Skip to main content
Topic: new public test: How Well Can You Hear Audio Quality? (Read 8188 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

new public test: How Well Can You Hear Audio Quality?

Reply #1
In my test results, less than half right, I was advised: "...you're unlikely to do better without an amplifier or a digital audio converter."  So obviously my low score has nothing to do with the chosen cuts themselves, the limitations of human hearing, or the quality of my headphones. The problem lies in the poor low level linearity of my notebook's D/A while listening to this (mostly)  heavily compressed rock music... Whew.

new public test: How Well Can You Hear Audio Quality?

Reply #2
Despite my best efforts, I failed miserably and even picked the 128K MP3 twice. 

Initially I noticed that certain tracks seemed to be significantly slower to load. Figuring that might allow me to "game" the system,  I switched to a wired network connection, which eliminated these differences, reloaded the web page and started from scratch.

new public test: How Well Can You Hear Audio Quality?

Reply #3
3 chosen as 320kbps, 3 chosen as lossless...

Quote
You got 3 out of 6 correct!

Despite its bad rap, the MP3 is actually a remarkable feat of engineering, and pretty good at filtering out mostly the sounds you can’t hear. Depending on the quality of your headphones, you might be able to distinguish between the two MP3s, but you’re unlikely to do better without an amplifier or a digital audio converter.

I'll have to do a ton of trials without throwing in random guesses after getting past the low anchors in order to keep intellectually dishonest, red herring-eating douche bags from trying to pigeon hole my results in order to fit their agendas.
Is 24-bit/192kHz good enough for your lo-fi vinyl, or do you need 32/384?

new public test: How Well Can You Hear Audio Quality?

Reply #4
What I never understand with people who claim to find mp3s really objectionable is that the problems on many commercially released CDs are far greater than those introduced by modern 128kbps mp3 encoding. (IMO. YMMV.)

Meaning that, if 128kbps is really objectionable, almost no commercially released CDs are even listenable.

(Though, again IMO, I would agree that most CDs sold in the last decade can't be listened to critically.)


How many released CDs, when encoded to 320kbps mp3, have more problems added to their sound by that encoding than they had already?


It's my opinion that, if you want to enjoy music, you just can't listen with the level of criticism that's needed to find mp3 annoying. If you're that easily annoyed by minor faults, then I think recorded (and probably live) music isn't for you.


I know some people can ABX these things. I'm not saying the differences are always imagined. I'm saying that I think some people like to exaggerate them to show off. If some of these "professional audio engineers" were really so offended by imperfect sound, they wouldn't release such horrible sounding CDs.

Cheers,
David.

new public test: How Well Can You Hear Audio Quality?

Reply #5
I would be surprised to find out that these were VBR and not CBR. Some of the tracks have artifacts at 128 that are simply annoying; more so than whatever was done to the tracks in the studio. Remember, I do not despise professionally applied multiband DRC as much as the rest of the community seems to, except when the master is driven into obvious (larger than just noticeable) audible clipping.
Is 24-bit/192kHz good enough for your lo-fi vinyl, or do you need 32/384?

new public test: How Well Can You Hear Audio Quality?

Reply #6
What I never understand with people who claim to find mp3s really objectionable is that the problems on many commercially released CDs are far greater than those introduced by modern 128kbps mp3 encoding...

I do second that. It happened that I bought CDs which had annoyiong distortions, i.e. a quality I would hardly accept if the mp3 encoder were to blame.
lame3995o -Q1.7
opus --bitrate 140

new public test: How Well Can You Hear Audio Quality?

Reply #7
I am afraid that the only correct answer to any listening test for me, now, would be "You might do better with a hearing aid."


(meant light-heartedly, but, none-the-less, true)
The most important audio cables are the ones in the brain

new public test: How Well Can You Hear Audio Quality?

Reply #8
I liked this comment:
Quote
I got 3/6 right, but I expected to do even worse ...

new public test: How Well Can You Hear Audio Quality?

Reply #9
There are three choices, so half right would be better than chance, provided there were enough trials.
Is 24-bit/192kHz good enough for your lo-fi vinyl, or do you need 32/384?

new public test: How Well Can You Hear Audio Quality?

Reply #10
What I never understand with people who claim to find mp3s really objectionable is that the problems on many commercially released CDs are far greater than those introduced by modern 128kbps mp3 encoding. (IMO. YMMV.)



I made a point along those lines in the comments yesterday (15 hrs ago). 

No replies to it.  I guess everyone there agrees.  ;>



new public test: How Well Can You Hear Audio Quality?

