redump.info Forum

OK, after no one replied to my question here:

/viewtopic.php?p=73948#p73948

I decided to research myself and found this article:

https://www.cdrinfo.com/Sections/Review … cleId=5095

where they praised "AOpen CRW2440A" as most capable to read such discs from long list of drives they tested including some very reputable brands like Yamaha, Plextor, etc.

In fact they achieved to get the exact same Hash ("Test CRC") on all, but only 2 tracks on the disc.

So, I have the same disc they tested and decided to try and reproduce their test - I was lucky and I found and purchased "AOpen CRW2440A" for only a buck (I must say it's in perfect, really like new condition, the original warranty seal wasn't broken and when I opened it - there was no dust inside or signs of use. I think it was hardly used if at all).

Long story short - I cannot reproduce their results with the same drive and the same disc and the exact same settings of which they included screenshot in the article - Every time I get different Hash for each track and not just that, but I've never get any of the Hashes ("Test CRC") they got and published in that article!

Later on, I found newer article by the same site, where they tested NEC ND-4550A drive:

https://www.cdrinfo.com/d7/content/nec-nd-4550a?page=4

in that article they claim:

CDS200
Ripping process completed, EAC reports no problems, Read&Test CRC comparison successful for all tracks

I remembered I had NEC ND-4550A in my storage, I found it and tested - it's the worst of all drives on Cactus I've tested so far (which are Plextor PX-4824TA, Asus BW-16D1HT, Samsung SH-D162C, LG GDR-8164B and of course the aforementioned AOpen CRW2440A). It not just cannot get the same TestCRC twice, but it reads such disc slower and with more difficulties than any of my other drives. One more time I get nothing close to their test results.

I don't know how that's possible, but at this point, I don't trust any review on that "cdrinfo" website.... It cost me nothing, practically nothing, as I had the same disc they used for testing, I had NEC ND-4550A and I found and bought AOpen CRW2440A for a buck, but the main point is that their test results just cannot be reproduced and how they got them at the first place I really don't know, because I am using 2 of their exact setups and getting nothing even remotely close. Anyway, I just wanted to share in case someone else is mislead by those reviews...

Still, I have no any solution what to do exactly with those discs or just trash them as garbage.

[EDIT] just to add that all of the above was with the earliest version of the protection with discs having newer revisions of the same protection is many many times worse.

Hmm, I wonder is it disc aging and that 15 years ago the tests and results from the reviews in my previous post were actually valid?

I believe that's valid hypothesis, because I found disc that is protected with one of the latest Cactus 200 revisions of the protection "5.10.090" and thus it's probably the newest I have (i.e. 2-3 years less aging than the other discs) - that disc even it's stronger protection, my Plextor PX-5224 can read consistently (always matching "Test CRC" Hash) on about 50% of the Tracks (other 50% of that tracks PX-5224 cannot read).

Even more interesting and important finding:

I have that same disc later release without Cactus protection. Of course, that unprotected disc can be read by any drive I have without any problems. However, the important finding is those same Tracks from the Cactus Protected disc that my Plextor PX-5224 can read property have the same Hash ("Test CRC") as the Tracks on the Unprotected disc.

So, that means (actually proves) on those tracks that Plextor PX-5224 can read consistently it really beats the protection and there is no any loss of quality or something (i.e. errors in the data). I wonder if that disc wasn't 15 years old, maybe Plextor PX-5224 can fully read it. Very interesting... at least it's some progress, but I've never read any confirmation that if you can read Cactus disc the data you get are guaranteed 100% good - that lucky coincidence of that disc I found having both protected and unprotected release confirmed that.

Wow, it's crazy, when the conditions are extreme, the difference between drives becomes so huge and really night and day - on the disc from the post above here are how 3 drives do:

* Plextor PX-5224 : 8 out 16 tracks, 100% properly read (and that 100% for sure, because they have the same Hash as those track from later unprotected release of the disc). The most amazing think is that those 8 tracks it can properly read it's really very fast like 20X speed. The other 8 tracks it can read, but with errors

* Plextor PX-4824 : 5 out 16 tracks, 100% properly read, but contrary to Plextor PX-5224 it cannot read anything from the other 11 tracks. Those 5 tracks it can read properly the speed is like 10 times less than  Plextor PX-5224, i.e. maybe average speed of 2X

* Asus BW-16D1HT : complete failure, it reads with 0.3X speed, really straggles, cannot read any track properly.

