CD v. ripped CD files - inferior Sound quality

[Desk Jockey]

I think there is some confusion here.

 

I have assumed the term "ripped" in this thread is referring to copying a file from a CD (source) to another location (destination), probably a hard drive in this case but it could be another CD, and playing that file from the destination. The complaint is about sound differences between the original file and the copy with the copy having degraded sound quality.

 

My point is the degradation is not caused by the data file. A computer is agnostic when copying files like executables, source code or multi-media (compressed or uncompressed) and manages file transfers the same way. This means the source file (CD) and the destination file, or copied file, are the same. This can be confirmed with a DIFF operation between the files.

 

Given the data in the source and destination files are the same, the cause of sound degradation is coming from elsewhere in the data/signal path. The data path from the CD drive to computer output is not exactly the same as the data path from the hard drive to computer output. That being the case, it's more likely that differences in sound quality start in the different data paths.

 

Cheers,

 

Greg

 

[John]

Greg, yes there is some confusion. You really should research how audio CD works. There are no files on an audio CD at all. None. (Don't get confused with the cda files. They are not on the disc.) One really cannot compare what goes on when ripping an audio CD to what goes on when copying a file. If we were talking about file copies you would be correct. However, this topic is nothing to do with file copying.

 

- John.

 

[ziggyzack]

'This means the source file (CD) and the destination file, or copied file, are the same. This can be confirmed with a DIFF operation between the files.'

 

CD reading is unreliable. That's what you have to grasp. Hence when you rip the CD, copy the music data from it, or however you want to phrase it, and diff it with another rip taken from the CD, this may well reveal differences in the rips, proving ripping is unreliable. Specialist software uses a number of strategies to successfully work around this problem (i.e. generate rips that are self-consistent and which agree with the master files used in the manufacture of the CD).

 

[Desk Jockey]

Ziggyzack,

 

Earlier in the thread, the case was made, not by me, that CD reading is reliable due to error correction and multi-pass reads. The point was that computers do make exact copies. How else would you ever successfully load or update the OS in your computer?

 

John,

 

I just loaded Elvis Costello's "Secret, Profane & Sugarcane" CD into my computer. Finder shows there are 13 AIFF Audio Files on the CD. I don't think you saying that I cannot copy those files to my hardrive unless I play them so are you saying that playing CDs is different than copying CDs? If so we are in agreement.

 

Based on Dvavc's system, which is impressive, my assumption is the CDs being ripped and played are in very good condition. However, in the case of damaged CD's it is certainly possible to get an inexact copy. I don't believe that is the point of this thread.

 

The point of the thread, I believe, is why do ripped files sound different than original files? There are many reasons for this to happen but file differences are much less likely than drive interfaces, jitter and the quality of the components in the data path.

 

Cheers,

 

Greg

 

 

 

[John]

Greg, that's exactly what I am saying actually. You cannot 'copy' those 'files' to your hard-drive. Finder reads the tracklist and makes up the aiff file names as a convenience. (This is the same thing your cd player does btw). It does not actually read the lpcm stream. That's why finder seems to work so quickly at this stage. If you drag one of the files to your hard-drive it will 'rip' (not 'copy') the track from within the continuous lpcm data and will encode it on the fly saving the result as a real aiff file, thus completing the illusion. (That's why it takes so long. It is effectively playing the audio cd quickly. It is not copying file clusters and comparing checksums &c). Note, this is not a question of semantics. You have the idea that ripping and copying are the same operation and that audio cd's have files like data files on a cd rom. Not so. This is the crucial reason why rips can differ in quality depending on the transport mech, the alignment, the software used etc...

 

- John.

 

[ziggyzack]

'How else would you ever successfully load or update the OS in your computer?'

 

As I mentioned earlier, CD-ROM and Red Book are different standards.

 

[Desk Jockey]

It seems there is consensus on the exactness of copying yellow book data so that issue can be parked from this discussion. It is apparent after "studying" the issue that the accepted solution in CA is to use specialized software, apparently dbPoweramp much to the chagrin of us Mac users, to achieve exact copies of rebook data.

 

For purposes furthering this discussion on data errors related to ripping CDs, can anyone quantify the error rates in terms of an average, a mean, and an expected range of bit failures in the process of ripping rebook data? Since we are ultimately talking about ones and zeros compared to two data files, data should be readily available. More simply put, what is the magnitude of the problem that 3rd party software is solving?

 

Additionally, I am very interested in learning where file copy errors fall in the overall problem of sound quality of ripped files verses original CD files. I am intrigued that the the sound difference on ripped files focuses on the data and not the data path as if all mother boards, chipsets and buffer devices are created equal.

