Ripping CDs....and quality CD drives... (a theoretical question)

There is now a special Melco product for ripping cd’s and experts call it a great improvement, so I guess that it will make an audible difference. However I will not re-rip my 4000 to 6000 cd’s even if it would bring an afvantage. My new music is for 95% download anyhow.

Melco usd to advocate a cheap Buffalo ripper, their new one is £800 which is pretty expensive. Naim use a relatively cheap Teac mechanism in the Core.....so I am a little confused.

On the Naim website you can find white-papers and under HDX there is one called Naim CD Ripping Engine. I dont know if it is applicable to Uniti Core but still interesting read and they describe the special requirements on the drives they use.

It is very complicated … CD's are sort of "digital"  but there is a lot of "analogue" going on in the reading process too … all that laser firing and reflection detection and alignment and conversion from laser return measurement to current to bits, etc. etc.

I remember complaining to my dealer that my spanking new Naim CD5X would not play CD's that happily played on my old cheapo gear and he explained that the CD standard has a lot of error events/codes defined in it but that sheapo players simply ignore these while the Naim gear was more strict in enforcing any read errors detected. While I cannot say this is definitively true, it does seem reasonable to me and also explains how when ripping on my UnitiServ, it sometimes slows down because it is re-reading the CD at different speeds to attempt to get an error-free rip.

This would also imply that it is not so much the quality of the actual transport mechanics that are important but more the quality of the software that is doing the reading and how much attention it pays to any read errors.

The white paper mentioned is well worth a read.

Allan

But it did strike me that when one rips CDs, at least part of the quality of the resulting file must surely depend on the quality of the drive reading the disc. And from what I have read, the drives built into most machines designed to rip CDs, are often cheap(ish) computer drives

I may be wrong here, much more learned people to give advice. I seem to remember asking the same question as it seems logical. But was told if one is ripping using dbpoweramps, say, there are CD ripper correction settings and the final rip is compared with other rips - there is lot on the web about AccurateRip https://www.dbpoweramp.com/secure-ripper.htm.

So you you could (errr I think) rip the CD with any old 'cheap' machine provided it passes the setting tests and you get an AccurateRip match.   

Yes, agreed. If the AccurateRip checksum matches, and dbPoweramp reports a reasonably high number of identical checksums, then you have a perfect copy of the CD. Better quality drives such as those used in high quality audiophile equipment are probably better a 'real time' reading and error correction, but with a decent ripper, it can have many attempts at reading the same section of the disk.

I tried a few different drives from in Laptop, PC & a heavy old desktop,  no difference that I can hear.    If there is a difference I suspect it might be with the ripping software.    Error detection programs such as Accuraterip where your rip is compared to others on a www data base,  or secure mode ripping where your CD is ripped twice & each compared,  both seem to be very effective.   

I suspect certain optical drives have superior prowess reading/correcting data from bad/scratched discs compared to others.

If a disc rips fine with a cheap drive with a matching EAC checksum it's done as good as you'll get, and I find most discs rip fine.  The question is whether or not a better drive will have a chance of accurately reading any discs that fail.

I have a copy of a soundtrack I bought a few years ago to replace one which was stolen and I'd only ripped in MP3 previously.  I'd forgotten about the replacement disc and all tracks except the last ripped accurately - well outside the window to return a faulty disc with no visible defects to the naked eye.

I did have a few cds that looked perfect which would not rip on my Core. I sent them to Naim for evaluation, apparently the Core uses a slimline optical rather than full size optical drive and just could not read/correct the few errors on these discs. I should add this is 6 discs out of 1300 ripped. Naim read and copied them onto new cd media and shipped back to me, so I was happy. That was a Flynn at Naim who was mentioned the other day.

As others have said, one of the points about ripping is that you don't have to error correct in real time and you can check bit for bit accuracy by comparing your file to other rips online. It makes life easier for the drive.

I simply don't buy the idea that drive quality need influence the accuracy of rips.

Beyond that, just thing of bluray drives - these have to accurately read at a much higher data rate than CD - the technical challenge is stiffer. So if a bluray drive can read blurays it surely much be able to accurately read CDs.

And of course you don't need to pay for a Krell-priced transport to accurately read blurays.

Essentially, I think the magic is in the DAC (and clocking the feed to the DAC) not in reading the discs. 

I did read very recently in a post on this forum, from a well known studio engineer (he was being applauded on the same thread by other Naim forumites for the quality of his work) that he uses a number of studios (I think these were in the UK...) and that even from the same master copy, the CD-Rs that are produced by the different studios are always different sounding. I can't find the thread at the moment, but I do remember that he said that he tends to use a particular studio which took longer to produce the CD....I seem to recall a comparison of 4 and 9 (seconds? minutes?) and he preferred the '9' every time.... This might be the time taken to read the original, or perhaps the time taken to write or to finalise...I just can't remember. But it does often seem that using a slower copying time produces more 'accurate' results presumably because the error correction mechanism is not 'tripping over itself trying to keep up'.

