
 

voodooless

Master Contributor
Forum Donor
Joined
Jun 16, 2020
Messages
6,123
Likes
10,151
Location
Netherlands
If it proves useful, who knows, perhaps Amir will adopt a similar approach as a standard test for DACs?
The key words here are “proves useful”. Perhaps having a larger sample set will shine more light on this indeed. The effort is definitely appreciated.
 
OP
manisandher

manisandher

Addicted to Fun and Learning
Joined
Nov 6, 2016
Messages
523
Likes
429
Location
Royal Leamington Spa, UK
A quick question for anyone who's looked into the PK Metric...

You are given two DACs to home test by a dealer. You take some captures with a good-quality ADC, and then analyse these with DeltaWave. You get the following (totally repeatable and consistent) results across a whole range of different musical genres:

DAC A:
- RMS Difference = -50dB
- PK Metric = -70dB
DAC B:
- RMS Difference = -70dB
- PK Metric = -50dB

Which DAC would you be inclined to keep for your long-term listening pleasure?

Mani.
 

solderdude

Grand Contributor
Joined
Jul 21, 2018
Messages
13,033
Likes
29,356
Location
The Neitherlands
Archimago also started doing this a while ago (but not with Deltawave) but abandonned it.
There is also or Russian guy on this forum pushing his difference files. I believe just before Paul came with Deltawave. I am sure @pkane knows who this is. He already had quite a collection of data on phones, DACs, MP3 players.

When it is not an official standard way of testing I don't think Amir will adopt it.

DAC A:
- RMS Difference = -50dB
- PK Metric = -70dB ?
DAC B:
- RMS Difference = -70dB
- PK Metric = -50dB

Which DAC would you be inclined to keep for your long-term listening pleasure?

The one with -50 is potentially audible different.
Whether it is RMS or PK metric it would warrant an investigation.
Would you be testing the original file vs a recording or compare 2 recordings and consider one 'reference' ?
 
Status
Not open for further replies.
Top Bottom