I'm not entirely sure I understand your question. Perhaps my clarification wasn't as clear as I thought?

What this means is that it's possible to scan a DVD in a consumer drive, e.g. in a LiteOn drive, and have the scans show PIE <= 280 and PIF <= 4 which is within ECMA specifications, and that same disc could still be unreadable in some other drives that don't read as well as the scanning drive or react negatively to some property of the disc that the scanning drive doesn't have any problem with.

If that isn't the question you wanted me to answer, you will have to rephrase your question.

Yes, that is what I thought you meant, though I'm not too sure if any consumer drive is truely reporting digital error rates which can be said to be 100% conforming to ECMA values, in the same way a calibrated professional analyzer can be said to.

It cannot be over-emphasized that existing BenQ drives and certain LiteOn DVD writers give false-negative results in (under-report) PI/PO error rates. Such evidence can be found in the following thread for BenQ 1650/1655:

http://club.cdfreaks.com/showthread.php?t=196294

I have observed that BenQ 800, 1620, and 1640 behave similarly.

Some cases for LiteOn DVD writers is in:

http://club.cdfreaks.com/showthread.php?t=98837

IMO the discrepancies between results obtained with many consumer drives and the CATS devices are mainly, though not solely, caused by such under-reporting behaviors of the former.

Obviously you are interested in a reader as picky as possible, but you won’t be able to find any normal consumer drive to meet your demand. Only when a consumer drive is out of order may it become pickier than the CATS. However, there are still some drives not far from your expectation. Scanning at 8X my Lite-On 166S and 167T DVD-ROM drives and 5232K combo give PIE/PIF rates at least approaching what CATS device reports at 1X. Some similar information is in:

http://www.cdrinfo.com/Sections/Revi...10033&PageId=8

Plextor 712 behaves similarly:

http://www.cdrinfo.com/forum/tm.asp?m=68880


AFAIK, Lite-On 166S and 167T DVD-ROM drives, 5232K combo, and Plextor 712 do not give false-negative results.

Looking briefly at that thread, it looks as though you claim the Benq is under-reporting PIE/PIF error rates by comparing its scan to a failed TRT. Comparing the two tests is irrelevant if you are comparing tests at different speeds. It would only be reasonable to make a conclusion by comparing tests performed at equal speeds. Discs can sometimes test with significantly worse PIE/PIF at maximum speed, and conversely a disc that fails a TRT at maximum speed can often read back perfectly fine at a lowered speed. Drives that report PIE/PIF levels, likely ALL of them, can sometimes report acceptable PIE/PIF error levels when the disc is actually unreadable particularly at maximum reading speeds, it's just the nature of testing media and that it's not a perfect science.

For proof that BenQ DW1655 drives can have reading errors without showing POF at the same location have a look in this thread, especially post #60, #63 and #64.

That particular disc has unreadable blocks at 3.47GB in two different BenQ DW1655 drives and tested with 4 different firmwares (BCDB, BCGB, BCHB, BCIB). The disc is unreadable at that location every single time it has been tested in those drives regardless of reading speed, and the only scan that shows POF is a 16x scan but even 16x scanning doesn't reliably show any POF at that location, or for most scans not even PIF at that exact location.

This means that the BenQ DW1655 doesn't always report all problems that it's experiencing, so the Disc Quality Scan is effectively lying.

BTW most other drives have no problem with that particular disc.

I am not suggesting it doesn't happen and I know that it can and does happen on rare occasions (and as I said in my post it is likely that ANY drive is capable of these situations). My point was that drives test differently at different speeds so it's a poor comparison in the link I was referencing. Your comparisons are completely valid since they show failed TRTs and Scandisc(s) at the same speed as the 'acceptable' PIE/PIF test.

Eheheh...I'm interested to the best reader as possible but I'm also interested to a drive that give PIE/PIF test similar to CATS as much as possible. But I also know that any consumer drive is a better reader than CATS.
Interesting the thing about the combo because I have a 5236V. Is there a big difference in comparison to 5232K?
Thanks.

This is not necessarily true, as the cdfreaks article comparing CATS (pulstec) vs. Plextor/BenQ/Lite-ON shows it is a capable reader and is not just 'picky' for no good purpose.

http://www.cdfreaks.com/reviews/CATS...st-Disc-5.html
http://www.cdfreaks.com/reviews/CATS...st-Disc-6.html

I was wrong!
CATS seems a good reader and it's able to read, like LiteOn, some terrible media with very high jitter.
I'm a bit worried for the second test with a very good TYG01. All Home tests give perfect PI/PIF/Jitter results while CATS show a quality problem in the outer part of the disc.
Yes I know that problem doesn't affect the playability of the disc...but with other (worse) discs what happen?

