AVS Forum banner

Mastered for iTunes

3K views 24 replies 7 participants last post by  mani 
#1 ·
 This is a fairly good look at the AAC encoding guidelines set forth by Apple.


The quote below discusses the subjective and empirical results of the process. It seems to have an affect for at least some material. The "mastered for iTunes" waveform was closer to the original CD waveform than the ripped AAC file.

Quote:
We [compared] the original CD master, the file generated by iTunes from that master, and our own "specially mastered for iTunes" version. Subjectively, our tweaks made the track sound better. Much of the "boxiness" was gone, and the song sounded more "alive."


To make sure we weren't hearing things, though, we repeated Shepard's null test experiment. As with Shepard's test, comparing the CD master to a straight iTunes rip produced quite a bit of warbled noise and static. But the null test done with our own specially mastered version was very nearly silent. A few audible bits of music and vocals remained, but the result was much quieter than the noise we heard in the previous comparison.

I'm not an iTunes user and own no AAC encoded songs, but I have subjected myself some testing against MP3 and found that 128k CBR is generally good enough for me. 256k is inaudible in all but the most killer of samples. It's my understanding that AAC is considerably better than MP3 and killer samples are more difficult to come by. If the mastering techniques outlined above continue to evolve and improve, the loss from 256k AAC should be inaudible to all but the most gifted listeners.
 
#2 ·
With memory being so cheap now , any sort of compression should be unacceptable . Unlike other forms of technologies which are spending money to advance the performance and up the bar (like display resolution, processor speeds, cars , home appliances etc etc ).. audio is the only industry which instead of improving the resolution /content quality is still doing research on how to compress the songs with a better algorithm .... unbelievable .....
 
#3 ·
There is a point at which compression will not be required. We're not at that point. Cost is but one important aspect of the equation. Physical dimensions play an important role as well.


My phone, as a converged platform, is not yet capable of storing my entire library of photos, videos, music, and movies in uncompressed formats.
 
#4 ·

Quote:
Originally Posted by jmichaelf /forum/post/21967137


There is a point at which compression will not be required. We're not at that point. Cost is but one important aspect of the equation. Physical dimensions play an important role as well.


My phone, as a converged platform, is not yet capable of storing my entire library of photos, videos, music, and movies in uncompressed formats.

This may be true but how critical is SQ for a listening session from your phone? 99% of critical listening on a high-end system is done in an environment where form/size limitations don't apply, and here the quality of compression is moot; just get uncompressed formats.
 
#5 ·
I agree for portable devices . But when you buy a song we should get the uncompressed file that we can download on our PC or MAC and can convert it to compressed format for a mobile device if we need to . But I refuse to pay for compressed audio files .
 
#6 ·

Quote:
Originally Posted by jmichaelf /forum/post/21966581



I'm not an iTunes user and own no AAC encoded songs, but I have subjected myself some testing against MP3 and found that 128k CBR is generally good enough for me. 256k is inaudible in all but the most killer of samples..

Why did you decide to post in this section of the forum?
 
#8 ·

Quote:
Originally Posted by edorr /forum/post/21967285


This may be true but how critical is SQ for a listening session from your phone? 99% of critical listening on a high-end system is done in an environment where form/size limitations don't apply, and here the quality of compression is moot; just get uncompressed formats.

I would say SQ from the phone isn't very critical. Where else would you listen to compressed audio?
 
#9 ·

Quote:
Originally Posted by jmichaelf /forum/post/21967665


I would say SQ from the phone isn't very critical. Where else would you listen to compressed audio?

No where else I would hope. But I thought you were making the point AAC compresses was good enough for the vast majority of applications, which I disagree with. Good enough for a phone / ipod yes. A high end system - hell no.
 
#12 ·

Quote:
Originally Posted by edorr /forum/post/0



AAC is mostly used on iDevices. So to make a post about AAC relevant for the 20K forum you need to find iDevice playback hardware more expensive than 20K, which much to my surprise actually exists.

So you are just helped him
knowing full well that we are more interested in 24/192. I saw something on this topic. Rush's next release "Clockwork Angels," is behind enhanced for iTunes. Once I saw that it will also be released on HDtracks, at 24/96, I lost all interest in ITunes. Besides the focus on mastering for iTunes appears to enhanced the sound for stuff like computer speakers. Personally I hope to avoid anything mastered for iTunes that has some kind of bass boost, to make computer speakers sound better.
 
#13 ·
#14 ·

Quote:
Originally Posted by Bulldogger /forum/post/21968763


So you are just helped him
knowing full well that we are more interested in 24/192. I saw something on this topic. Rush's next release "Clockwork Angels," is behind enhanced for iTunes. Once I saw that it will also be released on HDtracks, at 24/96, I lost all interest in ITunes. Besides the focus on mastering for iTunes appears to enhanced the sound for stuff like computer speakers. Personally I hope to avoid anything mastered for iTunes that has some kind of bass boost, to make computer speakers sound better.

Not me. Who needs high rez. I just discovered the marvels of compressed audio. All my gear is up for sale on audiogon. I'm saving up funds for the $500,000 docking station.
 
#18 ·

Quote:
Originally Posted by jmichaelf /forum/post/0



Here's an interesting analysis of that report:
http://people.xiph.org/~xiphmont/demo/neil-young.html

"Unfortunately, there is no point to distributing music in 24-bit/192kHz format. Its playback fidelity is slightly inferior to 16/44.1 or 16/48, and it takes up 6 times the space." OK. Think I understand clearly why you chose this section of the forum. I'm out. No time for this.
 
