Difference between MP3 and FLAC?
Jun 19, 2009 at 9:57 PM Thread Starter Post #1 of 93

Eternal187

New Head-Fier
Joined
May 30, 2009
Posts
13
Likes
0
Is there really a big difference between an MP3 320kbps (VBR) and a FLAC file? I mean can you really tell a difference or are you getting almost the exact same quality?
 
Jun 19, 2009 at 10:04 PM Post #2 of 93
Yes I can, not it's not the same as it throws away data. I deleted all mp3's even 320kps because they just sound bad. Treble sounds like a tin can/transistor radio, and bass is boomy like a midi system.
 
Jun 19, 2009 at 10:11 PM Post #3 of 93
well the plague in MP3 is Joint Stereo...so many 320kbits files are actually JS.

but yeah, MP3 is plain bad...AAC is far better, especially w/ the latest Nero encoder...but at least lossless is pure bliss, and APE at the highest encoding strength is pretty compact anyway.
 
Jun 19, 2009 at 10:35 PM Post #4 of 93
The difference in audio data is quite substantial. As one throw away quite a lot of audio data (MP3), while the other one are lossless (FLAC). There may not be an audible difference though, depending on several factors. Encoder, music, playback gear, and not least the listener...

There will always be someone who have can hear an audible difference between MP3 and FLAC, if you count the factors above..
 
Jun 19, 2009 at 10:58 PM Post #5 of 93
Quote:

Originally Posted by Eternal187 /img/forum/go_quote.gif
Is there really a big difference between an MP3 320kbps (VBR) and a FLAC file? I mean can you really tell a difference or are you getting almost the exact same quality?


The answer is...*drumroll* YES. Almost is the key.

That said, at 320, you are only saving 50% over typical FLAC, so it's not like a hardship, especially given how hard drive space keeps going up. "This partition that I can't fill up will now use double the space for music as it did before! Oh, no!"
 
Jun 19, 2009 at 11:18 PM Post #6 of 93
Reported hearing threshold
MP3, 1000 participants, and stereo systems ranging from 0-$5,000+.
Also, the test wasn't a blind test, so it doesn't rule out the people who imagined a difference, and doesn't rule out people who didn't want to admit to themselves that they couldn't hear a difference.
It also doesn't rule out the 25% who ended up not submitting their results.

I'm sure that there are a lot of people who can actually hear up to 320k.. and perhaps even 320 vbr against lossless, but I'm on the more skeptical side.

I use FLAC for peace of mind mostly on my computer-based setup, and for archival purposes.
MP3 isn't the best compression method, but I use LAME v0 anyways for portability since it's widely supported.
 
Jun 19, 2009 at 11:53 PM Post #7 of 93
I have yet to be able to be able to reliably identify FLAC from MP3 (-V0 or 320 CBR using LAME) in an ABX style test in Foobar. That is with headphones and my modest setup and my less than average hearing. My speaker setup is just M-Audio monitors in a nearfield setup. I don't know if the speakers would be more revealing of MP3 vs. FLAC or not.

Hearing the compression artifacts (pre-echo and such) in high bitrate MP3 files is something that you would have to train for. To learn what they sound like and how to identify them. That is something that I have not done other than proof-of-concept tests at low bitrates (I have heard artifacts at low bitrates).

Hearing the general lower sound quality of 320 CBR MP3 (the general overall sound rather than specific fleeting artifacts) is also something I would need to learn how to listen for. I'm not sure that my listening skills are up to that.

So currently in my case, I don't hear an audible difference between -V0 VBR MP3, 320 CBR MP3, and FLAC. Maybe sometime in the future I will with better equipment or better trained ears.

But all that aside, we are now past the tipping point where maintaining your listening library as lossless rather than lossy is practical. Hard drives are now more than big enough to maintain a big lossless library and a duplicate of the library in MP3 for the portable. Software has also gotten better at helping you manage such a library. I use J. River Media Center and it can automatically transcode my FLAC to MP3 for my iPod. It maintains a cache of the converted files so I end up with my FLAC files duplicated as MP3. Space is no longer an issue. I have a 1TB external drive for media, a second 1TB drive for backup of my media files, and a third smaller drive as a redundant backup that can be kept offsite (I don't want to lose my ripped CD collection).
 
Jun 20, 2009 at 12:14 AM Post #8 of 93
I can't tell the difference. 320 is just fine.

In an A-B test I could never distinguish a difference. For me this is the least important part of the audio chain, but to each their own.
 
Jun 20, 2009 at 1:03 AM Post #9 of 93
my usual reply: the source material usually is more the problem than the compression (at the data level) you might apply.

320k isn't a problem as much as BAD MASTERING is!

only on the 1% top recordings in my collection would I even hope to tell any diff. most music, once gone thru the chains of processing before it reaches you - your last stage is not usually going to hurt much more
wink.gif
 
Jun 20, 2009 at 4:31 AM Post #10 of 93
I have never really attempted any sort of scientific A-B comparison myself. However, I feel that, if I only have a check a box on my encoder to rip into FLAC, and the hard drive is large enough to hold the songs, I figure, why not? Not to mention, if you need to, you can transcode it into essentially any lossy format without the introduction of artifacts that would be there transcoding from one lossy codec to another. That's just my opinion.
 
Jun 21, 2009 at 3:22 PM Post #14 of 93
Quote:

Originally Posted by iriverdude /img/forum/go_quote.gif
Yeah but where's the lossless version? It's like comparing which poo tastes better, when you don't have lobster as well.


I'm sure you have verified this doing ABX in Foobar, right? Or even had a friend in charge of playing files for you alternating between lossless and 320k lossy, blind to you? Just making sure, because some people around here regurgitate things they have read but know nothing of personally.
 
Jun 21, 2009 at 3:26 PM Post #15 of 93
I've done tests between mp3, ogg of various bitrates, and lossless and they've picked out from best to worst (blind test, only I could see what file I was playing, on three people who are all into audio)

lossless > ogg > mp3

And I have also picked out mp3 with the most bloated, artificial bass notes.
 

Users who are viewing this thread

Back
Top