[blessingx]

Quote:

Originally posted by Moguta What worries me about the iTunes Music Store service is how they are selling 1) less than CD quality material as CD quality, and 2) they don't allow you to do anything but listen to the music on your computer or transfer it to the Apple-manufactured iPod player. And since you're not getting a CD, jewel case, booklet, or any physical shipping/distribution, you think they'd be able to knock the price of a song down pretty low. But no, a 15 song CD still costs $15. And modem users would be better served by just going out & buying a CD for the time it takes to download.

Just to keep the record straight (since this has be said on this site several times)- the songs are 99 cents, but most complete albums are $9.99, not $15ish (15 tracks or not). And since some reports have stated otherwise, I also wanted to say the aac tracks can be burned to cd-r for conventional CD playback.

Wish the quality was at least 192 acc kps though.

 

[FrostyMMB]

Quote:

Originally posted by Moguta What worries me about the iTunes Music Store service is how they are selling 1) less than CD quality material as CD quality, and 2) they don't allow you to do anything but listen to the music on your computer or transfer it to the Apple-manufactured iPod player. And since you're not getting a CD, jewel case, booklet, or any physical shipping/distribution, you think they'd be able to knock the price of a song down pretty low. But no, a 15 song CD still costs $15. And modem users would be better served by just going out & buying a CD for the time it takes to download. Oh, and just to throw a little twist at the unknowing... ...just because lossy compression like AAC & MP3 isn't full CD quality, doesn't mean a lossy file can't sound exactly like the original CD track to the human ear.

Well, the iTunes music store isn't much different than any of the other places selling digital music online. Plus, the iTunes music store a hell of a lot of easier than the others, which probably the biggest selling point. Sure it's not CD quality, but that sits fine for plenty of people, but not with many of us here on these forums because we prefer to have lossless audio, or at least audio with less loss than what online digitital music places offer.

It's not for everyone, and I'd say it's for few of us around here. But, plenty of other people don't notice or care that aac and mp3 aren't CD quality, and they'll agree anyone who says that it is, even when we know it isn't. And some of those people probably listen to mp3s exclusively. They won't have a problem with using them on their iPod or burning them onto a CD, and they'll be satisfied just fine. Thats why so far the iTunes music store hasn't been a fialure like plenty of people around here predicted.

I agree that the price per song should be less. I wouldn't pay a buck for compressed audio track. Plus, I want to the jewel case insert and the CD. But I understand that the recording companies still want to make some money, and Apple needs to make a profit to pay for bandwidth costs. There's a lot of things that that $0.99 pays for. So it's not that easy to just knock the price down.

 

[Moguta]

Quote:

Originally posted by FrostyMMB It's not for everyone, and I'd say it's for few of us around here. But, plenty of other people don't notice or care that aac and mp3 aren't CD quality, and they'll agree anyone who says that it is, even when we know it is. And some of those people probably listen to mp3s exclusively. They won't have a problem with using them on their iPod or burning them onto a CD, and they'll be satisfied just fine. Thats why so far the iTunes music store hasn't been a fialure like plenty of people around here predicted.

Yeah... and then you'll have people transcoding MP3s from their AAC-burned CDs, and on goes the cycle of bad quality compression.

The bad thing is that it does take so much work to rip & encode files well, that most people would rather buy or download any compressed file they come across.

What's needed is a one-step ripper & encoder that defaults to the best of EAC's and LAME's settings. Plus encoders that inform you what actions aren't good to take, so newbies don't go screwing up settings left & right.

That would be quite some work. =(

 

[blessingx]

Quote:

Originally posted by Moguta What's needed is a one-step ripper & encoder that defaults to the best of EAC's and LAME's settings. Plus encoders that inform you what actions aren't good to take, so newbies don't go screwing up settings left & right. That would be quite some work. =(

I assume this is a joke right?

 

[FrostyMMB]

Newbies don't care if they get 128k MP3 as opposed to a LAME alt preset extreme. A piece of software that defaults to the best settings will leave the newbies wondering why their MP3s have such huge filesizes. I have a friend who downloads 128kbps MP3s, the worst sounding ones at that, then burns them to CDs. He has a 100GB hard drive and throws away MP3s that are at higher bitrates because he only has 72.1GB left. He's lucky that I still call him a friend.

Anyway, in iTunes, it'll be pretty easy for me to re-rip to AAC. I'll stick the in CD and click import. The files will automatically be added to my library, complete with tags. It'll rip and encode a lot faster than the LAME script encoding with the extreme preset. They'll sound better and the files will be smaller. I'll plug in my iPod and my entire library of 8 gigs will transfer in a matter of minutes. That's not too much work for me.

