• 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!

Distortion Listening Test

JohnYang1997

Master Contributor
Technical Expert
Audio Company
Joined
Dec 28, 2018
Messages
7,175
Likes
18,292
Location
China
I did extensive tests on this site last year. I don't want to do it because it's tiring. What I learnt was, I can definitely hear better than what's on the test. Problem is our ears also distort so I always hear the same distortion on both samples. I suggest listening at low volume then gradually increase as it getting more difficult. If played too loud from the beginning, the test causes hearing fatigue and mixed audio memory. You want to take a rest after -30db. It's fairly easy to get -36db or below. It's basically conflict and trade off between audibility of the distortion and the distortion of our ears. The test music is horrible btw....
 

Blumlein 88

Grand Contributor
Forum Donor
Joined
Feb 23, 2016
Messages
20,522
Likes
37,050
I have several pairs including JBL 305p mkII. Also, headphones like AKG K701.
Assuming they are as good as the 305 mk 1(they might be better) they are close. Here is a 305 playing 95 db at 1 meter from the speaker in 34 ft x 19.5 ft x10 ft room. Which is really too large for them. It bounces around .4% to just over 1% mostly at 300 hz and above.

1570597511780.png
 
Last edited:

daftcombo

Major Contributor
Forum Donor
Joined
Feb 5, 2019
Messages
3,687
Likes
4,068
Assuming they are as good as the 305 mk 1(they might be better) they are close. Here is a 305 playing 95 db at 1 meter from the speaker in 34 ft x 19.5 ft x10 ft room. Which is really too large for them. It bounces around .4 to just over 1% mostly at 300 hz and above.

View attachment 35532
I don't listen that loud but you're right.

On the distortion test, I can't beat "real speakers" level with AKG K701.
 

Sedyn

Member
Forum Donor
Joined
Oct 20, 2019
Messages
25
Likes
10
Interesting test. I used the default settings of -- and Tracy Chapman and found the quality of the recording so bad I wasn't sure what was distortion and what was just bad audio when getting into the -'s. It's tiring and repetitive with distortion in both samples, so I have no idea which one is the one to pick past well, that one is horrible. Not sure I feel like I gained anything from it. Pretty sure I should upgrade my distortion laden soundcard and try again.

I also tried the Full Speaker. The lowest I got was -39 and the test result was -27. By the end I wasn't even listening to the full clip and there were pop distortions in both samples. Just don't have the patience to listen to it anymore. Gotta wake up in 5 hours for work.
 
Last edited:

dorirod

Active Member
Joined
Oct 9, 2019
Messages
249
Likes
249
Well, if these tests are accurate, it's great for my wallet (bottom of the rankings here I come) because I can't seem to tell distortion past -15dB on "full range speaker" playing "Music J. Stone". The "--" playing "Music T.Chapman - Bl(x) only " I only got to Real Speaker. Maybe I don't know what to look for though especially on such a short clip. Why such a big discrepancy between the tests? It was really easy to tell at -12dB in the full range speaker, but I had a really difficult time with the '--' option. I guess I'm lucky and can't tell distortion :)

-18dB in this one "guitar_speaker" playing "Music T. Chapman - all distortion" , but I was hearing it on the drum not the guitar parts.
 
Last edited:

Thomas_A

Major Contributor
Forum Donor
Joined
Jun 20, 2019
Messages
3,422
Likes
2,407
Location
Sweden
I heard down to -33 dB according this test and, assuming that this value is what I think it is, distortion levels below -40 dB/1 % should be reasonably safe for most music genres. Pure test tones are easier to hear distortion from, so perhaps some music signals (e.g. electronic music) may need even lower figures.