I really think 15 years ago with those discs without aging and in top shape, Plextor PX-5224 will read 100% of the tracks - the fact that it do so well on disc with latest strongest revision of the protection, but at the same time that disc is less older, i.e. less aging.

OK, a little progress... I found some more drivers in my storage, collected over the years. So, I further tested:

* Pioneer A03S : completely fails to read the Cactus protected disc,  i.e. recognizes it as data instead of audio disc

* Optiarc AD-7850S : cannot overcome the protection, struggles and goes down to 0.1X speed and generates errors

* LG GSA-H30N : cannot overcome the protection, struggles and goes down to 0.5X speed and generates errors

* NEC ND-1100A : similar to Pioneer, see above

and the Big Surprise:

* NEC ND-3530A: It was able to read 93% of the disc properly, beats in that regards PX-W5224, it failed to read only the last Track (probably due to disc aging). Amazing result and completely unexpected!

It's also the first time I get result consistent with the review here:

https://www.cdrinfo.com/d7/content/nec-nd-3530a?page=6

where they say ND-3530A is the first NEC model that is able to deal Cactus, because the previous ND-3520A and ND-3500A fail to overcome the protection.

Now, the most interesting thing is that my NEC ND-4550A drive (see my first post) is really the worst drive I have on Cactus and I get nothing close to the review here:

https://www.cdrinfo.com/d7/content/nec-nd-4550a?page=4

It's possible my NEC ND-4550A drive is faulty, i.e. the laser is worn out or something, but I really don't know. In any case ND-3530A demonstrates unbelievable performance!! I cannot believe it beats PX-5224 in regards to Cactus.

So,  ND-3530A cannot read the Lead-Out and even if it was able to break the protection for the last track (which it failed to do) still probably the end of the track won't be correct, but on my test disc PX-5224 can do the last track. However, since Cactus disc are multi-session and Plextor CD drives cannot really do the Lead-Out for multi-session discs, probably PX-5224 end of the track is not totally correct too.

In any way, currently for me ND-3530A is the way to go and try with PX-5224 only on the tracks on which ND-3530A fails.

Last, but not least, even I already explained in my previous posts, why I am sure I am doing proper testing, but to mentioned it again:

I have the same disc protected with "CDS 200.5.11.90" (according to ClonyXXL tool) and 1 year later reissued without a protection. So, the unprotected disc can be read with any drive and that's how I know the Correct Hash ("Test CRC") of each Track. So, if a drive is able to read the Cactus protected one, then the Hash of each Track matches the Hash of the unprotected one. That's how I am testing and know for sure how good a drive performance is.

[EDIT] just found another drive:

* LineOn SOHR-5239V: cannot overcome the protection, struggles and goes down to 0.2X speed and generates errors

[EDIT2] found 2 more drives:

* Teac CD-W54: it's exactly like Pioneer A03S - completely fails to read the Cactus protected disc,  i.e. recognizes it as data instead of audio disc

* LG GTC0N : cannot overcome the protection, struggles and generates errors

[EDIT3] found 1 more drive:

* Asus E612 : cannot overcome the protection, struggles and goes down to 0.8X speed and generates errors

[EDIT4] found 1 more drive:

* Toshiba SD-M1502 :  cannot overcome the protection, it struggles so much when reading such disc that it freezes the whole computer!

just to add few more drives I tested and are totally not working:

* Sony CRX230EE
* LiteOn DH52R2P
* Hitachi GD-7000
* Yamaha CDR200t

So, I don't keep count, but it feels that I already tested over 20 drives and only ND-3530A and Plextor PX-5224A can do the job, but together - none of them alone is able to extract all the tracks. However, ND-3530A is very close to be able to do all the tracks. Basically, ND-3530A is like over 90% of the disc and Plextor PX-5224A is like 50-60%.

BTW, mine ND-3530A has very strange firmware version 1.E0 and it refuses to be recognized by any of the ND-3530A firmware updates available online. I don't know if that is relevant, but my NEC ND-4550A is totally not working with Cactus (even some claim ND-4550A should work) and that ND-4550A is recognized by all available firmware updates available online. So, I don't know, but maybe my ND-3530A strange firmware is causing the big difference.

I wasn't familiar with this protection until I saw this thread, and I'm a bit intrigued now.

