It's probably true with Benq and LiteOn drives, but I don't think it is with some Pioneer drives (108, 109) and the last Nec family (3550/4550/4551).

Discs that show:
- a perfect TRT in these drives and in other drives
- very good to most excellent scans in Benq, LiteOn and the Nec 3540 (previous model),
Show out-of-specs and very funny-looking (not to say plain silly) PIE levels when scanned in these drives.

I experienced this (and saw reports from others) with many MIDS, including MCC, TY and Ricoh discs, so media quality is ruled out I think.

Luckily, the PIF reporting of these new NEC drives is very consistent and is definitly "usable".

Yet there is much confusion now in the NEC forums with guys having just bought a 3550 or 4550 drive, using it for scanning, seeing PIE levels far over 280 and not understanding what's happening to them. - I have a hard time explaining them that they shouldn't worry as long as PIF figures are good and the reading curves are perfect.

I had to through in my two cents when I saw this post. Excellent discussion.
And I think Francksoy nailed it,
Although I have had alot of fun messing with testing, trying different firmwares etc. I believe when you get so low in the errors it is useless to say the errors have doubled or are xx% higher, There are too many variables throughout the process to say "my errors have doubled between burns or even tests when the pifs are down to 25~100. I do think that it is a good way to look for trends in degradation, even though since I started using my NEC 3500 I have not seen any degradation with the TY 8x I use. I mainly look at the last of my older disks, A few CMCs, and RICOHs.

I get your point.

Actually I use scanning a lot because I like to play with many different MIDs in different burners. Honestly, if I wasn't so addicted to this DVDR game in general (*LOL*), and only used MCC004 (Verbatim-branded), TYG02, YUDENT02, RICOHJPRN01 (Ricoh-branded) or TTH02 (TDK-branded), I think I wouldn't scan more than a disc or two every six weeks, and would merely perform transfer rate tests to check my burns. Actually I think I'll come to this sometime soon, as my wife is getting kinda concerned with this DVR madness. And I can't blame her. (Also I've just, at last, finished to transfer my video collection, 676 burns and about as many scans and TRTs in 9 months, I need a BREAK! ¨*SIGH**)

My point being similar to rdgrimes', for many users, assuming they found a good MID/burner combination (either from good advice or personal testing) and buy only premium discs, scanning is not absolutely necessary. A TRT performed with the "accuracy" setting can most probably catch any possible new batch (or aging burner) problem. Anyway IMO it's better than scanning with poor understanding of the "big picture" and seeing only the numbers.

I'm still convinced that properly conducted and interpreted scanning can help in avoiding many problems, better understanding of how burners behave (like the poor -R re-linking of recent NEC drives for example), and in wisely choosing the right MID/burner/firmware combination.

[off-topic]
Oh and thanks for the compliment - BTW I'm interested in how your CMC discs are doing, degradation-wise. I'd like you to keep me informed, if possible? (take it to PM )
[/off-topic]

So whats being said is that if someone buys only premium DVDs like say Verbatims (which I use exclusively) and MIJs it isn't really necessary for them to use the disc quality testing?
I thought that was what I said? I'm sure that CD Speed has some decent testing uses but the more I learn about some of the tests and what it can be used for , I see less and less need for them as well as some of its reliability.

Now I haven't tried all the testing software but I ask why is it considered so important?

Use premium media, burn at a rate that your drive and/or system can handle, and take good care of them, why would testing be necessary?
If you use some crap media and you find out later that it is crap, maybe O.K.
But I haven't seen anyone come up with figures or test results showing the long term degradation. I think I'll take the NISTs' word on testing results anyway. When you start with crap you'll know in the very beginning, right?

*edit*

Quote:

Originally Posted by rdgrimes I'm pretty sure that I don't need you to tell me what I have said or not, or what I should say or not say.

That is exactly what you did in your first post in this thread. You started attacks even before I had talked about my observations.

Quote:

Perhaps if you would devote your time to reading ALL of a thread instead of just quoting what you think will support your own "cause", you contribute more to this discussion.