 

[mEtal]

AAC is the real deal and because of it's large backing, it has the most potential among current codecs to widely replace the mp3 format. It still has a few minor kinks to it, but you can still get better audio quality for lower bit rates than mp3. There are a few encoders out there already, the best one that's publically available is Ahead's PsyTEL AAC encoder.

 

[jessica00]

and on that note..
http://www.pcavtech.com/play-rec/Psy...109-winamp209/
Signal to Noise Ratio - Very Poor
strange i guess.

 

[Moguta]

Quote:

Originally posted by jessica00 and on that note.. http://www.pcavtech.com/play-rec/Psy...109-winamp209/ Signal to Noise Ratio - Very Poor strange i guess.

Only 41dB SNR? That seems odd. Are those the latest encoder/decoders?

And in response to Metal, AAC is poised to become the next standard only because it is made by the same developers of MP3, who now have massive attention & finances from the success of their first format. Nevermind if others might be better, or others might have a single encoder/decoder (excluding version numbers) so there is guaranteed uniform quality, not just good quality if you use a particular encoder. (Ogg Vorbis, MPC anwone?)

 

[Lando]

Aac has a lot of backing and less pre echo than mp3, but the quality is still inferior to Ogg at a lower bitrate. I'll also say the 128kpbs aac is anything but cd quality. With the exception of less pre echo, it sounds just as bad as a 128kbps mp3. The encoding is also extremely slow (around 3-4x realtime). Unfortunately, it will probably be a long time before a superior format like Ogg or Mpc sees any serious commercial backing. Big corporations rarely like open source efforts, and people in general don't care if their music sounds crappy.

 

[FrostyMMB]

Quote:

Originally posted by Lando Aac has a lot of backing and less pre echo than mp3, but the quality is still inferior to Ogg at a lower bitrate. I'll also say the 128kpbs aac is anything but cd quality. With the exception of less pre echo, it sounds just as bad as a 128kbps mp3. The encoding is also extremely slow (around 3-4x realtime). Unfortunately, it will probably be a long time before a superior format like Ogg or Mpc sees any serious commercial backing. Big corporations rarely like open source efforts, and people in general don't care if their music sounds crappy.

I'll agree with 128kbps AAC not being CD quality (of course it isn't), but I will disagree completely with your two statements that follow. 128kbps AAC is superior to 128kbps MP3, superior to 160kbps, and I'd say it's close to 192kbps, on which side I haven't figured out. I could do loads of blind listening tests to back this up, but I don't have the time. I can just listen and hear that AAC is better than MP3. On my machine, iTunes/Quicktime's AAC encoding is much, much faster than the LAME 3.92 extreme preset. I mean 2-3x faster than LAME encoding, and much faster than real time.

MacDEF and blessingx seem to have done some casual comparisons of quality, so I don't think they'd agree that it sounds worse that MP3.

 

[clintf]

Quote:

Originally posted by FrostyMMB I'll agree with 128kbps AAC not being CD quality (of course it isn't), but I will disagree completely with your two statements that follow. 128kbps AAC is superior to 128kbps MP3, superior to 160kbps, and I'd say it's close to 192kbps, on which side I haven't figured out. I could do loads of blind listening tests to back this up, but I don't have the time. I can just listen and hear that AAC is better than MP3. On my machine, iTunes/Quicktime's AAC encoding is much, much faster than the LAME 3.92 extreme preset. I mean 2-3x faster than LAME encoding, and much faster than real time.

I would agreet that at least Pstel encodes (VBR profiles, why use CBR?) are definitely better in the low bitrate range. Psytel -streaming is quite good for what it is producing files in the 120-130kbps range. I wouldn't say a 128kbps AAC is similar in quality to a 160 or 192kbps MP3 (although 192kbps is approaching APS bitrates and I would have serious doubts about a 128kbps AAC, especially Quicktime aac reaching --alt-preset standard quality). AACs in the ~128kbps still have problems, though they tend to sound less annoying than MP3s at such a low bitrate.

