Re: Think that your ears can detect distortion? Try this!
I was lazy and used the "Creative" satellite speakers hooked up to the PC; and reached -24 db reliably on the telecom 7-1 terz test. I would have posted a snapshot of the screen like others; except I couldn't figure out how!!
Re: Think that your ears can detect distortion? Try this!
Interesting tests. Poor test environment here, I ranged from average to top tier in the various tests. Bet I could do better without earbuds (one missing a foam cover) and work noise!
OK, I retook the tests and wanted to share the results. I encourage others to do the same. I did a screen dump of the results section (that shows where you stand by a red bar). This shows the distribution of the scores of all previous test takers as a bar chart.
The two tests I used were "6-inch octave" and "telecom 7-1 terz" the former an "easy" one and the latter supposedly more masking. I did two sets of trials, one using my stereo system and the other with a pair of closed (over the ear) headphones. There was little to no background noise.
A pdf file containing my results can be found here. Below is a summary:
Loudspeakers, 6-inch octave: -36dB
Loudspeakers, Telecom-7 terz: -27dB
Headphones, 6-inch octave: -33dB
Headphones, Telecom-7 terz: -12dB
It is interesting that when I switched to headphones I had much worse perception of the terz distortion. I am not sure that the levels were exactly matched, and maybe the (likely higher) level in the phones caused a greater level of distortion within my ear or perhaps the distortion of the phones is higher, making it difficult to make out distortion in the signal. But then why the vast difference only for the terz?
A pdf file containing my results can be found here. Below is a summary:
Loudspeakers, 6-inch octave: -36dB
Loudspeakers, Telecom-7 terz: -27dB
Headphones, 6-inch octave: -33dB
Headphones, Telecom-7 terz: -12dB
It is interesting that when I switched to headphones I had much worse perception of the terz distortion. I am not sure that the levels were exactly matched, and maybe the (likely higher) level in the phones caused a greater level of distortion within my ear or perhaps the distortion of the phones is higher, making it difficult to make out distortion in the signal. But then why the vast difference only for the terz?
-Charlie
Bump...
Maybe it has something to do with channel crosstalk...
Re: Think that your ears can detect distortion? Try this!
Interesting test. I used my desktop Z560 Logitech speakers 3" sats & 8" sub, and my headphone setup Beyerdynamic DT770s and Firestone amp DAC. I didn't try it on any other speakers.
One thing I've noticed as it gets harder the longer I listen to the samples the worse I do as the tests get harder. Even if I'm not sure on what I'm hearing if I just listen to 1 or 2 seconds of each sample and make a choice I tend to be right, anyone else have this experiance?
Its almost as if my brain knows the distortion is there and filters it out at the lower levels.
Re: Think that your ears can detect distortion? Try this!
I took it just now sitting at my desk with my noisy computer using my little sennheiser jogging headphones and did pretty decently in every test except for the single tone test.
Sometime I'll try it with my stereo or my HD600s in a quiet room, see how I do and post it. I'll agree with the previous poster tho. The more you listen the harder it is as you get down into very low levels. You're usually better off just listening for a second and going with your gut.
Re: Think that your ears can detect distortion? Try this!
Originally Posted by evilskillit
I took it just now sitting at my desk with my noisy computer using my little sennheiser jogging headphones and did pretty decently in every test except for the single tone test.
Sometime I'll try it with my stereo or my HD600s in a quiet room, see how I do and post it. I'll agree with the previous poster tho. The more you listen the harder it is as you get down into very low levels. You're usually better off just listening for a second and going with your gut.
I did better on the octave test with my computer speakers than I did with the terz test and headphones. I put damping around the computer to mask the noise from the fans which totally mask the distortion component at low levels. I did the best by listenening all of the way through and listening for the noise artifacts (modulation) imprinted on the tones.
Dave
"A man with an experience is never at the mercy of a man with an argument." - Hilmar von Campe
Re: Think that your ears can detect distortion? Try this!
I did pretty well on my current digital system, low-level listening (it's 24-hr quiet hours for finals) on my Fostex FE167E w/ P10 phase plug loaded TLs and tube amp, upsampling 24/96 Firewire DAC.
"It has been remarked that if one selects his own components, builds his own enclosure, and is convinced he has made a wise choice of design, then his own loudspeaker sounds better to him than does anyone else's loudspeaker. In this case, the frequency response of the loudspeaker seems to play only a minor part in forming a person's opinion."
L.L. Beranek, Acoustics (McGraw-Hill, New York, 1954), p.208.
Re: Think that your ears can detect distortion? Try this!
Originally Posted by nick29498141
I did pretty well on my current digital system, low-level listening (it's 24-hr quiet hours for finals) on my Fostex FE167E w/ P10 phase plug loaded TLs and tube amp, upsampling 24/96 Firewire DAC.
NK
It would be more instructive if you completed the test and then posted the final result instead of this screen shot, which is only part of the way through. The test proceeds like this: the level of distortion is lowered stepwise. When you cannot correctly detect distortion, the level in increased about 12dB and then decreased again until you again cannot detect the distortion. This is done three (or four?) times before the test is completed, at which point you get your "score", shown as a bar graph with your detection limit in red. The mean of the detection levels during the test is chosen to be the "score" you are assigned. Evidently the mean is a better indicator than average or the results of one run (which is what you show).
Also, you did not indicate which test you took. This makes a good deal of difference, e.g. the results can vary a lot between the different types of tests. Please include that in your post when referring to results and please indicate whether you are listening using headphone or with speakers of some kind. That would help to put your comments in to perspective with all the other preceding ones.
Re: Think that your ears can detect distortion? Try this!