Science is based on solid evidence, not on opinions. Observations made by others are also important, otherwise we need not listen to others.

Quote:

I've been aware of that for some years.

But obviously you did not know the reason.

Quote:

The point I'm trying to put froward is that it doesn't matter what is considered "Accurate" by anyone. The purpose of scanning is not to acheive "Accuracy" in error reporting.

Where have I said that purpose of scanning is to acheive "Accuracy" in error reporting? "Accuracy" is something very difficult to achieve, if not impossible. But it is much easier to pick out the obviously inaccurate ones.

There must be a full moon in the Blank Media forum at the moment. What's wrong with you guys. You're spending more time bickering than in open discussion. If it doesn't stop now, some of you are going to find yourself with a timeout.

Rocky Point, enough. Time to drop it. You have been warned.

I'll leave this thread open for now. If there are any more outbursts, it will be closed.

Posts edited and deleted...

@ Rocky Point

You may read the following thread first:

http://club.cdfreaks.com/showthread.php?t=161005

Let me know if that answers part of your questions.

I shall talk about reproducibility later, but my time is tight, so be patient.

Rock -would you mind me asking what DVD media you use ? what Burner ? and what Software ? Although I am also happy with my disks I'm always interested in seeing if there is a better way
Last edited by C0deKing : 1 Hour Ago at 10:36.

Just curious - what was edited and why ?

@ Frankensoy
My wife has been concerned also. For a while she thought I was spending all my time trolling around for Pron!
I'll PM with some info on the other disks later, Have to dig them up from the dry, dark recesses of my closet.

Now, I didn't see what was edited from Rocky's post(s) but his basic questions and/or conclusions still seem to me to in need clearing up... (from my perspective)

In my humble opinion, DQ testing using CDSpeed (or simmilar software) is of value, even for people using known good burners and media. Why? because nothing is good 100% of the time.

Case in point... I have a BenQ 1640, which by most accounts is a excelent burner, a respectable reader, and seems to report errors to reasonably well to software like CDSpeed. I never owned any Taiyo Yuden disks untill recently (about a month ago), and purchased them, largely because of all the rave reviews of Taiyo Yuden disks. I have also recently purchased disks made by MCC (Verbatim 16x branded)... Well my batch of Taiyo Yuden (Fuji MIJ 8X branded) are basicly flawed near the final 600MB of the disk. They sometimes play OK in DVD players, and proform Excelent when you put less then 4000MB of data on them. It would have been really hard, if not nearly impossible to have detected the problem with my batch of Taiyo Yuden disks, without scanning of some sort...

Now, here is where I beleive Rocky Point, is missing the point... Just because a movie DVD (or even a data DVD) plays fine... does not mean that it's 'flawless'. It merely means that it can play in some DVD players, and the error correction is able to correct the amount of errors that the player encounters as it's reading the disk. The players could be encountering 99% of the total number of errors that it's able to overcome, and if it encountered a small number more errors, it could become unreadable (at least in certian portions of the disk)... Just because a disk plays, doesn't mean it is flawless, in fact it could be full of flaws, just that the flaws haven't exceeded some level that would render the disk unreadable...

And... Can a disk degrade over time? YES, sure it can... for a variety of reasons... Exposure to exsessive heat, humidity, or lite can all cause degredation of the disk's ablity to be read, or to be written. Not to mention scratches... Do I have proof of this personaly? No... but if you've not seen any evidence of this, then you haven't looked very hard...

I will not vouch for the total accuracy of the following reports, but if you think even high quality disks can not deteriorate, you'd better read through the following...

http://club.cdfreaks.com/showthread....ight=Longevity

One final note... I'm not stating any of this to try and chastize, or to critize anyone for voicing their own personal opinions, but I do think Rocky Point is missing some critical points about DVD quality and/or if there is any value to doing personal testing (using what ever means they can afford)... I feel strongly that anyone using DVD to store any data of any value, should understand as much as they can about what exactly it is they are doing.

