[skamp]

I did try a double blind test comparing FLAC and (properly encoded!) MP3's, and found that I was unable to hear a difference. For comparison, I was quite clearly able to distinguish between FLAC and 300 kbps OGG files.

That's surprising. Ogg Vorbis at that bitrate should be fully transparent, especially if MP3 is transparent to you. What ABX software did you use? What was your score? What was the song?

 

[telecaster]

Difference is OBVIOUS in big orchestral passages. Try it m8!

 

[Mad Dude]

I was using Foobar with the ABX plugin. The songs used were from the Iron Maiden album "Powerslave" (non-remastered) - you may laugh, but I found that it was easier to hear a difference with well recorded rock/metal tracks, as opposed to "audiophile" acoustic or classical recordings.
I am not entirely sure what headphone I was using at the time, but I think it was the Hifiman HE-6.
 
Note that I did require about half an hour of "training" before I was clearly able to differ between OGG and FLAC. In the end, I got scores of 7/10, 9/10 and 8/10 in a row, which seemed to me adequate proof that there was indeed an audible difference*.
The same test, with the same time of preparation, yielded no conclusive results when using V0 LAME encoded MP3 files.
 
*I must say that the "audible difference" was indeed so small that under normal circumstances, I would certainly not have noticed them. But then again, someone with a trained ear may hear more of a difference.

 

[Mad Dude]

Quote:

Difference is OBVIOUS in big orchestral passages. Try it m8!

I'm having some trouble believing that the difference is "OBVIOUS". I must ask again, did you ever try a double blind test? Psychoacoustics can have a huge influence on the hearing.

 

[skamp]

The same test, with the same time of preparation, yielded no conclusive results when using V0 LAME encoded MP3 files.

That suggests a flaw in Ogg Vorbis. It would be nice of you to post about your experiment in the Ogg Vorbis section of hydrogenaudio.org, so that the developers can adjust their psy model. That's the place to report problems with lossy codecs, really.

 

[telecaster]

Quote:

I'm having some trouble believing that the difference is "OBVIOUS". I must ask again, did you ever try a double blind test? Psychoacoustics can have a huge influence on the hearing.

I have 100% blind test accuracy in a test MP3 at 320kbps vs FLAC(or CD) during fortissimo. During a Tutti (all instruments at once playing at the same time) they are so much instruments going on, their level is at their highest, and this is very demanding on the reproduction system. The difference is MP3 have harsh and undefined "mushy" sound during those passages while on CD its clear and liquid, impressive is the word.
 
I must say that this test was made on Meridian active digital speakers DSP3100, 5000$ a pair, at loud realistic level. On headphones difference is there maybe less obvious.
If you have an audiophile shop near you just defy my logic and try it for yourself with the song "Bring the boys back home" on The Wall Pink Floyd album!

 

[Gwarmi]

 
 Better the recording - the more blatant it becomes.
 
 Case in point this afternoon - I was converting a great recording by jazz bassist, Brian Bromberg 'It is what it is' (2009) 
 from FLAC to MP3 for some portable action on the Apple iShuffle.
 
 It's a very forward, close mik'd recording - very in your face - the 320kbps conversion just has a very nasty glare to it.
 
 Now if we're talking about the best of Cyndi Lauper or an early Green Day recording then sure - the difference is
 less obvious.

 

[Mad Dude]

Quote:

I have 100% blind test accuracy in a test MP3 at 320kbps vs FLAC(or CD) during fortissimo. During a Tutti (all instruments at once playing at the same time) they are so much instruments going on, their level is at their highest, and this is very demanding on the reproduction system. The difference is MP3 have harsh and undefined "mushy" sound during those passages while on CD its clear and liquid, impressive is the word.

Very interesting, I think only few people can tell the difference in a (volume-matched!) blind test, while a lot of people (especially "audiophiles") claim to hear "huge differences", but are unable to prove it in a blind test. I guess your hearing must be very good, and/or you know exactly what to listen for.
 
My own music library consists entirely of FLAC files, as there are less restrictions of disk space in a home setup than in a portable one. For portable use, I will stick with V0 MP3 files.
 
 
It must also be said that the MP3 format has evolved quite a bit since it has been introduced (some 15? years ago). I remember, with the first MP3 players, you could easily tell the difference between 128 kpbs and 256 kpps files even with the stock earbuds. Now, with LAME encoded files, the differences are strongly reduced.

 

[telecaster]

Agreed when you don't know what to look for most of us will never hear the difference. But in my case with my home speaker set which cost Cd player 4000$, preamp 5000$, 3 speakers 9000$, I hear the difference on loud specific passages ie. Orchestral Fortissimo! Otherwise even 128kbps of Eminem will rock your ass no problem!
 
Yeah Pink Floyd The Wall cost millions to record and was one of the most expensive album to record of its time! Nice reference for people who don't like classical!

 

[BASGTA]

So just like 10 mins ago, i had this revelation. I was playing a .flac file of Smooth Criminal by MJ using VLC media player through my HM5s(because i have nothing else to play .flac files)...and as the song came on, I heard breathing in the background. And I was like...the heck? I've never heard that before. So i went into itunes, and opened up a mp3 version of Smooth Criminal that I've had for ages and it played...and I heard no breathing in that .mp3 file. So i went back to the .flac file just replaying it over and over to make sure i wasn't hearing things...AND NOW

I think you're blowing things way out of proportion. I don't have a FLAC version of Smooth Criminal, all I have is a 128kbps. Just to see if I could hear the breathing in the intro and it's very, very clearly there.

 

[GL1TCH3D]

I notice the difference. I ripped my peter gabriel's new blood cd into 320kbps mp3 and noticed it didn't sound superb. I put the CD in my cd player and output to my receiver where I plugged in my headphones. Wow. Big difference.

 

[skamp]

So, is EVERYTHING an April's Fool today?

 

[syNRG]

Quote:

I think you're blowing things way out of proportion. I don't have a FLAC version of Smooth Criminal, all I have is a 128kbps. Just to see if I could hear the breathing in the intro and it's very, very clearly there.

Really? On all my other versions of smooth criminal i can't hear any breathing at all. In fact at that part...all i hear is silence. I've checked different copies I have of the song on different computers, and none of them have it

 

[telecaster]

Try my test, and DON'T believe me. But for god sake try it in an audiophile store!!!!
During fortissimo in symphonic orchestra, take the pink floyd track I mentioned, or Mozart Requiem, Marcia Funebre from Beethoven 3rd symphony or whatever, just try the fortissomo passages for god sakes. I'm fed up by the "there is no difference"... Don't BELIEVE ANYONE, TRUST YOURSELF!

 

[ImperialX]

 
Quote:

I think you're blowing things way out of proportion. I don't have a FLAC version of Smooth Criminal, all I have is a 128kbps. Just to see if I could hear the breathing in the intro and it's very, very clearly there.

 
I think he's talking about the times when MJ breathes in the track that's not audible on MP3s but FLACs only.