Have you tried ripping the audio tracks in CUERipper and then repairing the rip with CUETools? Assuming the disc is present the the CTDB that CUERipper / CUETools uses, you can correct your rip if it's "close enough" (i.e., there aren't too many audio samples that are corrupted). I've not dealt with this protection before, as I said, but I've been able to rip some really terrible condition discs that were close enough to be corrected with CUETools. If the disc doesn't return consistent results on re-reads, though, you'll probably find that CUERipper takes a long time. If the results are consistent but the drive reports C2 errors, I'm honestly not sure how CUERipper handles that. Might be interesting to find out?

It should also be possible to repair audio tracks ripped with DIC in CUETools IIRC, but it's not very straightforward. CUETools doesn't deal with raw CDDA, so you have to use something like ffmpeg to go from the raw audio to WAV or FLAC.

This doesn't really help in terms of finding a drive to image the data, though, and instead more relies on other people having been able to image the disc in the past. In any case, though, it's saved my ass many times when I had an audio disc with a defect or damage.

One other thought -- have you tried using Plextools Professional XL to rip the audio? It has some options like hiding the second session on the disc. I wonder if that allows reading the last track on one of the Plextor drives? I think the only reputable place to get Plextools Professional XL these days is via the Internet Archive's mirror of the Plextor files: https://web.archive.org/web/20101219101 … w.download

OK, update - I purchased (used from eBay, but again as mine in perfect shape - no any nasty scratches or any visible scratches, etc) another copy of my test disc protected with Cactus 200 revisions "5.10.090". So, lets call my own disc "TEST_DISC_MINE" and the one I've just purchased "TEST_DISC_USED". Here are the results:

* NEC ND-3530A:
---> TEST_DISC_MINE : Cannot read Track15
---> TEST_DISC_USED : Cannot read Track03

in both cases my NEC ND-3530A really fails with those tracks, i.e. hard fail, no matter how many times try to reread, etc.

in both cases though, I can get full content of the disc properly thanks to the (lucky) coincidence that my Plextor PX-5224 can do both Track03 and Track15 without any problem. In any case the performance of NEC ND-3530A is mind blowing to me, because it doesn't struggle at all - no noise, no desperate motor movement, etc and on 2 discs it's always do over 90% of the content.

Now, the biggest surprise - my Plextor PX-5224, which I purchased with the fault I described here:

/viewtopic.php?t=33976 … -to-close/

and I guess, because of that fault wasn't used for many many years, before I bought it, Initially it was really struggling with Cactus discs - very noise (like vacuum machine level), very slow at times, desperate motor movements back and forth and as result it was able to do only like 60% of the tracks on Cactus discs, it really looked like 2nd class citizen on those discs compared to my NEC ND-3530A.

However, yesterday, like magic, completely sudden, my Plextor PX-5224 starts working 100% on Cactus discs - the only explanation I have is that the laser "wake up" after it wasn't used for so long. I mentioned earlier in this thread that I believe Plextor PX-5224 problems with Cactus is due to disc aging and 15-20 years ago it was able to do such disc when they were fresh, but now it seems the laser of my Plextor PX-5224 needed to work some time to wake from all those years in the drawer.

So, both NEC ND-3530A and Plextor PX-5224 amazed me in their own way. Even now I can get 100% dumps two different ways:

* my preferred one:  NEC ND-3530A and use Plextor PX-5224 only on tracks that ND-3530A cannot do, as well as on the last track, because ND-3530A has no Lead-Out read and positive offset (see my comment at the end - it's possible Plextor PX-5224 is even more wrong on LeadOut, than no LeadOut at all with ND-3530A).

* use Plextor PX-5224 for the whole discs

Still those dumps are not DiscImageCreator approved, because all Cactus discs are Multi-Session and DiscImageCreator supports only Plextor DVD-drives for Multi-Session discs, because apparently on Plextor CD-drives, Lead-Out reading is not correct for Multi-Session discs.

@scsi_wuzzy

thanks for the suggestions!

what you're saying for Plextools - that option with hidden session is useful for another protection "Key2Audio", but that is perfectly handled by ExactAudioCopy too.

So, using things like CUERipper/CUETools is not an option, because the whole goal is to have 100% rip, not approximations.