Meanwhile... I'm not trying to say, that you or I should scan every single disk they own, or record... but periodic testing of some disks, can give you usefull data... If you understand what it's telling you.

agreed. I have a number of older DVDs which still play fine in any one of a number of players BUT fail when I try to copy to my hard drive. Scanning at the burn stage and again a few months later might have helped me save the content

Also regarding Taiyo Yuden - I used Fuji for a while and thought they were great. Then Fuji switched to Ritek in the UK and I started to use Unbranded Taiyo yuden. Now I wouldn't go back even if Fuji started to use Taiyo Yuden.

can we keep this discussion related to scanning and it's uses and/or value, please?

that's exactly the point and best value to scanning, verifying the
integrity of our data, not who can get the perfect burns, altho that
has a certain personal gratification

I could care less who can get perfect scans... Nor do I care if a little cleaning, polishing, your LG, and DVDfab has saved your damgaged disk(s). Simply because maybe your LG is above average, or your polish job was, or whatever... Scanning itself did not bring back your data... and if the disk hadn't been scratched to begin with cleaning and plolishing wouldn't have helped make the disk more readable either. I do however want to know more about what scaning may or may not be able to tell us about the quality of our DVDs(perticuarly those that aren't scratched).

dyno,
BTW I was talking to harry not you, and my disks are immaculate(with the exception
of those I let someone handle), and if frank thinks my post was off topic he should tell me not you! Can you add anything constructive?
I believe in consrtuctive hijacking.

Completely wrong. There are no "errors" present on a disc, nor does scanning measure anything "on the disc".

Theroretical numers of errors... (as near as I can tell... DrageMester had it right in his first post to this thread)

208 PIE per ECC (every single row in the ECC has a error). (1664/8ECC)

208 PIF per ECC (ever single row has at least 5 erronious bytes) (1664/8ECC)

Basicly the differance between a PIE and a PIF is that PIF is greater then 5 erroronious bytes per row, where a PIE is at least 1 erroronious byte per row. You should never have fewer PIE then you have PIF, since every PIF is also a PIE.


Theory vs. reality...

At scan speeds above 1X... I beleive it's entirely possible to have more then the commonly stated 280 PIE/8ECC and still be able to read the disk during 'normal' reading operations. That is because for a disk to be readable, it merely has to do so at 1X, and as you increase the speed of your attempts to read the data, you also increase the likelyhood of seeing erroronious bytes. Theoreticly, a disk scaned at speeds higher then 1X could show the maximum possible number of PIF(and PIE) and still be readable at 1X.

At 'normal' read speeds and operation... PIE in any 8 consecutive ECC blocks should not exceed 280. However this is not the same as saying having less then 280 PIE/8ECC is OK, since you could have 279 PIEat the end of one group of 8 ECC blocks, and anohter 279 at the begining of the next 8 ECC blocks... You'd need to keep 8 running tabs that compute the Sum8 value of PIE, and report those numbers... It's not clear to me if this is or is not what's actualy done with the scanning as being discussed here. (Perhaps somone else knows the answer to this) Perhaps it's reported differntly on differnt drives? or maybe with differnt peices of software?

There apparently is also a 'spec' that says no more then 4 PIF per ECC block... during a normal read operation... It's not at all clear to me that the typicaly used figure of 32 PIF/8ECC is at all usefull in determining if this spec has been exceeded... Since even 5 PIF in 8 consecutive ECC blocks could actualy be too much... On the other hand... more then 32/8ECC would surely mean this spec has been exceeded. I don't know with any certianty, but I suspect that CDSpeeds PO failure is actualy meant to document when this spec is exceeded... (can anyone say for sure if this is true?, if not what is a PO Failure as displayed by CDSpeed?)

Perhaps there are no 'errors' on a disk... but there must certianly be flaws in most disks that cause 'errors' that could never be avoided, no matter how good the burner is and scaning should be able to report these, should it not?

