• 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). Come here to have fun, be ready to be teased and not take online life too seriously. We now measure and review equipment for free! Click here for details.

Beta Test: DISTORT - audibility of distortions

MakeMineVinyl

Major Contributor
Manufacturer
Joined
Jun 5, 2020
Messages
2,053
Likes
3,067
Location
Santa Fe, NM
In my experience, once an ear gets sensitized to what distortion sounds like, its pretty difficult to "unhear" it. Older analog recordings - jazz in particular - regularly drive me up the wall with tape saturation distortion, yet those around me wonder what my problem is.
 
Last edited:

mitchco

Addicted to Fun and Learning
Manufacturer
Joined
May 24, 2016
Messages
556
Likes
2,088
...
The curious result to me was that the -75dB THD distortion seemed to be preferred by a statistically significant number of listeners, even those who correctly identified it as being at a higher distortion level. I didn't have such a preference, but the difference between -75dB and -175dB THD to me was nearly impossible to tell.

I'd certainly be interested in doing more tests, for example comparing odd to even harmonics, or low-order harmonics to high-order harmonics, etc. Thank you to @Archimago for such an excellent test and detailed analysis!

Paul, I find the -75 dB THD statistic interesting as well. In other listening tests of bit-depth comparisons and broad band PEQ compared to the original signal, the best I could do was around -72 dB. Here is how I performed this test, along with audio samples that folks can download and try for themselves: https://audiophilestyle.com/ca/bits...gital-audio-–-bit-perfect-audibility-testing/

Spending time researching a while back, it seems that 12 bits or 13 bits of resolution is about the best most peoples ears/brain can discriminate and seems to correlate with Archimago's test. I wonder is this is more about people's audibility threshold of just a noticeable difference that triggers the preference... I also wonder if it even matters if it were THD or IMD or bit-depth or any other artefact that at this level our ears/brain simply can't discriminate the difference. Looking forward to your tests if you decide to vary the distortion components.
 
OP
pkane

pkane

Major Contributor
Forum Donor
Joined
Aug 18, 2017
Messages
2,497
Likes
4,072
Location
North-East
Paul, I find the -75 dB THD statistic interesting as well. In other listening tests of bit-depth comparisons and broad band PEQ compared to the original signal, the best I could do was around -72 dB. Here is how I performed this test, along with audio samples that folks can download and try for themselves: https://audiophilestyle.com/ca/bits-and-bytes/fun-with-digital-audio-–-bit-perfect-audibility-testing/

Spending time researching a while back, it seems that 12 bits or 13 bits of resolution is about the best most peoples ears/brain can discriminate and seems to correlate with Archimago's test. I wonder is this is more about people's audibility threshold of just a noticeable difference that triggers the preference... I also wonder if it even matters if it were THD or IMD or bit-depth or any other artefact that at this level our ears/brain simply can't discriminate the difference. Looking forward to your tests if you decide to vary the distortion components.

That’s a good question, Mitch. You may be right, there’s just not much distortion audible below a certain threshold. With pure test signals, such as sine waves and pink/white noise, it may be a audible at lower levels but music tends to obscure a lot of sins.
 

Newk Yuler

Active Member
Joined
Mar 5, 2019
Messages
155
Likes
233
I'd certainly be interested in doing more tests, for example comparing odd to even harmonics, or low-order harmonics to high-order harmonics, etc.

Even/odd harmonics would be as important to poll as this test was. It would be a worthy followup test.

It's a shame there weren't more participants. Seems there should be many long time folks here in ASR would want to participate if they were aware. (I missed it.) That's my fault for not looking at Archimago's blog more frequently.
 
OP
pkane

pkane

Major Contributor
Forum Donor
Joined
Aug 18, 2017
Messages
2,497
Likes
4,072
Location
North-East

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
35,029
Likes
130,724
Location
Seattle Area
Even/odd harmonics would be as important to poll as this test was. It would be a worthy followup test.

It's a shame there weren't more participants. Seems there should be many long time folks here in ASR would want to participate if they were aware. (I missed it.) That's my fault for not looking at Archimago's blog more frequently.
I too did not see it till now.
 

maty

Major Contributor
Joined
Dec 12, 2017
Messages
3,928
Likes
2,568
Location
Tarragona (Spain)
Nope. Saturation, noise... I have tried many and none has convinced me these years. Only I want to add a bit of H2 when I play a small group of acoustic instruments (if voices, without Autotune). It would also be interesting to be able to subtract H3, thinking of those who have quite powerful SS amp and who usually suffer from being dominant H3.

