A MacCentral report on November 9, 2004 entitled “Apple supported technologies approved by DVB,” confirmed it’s now official! Here are key excerpts from their interview with Frank Casanova:

The Digital Video Broadcasting Steering Board (DVB) has approved a revision to its implementation guidelines for audio and video codecs over a broadcast Transport Stream. The revision includes two technologies supported by Apple Computer Inc., H.264 or Advanced Video Codec (AVC) and High Efficiency AAC (HE-AAC) audio codecs.

Excellent!

I hope they now really (hint, hint) implement gapless encoding and playback. That's all that is missing from iTunes, IMO.

Anything that helps push AACplus more into the mainstream is good to hear. I just noticed Club 977 is now on the Tuner 2 listing, and have to say for a 40k stream it definitely sounds pretty good, much better than mp3pro ever did. Sure hope we start seeing more and more streams convert over the course of the next year.

so iPod gen 6th(?) will support HE-AAC?

Why 6th gen? Why not just implement new standards with a firmware upgrade now? I don't own an iPod so I wouldn't really know anything about such feasabilities (FWIW, bought a Rio Karma and love it). My comment was really only about how with Apple backing, and the word of officiality, that AACplus is now set to become the new standard in lossy audio compression. Either that or the latest buzz word, lol (though I doubt that, it's quite good IMHO even though I don't use it).

sounds like you think "AACplus" is something that can replace anything else. The "plus" part is just an extension for low bitrates. It has its applications but it's not very attractive for a typical HA music archiver.
I for one wouldn't make use of it even If I owned an iPod which is capable of playing those files.

my 2 euro cents,
SebastianG

Hmm, maybe I was being a bit too broad in my last post. Honestly, most of my interest has to do with making internet radio better (lower bandwidth usage, better sound). FWIW I actually use FLAC for everything at the moment, and Vorbis at -q4.5 for my Karma which I'm very happy with. If I did own an iPod (especially one with a small hard drive) I would certainly have no problem making use of AACplus if that was what it took to get my entire collection on there.

It better not be the only iPod to support it. That would be a nasty trick on Apple's behalf. Surely, they could provide support for at least 3rd and 4th generations. Although, I wouldn't hold my breath.

Well, they added AAC playback to the 1st and 2nd generation iPods. Of course, they did have financial incentive to do so with the opening of their music store.

They've made AAC so confusing. I don't use it myself, my portable player (iAUDIO M3) doesn't support it, and there are (arguably) many superior codecs around.

What's the difference between high efficiency (HE), low complexity (LC) and AACPlus?

This just screws over newbies and makes the whole "scene" a little more complicated.

It's always a disaster when there's more than 1 option (think VCR vs Betamax, DVD+ vs DVD -, and on a lesser scale MP3 vs MP3pro <-- disaster)

LC AAC == basic AAC everyone knows
HE AAC == LC AAC with SBR extension, meaning it will be better at lower bitrates

aacPlus™ is the brand name of one particular HE-AAC implementation (don't ask me why people talked about it here, the press release doesn't mention it at all). This is the same as for example, talking about "LAME MP3".

I wonder what those "superior" than HE-AAC codecs are, btw.

We should just use WMA, Oh wait, they also made that so confusing with the WMA Pro thing. Let's use MP3. Oh wait, they also made that confusing with MP3Pro. Let's use...

*Cough* Points at Ogg and MPC

AFAIK there's no Oggpro or MPCpro

Heck there isn't even VBR, ABR, CBR options to confuse people :/

Haven't heard of WMA pro. The variations with AAC seem to be more visual however, as many encoders offer both "flavours" (LC and HE).

I thought AACplus is like MP3Pro (added some extra junk [SBR, whatever] to make "better" sounding files)

But HE-AAC sounds like MP3Pro too (both added SBR, right?)

SBR = junk?? WTF?

http://www.rjamorim.com/test/64test/results.html
http://www.rjamorim.com/test/32kbps/results.html

That just shows these codecs aren't being actively developed. Because "Oggpro" (Vorbis2) and "MPCpro" (SV8) were planned.



In MPC, yes. That's one of the reasons it's not usable in streaming.

In Vorbis: there is CBR, ABR and VBR.



And is that a disadvantage?



Yes. To make much better sounding files at low bitrates.



No, it sounds better.

PoisonDan: I didn't mean that SBR is junk, just that so many variations makes things:

a) More confusing for people in general; and
b) Problematic when it comes to compatibility and support (both hardware and software)

rjamorim: I meant to say HE-AAC is similar to MP3pro (in that both added this SBR thing)

The variations are a disadvantage when trying to "market" such encoders to the general public. The acronyms don't say a lot about what each type does.

As this article kind of points out:
http://msn.com.com/2100-9588_22-5409604.ht...09604&tag=tg_bz



And this from the same article



I thought Ogg was VBR only? (It is when using the -q settings isn't it?).

Its debatable whether one can say that neither Ogg or MPC are "actively" developed. Both are in development, and I would say that Ogg is actively developed, and growing support from corporations and hardware would only help to push this along.

I still think it's funny how Ogg and MPC have been made by "freelance" people/programmers, while AAC and MP3 (and WMA for that matter) have been made by paid "professionals", yet every test conducted puts them ahead of the proprietary codecs

I believe that starting a new format just to add a feature like SBR would be even worse.

And the other option would be not adding SBR at all...



Well, that is a sad state of affairs indeed, but what are developers supposed to do? Halt all development on new technologies just because people are too used to MP3?