 

Please use "data" in place of "files" if helps alleviate semantical concerns.

 

Cheers,

 

Greg

 

[ziggyzack]

accuraterip report a track error rate of 2%. If there are 10 tracks per cd, that works out an album error rate of 18% (i.e. 82% of albums are error free).

 

[The Computer Audiophile]

Hi Greg - I recommend reading through some of the information on the dBpoweramp site. There are some good "papers" on this stuff.

 

I don't disagree with you about the other possible causes of sound quality differences.

 

[vortecjr]

I scanned this post quickly and did I notice where you stated what your output from the computer is! How are you getting the music out of the computer?

 

Jesus R

www.sonore.us

 

[vicdom]

Hi dvavc,

 

I totally agree with your posts

 

I have ripped a lot of CD in WAV ( no data compression) to my PC´s HD using more than 1 cd/dvd drive and the end result was always the same.

Playing the same track from the HD and directly from the original CD(with the same dc/dvd drive used for ripping) sound quite different.

Useless to say that the original CD sound much better than playing from the HD. The differences I always notice are the ones you've already described.

The tracks on the HD usually sound a bit harsher, some lack of low frequency presence and detail which leads to some loss of stereo dimension. Once you compare to the original you realize that listening to the copy is a much poor musical experience.

The D/A path is the same for both cases as I am always using the same sound card. The digital stream processed by the sound card seems to be the same in both cases, as the digital data coming out of the cd/dvd drive while playing the original CD has no special encryption for any copy protection.

 

I have ripped audio CD´s using EAC and some others but the end result is always the same.

I know that reading data from audio CD is not the same as reading from a CD-Rom and so it is prone to some errors. Usually those error should be heard as some kind of pops.

Instead, this sound differences look more like some kind of jitter effect that is present while playing the copy ( some kind of out of clock sync).

 

 

By the way the same think happens to me with the DVD sound track when I rip a DVD-video. And here we don´t have those error problems as with a CD-A, because data in a DVD-Video is stored as files like in a CD-Rom. Once you descramble the data with DVDRipper you must get the exact audio and video content.

 

I haven´t found any reasonable explanation for this audio differences. I searched on the web, posted in some other forums...

I suppose the only way is to get access to the data stream the sound card is processing in either cases and try to figure out what is going on.

 

I live in Portugal, so sorry for my English.

 

 

 

 

 

 

 

[waas]

Hello. Ive tried a lot of different configurations with software and computers...I have found ripped music to be slightly inferior to the original cd as well. I'm using a Benchmark Dac 1 USB, Toshiba Laptop with JRiver and a Mac Mini with Pure Music. Then, I tried playing the original cd on a budget Pioneer dvd player, routing the spdif out to the Benchmark. Slightly smoother, fuller sound once again.

I'm quite puzzled by this, because I seem to be doing all the right things here...( using db poweramp for ripping, tweaking the o/s etc.)

 

[ludionisio]

I was directed to this thread from the other "Analog and Digital Comparison" thread and even though this seems to be already dead I have a question: Does burning an audio CD from the ripped problematic copy recover the details lost?

 

[PeterSt]

The best question from these whole two threads IMO.

 

It should ...

 

(sometimes things are just too simple to come up with - ok, by me at least)

 

[seta]

Kent Poon, a top mastering engineer out here in Asia and a consultant to Weiss engineering, did a test.

 

http://www.designwsound.com/dwsblog/hifi-computer-faq/cas-5-cd-ripping-for-mac-itunes/

 

Very good read.

 

 

[barrows]

"I'm using a Benchmark Dac 1 USB"

 

That is your "problem". You are not making a fair comparison as you are comparing the SPDIF input vs the USB input on your Benchmark. The rip is not responsible for the difference you are hearing. The Benchmark USB input uses the Centrance adaptive USB code running on a TAS 1020 chip-this is not equal in performance to a good SPDIF source. To make a good comparison use an asynchronous USB to SPDIF converter (like the Wavelink) then you are comparing both sources via the SPDIF input on the Benchmark.

 

[blueixus]

There was a time when many Audiophiles believed that ripped CDs burnt back again sounded better than the original CD. It was all the rage to use a Yamaha CDR burner/hard disc player.

 

I remember someone pointing that the reburnt CD's gave the average CD player less error correction to deal with.

 

Anyway this was all before 'Computer Audiophilia' existed, but the point it is usually the path out of the computer that is the issue not the file.

 

I believe the only really satisfactory way of working out if properly 'ripped' files are inferior is by reburning as mentioned by ludionisio, beyond that most comparisons will be even more apples and pears.

 

And what happens if the reburnt files sound better as many considered to be the case?