Good article, and kudos for pulling all this info together in one place. Your experience with Plextor drives mirrors my own, though I imagine you have tried many more drives than I have. It’s reassuring to find that my (less comprehensive) testing wasn’t a waste of time.

One minor point: AAC was *adopted* by Apple for iTunes, but it was developed by Dolby Labs as an improvement to AC-3. It stands for Advanced Audio Coding.

A few of big corrections for this article:

1. AAC was not developed by Apple; It was developed by Fraunhofer as a successor to MP3. This error is repeated throughout the article.  Apple simply licenses the codec from Fraunhofer.

2. “Music sold on iTunes (AAC 128) and Rhapsody (WMA 128): Small file size for downloads and portability, but sound breaks down significantly when played through stereo speakers.”

This is not accurate as well; most of the audio problems encountered by playing AAC files back are the result of the cheap electronics used in the headphone amplifier of iPods or computers, coupled with impedance mismatching.  If you do a true ABX test of AAC material that had been converted back to PCM and tested against the original, 99% of people can’t hear the difference.  I know because I have participated in the testing.

3. “Note that digital music sold online through iTunes and Rhapsody is 128 kbps, which is far inferior to CD quality.”

See above.  You can’t make this claim without proving it.

4. “Apple’s DRM, for instance, restricts the amount of times that a song purchased through iTunes can be copied or burned onto a CD”

Not true… Apple’s DRM limits the number of times a particular PLAYLIST can be burned (without change) but there are no such limitations on individual tracks.

Jim must be a PR rep for Apple. Noone needs scientific proof that these formats are inferior sound-wise. Jim you are an idiot. Apparently you are deaf (along with everyone else in your study).

I think you are splitting hairs on the A/B comparisons. His point I believe is that if you listen to a highly compressed file on something other than a pair of headphones the limitations of the file will be exposed, and you don’t have to be Bob Ludwig or Mutt Lange to hear those differences.

As for the claim that 128kps files are inferior to CD level sound (16-bit/44kHz or about 1411kps) that is a fact.

As for the DRM rules of iTunes’ files, it’s more compliciated that what’s expressed above. Here’s a more detailed explanation or Apple’s FairPlay DRM policies:

FairPlay-encrypted audio tracks allow the following:

The track may be copied to any number of iPod portable music players.[1]
The track may be played on up to five (originally three) authorized computers simultaneously.[1]
A particular playlist within iTunes containing a FairPlay-encrypted track can be copied to a CD only up to seven times (originally ten times) before the playlist must be changed.[2]
The track may be copied to a standard Audio CD any number of times.[2]
The resulting CD has no DRM and may be ripped, encoded and played back like any other CD. However, CDs created by users do not attain first sale rights and cannot be legally leased, lent, sold or distributed to others by the creator.
The CD audio still bears the artifacts of compression, so converting it back into a lossy format such as MP3 may aggravate the sound artifacts of encoding (see transcoding). When re-ripping such a CD one could use a lossless audio codec such as FLAC, however such files take up significantly more space than the original .m4p files
At this time, it looks like the restrictions mentioned above are hard-coded into QuickTime and the iTunes application, and not configurable in the protected files themselves.

An artifact of Fairplay is that it prevents iTunes customers from using the purchased music on any portable digital music player other than the Apple iPod, Motorola ROKR E1, Motorola SLVR, or iPhone.

Bob Archer

Cliff,

Glad to see you resort to ad hominem attacks and name calling instead of actually providing relevant material.  Truly the sign of someone who has nothing to offer.  How many recording studios have you worked in?  How long have you been in the business?

I have no connection with Apple, and I disagree with a lot of what Apple is doing in the music retail market, but I know what I am talking about.  I have worked as a producer of high resolution commercial music releases for years.  We’ve conducted multiple ABX tests on AAC encoded material with many people of all age ranges using accepted methodology, and most people cannot tell the difference between PCM and AAC->PCM tracks.  It is important to follow the methodology to eliminate variables like crappy headphone amp stages, poor cabling and interconnects, noise due to electronics, differences in DA converters, differences in source material, etc.  When you eliminate all that, you are comparing just the files, and there is virtually no audible difference that people can detect.

Bob,

“As for the claim that 128kps files are inferior to CD level sound (16-bit/44kHz or about 1411kps) that is a fact.”