I'm pretty sure that's not what they're implying. - the twist is the "error" word. I get your idea but the way you see it is flawed. That's the "measurment" part that you get wrong. The "errors" on the disc are not measured, they are evaluated. - most of the time very, very roughly. Why?

Because the only thing "measured" is the drive's error correction activity. There is nothing else to measure! No scanning device can analyse the low-level structure of a disc. The "errors" reported and that we see in a scan, are a combination of the disc's readability(not "errors") and the drive's error correction system applied. I quote rdgrimes so to bring it forward: ---
---
I'd strongly suggest everyone reads (again?) the post #31 on this page, personally I'd be very sad if this thread should be closed. Please, let's all try to be nice and on topic...
----

I actually think that one the main reason for the misconceptions around scanning is the word "scanning" itself...

The word gives the impression that we are "scanning" the disc for errors, when it's absolutely not what we're doing. We're just asking a software to allow us to read the applied error correction system of the drive.

Maybe if right from the start the expression would have been "error correction reporting" and not "quality scanning", we wouldn't have so muh misunderstanding between us.

I myself took a long time before actually understanding this...!

A little nit-picking:

The scans show what the drive's error correction mechanism is doing in terms of detecting and correcting parity errors. A parity error may be the result of the drive being unable to read the pits and lands on the disc properly, but it could also be the result of the wrong bit (0 instead of 1 or vice versa) having been written to the disc for some reason, i.e. data corruption somewhere after the PI and PO checksums were calculated in the burning process, or dropouts in the burning laser.

So the drive is in effect reporting both errors present on the disc (if there are any) and errors generated during the reading process. But most of these should under normal circumstances be of the second kind.

And just to make sure we are talking about the same thing when we are saying that the errors are (usually) not on the disc:

If there's something horribly wrong with the burning drive (e.g. laser isn't working), or something horribly wrong with the media (e.g. no dye in an area), then there are errors on the disc, they are just not binary errors - they are analog signal errors for lack of a better term.

I'm not sure your statement is accurate... at least not the way I understand things... That is, I don't beleive the PIE or PIF are a result of some 'error correction' process or system... It's my belief that these errors are merely the readers inablity to correctly read a byte... Of course to know if the byte was correct or not, you have to evaluate the a whole ECC block(I think). But in the strict sense, I belive these errors are counted and/or reported before any error correction takes place.

In fact I assume that error correction is only nessasary if you actualy want to 'read' the data... since when scanning, we are not really reading the 'data', there really isn't any need for error correction going on I don't think...

(Now I could be entirely wrong on this, so I won't be offended if someone sets me straight (as always))

I'm not sure I understand your point correctly, but I'm going to assume that I do; explain your point to me if I got it wrong.

The drive will read the disc by processing the reflection of a laser from the disc, and this is what I think of as "reading". This processing results in a stream of bits and bytes.

The drive has no idea if it has read the bits correctly or not, and that's why it uses error detection and correction circuits to process the bit stream, so that it can detect and hopefully correct reading errors. Luckily this is possible because additional ECC information has been written to the disc along with the data already, and this is done in such a way that it is extremely unlikely that 0's are interpreted as 1's and vice versa without being detected. In most normal cases the drive will also be able to correct these misinterpreted bits by applying the Parity Inner and Parity Outer algorithms.

So the drive cannot count any errors before this error detection and correction has taken place - it simply isn't possible!

Depending on whether the drive is in reading mode or scanning mode, and depending on how scanning mode is implemented on the drive, the actual data may or may not be sent out of the drive onto the ATA/SATA/SCSI bus, but the drive itself has to read the data and has to perform the error detection and correction!

There are no "errors" present on a disc, nor does scanning measure anything "on the disc" - I can not argue with that in the slightest.

BUT

Given the choice between scanning with a Liteon 1693S, a BenQ1620 or 1640, a Pioneer, or an NEC I am happy to forget all the theory and go with the Liteon.

Scanning with Kprobe, DVD info or DVD speed and the scans vary very little.

So I understand that all scanners are valid in that they report what they see but I really can't seem myself dashing out t