#19 ·

Quote:
Originally Posted by Bulldogger /forum/post/21974282


"Unfortunately, there is no point to distributing music in 24-bit/192kHz format. Its playback fidelity is slightly inferior to 16/44.1 or 16/48, and it takes up 6 times the space." OK. Think I understand clearly why you chose this section of the forum. I'm out. No time for this.

I wonder if he's related to DougWinsor???
 
#20 ·

Quote:
Originally Posted by Bulldogger /forum/post/21974282


"Unfortunately, there is no point to distributing music in 24-bit/192kHz format. Its playback fidelity is slightly inferior to 16/44.1 or 16/48, and it takes up 6 times the space." OK. Think I understand clearly why you chose this section of the forum. I'm out. No time for this.

Sorry to have offended your sensibilities.


Also, it seems I'm late to the party.

http://www.avsforum.com/avs-vb/showthread.php?t=1398397
 
#21 ·

Quote:
Originally Posted by jmichaelf /forum/post/21975382


Sorry to have offended your sensibilities.


Also, it seems I'm late to the party.

http://www.avsforum.com/avs-vb/showthread.php?t=1398397

Forget whitepapers. I trust my ears. In my experience, 48/24 is a big step up from 44/16, 96/24 sometimes sounds marginally better than 48/24, and I don't hear a difference between 196/24 and 96/24. DSD sounds better than 96/24 for most recordings.


Finally and most importantly, impact of recording quality trumps resolution any day. A well recorded 44/14 track, is good enough for me, and does not leave me wanting.


The biggest difference in SQ to me comes from going from 2 channel to MCH. No comparison. Biggest regret in audio to me is not lack of high rez content, but lack of native MCH content.
 
#22 ·

Quote:
Originally Posted by edorr /forum/post/21975568


The biggest difference in SQ to me comes from going from 2 channel to MCH. No comparison. Biggest regret in audio to me is not lack of high rez content, but lack of native MCH content.
Of course, it might be nice to have both hi-rez and multi-channel together!


There seem to be several rumored 'improved quality music initiatives' happening "now|soon", including this one, which has the virtue of at least appearing to be 'already in progress':
Quote:
Originally Posted by SoundChex /forum/post/21814717


It looks like the Galaxy Studios|Auro Technologies "Octopus Codec" technology in Auro-3D 'might offer' a solution: it would appear to provide for 2 channel PCM [CD] audio playable (with NO decoder) on legacy hardware to deliver a 'traditional' quality|mix, and capable also of playback through an [Auro-3D] decoder to deliver an alternative higher resolution|quality and|or multi-channel mix [. . .] This interview at Musikmesse 2012 seems to suggest that ["some 100"] Auro-3D encoded albums are 'planned for release' in the next year or so...?!





[Note: The first 7 minutes is an overview of Auro-3D mostly related to movies; after the 7 minute mark, the interview references the music, game, and auto sound industries.]
 
#23 ·

Quote:
Originally Posted by jmichaelf /forum/post/0



Sorry to have offended your sensibilities.


Also, it seems I'm late to the party.

http://www.avsforum.com/avs-vb/showthread.php?t=1398397

So...I read the article.. I don't see any reason not to use higher bit rate and sampling rate in that paper.

Please.... members who are more technically apt can chime in to educate us. That article indicates that the speaker transducers and power amplifiers will have to reproduce /amplify ultrasonic signals 96kHz and upto 192 KHz,.... Which will cause more distortion in audible range????? Why would they need to do that . 96 and 192khz is the sampling rate and not the frequency that needs to be reproduced. Once the signal is sampled, processed and goes through ADC , the analog amplifier and transducer will need to produce only the audible frequency that's contained in the recording. Who is saying that we need ultrasonic frequencies to be produced by transducer to get better sound??


Here are my experiences ;

- just because file is 192 KHz doesn't mean it will sound better . But if song is mastered in 192 KHz by a good engineer , it does sound better to me. The difference by no means is night and day, but it's there. I have few jazz albums and " Hotel California " in 192 KHz and 44KHZ and I prefer 192 KHZ in almost all the songs.

- the original recording and mastering is by far most important than the file sampling rate and bit depth.


So by that article we can conclude that we have already achieved the holy grail of digital audio capture and reproduction at 44KHz


- The article goes through videophiles wanting X-rays and infrared wavelengths to be included in the signal?? I have never met anyone one or have seen anyone post on this forum even remotely close to that?
 
#24 ·
And regarding the extra space needed for higher bit rate and sampling rate files....here are few numbers we can quickly crunch...(I have used approximations for ease of calculation , but they are not too far from reality)


2 TB hard drive can contain roughly 4000 CDs ( assuming each CD contains 500 MB worth of recording) which can be had for $99 or thereabouts on sale

NOw if you say the 192 kHz files require 6 times more stage ( which is not necessarily accurate . The 192 kHz FLAC albums I have downloaded fit in about 1.2 GB or so from what I recall which is twice as much space), you can still have between 600-700 CD,s on that hard drive . There are probably handful of people on this forum who have more than 4000 CDs or more than 600 192 KHz albums . We are still at 100 dollars worth of hard drive and you can get another as a back up ... Still$200....!! So space being an issue is non existent.

Let's assume apple has 10 million songs and we want regular 44KHz uncompressed CD quality files. Let's say each song is 50MB , which is quite generous, we would need 500 TB of storage space. And at $100 per 2 TB , we are looking at $25000. Now let's assume that cost will be higher as they will need enterprise grade drives and need for at least double redundancy, we multiply that cost by 10, we are looking at $250000, which probably is rounded off while preparing financial statements for a company like apple. Other costs for maintaining servers , manpower etc will be the same whether the songs are compressed or not...
 
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top