That wasn’t the claim that was made; the claim was that “the sound breaks down through stereo speakers” and that the files were “FAR inferior” to CDs.  I am saying that while there is a difference between the source and the encode, it is very small and most people cannot hear it except on specialize audio content that specifically breaks encoders.

I read your long list of Fairplay rules…did any of them contradict my comment that playlist burns are limited but individual track aren’t?  What was the point of that?

Also you said “I think you are splitting hairs on the A/B comparisons. His point I believe is that if you listen to a highly compressed file on something other than a pair of headphones the limitations of the file will be exposed, and you don’t have to be Bob Ludwig or Mutt Lange to hear those differences.”

No, that’s not what he said. He said that the AAC files are FAR inferior to CDs.  I agree listening to songs on crappy headphones and cheap playback electronics can be a problem, but that has nothing to do with the quality of file itself.  If you want to blame something for crappy sound, blame the $.05 amplifier and cheap DAs used in the iPod, not the AAC encode.

B

First of all I think you’re arguing samantics over the phrasing of his sentence. The point he is making is that compressed files don’t not sound as good as uncompressed files. I am sure the signal to noise ratio of an uncompressed file is higher than a compressed file. That would probably explain any technical differences in the ability to play an uncompressed file louder without audible distortion than a compressed file.
I also disagree with you the quality of IC parts used today to assemble even the cheapest electronics today. The chipsets of today perform better than anything anyone has ever experienced from previous product generations.
There are also a number of studies that can be found on the Web that have found that in A/B test consumers choose the higher bitrate file.
There are differences in the quality of files, and if there were no obvious differences I am sure the RIAA and its members would cram more songs onto a CD to charge people more money since no one would be complaining about the sound quality of their CDs.
Moreover, it’s bad enough that there are plenty of people that complain about the compression used at the mix/mastering level of the recording process to make things radio/iPod friendly. People would really complain if the CDs they purchased were encoded at 128kps.
There is a difference between redbook CD standards and compressed files and you don’t have to be a recording engineer or a mastering professional to hear it.
Lastly, there are limitations on the usage of Apple’s DRM files, and it’s stated in the above email.

Bob Archer

Fortunately some of you have already mentioned all the errors in this article. I have found technical inaccuracies/mistakes in almost all of CEPRO’s articles and I don’t know why AVSForum lets them spam their site with nonsense.

Some of the “facts” in this article: Ugh.

Fix your factual mistakes. And yes data is required to make rash claims about something sounding better than the other. Jim is right. The end.

“An audio CD player reads data in a continuous manner, with its laser following a smooth track. Computers, on the other hand, read information in blocks.”

Block-reading is actually an advantage when it comes to producing high quality rips, not a disadvantage.  Most CD-ROM drives for PC’s can read a CD in one continuous stream if your software tells them to do so.  This is fine for one-time playback, but it’s not desirable at all if you’re trying to make a perfect copy. 

As with any physical media, CD’s are prone to reading errors.  These may arise from many different possible sources, just one of which is damage to the physical media.  However, most of these errors are detected right off the bat since Red Book Audio CD’s contain parity information.  However, parity checking does not catch all errors. 

(If you don’t understand parity checking, think of it this way.  Take a number.  Add up its digits and then tack a 1 onto the end of your number if the sum is odd, and a 0 otherwise.  That 1 or 0 is a form of parity information.  Say that somebody else reads that number and notes that the last digit, the parity digit, is 1.  If they check the rest of the number and find it to be even they know there’s a problem.  However, there is still a high probability that they’ll read the number wrong and get something odd.) 

If you were to read the CD in one continuous stream, you would almost certainly wind up with several wrong numbers (i.e. samples) that went undetected.  Only some of those errors will produce obvious clicks and pops.  This is why a good ripping program will read the same “block” repeatedly to ensure that errors do not go undetected, and why simply listening to a track that was not properly ripped will not ensure that it is free of errors.  (i.e. Some errors may be rather subtle)

There are a lot of sweeping generalizations in this article, and I concur with a lot of what has been said by previous posts - quality of formats is only good on an a/b but in addition, there must be considerations given to the needs of the target device - a 30GB ipod is going to fill up quickly with lossless files and may not benefit the user to any real degree.
In saying that the idea that the Custom installer needs some additional advice on formats, compatibility and the ‘real’ issues and ‘real’ comparisons of formats is important. It is therefore vital that articles such as this accuratley protray the needs of the market and can address questions that a CI will get at all levels.

