best brand of cd-drive for ripping

[Spoon]

I only tested the Premium 1 very quickly, it seemed average.

[Dude]

I only tested the Premium 1 very quickly, it seemed average.

Still my original question remains: how can you actually see how good a drive is?
Where do you look at when ripping to determine it's good, average, not good, ...?

thx

[Spoon]

Testing is not too easy, the best test uses the same CD which is just on the edge of recoverability and rip on multiple drives.

[pls1]

An update here.

Using dbpoweramp I've ripped about 500 classical CD's (which is 10% of my collection). These are from my personal collection and many are physically from the dawn of the CD era. The CDs have been ripped on new Plextor PX-760 or PX-755 drives with secure dbpoweramp settings.

Of these 500 CDs, about 50 had track(s) that would not rip for a total of about 100 tracks. Most CDs had visible scratches but some seem to be deteriorating from age. Doing a re-rip on my "best" PX-760 drive (where best was defined as highest number of C2 errors detected using Plextools Pro) I could get about 10% to rip on a second try.

Yesterday, I finished a dedicated IDE machine with a fresh XP install and PX-230A drives on the main IDE and an IDE/ATA card in th PCI bus with a second PX-230. These PX-230s are new surplus OEM.

The PX230 drives have been able to rip about 90% of the tracks that the PX-7xx could not. I re-ripped ten of the CD's that were successfully ripped on the PX-230 again on the "best" Plextor 760. Not one of them would successfully rip.

Phil

[bhoar]

Looks like the PX-230 is a good drive for damaged discs, then.

-brendan

[pls1]

yeah, looks like I don't need to buy an automated disc polishing machine.
Phil

[tourrilhes]

Hi,

My personal experience, with around 300 rips. I'm currently using Plextor PX-230A, a Samsung SH-S162L, a Lite-On 20A1P. I've tried quickly a Benq DW-1625, a Benq DW800A and a LG GCE-8160B.

I currently have not been able to fault the PX-230A, it just works great and its fast.

My opinion of the SH-S162L is that it is superb on scratches, maybe even better than the PX-230A, but it will give the occasional random undetected error on perfect CDs. Burst speed slower than the PX-230A, EAC secure speed about the same. It does not support lead-out, but that does not matter much as the offset is only 6.

My opinion of the 20A1P is that it's pretty good on scratches, just a notch below the other twos. But, like the SH-S162L, it it will give the occasional random undetected error on perfect CDs. Burst speed is faster than the SH-S162L, EAC secure mode twice slower. It does not support lead-out, but that does not matter much as the offset is only 6.

My opinion of the LG GCE-8160B is that it's not very good on scratches, and may give undetected errors. It was probably as fast as the SH-S162L. Newer LG may be different.

My opinion of the two Benq is that they are not very good with scratches (comparatively) and slow.

My personal opinion is that the key to accurate ripping is diversity. If the CD is not in the AccurateRip database, I rip it again on a drive from a different manufacturer and compare the CRCs (i.e. my personal AccurateRip).

Jean

[Spoon]

Yes verifying on a 2nd drive (different chipset - easy to spot if the drive has a different offset), is very good as secure rips are concerned.

The 230a does has a minor fault, that is, sometime on a pristine disc with no scratches it can struggle, now it could be the disc is not quite within the 'CD' manufacturig guidelines, ie has a manufacturing defect, where as other 'poorer' drives might rip the CD fine. The 230a could rip 99.5% fine then other 0.5% you might need a 2nd drive.

[Dude]

My opinion of the SH-S162L is that it is superb on scratches, maybe even better than the PX-230A, but it will give the occasional random undetected error on perfect CDs.

Can you explain what you mean by 'occasional random undetected error' on perfect CDs? How can there be an error if your CD is in perfect state?

[Dude]

The 230a could rip 99.5% fine then other 0.5% you might need a 2nd drive.

I'm still wondering how you can actually see the drive ripped a track fine? If dbpoweramp says 'secure' (when the cd is not in accuraterip database) can you assume that the drive ripped it securely? Because how trusty would this result be if the drive reports wrong C2 error pointers or is this not possible?

thx

[Dude]

I currently have not been able to fault the PX-230A, it just works great and its fast.

How do you know there are no errors? Suppose dbpoweramp says it's securely ripped but the drive just passed wrong data on scratches in some tracks even with C2 pointers enabled and drive's cache disabled?

[tourrilhes]

Can you explain what you mean by 'occasional random undetected error' on perfect CDs? How can there be an error if your CD is in perfect state?

Note that I personally use EAC, but I think most of what I experience would apply to some degree to dbpoweramp.

One less than 1% of the track I rip, EAC indicate a secure rip but the track either does not match with AccurateRip or with a rip on my other drive.

Why would there be such error ? Well, between the CD and the final rip, you have an awfull lot of software and hardware, those things are not perfect.

