Kolya on 17/5/2007 at 11:57
Following the discussion in (
http://www.ttlg.com/forums/showthread.php?t=113469&page=4) this thread I have prepared a little blind listening test. The purpose is to find out if you really can hear the difference between a MP3 file encoded in 128kbps and one in 320kbps.
Of course you can find out the truth easily but if you're interested to find out if your hearing is as good as you think, take the test first before you cheat.
(
http://www.kolya.mynetcologne.de/1.wav) File 1
(
http://www.kolya.mynetcologne.de/2.wav) File 2
So which of these files is the lowly 128kbps encoded one?Both are encoded with MPEG1, constant bitrate and normal stereo btw.
EDIT: I decided to decode both files to WAVs again after encoding them to the MP3 formats. That should make it a bit harder to cheat.
Nameless Voice on 17/5/2007 at 12:10
<s>In the interests of fairness for this test:
If you're using WinAmp to play these, make sure to 'windowshade' the main player window by double-clicking on it so that you can't see the bitrate displayed there. Then, load the files into the playlist by using 'add URL' so that you don't have to download them and see the relative filesizes.</s>
EDIT: This no longer matters since Kolya's evened the playing field by uploading them both as WAV files now.
EDIT 2: I almost voted for the wrong file because I'd been listening for the high-quality one and the question is which is lower quality. It's somehow human nature to assume the question is what you were listening for. Must pay more attention...
D'Arcy on 17/5/2007 at 12:14
I'm opening the files in the browser.
EDIT: Wait, I actually got it right (voted correctly, just what I wrote in the post is wrong).
EDIT2: I deleted the spoiler, I think that's better for the test.
Bjossi on 17/5/2007 at 16:07
It is much easier to hear the difference in a mp3 than a wave file, but I think I hear file 2 as the better one, not hundred percent sure though.
Nameless Voice on 17/5/2007 at 16:28
Okay, so, let me get this straight.
You're saying that you can tell the difference between two MP3s, one at 128k and one at 320k, much more clearly than the difference between those same files converted back to WAV?
Are you implying that the 128K version will magically regain quality when it's converted back to WAV?
Bjossi on 17/5/2007 at 16:36
Well, magically I can tell the difference between 128k and 320k files that I encode. The 128k version always gives me this audible background noise that is not present in either of those files Kolya linked to.
ZylonBane on 17/5/2007 at 16:57
Then you're not doing it right.
EDIT: I voted for File #1 as the 128kbps one. #2 seemed to have slightly greater frequency response. The sample selection isn't a very good stress test for MP3 though, so naturally both files are going to sound very similar.
Al_B on 17/5/2007 at 18:39
I listened to both of those files using headphones, and although they were very close one of them had a better quality to the instruments and the other one sounded a little 'harsh' in comparison. I very much doubt that I would be able to make a real judgement if I were listening on speakers without having a very good quality audio setup.
Of course, Koyla could be playing a trick and they're BOTH 128kbps ;)
Kolya on 17/5/2007 at 19:07
Nah, no tricks involved.
In regard to this being a bad test example: I tried a bunch of samples: rock, electronic music, etc. I couldn't hear the difference in any of those (to my chagrin). I was sure I would hear it as I could swear I had done so in the past. But it must have been even lower encoded files or simply bad sources. :erg:
Bjossi on 17/5/2007 at 19:17
Well, I don't know how you do it, but when I encode (tried more than one codec) I hear an audible difference no matter what song I convert and compare. I tried LAME and I think something called FSH which GoldWave uses.
