Surely there is a loss of sound quality when ripping?

[Gazzip]

Hi everybody,

This is my first post on here and I hope this question isn't too dumb. Apologies if it is.....

I was wondering how much quality is lost during the ripping process from CD to PC or Mac? I don't mean in respect of lossy vs lossless files etc. My query is more aimed at the transport mechanism in home computers. How can a £10 OEM PC/Mac CDROM possibly extract anywhere near the same amount of info and detail from the CD as a Hifi CD player does? It surely follows therefore that the resultant file is lacking much of the detail that was on the original CD?

Just doesn't make any sense to me so hopefully somebody can explain?

Thanks

Gazzip

Cyrus 8VS2; Cyrus 8SE; 2 PSX-R's; EB Acoustics EB+2's on Dreadnoughts; Chord Crimson; QED Silver Anniversary

 

[The_Lhc]

Ask yourself this, how could your pc install software from cd if it couldn't read every bit from the disk? Far from being inferior a pc is probably better at reading a cd than an audio cd player is. It doesn't need to do it in real time for a start, so can re-read the disk as much as it needs to, something an audio player doesn't have the option of doing.

 

[manicm]

Ask yourself this, how could your pc install software from cd if it couldn't read every bit from the disk? Far from being inferior a pc is probably better at reading a cd than an audio cd player is. It doesn't need to do it in real time for a start, so can re-read the disk as much as it needs to, something an audio player doesn't have the option of doing.

Data and audio shall are two completely different things. Firstly, in 99% of ripping, CDs are first ripped to WAV and then to FLAC - EAC does this. And WAV is not stored on a CD, even though they're both PCM based.

Which is why to my mind the 'bit-perfect' rip is complete bunkum, there is no such thing. There are excellent rips, sure, but no such thing as 'bit-perfect' - it's a damn myth. If a recording is done directly to hard disk in whatever digital format - then it's bit perfect. And also why to me ultimately things like AccurateRip are a red herring.

And, in ham-fisted irony, to answer the OP, I think RIPNAS at first boasted about their TEAC optical drive to read CDs.

 

[Anonymous]

How can a £10 OEM PC/Mac CDROM possibly extract anywhere near the same amount of info and detail from the CD as a Hifi CD player does?

Over 90% of the variance between CD players is because of the analog part. The "transport" part is usually a stock Sony or Philips design. Most SACD players use a bog-standard DVD unit as transport. And a CDROM drive doesn't do the analog part.

Which is why to my mind the 'bit-perfect' rip is complete bunkum, there is no such thing. There are excellent rips, sure, but no such thing as 'bit-perfect' - it's a damn myth.

A myth? Ok, then take this challenge: convert some of your favourite files to 16-bit wav. Fire up your favourite burner, and create a new audio disc from the files. Rip that disc, and compare the resulting files to the original. Do they match? Congratulations, you have just witnessed a miracle!

If a recording is done directly to hard disk in whatever digital format - then it's bit perfect. And also why to me ultimately things like AccurateRip are a red herring.

May I refer you to http://xiph.org/paranoia/faq.html#play? Although a bit dated, that should explain nicely the difficulty in extracting digital audio from an audio disc. But difficult and impossible are two different things.

 

[Anonymous]

Do not worry. Rips are just fine, try it, burn the rip back to cd and listen. Unless your cd player has a problem with readng CDRs they will be the same. Do a bit by bit comparison (must be programs for that) and they are the identical, unless there was a real problem with the rip. That can happen, expecially for scratched cds. That is where accuraterip comes in. If in the database 20 people produced exactly the same rip, bit by bit, than it is safe to assume that it is the rip as intended by the maker, or at least what the cd presser in the factory made of it.

In addition to lhc: cds are an ancient invention, in a time where it was difficult enough to get the info in realtime off the cd. No error checking and rereading on errors was implemented. No losless compression because lack of computing power for the codec. And limited to 44.1k samples of 16 bit values per channel.

Just listen to how good this old invention sounds, as well as the ripped versions. Don't worry be happy.

 

[Alec]

Ask yourself this, how could your pc install software from cd if it couldn't read every bit from the disk? Far from being inferior a pc is probably better at reading a cd than an audio cd player is. It doesn't need to do it in real time for a start, so can re-read the disk as much as it needs to, something an audio player doesn't have the option of doing.

Data and audio shall are two completely different things. Firstly, in 99% of ripping, CDs are first ripped to WAV and then to FLAC - EAC does this. And WAV is not stored on a CD, even though they're both PCM based.