How do you know there are no errors? Suppose dbpoweramp says it's securely ripped but the drive just passed wrong data on scratches in some tracks even with C2 pointers enabled and drive's cache disabled?

The reason I know if there is an error or not is because I spent time to work on my workflow and validate it. I still can not be sure that my workflow will detect all errors, but my goal is to minimise the probability to a level where it does not matter.

As as say, the key is diversity. Basically, I trust nothing, and assume everything has a probability of error. But if different method which have no correlation aggrees, then you get better confidence. Let say method A has 1% error rate and method B has 1% error rate and are not correlated, if both agrees you are close to 0.01% error rate.

My workflow is as follows :

1) Rip in EAC secure mode with the SH-S162L or 20A1P.
2) If AccurateRip match with high confidence, done with it.
3) If a single track mismatch, re-rip that track (i.e. go to 1)
4) Test in EAC burst mode with the PX-230
5) If CRC of test matches the rip version, done with it.
6) If only last track mismatch, check if due to offset difference (i.e. difference is only in last samples).
7) If CRC mismatch, go back to 1, using the other drive.

Note that I have a very short stack of CDs which are problematic. Most of them because on one track my results agree with themselves but disagree with AcurateRip (i.e. they came out of my workflow at step 5).

Jean

[MusicTrax]

I have an older Plextor 230 that was getting flakey, and opted to try the PX-750A, because it was highly rated by CDR Labs. That site does some extensive tests on ripping audio files, data files, etc., and charts how the drives do in ripping from damaged discs. They stated that the 750 handled damaged discs "surprisingly well," coming in much better than comparable drives from LG, Asus and NEC. It also supported C2 and AccurateStream, which a lot of newer drives don't anymore. I think its performance is actually better than most of the Plextors that followed it, which appear to be more cheaply made.

So for the above reasons, I opted to go for several banks of PX-750A drives in two ripping stations I set up. So far, after going through a couple of thousand CDs, it's been pretty much flawless. My only complaint with it is that it tends to slow down quite a bit on damaged discs -- roughly about 5X or 8X -- but I still have yet to find a single file with any audible problems. I'm satisfied that it's doing a great job.

I think it helps that I have 8 different PX-750s, so I'm not beating the crap out of one drive over a long period of time. I was able to find a local supplier that had a dozen of them as NOS, and they were only about $50 each, which I think was a reasonable price. Once I finish my ripping project (2000 down, only 10,000 to go), I hope to be able to re-use the drives in other systems for DVD viewing and writing.

[pls1]

I opted to go for several banks of PX-750A drives in two ripping stations I set up. So far, after going through a couple of thousand CDs, it's been pretty much flawless.

What do you do about discs not in AccurateRip where it took more than one pass to get a clean rip or required frame re-ripping? How do you determine that you have a bit perfect rip. Do you just check "interpolate unrecoverable frames" and not care about getting a bit perfect rip as long as there are no noticeable sonic artifacts?


Similarly, as per tourrilhe's post above, I am getting a little less than 1/2% of my tracks that are:

1. NOT in Accuraterip,
2. Required multiple passes or frame re-ripping
3. The rips are marked as "Secure" by dbpoweramp.
4. Can re-rip them on a different model drive, again get a "Secure" rip
5. Get two DIFFERENT CRCs.

This is based on about 700 disc rips so far. I've viewed all these rips rips as "testing" so I do not have a formal Quality Control log or consistent "Secure" settings. Therefore, I'm not completely sure of the actual error rates. but yesterday I re-ripped 10 problematic tracks and definitely got "Secure" but different CRCs.

Since Spoon said it can take up to two months for a CD to get into Accuraterip, I can't use my own first rip in accurate rip to verify my rip using another drive. I'm now going try a statistical test using a variation on tourrilhe's work-flow without interpolation.

Anyones suggestions on this would be welcome.

First I'm going to re-rip my several hundred test CD's. I'll use banks of Plextor 760s as the first rip station with Secure settings of one Ultra passes, no interpolation, re-rip frames and mark track as error. This establishes my base CRC.

Every CD not in accuraterip gets ripped a second time on a second bank of Plextor 230s set at First Pass + 1 Ultra. If the CRCs match I'm done.

All CRC mismatches can be re-ripped on a different 760 or 230 at higher ultra-secure settings or I can use one of my other available drives including Plextor B900, PX755, PX-708, Premium or Samsung SH-W162 until I get a matching CRC. Then I'm done.

I'm only doing this because I have so many irreplaceable CDs, I want to know whether I have an exact file extraction. I have already found 1% have rotted in some way as to have unrecoverable tracks.

Phil

[EliC]

The PX230 drives have been able to rip about 90% of the tracks that the PX-7xx could not. I re-ripped ten of the CD's that were successfully ripped on the PX-230 again on the "best" Plextor 760. Not one of them would successfully rip.

Phil

I think you have to compare to the 708, 712, or 716 as the rest of the 7xx series is not made by plextor (there are some questions about the 750-760 though).