I can't recall Tower Records, HMV, Newbury Comics, Strawberries, etc. ever making a stand on an issue that would IMPROVE the listening situation of the consumer.

I would agree that there is a trend towards higher fidelity... but I think that Apple is actually participating in that trend... and in this case specifically driving at a better listening experience.

I'm unclear what "upstream music production" means? They do manufacture a leading DAW.

t

Wow, after reading that I'm surprised my grocery store can still sell Apples, or do they have to pay the Beatles a licensing fee for each apple sold? Maybe that's why they cost so much.....

exactly...

sheete!

damn!

doh!




Got me.

Hi Res audio from Apple
 

Update: High def audio on itunes

Dude! A little respect! We are damn lucky to have the people who actually make this stuff post here. Come on. Yes, he is an expert!


That being said I still don't like ACC. But I don't like MP3 either so...

Guardian Reports possible hi-res files on way for iTunes
 

Apple is working on a new audio file format that will offer "adaptive streaming" to provide high- or low-quality files to users of its iCloud service.

The new format could mean that users can get "high-definition" audio

Full article at:

Apple developing new audio file format to offer 'adaptive streaming' | Technology | guardian.co.uk

Well you DID notice I said I use lossless if I use encoding at all nowdays, didn't you? :)

Perceptual coding is for WHEN YOU HAVE TO USE IT and no other time.

They'll let anyone into Carnegie-Mellon these days.

This is sad. It's typical of the audio industry, and of the general USA's anti-science political climate in general.

But I have to say that the audio production community went anti-science rather before the republicans.

For the record, my comments were based on MPEG tests run with good encoders, on difficult material, and the results are very, very hard to argue with, at least from a factual basis.

I did notice somebody else said something about m4a having smaller files, they did notice, I hope, that I said "at the same bit rate" which would give you the SAME sized file, give or take a tiny bit.

Finally, people use AAC (and MP3, and WMA-pro, and most other compression algorithms) at rates they should simply not use them at.

That quite aside from there being some rather unpleasant encoders out there. With any of the perceptual coders, the encoder decides all, and if it's wrong, you're scrawed.

I think you meant "in 1970", yes? :p

jj - since you are here, maybe you could shed some light on variable bit rate encoding. I noticed the new Apple AAC is variable bit rate. In my own experience with mp3 VBR, I find it lacking in low midrange detail - particularly in the "body" of bass guitar tones. I have actually surprised myself several times; noticing this phenomenon while listening, only to find out later that it was a VBR encoded mp3. Of course, it could be just the encoder used or just not a high enough rate. But, can you speak to the comparative quality of VBR as opposed to constant bit rate?

If the VBR is doign this, it means that the threshold calculation in the encoder isn't providing the right masking thresholds, or that something that impliments them is fubar.

Hi JJ, I have learned and learn a lot from you, THANKS!

I think, in this case, the problem comes from the decoder and not the encoder.
For example, iTunes (the computer software), reconstruct audio clipping at 0 dBFS.
This is wrong. It would be much better to reconstruct to 32 float, normalize to 0 dBFS (maybe taking into account ISP's), and then going to 24 bit for the DA converter.
I have tested the new iTunes Mastering suite (except afclip because it does not work in my computer) and you can encode AAC files up to +12 dBFS without distortion (AAC from 32 bit float, file reconstructed with Wave Editor). A 1 kHz sine wave starts to distort at +14 dBFS (realtime @ 44.1 kHz).
So the problem, at least with the iTunes AAC encoder, is not on the input.

Clipping is fatal for lossy audio. But clipping is not 0 dBFS. Many masters clip at -0.1 dBFS or -0.3 dBFS. So the solution, in my opinion, is not in the final level of the master but in the absence of "square wave" clipping. Clipping is not bad for linear PCM but it's mortal for lossy audio, so I think it's necessary to do a different master for lossy.
If you go to 0dBFS without clipping you will have better sound than clipping at -1 dBFS and encoding to lossy.
The ideal scenario will be if all decoders, and lossy audio players, reconstruct to 32 float with ISP normalization to 0 dBFS.
I think it's time to update the iTunes software... IMHO, of course.

yingyang

-------------------

A side note from my iTunes Mastering suite testing.

afclip: it does not work in my computer. The wave picture in the manual looks very dangerous for the speakers. Maybe it needs a WARNING!: don't play this file.

afconvert: works OK.

Droplet: fast and easy. Great.

RoundTripAAC: absolutely fantastic!
It works very well as intended and it could be used also for creative purposes in a mixing environment:
1. Put the same soundfile in two tracks.
2. First track Dry, channel fader -1 dB.
3. Second track plugin order: Volume plugin #1--> RoundTrippAAC (with delay compensation) --> Volume plugin#2. Channel fader about -15 dB.
Params:
Volume #1: +30 dB (from normalized 0dBFS audio). (AAC input distorts).
RoundTrippAAC: test different encoder type and bitrate (each one needs its own delay compensation). The AAC-HE is really funny at very low bitrates...
Volume #2: -30 dB.
Name for this processing chain: Warmaac (AAC Spectral "Warmifier").
Introducing lossy audio as an effect...? :cop:
sorry...


SRC is phase linear.
At 96 kHz I have discovered some non linearities. At +18 dBFS 3rd harmonic predominant... tape/console emulation prior to SRC at 96 kHz?
...mystery...

:cool:

JJ...OK so you're not an Apple plant...I think anyway...but seriously now...the only thing sad is how a multi billion dollar company stood on the backs of creative folks and their support systems (free content) to sell their gadgets...now is that on topic?..nope? does that have anything to do with AAC at all?...nope...

so yeah a bit of a winge...but very realistic.