BTW I think when people are talking about slow AAC encoding speeds, they're talking about Psytel aacenc which is slow, but from what's reported at HA may still be the best public AAC encoder. They could very well use Psytel fastenc which is far faster, but with a compromize to sound quality, which is presumably why Apple's Quicktime encoder is so fast, not to mention it's likely heavily optimised for those RISC CPUs. BTW if you really wanted Lame to haul-ass, you could, but it would sacrifice quality to speed it up so it's not really fair comparing --alt-preset extreme to AAC encoder with speed as its priority. A lot of Apple users are on machines with horribly slow CPUs <450Mhz, the AAC encoding in Quicktime needs to be able to work at a reasonable speed with those processors so you can imagine compromises must be met.

 

[Moguta]

Quote:

Originally posted by clintf A lot of Apple users are on machines with horribly slow CPUs <450Mhz, the AAC encoding in Quicktime needs to be able to work at a reasonable speed with those processors so you can imagine compromises must be met.

Compromises that wouldn't need to be made if MPC were used instead. Fast encoding w/o a sacrafice in quality.

And while MPC is open-source, it's far from being patent-free as Ogg Vorbis is. Suspected MPEG-2 patents could apply to MPC, which is why there is no hurry for any corporate support at all. They don't want to get sued for MPC using what could turn out to be a Fraunhofer-patented algorithm somewhere in the code.

At least Ogg Vorbis has two portable companies developing support. For me, public-interest formats like Ogg have a lot more significance than corporate-interest stuff like MP3/AAC. Of course, people downloading music (some for free) could care less about whether the format is corporate-created or not. Maybe some will care more about the impact of corporate interests when most CDs are more copy-protected & Microsoft's Palladium makes its debut.

 

[FrostyMMB]

Hold on. Where's this documentation stating that the AAC encoder used in Quicktime/iTunes is compromising sound quality for encoding speed? You guys are talking like you know that corners are being cut to make encoding faster. If you're basing this off of old HA threads, maybe a year ago what you heard from a Quicktime AAC vs Psytel AAC would be what you hear from a comparison today. The encoder hasn't just sat in stagnation. Another one of those 'Macs are slow and everything Apple is inferior' attitudes.

 

[clintf]

Quote:

Originally posted by FrostyMMB Hold on. Where's this documentation stating that the AAC encoder used in Quicktime/iTunes is compromising sound quality for encoding speed? You guys are talking like you know that corners are being cut to make encoding faster. If you're basing this off of old HA threads, maybe a year ago what you heard from a Quicktime AAC vs Psytel AAC would be what you hear from a comparison today. The encoder hasn't just sat in stagnation. Another one of those 'Macs are slow and everything Apple is inferior' attitudes.

I think I remember Ivan Dimkovic saying he thought Apple's Quicktime's AAC encoder was a type of fastenc and I have yet to see anyone suggest it produces superior quality encodes to Psytel aacenc (not one year ago, not today). Remember that it's not even accepted on HA that Psytel's successor Nero AAC plugin is superior to Psytel.

"Another one of those 'Macs are slow and everything Apple is inferior' attitudes."

Where do you get that from? Are you getting defensive over an aac encoder?

 

[Pepzhez]

I for one am not going to get defensive over an AAC encoder, but I will add that, according to Apple, the new version of QT features an overhauled AAC encoder. What exactly the changes are is unknown. (Apple does have an annoying habit of not revealing very much.)

I read Ivan Dimkovic's post concerning the "compromised" QT AAC encoder at the Hydrogen Audio forum, but keep in mind that his post predates the latest QT release and its "new, improved" AAC encoding engine, so Dimkovic's comments may or may not be of relevance today.

For what it's worth, I did not bother much with AAC until the latest QT release, so I'm not qualified to say that the current QT AAC encoder is superior to the older one. However, my listening tests have made me a convert: 192 kb/s AAC (QT) compares very favorably with (and in some cases, exceeds) LAME alt-preset-extreme. I do agree that 128 kb/s AAC is hardly at the quality level Apple claims it is, but AAC at 192 kb/s is damn near transparent to these demanding ears. In fact, I find AAC @ 192 to be superior to the aforementioned LAME settings, and I appreciate the disk space saved. I also appreciate the speed of Mac AAC encoding (as compared to LAME). And, yes, I am also fairly certain that AAC encoding in QT is optimized for PowerPC processors as well.

One more thing: most Mac users today are NOT using processors less than 450 mhz! (And I'm not going to get into the old familiar story that a 450 mhz PowerPC processor and a 450 mhz Intel/AMD are not comparable in any real way- you are comparing apples and oranges here - pun not intended.) Those using older Macs, yes, but Apple itself has not sold a machine that slow in at least a couple of years. I really don't know where people get such crazy notions.