Secondly, dbPoweramp comparing a rip of a CD with an online version....that does sound as if it could help the situation, but I'd hate to think that I spent a lot of money on equipment to rip CDs successfully, only to have the rip compared and perhaps amended by comparing it with potentially different data, with faults, from other lesser sources.

Fascinating, this.....

Thanks all for contributing....it makes interesting reading.

Please continue....!

Vintageaxeman,

Although you say you are not a 'ripper', you already are ripping your CDs into iTunes. iTunes may default to creating a compressed file which is probably not an issue for the type of listening you do with these particular ripped files.

As a number of other people have stated, dbPoweramp (along with Accuraterip) produces a perfect ripped copy of the original CD. It checks a database to determine whether or not it believes your rip has been perfect and informs you of the result. It does not amend the rip. It is basically flagged as either secure or not secure as the case may be.  

Hmack posted:

Vintageaxeman,

Although you say you are not a 'ripper', you already are ripping your CDs into iTunes. iTunes may default to creating a compressed file which is probably not an issue for the type of listening you do with these particular ripped files.

As a number of other people have stated, dbPoweramp (along with Accuraterip) produces a perfect ripped copy of the original CD. It checks a database to determine whether or not it believes your rip has been perfect and informs you of the result. It does not amend the rip. It is basically flagged as either secure or not secure as the case may be.  

Thank you Hmack!

When I said I 'am not a ripper', yeah, I realise I am to some extent by using iTunes. What I really meant is that I am not ready to rip 1100 CDs and then store them in the loft! (so...I am not a serial ripper...)

Thank you for the extra info about dbPoweramp. That's very useful. Cheers.

Interesting, when I bought my Melco N1a it was being demonstrated with a generic £20 external CD/DVD drive and I was clearly told it's not worth  spending more. So I spent about that in Argos and it lasted about 700 discs. I replaced it with a slightly more expensive own brand from another store, but it is clearly labelled Hitachi - LG.

There have been odd discs that will not copy but do so easily onto my Packard Bell netbook and an LG branded drive using the packaged media software.

I was told that as ripping involves interrogating rather than reading the disc(?), the drive is no so much consequence. Confusingly the Melco offers an option to play from the ripping drive and in my opinion, it sounds better than my £1000 CD player.

Gazza posted:

Melco usd to advocate a cheap Buffalo ripper, their new one is £800 which is pretty expensive. Naim use a relatively cheap Teac mechanism in the Core.....so I am a little confused.

Quite. The CD transport is a transport... it really matters not these days with low cost precision transports... the benefit of better CD players, streamers occurs when the data serilation occurs... and here the clocking of the data becomes important because lack of stability here will add distortion into the serial signal. If the was a difference from different transports when statically measured or assessed the files would not be equal and fail check sums which is clearly not the case..(excluding gaps between tracks where differences in terms of where to agree the break between tracks can vary)

Thanks Simon

Hmack posted:

Vintageaxeman,

Although you say you are not a 'ripper', you already are ripping your CDs into iTunes. iTunes may default to creating a compressed file which is probably not an issue for the type of listening you do with these particular ripped files.

As a number of other people have stated, dbPoweramp (along with Accuraterip) produces a perfect ripped copy of the original CD. It checks a database to determine whether or not it believes your rip has been perfect and informs you of the result. It does not amend the rip. It is basically flagged as either secure or not secure as the case may be.  

It would be interesting to compare, and far from impossible to do so, but I suspect that the vast majority of the times with a good condition CD, provided the PC/Mac's optical drive is not faulty iTunes will get a perfect rip, just you won't get the EAC checksum comparison to tell you.

Naturally iTunes may compress to AAC by default but that's easy to change.

I suspect you never hear of anyone primarily ripping with iTunes complain the their music sounds bad and to be honest apart from the EAC checksum providing some reassurance, unless bad discs caused stuttering/clicks in audio rips I doubt many of us could really tell the difference.

Alley Cat posted:

Hmack posted:

Vintageaxeman,

Although you say you are not a 'ripper', you already are ripping your CDs into iTunes. iTunes may default to creating a compressed file which is probably not an issue for the type of listening you do with these particular ripped files.

As a number of other people have stated, dbPoweramp (along with Accuraterip) produces a perfect ripped copy of the original CD. It checks a database to determine whether or not it believes your rip has been perfect and informs you of the result. It does not amend the rip. It is basically flagged as either secure or not secure as the case may be.  