Vorbis has managed bitrate modes to output ABR and CBR streams. The -q settings are VBR.



Well, Vorbis is being developed, but not in a nearly fast enough fashion to keep up with competition. AAC developers are working quickly and furiously. Meanwhile, on the Xiph camp, noone is working on Vorbis anymore because Monty reinvented himself as CEO.

For all that I can see, MPC development is completely halted, at least at the encoder front.



That's partially because these "made by professionals" codecs have other stuff to worry about than audio quality. Streamability, complexity (matters for hardware implementation), good quality at the entire bitrate range (unlike MPC that only cares about mid-high bitrates) and so on.

Besides, AAC & al are harder to tweak because you are constrained by interoperability. With formats like Musepack, the developer can go nuts since he can break compatibility and just release a new decoder to fix a quality issue (since all decoders are software based, there's no need to worry about breaking hardware playback). With AAC, MP3 and WMA, you need to do workarounds to ensure the format stays the same (otherwise, even if you update the software decoders, the hardware ones will still be broken)

I've seen this argument before, it was bull then, and it still is bull now. Someone already pointed out just how wrong your "every test conducted" statement is.

In the area's where it matters, that currently being multichannel and low bitrates, the properietary codecs far outperform the free ones.
The free codecs are in the questionable position of being better where no normal user can hear the difference, and being worse where people *can* hear the difference. Yes, clearly those "professionals" have not been doing their job well

At the low bitrates were talking about here, they sound very different to me. Plus mp3pro has always mildly hurt (yes you read that right) my ears for some reason, usually after listening to it for a little while (can even give me a headache if I listen too long). I've always guessed it had something to do with the higher frequencies (SBR perhaps?) but don't really know enough about biology and these codecs to make even an educated guess. So I generally try to avoid it when I can, hence my comments about radio stations switching. As I've recently discovered, HE-AAC (ACCplus) isn't nearly as irritable to me, and again no clue why.

i would say that in the good old 128kbps range (doesnt this matter the most all in all? at least for me it does and i assume most people encode their stuff to that bitrate range) i think robertos tests showed that proprieatary implementations (aac and wma) stand a hard time against open source codecs, like vorbis and even old mp3

I agree. The area that actually matters to the target audiences of my tests is 128kbps. Other than that, they might care about 64kbps for 1-CD DVD ripping. And I still wonder what was the use of my 32kbps test.

>>And I still wonder what was the use of my 32kbps test.

I-net radio, may be ?

Yeah, but I don't think radio stations are taking my test results into consideration. :B

For them, format support is way more important than quality.

I thought ya did it just to know(on the 32kbs test)?

I imagine sooner or later i-net stations will use it. Right now MusicMatch uses Mp3Pro as their CD quality streaming codec(not that it is CD qua).

I wonder if Mp3Pro used Lame's encoder for the Mp3 encoding portion of the Mp3Pro files would end up sounding better? Usually I hear more problems in the midrange than the high end of Mp3Pro encoded files, that is why I wonder this. It is easy to ABX a Mp3Pro file though, for me the high end is slightly lower volume than the original(so it is almost always noticable).


I don't think anyone ever answered someone above about HE-AAC, AACPlus, and Mp3Pro.

The deal is all these codecs use SBR.

HE-AAC=AACPlus

Mp3Pro was the first to use SBR though. The name they gave that version of Mp3 was Mp3Pro.

What is going on is the codec only encodes half the spectrum of frequencies. So AAC and Mp3 encode up to say 10khz and the high frequencies are encoded using the SBR technique. SBR is added back to the file with the Mp3/AAC data. I believe SBR takes around 2kbs worth of data, if I remember right(might be 4kbs).

I didn't mean to offend you in particular as a developer (how come Ahead are developing/interested in AAC anyway?)

But numerous tests (albiet many use mid-high bit rates for musical compression), show that the "free" codecs are performing rather impressively against the new wave of proprietary ones.

What sort of applications are AAC targetting in particular then? The main arena for multichannel which I can think at the moment of is DVD's and movie theatre audio (which use other formats, like AC3 right?)

While low bitrates can be used for business conferencing etc. (which as far as I can see is still relatively limited to larger corporations/mulinationals etc. in the real world). Use of video conferencing using MPEG codecs would make AAC usage unneccessary



I thought they're separate. I read an article saying that AACPlus was different and that it was aimed at either replacing AAC or just producing overall better sound for other applications:

http://www.codingtechnologies.com/products/aacPlus.htm

So is AACplus old news then, or was it recently ?

What I mean't to say before that: "HE-AAC IS like MP3Pro too (both added SBR, right?)"

I mis-typed that sentence in my earlier post (didn't pick up on it until someone quoted it. I understand that HE-AAC should indeed sound better than MP3Pro

EDIT - BTW, is HE-AAC limited to lower bitrates like MP3Pro seemed to be? It seems weird to do that if it does make better sounding audio files. (MP3Pro couldn't go higher than around 98-112kbps or so I think, while "standard" MP3 can go to 320kbps (or even 640kbps if anyone wanted to try it)

Read the first sentence of your link:

And I still wonder what was the use of my 32kbps test.

--> some amateur "ham" net radio want to start up his/her own radio station, maybe

or [embed] song in blog/forum

or better sample song ( for example tranceaddict.com and their ID section)

etc

I already acknowledged and quoted that very line in my post. Thanks for reading