Which is why to my mind the 'bit-perfect' rip is complete bunkum, there is no such thing. There are excellent rips, sure, but no such thing as 'bit-perfect' - it's a damn myth. If a recording is done directly to hard disk in whatever digital format - then it's bit perfect. And also why to me ultimately things like AccurateRip are a red herring.

I'm completely lost, not least by your point about data and audio, but by the rest too.

 

[amcluesent]

>There are excellent rips, sure, but no such thing as 'bit-perfect' - it's a damn myth<

Try reversing your SATA cables. When I did this, a veil was lifted, even the wife came in from the kitchen and asked what had happened to the hi-fi

 

[hammill]

I recently ripped about a dozen CDs using dbpoweramp. I had to re-rip one when a problem was found with the accuraterip comparison. On this small sample, I deduce that accuraterip is not a red herring but that most rips are fine using the cheap drives provided with PCs and accuraterip will point out when they are not.

 

[Anonymous]

For all intents and purposes, the quality should be pretty much the same.

Get listening to some decent 24/96 and 24/192 source music and you will soon realise that the quality of CD based material is so inferior that even if a couple of bits of data differ slightly on your rips (which you would never actually be able to hear) it does not matter one bit (no pun intended).

Currently listening to Metallica's Black Album in glorious 24bit/96khz. CD has had its day.

 

[Gazzip]

Thanks for the answers and opinions guys. An intreresting debate indeed.

I'm not sure however that there is a direct comparison to be made between the way that a PC or Mac reads audio CD's and CD-Roms.

When data/software is recorded on to a CD-Rom it has a series of readable error correction levels burned on to it at the same time as the data. This is precisely because every single bit from the disc needs to be read in order for the data it contains to be accessible or for the software to install correctly. This technology is so sophisticated that it can entirely correct errors caused by scratched to the disc's surface up to a couple of millimetres in length. Clever stuff!

However, an audio CD is not recorded with the same level of error correction. Although there is a degree of correction built in to Audio CD recordings, they rely far more heavily on error concealment. This a process whereby a CD player's or CD-Rom's transport interpolates where information is missing and replaces it. In other words it has a best guess at substituting the missing info with something that it considers will "sound" right.

With the above in mind I am convinced that a PC's CD-Rom transport cannot possibly be doing an equivalent job to a high end Hifi CD player's transport. If it is doing the same job as a Hifi's transport then the whole CD player industry for two decades or more has been built around the lie that the more you pay for the transport element of your CD player the better the information that will be sent to the DAC.

That conclusion scares me and I simply don't buy it.....

 

[altruistic.lemon]

Gazzip, tremon is right. Philips and Sony supplied most of the transports for CD players when in their heyday, whether expensive or not, and they certainly didn't have a vast range of transports. The difference was how the different manufacturers handled the rest of the stuff in the CD player.

Could be wrong, but isn't error interpolation part of error correction anyway? Could be wrong, I'm not a computer whiz, just a humble kitchen hand.

 

[hammill]

Thanks for the answers and opinions guys. An intreresting debate indeed.

I'm not sure however that there is a direct comparison to be made between the way that a PC or Mac reads audio CD's and CD-Roms.

When data/software is recorded on to a CD-Rom it has a series of readable error correction levels burned on to it at the same time as the data. This is precisely because every single bit from the disc needs to be read in order for the data it contains to be accessible or for the software to install correctly. This technology is so sophisticated that it can entirely correct errors caused by scratched to the disc's surface up to a couple of millimetres in length. Clever stuff!

However, an audio CD is not recorded with the same level of error correction. Although there is a degree of correction built in to Audio CD recordings, they rely far more heavily on error concealment. This a process whereby a CD player's or CD-Rom's transport interpolates where information is missing and replaces it. In other words it has a best guess at substituting the missing info with something that it considers will "sound" right.

With the above in mind I am convinced that a PC's CD-Rom transport cannot possibly be doing an equivalent job to a high end Hifi CD player's transport. If it is doing the same job as a Hifi's transport then the whole CD player industry for two decades or more has been built around the lie that the more you pay for the transport element of your CD player the better the information that will be sent to the DAC.

That conclusion scares me and I simply don't buy it.....

You seem to have missed two points.

When ripping, the data can be read multiple times - it does not have to provide the data in real time and does not need to necessarily meet the same standards as a HiFi CD player (although one should not forget a lot of expensive players use cheap transports from the mass manafacturers anyway).

Accurate rip technologies show that people all over the world manage to rip the same information from different copies of a CD using different (but cheap) hardware. Either they are all getting it wrong in exactly the same way or the cheap transports are perfectly adequate (PS if you chose answer one, may I get you to buy shares in my perpetual motion machine company).

