http://mp3ornot.com/

Couldn't find the site with the search function, so probably hasn't been mentioned here yet.

I'm not sure what the goal of the site is. Showing that 128kbps MP3 is bad and we should all switch to 320kbps, or that using 320kbps is a waste of space...
(I got the answer right, though, but I guess 50% of people would...)

Couldn't tell the difference

You did hear no difference?

Just did the test by myself. I can tell the difference by listening to the speakers of my notebook computer.

Voted 320 kbit/s.

A really obvious difference. But I see what the point beyond that is: sadly enough, for the vast majority of music consumers out here anything beyond 128kbps is a waste of space.

I wonder how big and cheap the hard drives have to become to stop people complaining that the files are too big. In my opinion, only unused space is wasted space

Not if you care to fill your flash-memory DAP with tunes you're not going to listen anyway

(Well, since solid state memory is becoming cheaper I should simply restate that as overkill...)

I can distinguish between the two fairly easily but fail to see the point in the test. Why don't they offer a VBR file for comparison too?

I've seen similar comparisons made between low and high bitrate MPEG-4 encodings where no mention is made of the fairly obvious choice of using VBR (1 Pass Quality mode in the case of XviD). All it does is to give the impression to those not so knowledgeable that there is no alternative but to use high bitrates with the resultant files being large to obtain high quality results.

Is this down to stupidity on the part of the people setting up the tests?

Cheers, Slipstreem.

mp3ornot.com.

Is it worth pointing out the flaw in that domain name?

the bad clip sounds like someone would add way to much artificial noise, never heard that before in such obvious manner.

for anyone saying it was obvious, can you repeat multiple times under blind conditions? I knew my hearing was bad but I'd have thought I could discern the blindingly obvious

I have to say: I did select the 320kbps file, but I did not find it easy, and it may have just been luck.

I know my hearing isn't brilliant though. It doesn't concern me.

The site wouldn't be as stupid if the options were randomly swapped each time.

Some people will hear the problem and get it right, some will just guess the right answer. It should have been created with a randomiser and a minimum number of trials.

That being said, it wasn't too hard to differentiate. What encoder was used? I'm willing to bet LAME 3.97 VBR @ ~130kbps would be much better.

edit: There are some right numpty's on this planet. If you want a good laugh: Digg

What are the problems with these samples? I cannot hear a difference. A lossless sample might have helped to differentiate.

direct links;
http://mp3ornot.demandstandards.com/A.mp3
http://mp3ornot.demandstandards.com/B.mp3

The difference is quite obvious.
The encoder used is FhG. I wonder if the difference would be so obvious with LAME...

I was surprised that I could, because my hearing has been abused (10+ years in metal/noise bands), but the difference really seemed to show in the instrumental section between the first "t" and the second "t". It was noticeably thinner.

- edited for a bunch of errors in the posting process

I guessed correctly, but couldn't really hear a difference.

Do I have to leave HA now?


I loved the Steve Martin quote on digg.

Cheers,
David.

Could hear difference. Passed test

I voted correctly, although I'll admit I wasn't over confident. A poorly done ABX test anyhow - samples not randomised making it a single-trial test, hence useless as a proper ABX test.

Arite.

...no specification of encoders or settings.

I have a feeling the test would have been harder had Lame -V4 and -b 320 been used instead.