Skip to main content

Notice

Please note that most of the software linked on this forum is likely to be safe to use. If you are unsure, feel free to ask in the relevant topics, or send a private message to an administrator or moderator. To help curb the problems of false positives, or in the event that you do find actual malware, you can contribute through the article linked here.
Topic: AAC vs MPC and others (Read 5218 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

AAC vs MPC and others

The general concensus seems to be that MPC has no real competitors based on quality for the bitrate in its "standard" and up settings.  Which is its design.  However, for producing high-quality audio files in the 128 range and below, would AAC then be the best?  My experience has been that wma sounds better at 128 and 96 than does ogg, and of course the mpc "thumb" setting isnt' too hot.  I've figured that the lack of discussion of wma in this form is due to the format's ownership and likely lack of associated freedom of usage.  I've been encoding most of my stuff using ogg, because I don't care about quality enough to go up in bitrate to the MPC standard setting.  Would AAC be a better way to go?  Or wait for future ogg releases?
God kills a kitten every time you encode with CBR 320

AAC vs MPC and others

Reply #1
At 128kbps, I believe that OGG, AAC, and MPC are all roughly equal, or very close in quality.

You should check out these links for more information:

First Test:
http://www.ff123.net/128tests.html

First Test Results and Comments:
http://www.ff123.net/dogies/dogies_comments.html

First Test Analysis:
http://www.ff123.net/dogies/dogies_plots.html

Second Test:
http://ff123.net/128test/instruct.html

Results for Second Test so far (according to ff123, may not be significant yet):
http://ff123.net/128test/interim.html

AAC vs MPC and others

Reply #2
hmmm... these tests implied that mpc and aac are the best at 128 kbps, while ogg, lame and wma are a bit lower.  In my experience, the "radio" setting in mpc generally outputs files around 140 kbps.

I'm generally curious about why ogg gets a lot more attention here than does aac, if aac is higher quality.  Does the ogg framework provide greater potential for quality than does aac?  Or is the open-source factor a big reason for ogg's popularity?  And is wma paid little attention because it is optimized more for lower bitrates and streaming media than for audiophile quality, or more because of Microsoft and the liscensing issues with the codec's usage?
God kills a kitten every time you encode with CBR 320

AAC vs MPC and others

Reply #3
Note that comparing VBR codecs (MPC, Ogg) with codec set to CBR (AAC streaming 128) is, at least, unfair - Since CBR is very restrictive mode of coding and CBR codec will do anything to keep not just that bitrate, but even bitrate fluctation within certain limits to allow internet streaming (bit reservoir of 8184 bits for 128 kbps).

VBR codecs have freedom to have unlimited bitreservoir and to allow free fluctations in bitrate.

Better idea would be to compare MPC -radio or Ogg -b128 with some AAC VBR preset (Liquifier Transparent 128, or PsyTEL -streaming / -internet mode) - otherwise we would have unfair test.

In my opinion, AAC has much greater framework and quality abillity than Ogg, but it is protected by lot of patents. AAC has every bit of the "state-of-the-art" technology incorporated into it, since it was developed by biggest names in the audio coding field. However, Ogg is much more "developer friendly" and it is completely free and open. It all depends on what user wants - if he wants ISO compatible and highest quality standard, he will go for AAC. If patent-free technology is wanted, Ogg is the right choice.

AAC vs MPC and others

Reply #4
Quote
hmmm... these tests implied that mpc and aac are the best at 128 kbps, while ogg, lame and wma are a bit lower. In my experience, the "radio" setting in mpc generally outputs files around 140 kbps.


A few qualifiers:  not just any AAC -- Liquid Audio Liquifier 5.0.  Psytel still needs some work on samples like applaud.wav at 128 kbit/s.  Also, if I had included WMA7, it may have performed better than WMA8 did at 128 kbit/s, at least that's what people like Hans Heijden and Filburt say.  And the MPC setting wasn't just -radio, it was tweaked to get the bitrate down a bit.

Quote
I'm generally curious about why ogg gets a lot more attention here than does aac, if aac is higher quality. Does the ogg framework provide greater potential for quality than does aac? Or is the open-source factor a big reason for ogg's popularity?


I think the latter is definitely a reason.  And ogg's potential is high, considering the tremendous progress it's made so far in such little time, and the ideas that remain to be implemented.  If you're looking for formats which don't get attention, try mp3pro, vqf, and realaudio, in addition to wma8.

Quote
And is wma paid little attention because it is optimized more for lower bitrates and streaming media than for audiophile quality, or more because of Microsoft and the liscensing issues with the codec's usage?


Yes and yes.

AAC vs MPC and others

Reply #5
Quote
Originally posted by timcupery
I'm generally curious about why ogg gets a lot more attention here than does aac, if aac is higher quality.  Does the ogg framework provide greater potential for quality than does aac?  Or is the open-source factor a big reason for ogg's popularity?


A few reasons.  First of all, Vorbis isn't even finished yet (as in 1.0).  With each release quality has been significantly improving.  Those tests were done with older versions of Vorbis, (RC2) and in my experience RC3 performs quite a bit better.

So in a sense, Vorbis is already "good" and still has a lot of potential left.

The Open Source factor does play a role here also though.  Vorbis is free and is very accessible so it naturally garners more attention and use.  AAC on the other hand is a very restricted technology.  If it wasn't for PsyTEL, high quality ISO AAC wouldn't be available to the end user.  For that matter, even with PsyTEL, AAC is still not very accessible.  Most people outside of some of these communities don't know where to get this encoder.  I believe distribution of this encoder falls into a bit of a "grey area".  So basically, with AAC it's totally the opposite of the situation with Vorbis.  The quality of AAC is pretty high, and perhaps with the Fhg encoder (not available to the masses) it may be the best low bitrate format so far.  However, Vorbis is rapidly approaching the quality of PsyTEL and is much more accessible to end users.

That's not to say I don't see merit in AAC or PsyTEL, but for many people who are really interested in personal use of a high quality codec, AAC can be a little difficult to work with.

Quote
And is wma paid little attention because it is optimized more for lower bitrates and streaming media than for audiophile quality, or more because of Microsoft and the licensing issues with the codec's usage?


Both.  WMA doesn't offer very high quality and it is completely proprietary.  To my knowledge no publically available sources exist for either an encoder or decoder.  It's also not cross-platform.  In addition to all of this, Microsoft is pushing it as a "secure" format, and most people around here are actually against such a thing.  There's really not much of a reason to use it at all IMO.

AAC vs MPC and others

Reply #6
Quote
Better idea would be to compare MPC -radio or Ogg -b128 with some AAC VBR preset (Liquifier Transparent 128, or PsyTEL -streaming / -internet mode) - otherwise we would have unfair test


I did try to find a VBR setting in Liquifier which might work.  Perhaps I should have chosen VBR but set the lowpass at 16 kHz.  That would have kept the bitrate down near 128, I think.  Which is the harsher penalty:  restricting the mode to CBR or restricting the lowpass to 16 kHz?

ff123

AAC vs MPC and others

Reply #7
Restricting cut-off to 16 kHz will do less harm than restricting bitrate to 128 Kbits/s.