Let us not forget that one can buy a Blu-Ray player from Sony for about £100 which produces excellent sound and picture, reads CDs and SACDs and does other fancy things. How much do you think the transport on that costs?

 

[Gazzip]

altruristic.lemon,

Interpolation and correction are part of the same process but are different beasts. One involves the harvesting of duplicate data from the multiple levels of a CD burn in order to replace, verbatim, what is missing. The other is a best guess at the data that is missing.

You are partly right when you say the difference was how the alternative manufacturers handled the rest of the stuff in the CD player.

Where I disagree with you slightly is that alot of that "other Stuff" is dealt with by the transport element of the CD player. If you consider a Hifi CD player in its constituent part you get a DAC and a Transport.

The DAC only converts what is sent to it by the Transport, and it is within the Transport that all of the interpolation and correction is done.

 

[AnotherJoe]

As far as your computer is concerned the audio stored on a cd is just data.

It is made up of a series of 16 bit (signed) two's complement which is just binary with values

-32768,16384,8192,.......4,2,1

(ie 1000000000000011) would give you -32765.

Eg. You are taking samples every 1/44100 of a second, and the audio in that instance is encoded to LPCM giving a value of -32765 (note this is not 100% accurate to the source - due to the nature of encoding to LPCM). This binary version is then written to the cd with/without error correction/encoding as required.

So for stereo cd, u have 2 channels X 44,100 of these 16 bit twos complement numbers a second.

When you read these binary values off the disc - the values you get are 100% the same as those burnt to the cd assuming no media defects,scratches etc. The encoding that was added can help recover from small errors as long as they are below the correction threshold.

The term lossless only applies to comparing what is on the cd and the resultant that is read off the disc - not the original source.

 

[Gazzip]

I'm not saying anybody is getting anything wrong so sorry if that is how I am coming across.

I know I sound just like my dad did twenty odd years ago, harping on about LP quality being unsurpassable when I got my first CD player.

I am however genuinely very interested in this because the way forwards for Hifi is obviously to use computers for data storage and serving. I'm just a little reluctant to take that step without knowing all of the facts. Spending 100's of hours ripping CD's to my PC aint my idea of a hobby so I'd like to get it right first time.

I am however a little surprised that a high end Hifi manufacturer is not pandering to ludites like me and producing a CD-Rom transport to do this job in lieu of an OEM PC CD-Rom.

Unless they have and I have missed it?

 

[Anonymous]

I havent found one. I went down the Olive HD route and soon realised that there is a whole new bandwagon out there based around snake oil ideas about digital music. I don't have the Olive anymore. The CD drive bust for a start and editing the tags on the files is a nightmare. Coupled with the backup files it produces which are inaccesible I went home made.

Use my laptop and DBpoweramp to rip the discs, MP3tag to edit any wayward tags, shove it on a NAS drive and feed it through a Sonos to a Peachtree amp. 100% digital path.

Consider the latest such exotic CD ripper in this months issue, the Meridian Sooloos Control, what you get for £4750 is oustandigly very little. A CD drive, some ripping software, a palty 500GB hard drive a tough screen LCD and a fancy aluminium case. You can put the same togather yourself for a tenth of that cost. Buyers beware as the Hi Fi industry has just realised a new cash cow out there. Once it was (and still is) cables, then Hi Fi Racks and now digital music.

I have tried 24/192 stuff, on the Olive and yes it did sound good BUT the sound you hear is as much a function of the care the recording engineers in making the original recordings. You can have a 16bit CD sounding every bit as good. Try Wagners Ring from the 1950's, Solti recordings, unbeatable.

By the way I rip everything to Flac, then convert them also to MP3 for my good ladys Galaxy phone.

Have fun ripping!

 

[iMark]

I have ripped our entire CD collection to ALAC using iTunes with error correction on. Playback is through iTunes->Airport Express->DACmagic. Our music sounds better than any of our CD or DVD players. Our old DVD player does sound better through the DACmagic, but not as good as the ALAC files from iTunes.

I think the clue is the error correction. When you rip a CD with error correction you get a very good file. The processor in the computer only has to stream the content. When you use a standalone player the music has to processed on the fly in the machine.

 

[Craig M.]

I am however a little surprised that a high end Hifi manufacturer is not pandering to ludites like me and producing a CD-Rom transport to do this job in lieu of an OEM PC CD-Rom.

Unless they have and I have missed it?

i really can't see the point of trying to find something 'better' then a oem cd-rom drive for ripping, the standard drives have no problem getting a bit perfect rip from a cd. as has already been said, it's easy to test for yourself. also, as others have stated, the vast majority of hifi cdplayers use a pretty cheap and cheerful transport anyway. i've compared several cdp's as transports against my mac mini and macbook, from an old pioneer stable platter to a cyrus cdxt-se, there is no audible difference. i seem to remember that rega use a standard pc cd-rom transport for their apollo and saturn players.