Distorsion of my speaker, SPL unknown in this case (have not calibrated the microphone). 3rd harmonic distortion goes above -50 dB around 1-2 kHz.
distortion.png
 
Last edited:

DonH56

Master Contributor
Technical Expert
Forum Donor
Joined
Mar 15, 2016
Messages
7,835
Likes
16,497
Location
Monument, CO
Years ago I created a series of files with different amounts of distortion just for fun. With single tones I could "easily" hear 1%, much harder to pick out 0.1% to the point of guessing much of the time. With a two-tone test 0.1% IMD was reliably distinguished. When I switched to music, 1% HD was hard and 5% was actually acceptable much of the time (depending upon frequencies and such). IMD was still easier to pick out but I had to listen more closely, usually to certain passages, to hear it. But I am far from being a trained listener at this time; too many years since I did a bunch of serious listening and went through all the DBTs and such. Just average now, hmph.
 

anmpr1

Major Contributor
Forum Donor
Joined
Oct 11, 2018
Messages
3,722
Likes
6,406
From the Boston Audio Society Speaker [August 1976] archives:

With hi-fi advertisements boasting equipment distortion levels of 0.05% and less, many people would be surprised to learn that their threshold for perception of distortion in a complex waveform is 2% to 3%. This was demonstrated at AR [Acoustic Research] with a standard psychoacoustic testing technique. The subject, listening to a signal, controls the level of distortion with a switch, causing it to rise or fall gradually at will. He is told to decrease the distortion until it is no longer audible to him and then to increase it until it can again be heard...

One of the audience, eager to demonstrate his aural acuity, scored right along with the majority, detecting 0.2% sine-wave distortion but falling back to a 2.5% distortion threshold on a sawtooth waveform, where the rich spectrum of harmonics tends to mask harmonic distortions.


Results similar to an earlier test Stereo Review did with one of Bob Carver's modified amps (or maybe it was a preamp) that Bob modded to inject known quantities of distortion into the signal.
 

Thomas_A

Major Contributor
Forum Donor
Joined
Jun 20, 2019
Messages
3,422
Likes
2,407
Location
Sweden
The problem is that music signals can vary so much, so being on the safe side is just a good thing.
 

anmpr1

Major Contributor
Forum Donor
Joined
Oct 11, 2018
Messages
3,722
Likes
6,406
The problem is that music signals can vary so much, so being on the safe side is just a good thing.
Absolutely. I in no way endorse amps having distortion more than is design-practical at a given price point. 0.1% seems to be about the most anyone ought to accept in a solid state electrical circuit. Besides, one's loudspeakers (and other analog sources if you use them) will provide enough distortion to satisfy. :)
 

Armand

Member
Audio Company
Joined
Jun 25, 2018
Messages
93
Likes
321
My result. Active speakers with miniDSP 2x8 and UcD400 in a three way system. Pretty happy with the result considering. My tell was the click around when she sings "babe" in the beginning.
I could not hear any anomaly at the end. Have they fixed the test?
I don't beleive the -54 result. Pretty easy to analyse this if one should bother. (And I am sure someone will )
1578168761175.png
 

ConnorG

Member
Joined
Jan 7, 2020
Messages
11
Likes
43
Location
MA
I managed to get a perfect score using STAX SR-001 Mk2 IEMs and an RME ADI-2 DAC as a source.
I found that some of the "lower" distortion tests were easier to decipher than some "higher" distortion tests. Very strange.
I think this is a great ear trainer nonetheless!

Klippel.JPG
 
Last edited:

MZKM

Major Contributor
Forum Donor
Joined
Dec 1, 2018
Messages
4,240
Likes
11,462
Location
Land O’ Lakes, FL
I managed to get a perfect score using STAX SR-001 Mk2 IEMs and an RME ADI-2 DAC as a source.
I found that some of the "lower" distortion tests were easier to decipher than some "higher" distortion tests. Very strange.
I think this is a great ear trainer nonetheless!

View attachment 44895
Very impressive, did you try the Chapman track?
With that two-tone test, I was able to get -51dB with Fidelio X2HR headphones with an Apple dongle; only to the -20’s with the Chapman track.
 
