BLIND TEST: 128kbps mp3 vs Lossless - Page 8

Same here, first couple of seconds of track B gave it away.  This was a little fun exercise with my coffee.

 

I wonder if it would have been even more apparent using my home desktop rig rather than my crappy laptop headphone out.

 

Good thread, I can keep my audiophile card another day.  But tomorrow, who knows.
 

 

 

Logs?

Be careful not to confuse salespersons' recommendations to customers with personal opinions of sound quality. The incentive to sell one product over another could be fairly significant.

Got it right.  It took me 3-4 back and forths.  For me, it was the percussion just past the half way point.  MBP > DEQ2496 > Valhalla > DT880

 

Oh, and it should be obvious that people are going to discuss results even if you ask them not to.  So I just took the test first, and then read down.  I'm glad the OP chose this song- I like it.  I found a while back that building my library primarily by recording quality resulted in having a shelf full of 'reference recordings' that I don't often listen to.  Music comes first for me; then the technical stuff.  That said, it would be interesting to see the results from different material.  Luckily there are a few more of these threads floating around here.

 

I would imagine that if such a test were administered to the general public, the results would be closer to 50/50 or pure guessing.  When we listened to these samples, we knew we were taking a test.  I doubt people even around here listen in this way all the time.  I also wonder- if 128k mp3 is really as awful as some say, then they should be able to identify it without a reference file- right?  Don't get me wrong.  I like good stuff and I even like to 'pad' things just in case.  But I don't need to see a picture of the sun to tell whether or not it's nighttime.  ;-)

yay I got it right, out of my Crappy headphone out on my laptop. Maybe it would have been more apparent with a dedicated DAC and amp.

 

I was surprised at the lack of stark contrast. I know 320 and Lossless are very very very similar, but I have heard 128bit tracks that sound absolute crap and its immediately apparent that its 128 or lower. So much depends on the recording process used by the studio too.

It took me one playthrough through my computer. My headphones aren't great right now [I've shelved my V6's because of trebleness so am currently using a pair of AKG K-44's] but there was just an obvious ... I guess crisping(?) of the cymbols in A where there was less of it in B. And B just sounded a little better.

Difference wasn't massive, though.

I listened to this out of my Audiotrak Prodigy Cube, with ATH-M50s and Westone 3. I actually did scroll down and see the result someone had posted, but I only glanced at it, and quickly avoided the placebo effect by repeatedly thinking "The lossless track is A, B, A, B, A, B, A, A, B, B, A, B" etc. until I didn't know what left and right were anymore. :P

The track I took for lossless did sound more natural to me, and I noticed it most in the low or perhaps mid-low range right in the first few seconds of the song, and the track I took for compressed just sounded less "real" to me. 

This was fun, because I was wondering if all this audiophile stuff was for me, or if I was really hearing any difference. x3

If most of the HeadFi'ers on this thread are correct, and voted, then it seems I was correct. :D

Warning: Spoiler! (Click to show)

 

 

It was interesting. At first listen, about ten seconds each track, I had come up with a conclusion right off the bat, which is the conclusion written in the spoiler box. The more I listened, though, the less of a difference I could hear. It was really a visceral response, and I had no way to quantify it. I tried to quantify it, but was unable to.

its gotta be b, b is the original file, the drums sound flatter on a

 

 

 

lol man im going to look like an ******* if im wrong

The differences are very obvious on my Stax, even if I am temporarily on on-board sound. A sounds a tad muddier, while B has a more 'real' tone to it. It's most obvious in the vocals with these headphones.

I have Fiio E7 and AKG242

and...

B is lossless.

I figured it out when I compared guitars and drums on both versions.

Also went with B... (before scrolling down)

 

Just generally a richer more engaging sound.

 

It would be fun to try a few different tracks ranging from high quality to loudness war debris and see how consistently people can identify the FLAC copy.  Personally I think the sample only needs to be 10-20 seconds.  FLAC vs 256K AAC (aka iTunes) would be perhaps more interesting as that is one of the biggest "it's obvious" vs "I can't hear it" quality fights I hear.  I think its good to through in tracks that people may not be familiar with.

 

Much higher though would be tough, I personally have a very tough time telling 320K Ogg vorbis from FLAC except in very good recordings.

 

 

I voted for "they sound the same". Couldn't detect the difference. Played both files 2 times.

 

Gear I used:

-Integrated onboard audio ---> Genelec 6010's (which are very badly placed at the moment, directly on table etc.)

-Brains which been awake for 20hrs+ now; xtremely tired.

 

 

More blind tests please! Next time I'll use some better gear. Also the song sucked, really harsh sounding recording imo.

 

So has the OP approved which is the right answer?

 

 

 

 

 

Answer (Click to show)

 

It has been already mentioned earlier, but note that B.wav is louder than A.wav by about 0.36 dB. If you do not correct the playback for this, it makes the test easier than it should be. I think it would be best if the OP included a note regarding level matching.