Strictly speaking those links only show that the CATS device can *scan* those two discs - it doesn't show that it can *read* them, because the existence of POF in the last part of the CATS scans shows that the Pulstec drive couldn't read those sectors correctly on the first try, and it might not be able to read them correctly at all.

It's possible that the LiteOn 1633 drive cannot read those discs, but the evidence presented in those links doesn't prove that one way or the other; I would guess that Test-Disc-6 isn't readable in the LiteOn drive given that PIF reaches 209 which is higher than the theoretical limit of 208 (probably a scanning glitch).

Being able to scan and being able to read isn't quite the same thing, and a drive will usually produce reading errors before it produces scanning errors.
A Plextor Beta/Jitter test would be an exception to this rule of thumb, however, as it has been known to fail on readable discs.

Well worth repeating, again and again and again.

The big problem is, how many of us have access to professional scanning equipment?
I would guess less than 1% of us. Hence this article was put together.

Some background to this article which may interest some of you.
Back in January 2005 i was asked to participate in an article. I was pretty new on CDF and had recently started modifying NEC firmware (with help from Liggy).
The article was basically a shootout between my modified firmware and NEC's stock firmware. The article is here for anyone who is interested.
Anyway, i was saddled with a NEC ND-3520A drive which didn't burn very well. But i don't like to lose either.
I had recently bought a BenQ DW1620, basically because i thought i could use it for scanning the disc while developing the firmware. The problem was the BenQ drive didn't help, in fact quite the opposite. I was basically getting nonsense scans. I gave up on scanning on the BenQ and returned to scanning with a Lite-On and Plextor PX716.

After the article was finished i had many discussions with various people around the forum about BenQ scanning habits. It was at that time i had the idea of putting the "Home Scanning Article" together. During the rest of 2005 i basically ran 100's of tests on different scanning drives. It became clear to me that while the BenQ drive could scan the media it had burned quite well, it was in most parts unable to give meaningful results on media burned on other drives, where the burn speed was higher than 8x. To make matters more confusing, along came PI/PIF scanning for NEC drives. After a short honeymoon period of scanning discs with the NEC drives, it become apparent that the NEC drives were giving even less meaningful results.

Basically this article cannot be compared to CATS scans. The idea is, without to much expense ( Home ODD drives are very cheap) and without taking a lot of time. Home users would be able to run a PI/PIF scanning test and least have an idea what kind of results to expect if they were using one of the brands used in the article for scanning.

I'm still convinced that by running a Disc Quality Scan and backing this up with a Transfer Rate Test. The average home user will have a good idea of how their burning drive and media is performing.

Another thing that was important in the article. Each of these tests can be run on most CDF members own drives and they can compare their own results with the ones in the article. This is something that CDF thinks is very important. The whole idea of CDF is you have people sharing knowledge and we believe members must be able to participate and be able to run the same tests we do in our articles and reviews.

Their are a number of other ODD sites around which now use professional scanning equipment to test their review media. That in itself alienates most of their members, as they cannot in most cases run the same tests themselves to compare.

Amen

DrageMester:

Well in my experience with degraded/badly manufactured recordable media, there is correlation between ability to complete a scan & readability of the disc in the same drive. If the article had included TRT, we wouldn't need all this conjecture

And I, on the contrary, came to think that scans performed in my Benq 1640/1650 and NEC 3540A wer far more useful than Litey scans, in terms of predictability of real-world reading behaviour in several drives.

Actually, now, after hundreds of scans, I couldn't care less what drives would show more 'accurate' results. It's no use to me.

To compare different burns, any consistent scanner will do, whatever the actual numbers. And my Benq units are/were consistent (heavily tested for consitency!), contrary to what's been reported.. - and so are my two 3540 units. On really good discs (= real-world readability in a wide number of drives, including picky standalone players), even my 'looney' 4550 gives surprisingly consistent results.

Actually, I'm only experiencing something that has already been emphasised on this forum long ago, in the 'basics' of scanning: really good discs/burns will tend to show good PIE/PIF scans in most testing drives, while mediocre or poor burns will show wild variations among testing drives (or between several passes in the same drive). Except that in Liteys @4X or @8X, these variations will be much smaller than in Benq drives @8X. Can't say about Plextor units.