Now I have managed to have more depth even though listening is in the near field. Playing with H2 maybe I could increase it even for complex recordings (orchestral, electronic, ...).
 

maty

Major Contributor
Joined
Dec 12, 2017
Messages
3,928
Likes
2,568
Location
Tarragona (Spain)

Attachments

  • GreatNrews-RSS-reader-nauscopico.png
    GreatNrews-RSS-reader-nauscopico.png
    72.7 KB · Views: 26

Dimifoot

Senior Member
Forum Donor
Joined
Mar 27, 2019
Messages
458
Likes
633
Location
Greece
I am copying from @Archimago findings:
“Let's now summarize the main "take home" points from this study......
.......Just as "extreme subjectivists" might erroneously claim measurements make no difference, we need not be "extreme objectivists" either thinking that a -120dB THD DAC "sounds better" than one at -100dB beyond just being able to appreciate the engineering efforts to achieve excellent performance!”

That’s something to re-consider in the AV products discussions

:)
 

DDF

Addicted to Fun and Learning
Joined
Dec 31, 2018
Messages
565
Likes
1,070
In my experience, once an ear gets sensitized to what distortion sounds like, its pretty difficult to "unhear" it. Older analog recordings - jazz in particular - regularly drive me up the wall with tape saturation distortion, yet those around me wonder what my problem is.

What makes these tests tricky is that without tight controls on the playback chain distortion, its difficult (if not impossible) to determine what perceptual effects are actually being tested: absolute perception thresholds of added distortion or thresholds of how the added distortion interacts with the playback chain. This adds significant variability to the outcome making it almost impossible to make any statistical conclusions.

For example phase addition, cancellations and beatings of added harmonics with signal chain distortion, especially those that change how the distortion components are masked or unmasked by the signal envelope, which significantly affects audibility thresholds.

There are dbt papers in the audition field where this has been studied in a limited way.
 

MakeMineVinyl

Major Contributor
Manufacturer
Joined
Jun 5, 2020
Messages
2,053
Likes
3,067
Location
Santa Fe, NM
What makes these tests tricky is that without tight controls on the playback chain distortion, its difficult (if not impossible) to determine what perceptual effects are actually being tested: absolute perception thresholds of added distortion or thresholds of how the added distortion interacts with the playback chain. This adds significant variability to the outcome making it almost impossible to make any statistical conclusions.

For example phase addition, cancellations and beatings of added harmonics with signal chain distortion, especially those that change how the distortion components are masked or unmasked by the signal envelope, which significantly affects audibility thresholds.

There are dbt papers in the audition field where this has been studied in a limited way.
I once pondered what effect, if any, recordings of instruments using the equally tempered musical scale had on electronic distortion perception verses recordings of instruments with just intonation. With equal temperament most notes of the scale are in fact "out of tune" compared to just intonation. Does that mess with distortion perception in that electronic distortion components themselves follow natural harmonic progression while the musical scale is equally tempered (with those "out of tune" notes's own harmonics following the natural harmonic progression)? I stopped thinking about it, blamed J.S. Bach, and had a drink.
 

RayDunzl

Grand Contributor
Central Scrutinizer
Joined
Mar 9, 2016
Messages
11,509
Likes
12,154
Location
Riverview FL
With equal temperament most notes of the scale are in fact "out of tune" compared to just intonation.

Out of tune? Try different scales - where there are more than 12 notes to an octave...

Click on through to YouTube for the tracklist and descriptions below that.

 

MakeMineVinyl

Major Contributor
Manufacturer
Joined
Jun 5, 2020
Messages
2,053
Likes
3,067
Location
Santa Fe, NM
Out of tune? Try different scales - where there are more than 12 notes to an octave...

Click on through to YouTube for the tracklist and descriptions below that.

Oh, I listen to that stuff! Used to drive my parents crazy because I didn't listen to 'normal' music.
 

RayDunzl

Grand Contributor
Central Scrutinizer
Joined
Mar 9, 2016
Messages
11,509
Likes
12,154
Location
Riverview FL
For those who may not be familiar with the pitches the harmonic series adds to a fundamental, here's a nice example.

 

MakeMineVinyl

Major Contributor
Manufacturer
Joined
Jun 5, 2020
Messages
2,053
Likes
3,067
Location
Santa Fe, NM
That's a great demo, but my ear drove me nuts wanting it to resolve up to the octave when it had the maximum harmonics - kind of like the old THX "tone" but not having it resolve, leaving you hanging.

Getting back to distortion, one can think of those harmonics as literally the same thing as harmonic distortion. Instruments are awash with distortion, your voice is a distortion generator, otherwise everything would sound like sine wave generators. So our electronics not only have to deal with passing all the baked-in "distortion" from musical instruments, but not add its own harmonics to the mix.

And then intermodulation distortion.......
 
Last edited:
Top Bottom