Head-Fi.org › Forums › Equipment Forums › Sound Science › 320k MP3 vs. 448k AAC
New Posts  All Forums:Forum Nav:

320k MP3 vs. 448k AAC

post #1 of 32
1/17/10 at 11:10pm
Thread Starter 
Recently some debates on other forums are focused around this.
They say, with no solid proof, that AAC is highly superior than MP3 that, quote (translated):
Quote:
...If FLAC is 10 points, then 448k AAC is 9.5 points while 320k MP3 is only 6.5 points.
Really? Experienced users please help.
Reply
post #2 of 32
1/17/10 at 11:25pm
Quote:
Originally Posted by 3602 View Post
Recently some debates on other forums are focused around this.
They say, with no solid proof, that AAC is highly superior than MP3 that, quote (translated):

Really? Experienced users please help.
There may be a measurable difference, but IMHO, not an audible one.

USG
Reply
post #3 of 32
1/17/10 at 11:38pm
AAC might be superior, it certainly makes sense looking at the increased bit rate. Who cares? If you're that sensitive to the existence of compression artifacts then it doesn't make any sense to use 448 when lossless can be had for a mere few hundred more kbps.
Reply
post #4 of 32
1/18/10 at 1:21am
AAC does not measure much better or test much better in blind testing. It may, along with ogg, measure marginally better than MP3, but a properly encoded 3.97 LAME MP3 should come very close. The only codec I know of that substantially outperforms MP3 is Musepack, but that has limited support outside of the PC environment. This entire discussion is basically irrelevant because MP3 is audibly transparent beginning below 200 kbps in 90%+ of recordings.
Reply
post #5 of 32
1/18/10 at 1:41am
Don't know why it would matter, if you are encoding at 448kbps AAC, just go for FLAC and call it a day.

Otherwise at those high of bitrates it kinda gets silly to discern one from another.
Reply
post #6 of 32
1/18/10 at 7:02pm
Thread Starter 
Does anyone know how to gain ALAC? MP3Gain, of course, doesn't do it.
Reply
post #7 of 32
1/18/10 at 7:28pm
If you're doing 448 kbps you might as well be doing lossless cause you're approaching or even equaling or even exceeding the file size of FLAC files. I have FLAC files that are under 448 kbps (classical and jazz). Rock music that is loudnessed is around 1000 kbps.

The primary purpose of lossy compression is to save space. If you aren't saving space then might as well do lossless. 448 kbps is crazy high for lossy. What do you do when you reach the point that lossy files are bigger than lossless? Do they suddenly sound better than lossless?
Reply
post #8 of 32
1/18/10 at 7:51pm
Well, there are some DAPs that do not play any kind of lossless but do play AAC.
Reply
post #9 of 32
1/18/10 at 10:03pm
tbh, AAC @ 448Kbps is extreme overkill and is tbh would be 9.97 of a FLAC file.

AAC @ ~192 kbps is better than VBR -V0 / 320kbps CBR mp3.

Solid proof?

(I converted the AAC file to WAV to work with Spectro but of course it's still a AAC file [just in a WAV container]).



Reply
post #10 of 32
1/18/10 at 10:05pm
Quote:
Originally Posted by 3602 View Post
Does anyone know how to gain ALAC? MP3Gain, of course, doesn't do it.
ReplayGain. I can easily do it via F2K.
Also AAC is technically much superior to mp3.
Also the theory of DBT is sound, how it's conducted by most people = extremely flawed as there are many factors to account for.
Ogg Vorbis offers better compression but less accuracy than AAC. mp3 is a distant third on both counts.
mp3 is an outdated, inefficient, inaccurate codec. I cannot test .wma properly as there are no third party encoders in which I can tweak the bitrate accurately with (WMP11 fails in this aspect). I used Robot Rock by Daft Punk as it's a highly complex song.

I use Vorbis @ ~209.6 kbps to counter the inaccuracy + unfortunately doesn't take AAC.
Reply
post #11 of 32
1/19/10 at 2:44pm
Thread Starter 
Oh wow. Could you do a 448kbps AAC vs. 320k MP3 CBR comparison?
And is it worthwhile to convert my APE/FLAC to AAC? Already LAME V0 fast.
Reply
post #12 of 32
1/19/10 at 3:13pm
Could be preferable to lossless in an mp3 player, probably audibly imperceptible and good for battery life and space.
Reply
post #13 of 32
1/19/10 at 3:14pm
Assuming you mean conversion for portable use, ~200kbps VBR is a sweet spot for the file size/audio quality ratio. Although it doesn't make sense to convert at all if you don't have to deal with strict storage limitations. Stick with flac
Reply
post #14 of 32
1/19/10 at 6:54pm
Thread Starter 
There is a player called iPod with no FLAC support.
Well, I'll convert to ALAC and gain it down. All tracks are now at 89dB.
Reply
post #15 of 32
1/19/10 at 11:59pm
Quote:
Originally Posted by chinesekiwi View Post
AAC @ ~192 kbps is better than VBR -V0 / 320kbps CBR mp3.
What? How can you have a constant bit rate, using a variable encoder? The only way you would get that is with a sine wave.
Reply
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Sound Science
Head-Fi.org › Forums › Equipment Forums › Sound Science › 320k MP3 vs. 448k AAC