Now AAC...have to still say...even though I have my foot in my mouth regarding your expertise....

it simply is not good sounding...sorry man...but it doesn't handle transients well...from my experience...Logic mostly for AAC...and downloads from iTunes...soft soft softy soft...changes the shape of mixes too much for my liking...decent 256 MP3s are just fine...and much better than how cassettes represented...I might add.

yes/no/both

The problem is that any perceptual coder is going to toss the parts it thinks are inaudible. This means a lot of upper harmonics that are masked. This means that the waveforms are going to acquire ringing. So the damage is done in the encoder, but won't manifest until the decoder.

And of course, clipping below fs is going to create a (I*(**&(&U load of harmonics, many of whicha re going to get removed, and there you go, EWWWWW.

The 3rd harmonic is, um, 'interesting' in that second thing. Dunno offhand.

Nulling an mp3 or AAC against the original wav can be very informative as to what is actually being eliminated from the signal to achieve data reduction. With the better encoders and their highest rates, it's not all that much.

Hmm, lemme think, how much did Steve make on AAC and itunes?

Now, how much did I make on AAC?

Yeah, on the backs of the inventors, you bet your not too sorry ***.
Sounds like a bad encoder. A good encoder at 256kb/s with AAC is pretty much bulletproof. Most people fall off the ability to detect it (meaning trained listeners, not people off the street) circa 160kb/s. You can see the results on the old CRC test for the US DAR project.

But, yes, that was my research encoder, which remained with Lucent/AT&T and that was not picked up by anyone else, because Fraunhofer disagrees with how to make encoders, mostly.

But what you said suggests a real problem in the block switching detection to me, or maybe a faulty TNS implimentation. I'd have to look, and I'm not likely to, since the companies that make money from it don't give any to me.

How does "figuring out" what is the least possible quality that you can slip by most humans [without them noticing the degradation in the moment], so that you can surreptitiously remove the other true, original information [information that the conscious, subconscious, and / or superconscious mind may have perceived, or worked toward perceiving], help the humans to become more highly evolved with respect to perception and / or otherwise?

Think about what you do.

When, if at all, is it truly appropriate and beneficial, not just expedient, to employ such methods? When are they anything more than simple deception?

Who determines that the few who are perhaps capable of identifying the degradation in the moment are unimportant?

nonsense...it's a lot

Apple made Billions of the music industry...make no mistake about that...what you got paid for developing AAC...well that is none of my business.

AAC? I don't like it.

Thanks for the info. I didn't encode them, so no telling really. A few of them were actually from Amazon.com.

Of course there could be something other than VBR causing this, but then it' a strange coincidence that each time I detect it, the file is VBR.

The same idealist criticism could be applied to any audio release medium from the wax cylinder on down.

Have you tried it? What was left?

Maybe you should use something else.


DC

Oh, come on, not another one of those luddite rants.

At any give frequency (inside of a critical band around it, as well) your ear has at most a 30dB signal to noise ratio.

That's a simple fact. If you do, in fact, always keep everything below that, taking into account the time-domain issues of a varying bandwidth with frequency, the stimulii on your auditory nerve will be equivalent.

That's a simple fact.

If you want to know how that 30dB gets mapped over 90+db, pick up my hearing tutorial at www.aes.org/sections/pnw/ppt.htm and study it.

Perceptual coding is for when you MUST use it. How is this hard to understand?

Straight salary, like if I hadn't done it.

Oh, and Bubba, as someone who has spent a lot of time analyzing the difference between AAC and PAC signals and the original in my time, they don't have to be very different.

The whole key is in HOW they are different, or not.

If you ever find yourself at an AES meeting that I'm at, I have a couple of files to play for you.

How does one go about uploading 24 bit masters to iTunes?

Major labels tend to use the 16 bit masters ripped from a disc for iTunes uploading in my experience. Tunecore and ReverbNation don't seem to accept 24 bit wavs.

How does a mastering engineer go about getting his clients' 24 bit masters to Apple?

How does an artist go about getting the "Mastered for iTunes" designation on their album once it's uploaded?

Not by me, man. Because I would find it ridiculous.

Striving to reproduce sound with the best possible fidelity, given current technology, and an attempt to devise, or to determine, a means of "fooling" people with carefully disguised sub-optimal fidelity, are different things, in my view.

I can't believe you think the two different things would somehow be indicted by the same argument.

I can't believe someone like you would say something like what you did, quoted above.

Also, I don't understand why you speak of idealism as if it were something bad. Idealism is good, imo.

I have absolutely no control NOR do I have any pull in this area, other than to ridicule people who use ridiculous bit rates, in which case I'm just another voice in the crowd.

But why would you suppose I've been saying that since 1989? Here, need a clue-by-four?
Since nobody is "dumbing down the public" with a PROPER application of the science of hearing, I'm not sure where you got that. There is no insult to anyone's intellegence, unless you want to get into the clipped, hypercompressed (in the level sense) crap that is all that seems to get turned out these days. I mean, what maroon decided to clip the remaster of the Tremelo's "Silence is Golden?" Geemany dogs allbarking.

You may not think my explanation is comrehensible, I submit that perhaps you can tell us all, right here and now, the difference between loudness and intensity. From there we'll move on to critical bands, the cochlear filter bank, and thence to masking.

Oh, and your scare quotes constitute an accusation. Care to try to back that one up, fella? I really have had it with this profession, it's chockablock full of people who like to be rude, nasty, and insulting, and who would rather get into a fight than learn.

So, seeing we still can't generate the actual assets ourselves, are any MEs here at this point confidently handling the labels 24/96 files for iTunes-ification? Sounds like a potential can of worms as long as procedures aren't tried & tested on the label side... Do any labels have provisions for this already?