Then to predict real-world behaviour, I choose to give more credit to scans from drives that are magnifying error levels and variations, rather than drives (Liteys) that give OK PIE/PIF results with about anything except with downright awful burns that won't be read properly in many drives (and, notably, in older or cheap standalone recorders, in case of video discs).

Put differently: in my opinion, testing for PIE/PIF in drives that happen to be the best readers around is just like having a graduation test with the most forgiving examinator around. All students will do well except the most ignorant ones. Need I say that I prefer to have a very picky examinator if my goal is to select really good students? It sounds like a truism, though only a handful of CDFreaks express the same feeling!

Yes, being able to compare one's burns with other members burns is fine, fun, and is an asset to the forum. But if it ain't really useful to sort out really good burns from so-so ones, isn't it kind of "alienating" too...?

I strongly advocate for a change in paradigm concerning PIE/PIF scanning with Liteys. For al purposes except comparing burns, I dare say that only scans performed @12X have any significance in LiteOn drives.

Of course, I don't expect this change in paradigm to happen anytime soon... Aaaah, my friend, we can only agree on this one. - transfer rate tests are wrongly regarded as useless by too many members here.

Kind regards,

Franck

@Francksoy

Are You Saying When Using A Lite-On To Scan At 4x-8x Only Shows In Most Cases How Good The Lite-On Is At Scanning But Not Nessicery How Good Or Bad The Burn Is.

Ouch Am I Making Sense.

If that disc Mciahel tested is completely readable, most drives will slow down in TRT when encountering difficult sectors rather than abort the test, and even reread in read test. I am not aware of any disc with aborted TRT at high speed to finish that test at lower speed in repetitive scanning. Can you give such an example (please do each kind of test at least three times)? I myself have always conducted such comparisons at the same settings to avoid disputes.


Not every drive does so AFAIK.



It is well known that many drives tend to report higher error rates at higher speed, if that speed can be maintained. However, in actual reading of files, the drive will slow down and even reread the bad blocks. So if a drive reports acceptable PI/PO error levels when the disc is actually unreadable, then that drive under-reports PI/PO errors IMO.

Optical discs are based on hi-tech, and disk quality test is well within the scope of science; you need to know how the binary information is processed, which is very complicated, to understand that though. PI/PO errors vary in part with many parameters pertaining to the testing device, so we cannot say that the error rates measured at only a particular speed is correct, but it also does not follow that PI/PO errors given by each drive is correct.

My Sony 320AE, which is a rebadged Lite-On 5236, tends to give lower PIE/PIF counts than 5232K; I have not done many comparisons though.


Since the scanning speed with the CATS device is 1X, but is 2X or higher with others, it is not possible to draw conclusions about relative pickiness from those tests.

Very roughly, yes. - a little oversimplifying, though. Seems like you are... I'm not sure what you call "aborted" TRT here, but examples have been posted by DrageMester and myself of discs scanned in Benq drives @8X with nothing wrong showing in the scan, though a TRT fails (or shows extreme slowdowns).

It's been 8 months now that I'm using 16X scanning only (which explains why I don't post scans anymore in the various reporting threads I used to contribute to), and such a discrepancy between scan and TRT has almost disappeared from my testing routines.

An example of a disc showing such a behaviour (I have kept about a dozen if you like).

First the TRTs, showing that ALL drives have a problem with this disc (so excluding drives 'sensitivity' as a variable).

1. Bad TRT in Pioneer 109
2. Bad TRT in NEC 3540
3. Bad TRT in NEC 4550
4. Bad TRT in BENQ 1650

scans at different speeds following (posting limit)...

continued:

1. OK scan @5X in NEC 3540
2. OK scan @8X in Benq 1650
3. OK scan @12X in NEC 3540
4. OK scan @12X in BENQ 1650

to be continued...

And ONLY @16X the problem annouced by the TRTs gets detected in the scans...

1. Scan NEC 3540 @16X
2. Scan BENQ 1650 @16X

As you can see, scoobiedoobie's comments definitly make sense. That's been my own everyday experience since I started stopping trusting scans at "standard" speeds and perform them @16X.

Several times I've had failed tests at high speeds in TRTs that passed fine, often without so much as a slowdown, at slower speeds.

Thanks for showing those examples, saved me the effort.