Last edited:

despoiler

Member
Joined
Jan 8, 2019
Messages
45
Likes
50
I managed to get a perfect score using STAX SR-001 Mk2 IEMs and an RME ADI-2 DAC as a source.
I found that some of the "lower" distortion tests were easier to decipher than some "higher" distortion tests. Very strange.
I think this is a great ear trainer nonetheless!

View attachment 44895

What do you get with the same test Amir used? Real music was harder than the two tone test for me.

Here's mine.

Klippel.jpg


Klippel Test Process.jpg
 

ConnorG

Member
Joined
Jan 7, 2020
Messages
11
Likes
43
Location
MA
I didn't bother with the music tests, as they aren't as linear as the sine wave tests. There would be no way to get an "accurate" self measurement IMO.
On the flip side, the music tests are more "realistic" for an every day listening test
 

solderdude

Grand Contributor
Joined
Jul 21, 2018
Messages
15,891
Likes
35,912
Location
The Neitherlands
The highest danger in this test for scoring low is the used stimulus (tones vs music) and above all the used transducer/gear and experience of the tester.

You can see the higher pole on the left which means more people ace the test compared to folks reaching -48. This is no fluke. The test can be passed with excellent gear (ADI-2 and stats qualify).

None the less it is important to test ones own abilities with gear they own to see how far they get. It could also tell that person he should buy better gear or just sucks at this.
In all cases the owners of said gear most likely fully enjoy music from their systems.
 

xr100

Addicted to Fun and Learning
Joined
Jan 6, 2020
Messages
518
Likes
237
Location
London, UK
Ran into this interesting listening test by Klippel with regards to how much distortion you can hear.

This is the link: http://www.klippel.de/listeningtest/lt/

I tried "full range speaker -- Joss Stone."

Alas, for some reason I couldn't get the audio routed to my "proper" "home office" system (EQ/active crossovers using plug-ins in Reaper) so I ended up using my "desktop" speakers (Bose Computer MusicMonitors, normally used for system sounds, random YouTube videos, etc.) fed by on-board sound. (ESS 9023P on an Asus motherboard.)

I got to -21dB at the "second turning point" and called it a day (for now) at that point.

Useful test, but I think it would be far better if the user's choice of music could be loaded, as well as being able to customise distortion type/process.

Also worth noting, I think, that if this test uses static "waveshaping" distortion, then this may well be useful in investigating the threshold of audibility, but more sophistication can be needed to properly "emulate" the real-world behaviour of hardware. (Power compression etc. being an obvious one with loudspeakers.)
 
Last edited:

xr100

Addicted to Fun and Learning
Joined
Jan 6, 2020
Messages
518
Likes
237
Location
London, UK
I didn't bother with the music tests, as they aren't as linear as the sine wave tests. There would be no way to get an "accurate" self measurement IMO.

Distortion is far more audible on sine waves than complex music signals. You can add what sounds like pretty severe distortion to sine waves, but the impression on music might well be relatively subtle. Distortion may mimic the ear's own non-linearities and be "ignored" by the brain, perhaps being perceived as "louder." But this does point to the problem in terms of what kind of sounds are being distorted and the extent to which masking occurs etc.
 

xr100

Addicted to Fun and Learning
Joined
Jan 6, 2020
Messages
518
Likes
237
Location
London, UK
- I love Tracy Chapman's music and found her recording to be decent back then, but I don't understand how it can be used as reference these days. In a way, it is the Lenna (https://en.wikipedia.org/wiki/Lenna ) of audio. I can see someone publishing a paper about TC's recording as being among the best in the early 90ies, others quoting that paper and using is as a standard, but I just find it OKish these days.

"Fast Car" was released in 1988. The 1980s was the era of "clean" recording, and there are many tracks which sound far better.

Avoiding embarrassing myself with a selection of "disreputable" synth-pop examples, an obvious well-known example is Donald Fagen - "The Nightfly." (Vocals still sound quite coloured though.)
 
Last edited:
Top Bottom