Originally Posted by charlielaub
It would be more instructive if you completed the test and then posted the final result instead of this screen shot, which is only part of the way through. The test proceeds like this: the level of distortion is lowered stepwise. When you cannot correctly detect distortion, the level in increased about 12dB and then decreased again until you again cannot detect the distortion. This is done three (or four?) times before the test is completed, at which point you get your "score", shown as a bar graph with your detection limit in red. The mean of the detection levels during the test is chosen to be the "score" you are assigned. Evidently the mean is a better indicator than average or the results of one run (which is what you show).
Also, you did not indicate which test you took. This makes a good deal of difference, e.g. the results can vary a lot between the different types of tests. Please include that in your post when referring to results and please indicate whether you are listening using headphone or with speakers of some kind. That would help to put your comments in to perspective with all the other preceding ones.
"It has been remarked that if one selects his own components, builds his own enclosure, and is convinced he has made a wise choice of design, then his own loudspeaker sounds better to him than does anyone else's loudspeaker. In this case, the frequency response of the loudspeaker seems to play only a minor part in forming a person's opinion."
L.L. Beranek, Acoustics (McGraw-Hill, New York, 1954), p.208.
"It has been remarked that if one selects his own components, builds his own enclosure, and is convinced he has made a wise choice of design, then his own loudspeaker sounds better to him than does anyone else's loudspeaker. In this case, the frequency response of the loudspeaker seems to play only a minor part in forming a person's opinion."
L.L. Beranek, Acoustics (McGraw-Hill, New York, 1954), p.208.
Re: Think that your ears can detect distortion? Try this!
Originally Posted by nick29498141
Is that what you needed?
HTH
NK
Yes, thanks, perfect! Showing this graph is better IMHO because it shows the type of test taken, your own score (in red), and puts that in perspective with all the other scores to date.
Looks like your sensitivity to distortion is way above average!
Re: Think that your ears can detect distortion? Try this!
Originally Posted by charlielaub
Yes, thanks, perfect! Showing this graph is better IMHO because it shows the type of test taken, your own score (in red), and puts that in perspective with all the other scores to date.
Looks like your sensitivity to distortion is way above average!
-Charlie
I looked at the graph and I saw a red flag. A normal distribution curve wouldn't have so many people pegging out the scale. I wonder what the reason for all of -45db scores are and I doubt that it is a thing of golden ears.
Re: Think that your ears can detect distortion? Try this!
Originally Posted by brianpowers27
I looked at the graph and I saw a red flag. A normal distribution curve wouldn't have so many people pegging out the scale. I wonder what the reason for all of -45db scores are and I doubt that it is a thing of golden ears.
Here is something that could explain this: the test stops at -45dB. The "-45dB" category is, therefore, the sum of all categories to the right, e.g. it also contains everyone who could have heard less than -45dB.
Also, there is nothing to say that the distribution does not have more than one peak. There could be many "golden ear" types that make up another peak off "to the right" e.g. who can hear distortion at less than the -45dB level. Weak evidence for that is the fact that several people have commented that they can perceive IM distortion at the -60dB level.
Re: Think that your ears can detect distortion? Try this!
Originally Posted by charlielaub
Here is something that could explain this: the test stops at -45dB. The "-45dB" category is, therefore, the sum of all categories to the right, e.g. it also contains everyone who could have heard less than -45dB.
Also, there is nothing to say that the distribution does not have more than one peak. There could be many "golden ear" types that make up another peak off "to the right" e.g. who can hear distortion at less than the -45dB level. Weak evidence for that is the fact that several people have commented that they can perceive IM distortion at the -60dB level.
-Charlie
I'm 19 and do have very sensitive ears. So I did well? I'm happy with that.
"It has been remarked that if one selects his own components, builds his own enclosure, and is convinced he has made a wise choice of design, then his own loudspeaker sounds better to him than does anyone else's loudspeaker. In this case, the frequency response of the loudspeaker seems to play only a minor part in forming a person's opinion."
L.L. Beranek, Acoustics (McGraw-Hill, New York, 1954), p.208.
Re: Think that your ears can detect distortion? Try this!
It is a fun test. The first time I took the test I did not score well (-20's dB) with cheap headphone. With some better speakers it improved (-36 dB). But for me there is a fatal flaw; at some point I discovered that there was a glitch about 2/3'rds of the way through the undistorted file. Now I can get a perfect score, although I am very challenged to hear the difference at -45 dB.
Re: Think that your ears can detect distortion? Try this!
Originally Posted by brianpowers27
I looked at the graph and I saw a red flag. A normal distribution curve wouldn't have so many people pegging out the scale. I wonder what the reason for all of -45db scores are and I doubt that it is a thing of golden ears.
I agree, maybe. The first test I took was the 6" driver terz and I scored very high but as I posted earlier, beyond -21dB to -24dB, I really couldn't hear any difference other than the volume. I followed that clue and hence scored high. I say I agree, maybe, because others might be doing the same or maybe they really can hear a difference other than volume.
The second test I took after going back and reading the original post where I discovered the test to take was the telecom 7-1 playing terz. I did poorly and I didn't like the sound. It does sound like a telephone. Hmmm, maybe that is why my hearing is worse in my left ear than in my right ear..too much time spent on the phone during my working career.
Re: Think that your ears can detect distortion? Try this!
Originally Posted by whatatrip
![$vboptions[bbtitle]](images/techTalkHeader-leftGray2.jpg)
![$vboptions[bbtitle]](images/techTalkHeaderRepeatGray3.jpg)
Reply With Quote
and used the "Creative" satellite speakers hooked up to the PC; and reached -24 db reliably on the telecom 7-1 terz test. I would have posted a snapshot of the screen like others; except I couldn't figure out how!! 
