Joe Coffee is Angry Coffee's resident Internet audio expert. Angry Coffee fans can catch Joe's wisdom here in the Audiofile section. When he's not here, he's hanging out with his pal Too Much Coffee Man. Look for their upcoming Power Latte Tour.

MPEG 2 AAC - Experts Agree It Beats mp3

Think of all those mp3-related bumperstickers and coffee mugs that would have to be scrapped if some new "mp4" came out that cost the same, worked as quickly, and sounded better than mp3. "That's crazy talk," you say. Sorry, it's not that crazy. It's called the MPEG2 Advanced Audio Coding, and it's been around for a while.

Almost all popular music-related websites have free or inexpensive mp3. To date mp3 is the most popular file format for delivering "near CD quality" digital audio. How did this come about? Why can't it sound better? You could easily compare the rise of the mp3 file format to the battle for high ground between the VHS and Beta video cassette formats. This just in: VHS won. So did mp3, aka the MPEG I Layer III codec.

Mp3 has been the most popular way of distributing music over the Internet since early-ish 1999. With a thriving mushroom patch of free mp3 software, processor speeds going up and over 600 mHz, and the introduction of high-speed Internet connections the mp3 craze took off. But mp3 isn't the only codec, and some credible experts have pointed out that better sounding alternatives are ready to go.

The MPEG Organization were not thinking of their codec being applied to the Internet when they started working on its standardization back in 1992. Back in the day, the International Standards Organization formed MPEG, or the Moving Pictures Expert Group, from the existing JPEG (Joint Photographic Experts Group). When some companies in 1995 were trying to figure out a way to deliver audio over the Internet they knew they had to deal with a few end-user realities. The c odec had to have attractive features to be accepted; it had to be fast and easy to encode and decode with the contemporary processor speeds; and be of the smallest file size for delivery over those old Internet broadband speeds (remember that 14.4 modem y ou gave to your kid sister?). Oh, and it has to sound good.

Way back in 1998, mpeg.org tested all the codecs for sound quality using different audio profiles. They figured that it might be wise to author a non- ahem Redmond-based study of these standards recommendations, whether they were ready for mass consumption or not. Each of the thirty experts involved in the study was an audio professional -- or at least they could tell what an "audio artifact" is.

The truth is that right now there are file formats that sound better than mp3. One such codec is MPEG2 AAC, or Advanced Audio Coding. At the time mp3 caught fire, the processor demand required by AAC for encoding and decoding was unrealistic, which is why mp3 won out. But now personal computers are ready for AAC, and depending on how fast standards bodies like the ISO and the IEC (or International Electrotechnical Commission) move, AAC wi ll be part of the upcoming mp4 revolution -- um, unless someone dethrones the MPEG compression format with some other brilliant idea. Also, keep in mind that future decoders will be backwards compatible. They'll be able to encode and decode all mp3 files, as well as being able to decode all files encoded in the MPEG II AAC compression codec.

The other issue with these compression algorithms isn't processing cost but actual money-type cost. When I look at the licensing terms and royalty rates one has to pay to develop, market, and sell any technology using MPEG codecs included in the licens ing program of THOMSON multimedia and Fraunhofer IIS-A, I have to wonder how anyone can really make any money. Thus the heap of terrible sounding encoders from code miners worldwide and all those virtual burnt-out buildings on the Internet's backstreets f illed with encoders that no longer have corporate support. Keep in mind that the licensing fees, which are very different from all the marketing dollars, are pretty much the same across the board when we talk about all these codecs.

To folks who think about this type of thing, a most strenuous search was worthwhile for an efficient coding method for surround signals, like the 5-channel signals (left, right, center, left-surround, right-surround) being used in movie theaters today. AAC can encode data in mono and stereo mode and multiple channels of data up to 48 channels. [48 channels! Wow! Two phat 24-track recorders all in one! (Sorry, sometimes I drink too much coffee when I write Angry Coffee...)] Simple attributes make it att ractive for applications that need high quality while having limited transmission bandwidth.

There are 3 different "profiles" in the AAC standard:

AAC Main - This is the deluxe model that includes all the tools available and delivers the highest quality audio.

AAC LC - LC means "low complexity." It reduces audio quality while also reducing processor demand.

AAC SSR - SSR means "Scalable Sampling Rate." It's the same as LC with a different filter bank and a special gain control.

Through rigorous testing of every imaginable way by The MPEG organization's tested the MPEG family of compression codecs and published the MPEG-2 AAC Stereo Verification Tests. If you just absolutely have to read this thrillride document, it's availabl e as a .zip file at www.mpeg.org/MPEG/aac.html

The study found this ranking of the best sounding codes:

1) AAC Main encoded at 128 kbps

2) AAC LC encoded at 128 kbps

3) AAC SSR encoded at 128 kbps

4) AAC Main encoded at 96 kbps

5) AAC LC encoded at 96 kbps

6) MP2 encoded at 192 kbps

7) And in 7th place the most popular format today! MP3 encoded at 128 kbps.

The study was completed and documented in February of 1998. I know, it's some old news, but it's still interesting. In fact it gets more interesting with each new computer you buy that has the power to play music that sounds better than the mp3.

As a fellow audiophile, I know that you are probably asking the question, "Whaaaa? What happened?" Perhaps this is another lesson in the inertia of big business dealings. There's so much money bet on mp3 that those who made the wagers have to be in pos ition to roll out the best sounding codec or lose their customers. But don't blame it all on poor Wall Street. It's just as likely that a better sounding codec will happen as soon as the oh-so-thorough international standards process has done it's necessa rily rigorous job. Whatever the reason is for the hold-up, keep an eye on this fine-sounding, if not yet broadly-used technology.