 

[eggontoast]

I was wondering how much quality is lost during the ripping process from CD to PC or Mac?

None. If you use the correct software (something like EAC) you will get a perfect rip of the cd without any errors, providing there is no damage to the disc of course. It might be argued that the rip could be better than playback from a cd player for reasons already mentioned in this thread.

I recently re-ripped my entire cd library and it is quite an eye opener to how little damage a cd can have to throw up errors, they are certainly not as robust as marketed back in the 80's. It is also surprising how many cd's don't play at all and have suffered from disc rot, I believe a majority of these discs were manufactured in one plant in the UK over a finite period.

 

[Gazzip]

So I guess that high end CD player transports are very, very good at extracting maximum info on the fly, whereas computer CD-Roms can take their time and "re-visit" parts of the CD to ensure that all available info was extracted?

This now makes sense to me. Thankyou all. I am now seeing the light.

One more question: Does anybody use I-Tunes to rip their CD's losslessly or is that a no no? I only ask 'cos that is the easiest way for me to do it as I have it installed on my MAC and some of the programs mentioned on this thread don't seem to have a Mac version available.....

 

[altruistic.lemon]

I do, mate. It's a piece of p*ss, meaning as easy as anything. ALAC on my hard drive and AAC on the phone.

 

[eggontoast]

So I guess that high end CD player transports are very, very good at extracting maximum info on the fly, whereas computer CD-Roms can take their time and "re-visit" parts of the CD to ensure that all available info was extracted?

Not really, but close enough

One more question: Does anybody use I-Tunes to rip their CD's losslessly or is that a no no? I only ask 'cos that is the easiest way for me to do it as I have it installed on my MAC and some of the programs mentioned on this thread don't seem to have a Mac version available.....

If you are going to use apple lossless files (ALAC) then you would be best staying with iTunes. If you were going down the FLAC route and being a mac user I doubt you will, dBpoweramp or EAC would be preferable.

 

[Anonymous]

Interpolation and correction are part of the same process but are different beasts. One involves the harvesting of duplicate data from the multiple levels of a CD burn in order to replace, verbatim, what is missing. The other is a best guess at the data that is missing.

Well let's at least call it an educated guess. But essentially, you are correct.

The DAC only converts what is sent to it by the Transport, and it is within the Transport that all of the interpolation and correction is done.

Sure, that's correct. But that's still missing the point hammill made: digital audio extraction does not have real-time requirements. When a read fails, the drive can simply re-read the same sector. It doesn't have to guess immediately.

So, IMHO, no, there is not surely a loss of sound quality in ripping vs playing. Either the disc is flawless and a rip will unearth just as much data as a hifi transport can, or the disc contains errors in which case the rip will likely contain the same (masked) errors that a hifi player will produce. Either that. or the ripper will spend a few minutes reading the exact same sector over and over again, and produce a corrected sector from the re-reads -- in which case the ripper is possibly more accurate than a hifi transport can do.

That said, there is one other problem with rips: fluke errors are immediately set in stone. A transient error while playing is not a real problem, but hearing the same error again and again in the same song is a nuisance. I've had two such errors out of a few hundred (or so) rips. Then again, I've also been able to rescue a disc that was virtually unplayable in every cd player I've tried. That rip did take more than 12 hours to complete...

I do understand your concern though: I have used Plextor drives for all my rips, because reports at the time (ten years ago) suggested they were the most reliable manufacturer for DAE. My current ripping drive is still an LG cdrw (they used to source their drives from Plextor) -- but that's mainly because it rips discs almost twice as fast as my NEC dvd drive.

 

[manicm]

AccurateRip, for all intents and purposes, is a red herring I maintain. All it does is compare the checksums of rips of a recording worldwide, but as Naim concurs (go over to their forums), what scientific method proves it's necessarily correct? Zip, nada, zilch. It has no scientific basis.

I must stress again, CDs don't contain WAV files, so when some poster here said you can compare - how are you comparing? You cannot compare any rip to a CD - simple as that my friend.

 

[Anonymous]

So lets take a popular CD, one which as been ripped 1000's of times in accuraterip, of those 1000's 99% might rip it exactly the same (those other 1% get an error which AccurateRip reports), that is different condition discs, ripped on different drives all giving the same result, how can that not convince you?

You are stating that these 1000's of rips are somehow wrong (ripping using multiple programs: dBpoweramp, EAC, XLD to name a few), so how would they be all wrong? 100's of different makes of drives all reading the CD with the exact same error?