Reply #11
Let me back up a bit.  Two of the samples do in fact sound like shit to me because of poor EQ and loudness maximization.  One just has over the top compression. I didn't much care because I don't like any of those three, and still wouldn't even if they weren't butchered (well maybe I might better like one of them). Of the six only one sample can be found in my library, and I rarely listen to it.

The track that sounds the worst at 128 to me was by Cold Play.  Ironically, I suspect the artifacts wouldn't have been as bad had it not received the heavy compression treatment. I would still have zero love for much of anything by that band, however.
Is 24-bit/192kHz good enough for your lo-fi vinyl, or do you need 32/384?

new public test: How Well Can You Hear Audio Quality?

Reply #12
There are three choices, so half right would be better than chance, provided there were enough trials.



I'm not sure how the stats work.  When there are three choices, and if it's possible to confidently rule out one, and then  the probabilities change for the remaining two, don't they?

I wonder if  this is related to the 'Monty Hall' problem.  Which I always have to review, to understand. 

new public test: How Well Can You Hear Audio Quality?

Reply #13
Well yes, though I don't know that being able to identify the 128 is a given for everyone.
Is 24-bit/192kHz good enough for your lo-fi vinyl, or do you need 32/384?

new public test: How Well Can You Hear Audio Quality?

Reply #14
There are three choices, so half right would be better than chance, provided there were enough trials.
My bad. My Android phone only displayed 2 options since the browser doesn't support lossless. Sigh

new public test: How Well Can You Hear Audio Quality?

Reply #15
^Oops. I made the exact same mistake as Kees. My dumb browser only showed me two choices, not three. Disregard my earlier post.

new public test: How Well Can You Hear Audio Quality?

Reply #16
Well yes, though I don't know that being able to identify the 128 is a given for everyone.



I'm sure it's not.  In my fairly cursory run through, I couldn't confidently say  128kbps *ever* sounded dstinguishable to me.  But look at all the bragging going on in the comments, '6/6, it was easy, I'm a musician!' You believe that?


(meanwhile it seems the lossless tracks load slower on several platforms....gee, that a couldn't be a cue, could it?)

new public test: How Well Can You Hear Audio Quality?

Reply #17
Wouldn't play at all for me in Firefox  Works in Chrome though.

Is half actually that good?  Assuming you can reliably spot the 128 sample, that means you're guessing which is WAV and which is 320 MP3.  I got 3/6, although I wouldn't claim the 128 samples were obvious.  So really I was probably guessing.  Cold Play sounded so horrible I wouldn't play it all the way through, let alone listen several times so that was definitely a guess

new public test: How Well Can You Hear Audio Quality?

Reply #18
Quote
'6/6, it was easy, I'm a musician!' You believe that?

If thousands of people take the test, quite a number will get 6/6 even if they're guessing ... Or maybe they have better equipment and better ears than me, wouldn't be hard.

new public test: How Well Can You Hear Audio Quality?

Reply #19
Quote
'6/6, it was easy, I'm a musician!' You believe that?

If thousands of people take the test, quite a number will get 6/6 even if they're guessing ... Or maybe they have better equipment and better ears than me, wouldn't be hard.


Seems a fair few folk are getting 6/6 -- simply by noticing the additional delay before the WAV starts playing over mediocre internet speeds.

new public test: How Well Can You Hear Audio Quality?

Reply #20
3 of 6 for me.  I was able to identify the 128's pretty quickly, but was guessing between the 320's and lossless.  As other posters mentioned, I too had little patience for "torturous" listening on 3 of the tracks.  Chrome with good internet speed gave me no track loading delay hints.

new public test: How Well Can You Hear Audio Quality?

Reply #21
It worked fine for me on the Opera browser, which shares some elements with Chrome, so that's not so surprising.

I picked WAV correctly twice out of four and the 320k mp3 twice. But I couldn't bring myself to listen to Katie Perry or Coldplay...

new public test: How Well Can You Hear Audio Quality?

Reply #22
2 out of 6 (okay, 5 ou of 6 if you consider 320kps MP3 harder do recognize from lossless, but then I was just guessing, mostly).

Even though it was a pair of respectable headphones, (Senns HD-580) they were connected to the laptop's 3.5mm jack, as DAC seems to have given up the ghost lately.
No noticeable delay on loading the songs.

I said it before and will say it again: I'm such a happy bunny with my pair of non-discerning ears. 
Listen to the music, not the media.
Qualidade em MP3

new public test: How Well Can You Hear Audio Quality?

Reply #23
Btw, 5/6 is statistically significant (p=1.8%). Mean is 2.

It's too bad that there's the problem with loading times..
"I hear it when I see it."

new public test: How Well Can You Hear Audio Quality?

Reply #24
0 out of 6

Science wins.
Ad hominem attacks are not Science.

 
SimplePortal 1.0.0 RC1 © 2008-2019