As a company who has been producing a premium quality auto ripping product range for many years , i also dispute the need for a ‘Million dollar cd ripping machine’. There is no such thing, nor is there a need for one. PC’s were made for ripping and the way in which cd-roms read discs for ripping purposes can result in a superior digital copy on a good digital music player vs a CD player. If a solution (automated) is bought from the right partner, software and hardware should work seamlessly, with a low investment ($2000 - $7000) and not a million dollar solution.

Select your drives and your software correctly and you can have a fantastic quality of rip. Service providers make a good job at their business because they have been through this process and even more so have an attention to detail for quality and metadata.

Metadata does not need a team of musical experts to sort, but it does need a company who uses metadata from a source that allows edits. take GD3 for example - a ripping company using this can make edits and therefore make the end result better for all - even the casual ripper.

A sweeping statement such as ‘plextors are best’ is inaccurate at best. Yes, plextor make great drives, but not all equally. Pioneers, NEC’s Sonys and LiteOns all have models that make a superb job of extracting digital audio.

Lastly, it is worth noting that manufacturers should be working harder on caring for cross platform support and education to dealers and consumers on the best way to enjoy digital audio.

I thank the author for writing this because it will help me sell more digital music servers.  My clients are always looking for “the best” and I am not going to overwhelm them on a discussion of parity checks and GD3.  I need to have real reasons for selling them expensive hardware, NAS drives, lossless formats, and charging them for processing their CDs, since god knows I don’t want to process them. CE Pro and this article has given me those reason: format and CODEC are important, processing hardware and playback hardware matter, clean information matters, etc.  Making sure it all works with the hardware I install is more important then the ins and outs of Apple DRM.
Bob

A very simple little web page can be found here:

http://www.ripfactory.com/quality.htm

This is a simple and basic calculator which will let an installer pop in the size of hard drive on the unit and will tell you the highest quality format you can rip in for a given number of cds. Of course, you will need to ensure that the media server allows that codec, but it provides a really handy quick reference tool.

Jim, my practical experience is when trying to get Apple songs purchased on iTunes to play on my client’s media center, I have to burn them to a CD and then re-rip them.  I think the the article makes it pretty obvious “garbage in, garbage out” that if you start with a lossy format, expand it, and rerip it, the end product is even worse (plus the album art and metadata get messed up). Try it yourself and you will see that the “facts” of this article are on the money.
I just can’t afford to load bad quality songs on high end equipment, so I tell my clients: give me your CDs, or buy new CDs, and I send them off to get clean lossless digital versions, and everyone is happy.

If I had a $100,000 system I’d put some money into a quality CD player - like a Sony CDP-X77ES.  It’s vintage but they’ll never make ‘em like that again.

If I had a $100,000 system I’d put some money into buying CDs.  A lot of money.  CDs may be the last “lossless” DRM-free format adopted by the music companies.  Buy them while you still can.

Listen and study.  Not all CDs are mastered equally.  With a little work one can find better quality CD masterings and improve sound much more than a room full of tweaks.  Quality in—Quality out.

Have fun!  This all is supposed to be fun.

My concern with this is article was that, in one important area, it missed the “big picture”, as follows: 

I believe a naive reader would be given the impression, from reading this article, that CD-quality represents high fidelity.  But it doesn’t.  In my experience, there is a striking difference in quality between CDs and SACDs; the latter simply sound more like real music.  [The 24 bit DVD-audio standard, currently available in Blu-Ray through either DTS-HD Master Audio or Dolby True HD, likely provides a similar increase in fidelity over CD.] 

And to me, this is the interesting question: not CD vs. MP3 (which, to my mind is a debate over two inferior formats), but rather the question of how much fidelity is achievable through digital audio, and how best to obtain it.

DVD-Audio and SACD are dead on the vine.  It would have been spectacular had either format suceeded.  I wasn’t to be.  For now both formats remain for demo purposes.  The market wasn’t ready and MP3 and other compressions zoomed passed it faster than an F-16.  It isn’t about “how good” anymore it’s about “how much”—whether it be price, quantity or size.  Whatever gets bragging rights.