In fact I believe I am the first one that really confirmed there are drives, that can no just read such Cactus protected discs, but read them 100% correctly, i.e. bypass the protection. That is due to very lucky coincidence, as I mentioned in the previous posts, that i happen to have the same CD protected with one of the latest (and thus hardest to break) revisions of the Cactus 200 protection and that same discs re-issued few years after without the protection. That allowed me to know the exact correct Hashes of the tracks, because the unprotected CD can be read perfectly by any drive. So, if when I read the protected CD the Hash doesn't match - then the drive cannot do it.

BTW, at least in the UK, almost all discs in 2001-2004 period were protected with Cactus...

Anyway, see the update I wrote in the post above.

matura713 wrote:@scsi_wuzzy
So, using things like CUERipper/CUETools is not an option, because the whole goal is to have 100% rip, not approximations.

I assume by 100% rip you're referring to having a "proper" DIC image of the disc as opposed to just audio tracks, and I definitely agree that it would be preferable to have a proper DIC image of the disc for archival. But, just in case I was unclear in what I said about CUETools, I want to say that, when CUETools corrects an audio disc, it makes the audio samples on the disc bit-for-bit identical to a existing rip of that disc. It's not like with record albums where there's software that just tries to find pops and clicks based on audio signature and mindlessly smooth them out. The database used by CUETools (CTDB) is, in part, a big database of data parity information that can be used to correct rips such that they match other known rips of the same disc. Essentially, you can go from a bad rip to a perfect one by correcting the bad audio samples in your rip to match an existing exact rip that someone else made of the disc. It doesn't simply clean up the audio to mask any audio artifacts that might be present -- it makes the audio samples match another known (presumably good) rip.

But, yes, it's definitely not as "archival" as getting a proper rip with DIC -- especially since it doesn't archive the data session. (Though CUERipper might record the existence of the data session in the CUE sheet -- I can't recall.) Fortunately, I don't think any of my audio discs have this protection. In fact, very few of my audio discs have any data tracks present at all, so I've mostly just been using CUERipper for audio discs (so I can use AccurateRip / CTDB to verify my rips) and DIC for anything that has data.

On another note, I kind of wish we had a parity database for the Redump rips. I have some discs I've acquired over the years that have a few bad sectors. I could have perfect rips of those discs with probably just a few KB of parity data...

scsi_wuzzy wrote:I assume by 100% rip you're referring to having a "proper" DIC image of the disc as opposed to just audio tracks..

no, by 100% rip, I mean one that has 100% correct hashes of the tracks and that is for sure.

So, maybe it was lost in my long posts here, but until now (at least I cannot find any such confirmation made in the past), there was no way to confirm that a track ripped from Cactus disc is proper, i.e. if a drive can read the disc that what you get is really proper data with protection bypassed - for example "AOpen CRW2440A" can read such discs, but Hashes of the Tracks are totally wrong - of course, you can tell that only if you know the real Hash. So, what made the breakthrough for me and allow me to test and confirm for sure was the following:

matura713 wrote:I have the same disc protected with "CDS 200.5.11.90" (according to ClonyXXL tool) and 1 year later reissued without a protection. So, the unprotected disc can be read with any drive and that's how I know the Correct Hash ("Test CRC") of each Track. So, if a drive is able to read the Cactus protected one, then the Hash of each Track matches the Hash of the unprotected one. That's how I am testing and know for sure how good a drive performance is.

So, build "5.11.90" is one of the latest of the Cactus protection (CDS200, aka Cactus Data Shield 200) and thus is one of the hardest to overcome. It's real luck that there is disc with it and the same one reissued without the protection - that gives the sole ability to test a drive and be sure that test is correct. Also, if a drive can read one of the latest "CDS 200.5.11.90" revisions (in fact the latest at least that I know), you can be sure it can do older version of the protection, as well.

As far as DIC image is concerned - all those discs are Multi-Session and DIC supports such Multi-Session discs only with Plextor DVD drives - I don't currently own working one to test it. So, the  only way to get DIC image is if Plextor DVD drives can handle the protection like PX-5224 can do it.

It was long journey for me, because initially my PX-5224 cannot do it (as I mentioned above, now I believe the laser needs some hours of work to awake up) and I don't know, if it's bad luck or it's that rare, but none of the tens of other drives I tested and listed here was working, except that lucky find of NEC ND-3530A, which I don't believe is an ordinary NEC ND-3530A, because firmware update for NEC ND-3530A doesn't recognize it due to its very strange 1.E0 firmware version.

Anyway, currently, the open question for me is - can Plextor DVD drives really overcome CDS200 protection, because that's the only case in which DIC image will be possible.