It would be interesting to compare, and far from impossible to do so, but I suspect that the vast majority of the times with a good condition CD, provided the PC/Mac's optical drive is not faulty iTunes will get a perfect rip, just you won't get the EAC checksum comparison to tell you.

Naturally iTunes may compress to AAC by default but that's easy to change.

I suspect you never hear of anyone primarily ripping with iTunes complain the their music sounds bad and to be honest apart from the EAC checksum providing some reassurance, unless bad discs caused stuttering/clicks in audio rips I doubt many of us could really tell the difference.

Some audiophile will always find away. (subjectively of course)

It would be great if the drive would not make a difference as long as the accurate Rip checksum is perfect, but unfortunately in real world it can make a big difference.

This past weekend I have heard that at my dealer. The Rips done with the Melco Drive were much more natural, Piano was so much better. It was really astounding how big of a difference it made. And we did not need a big bucks system. We were listening with a Uniti Atom using Russell K Red 50 speakers.

Judge for yourself using your own ears. Then you can discuss how that may be possible.

In the real world the drive makes no difference if reading the disc with no errors. It’s the serialising of the data that can. The checksum confirms the data has been read correctly, but of course that says nothing of the clock stability of any serialisation.

if you go back a few years some of us did some analysis from different rippers and drives on this forum and compared files from different rippers without letting on which. The sample pcm  file data was identical, the WAV file formats  produced did vary however depending on ripper .. but the payload data was identical. iTunes rips sometimes had some offset differences based on cd drive used. This effected variation of a few mS at the start and end of tracks. The tracks played back identically on a given media server.. no audible difference we’re discernible. However different media servers sounded different when streaming over Ethernet... and the only visible difference was their in frame timing consistency at the network layer.

it was an interesting exercise and dispelled a few  myths at the time over differences between different cd transports and rip quality/bit perfectness. It’s good to trust your ears and correlate with evidence rather than base ideas on hunches, sales talk and pseudo science..as you can convince yourself of anything that way... just like if you concentrate hard enough you can see dragons in cloud shapes.

If you can achieve a bit perfect rip, and you know it's a bit perfect rip, then it's a bit perfect rip. End of. A decent streamer will play it back perfectly every time. No jitter, no interpolation, no error correction, no motors and servos spooling up and down and swinging about all over the place.  Rippers have access to reference databases where the results can be compared, bit for bit, and a bit perfect rip can be verified.

I don't doubt that that's what you heard in your system with the drives and software you were using. I only wanted to give a different experience and opinion about the Melco D100 dirve ripping CDs. The difference was very clearly audible for everybody listening. Somebody compared it to going from 16 bit to 24 bit when listening to Qubuz for example. The music was stored and played from the same NAS.

Everybody who is wondering about a possible difference should go and listen for themselves before wondering if it could even be possible. Or borrow the drive from their dealer and try it for themselves at home if they don't trust the dealer in this comparison.

I am not going to argue about what I heard and most of us or maybe even everybody here does not not have a deep eough understanding of the technical differences so that is mute in my opinion.

This whole discussion is the same everytime a new digital technique for music reproduction comes around the corner. When CD was released, there could not be a differnce between drives or players. When Streaming started, all network cables must sound the same. Now all Rips must sound the same as long as they are bit perfect.

Once you listed to some rips on a very high quality drive like the D100 compared to a cheapo PC drive, you know there are differnces and start wondering how that can be possible. Obbiously there's more to digital than just 0s and 1s. These discussions are almost everytime between pleople who have heard the difference and those who do not bother to try it for themselves because in their understanding of the world, there can not be a difernce and therefore there is no need to even try it.

That's too bad. Spending a ton of money on a high end system and not trying to get the best out of it by optimizing the ripping, storing and cabling aspect.

If you have achieved a bit perfect rip you won't improve on it. How you store it and play it back will produce variable results. As will playing a CD. What's actually in the pits of the CD doesn't vary, although how it sounds will vary considerably depending on how well the data is pulled and processed, and what it is played back through. This is why I prefer streaming. Pulling the data and verifying its accuracy is a one time process.  One less point in the chain to potentially produce errors.

meissmar posted:

Once you listed to some rips on a very high quality drive like the D100 compared to a cheapo PC drive, you know there are differnces and start wondering how that can be possible. Obbiously there's more to digital than just 0s and 1s. 

Back in the day when you purchased Microsoft word, it came on a CD / DVD, there was no "you get a better version of Word if use a more expensive CD drive to install it from", the data is either read 100% correct or not. Audio CDs are still digitally stored data CDs, they have less error detection built in than a computer data CD, however that is where AccurateRip comes into play, in verifying the rip against someone who has a different version of the same CD and a different CD drive.