Quality Scan vs Transfer Rate Test

The transfer rate test is much better it can tell if the disc is readable or not and if it can keep up at good speed. A poor quality disc will have problems above 8x. Quality scan is good for comparing between 2 good media but cannot tell if a disc is readable or if it will coaster. That's why I find it less useful. I don't bother with quality scan for this reason, I rarely use it but I only did it here cause others request it. A bad burn or poor quality media will read slow 4-8x so I know that a quality test will be bad like 0. no need to waste my time. If it pass TRT with flying colors, then I know this is good quality. Usually 5 min with TRT at max 16x, quality test 5x-8x 10min to 14min. Only to have to do TRT anyway, a pain.

I go for "speed" rather than "accuracy" when I scan these days, as it's good enough for what I'm trying to get a feel for. Knocks it over in less than two minutes on my Pioneer A10XL, for a "good" disc.

Mmhhh. Seems like the wind is turning at CDFreaks about testing, at least more and more users are questioning the relevance of PI/PIF scanning to sort out good burns from coasters. Not a bad thing, if you ask me: this whole scanning subject has been far too oversimplified IMO and too many points of view are taken for granted.

IMO there is enough empirical evidence to support the idea that PIE/PIF scanning is unreliable to sort out good discs from coasters, but it's not as simple as replacing PIE/PIF scanning with transfer rate tests, as the latter are just as dependant on the drive used than PIE/PIF scanning is. And I think that PIE/PIF scanning is under suspicion mainly because of the discrepancies in errors reporting among drives.

There is much room for debate in this area. Depending on the drives they use and the way they interpret the results, different people can have different opinions and can, paradoxically, all be right. (and probably sometimes all be wrong as well *LOL* ).

The most interesting point that you rise, sugarmommyst, is this one IMO: Quality scan is good for comparing between 2 good media but cannot tell if a disc is readable or if it will coaster . This is something that numerous users don't know about: a drive doesn't actually READ the data while performing a PIE/PIF scanning. It's a point I've personally stressed several times on this board. There are even guys thinking that the reading curve in a PIE/PIF scan is the same than a TRT, which is, of course, wrong.

Actually it's been common practise among advanced users and mods, here, to perform BOTH tests. That's also what reviewers (at least the serious ones) do when testing drives and media. Both tests give indications.

@Colin: With the "speed" setting, the drive drops data blocks to achieve a shorter reading time. It doesn't actually read faster! I wouldn't do that if I were you, if your goal is to check the integrity of the disc. A single block with unrecoverable errors may get unnoticed!

Suggarmomyst you still here? the reason you dont like PIE/PIF testing is because you buy cheap crap like AML that more often than not gives very poor scans because *drum roll please* its sh*t

please see
http://club.cdfreaks.com/showthread.php?t=163020

whe using the correct drive e.g BenQ or LiteOn a good disc will more often than not scan well, whereas a poor quality disc will more often than not scan poorly, i agree PIE/PIF testing isnt the be all and end all but when using the right drive its a fairly accurate test.

As far as im concerned there is no TRT versus PIE/PIF as they both test different things, if you want a good idea on the discs quality then you need to run both.

I do agree with the implicit idea, just that the word "accurate" doesn't seem.. accurate to me

But yes, too many people come to the wrong conclusions because they perform tests with unreliable drives.

People always "like" the test that gives them the results they want. Scanning and TRT are 2 completely different things, each providing a different piece of a puzzle. Unfortunately, the puzzle also has many other pieces that we do not really have access to. But until some other end-user tool arrives, I'll be scanning and not TRT. Not because one is "better", but because of time considerations and a scan provides more information.

The way i see it is if you use known good media e.g Taiyo Yuden/MCC Verbatim, then if the disc has a good TRT then the PIE/PIF scans are a good way of seeing the differences between firmware/burn speeds and different batches and also giving an idea on the actual quality of the disc, however lets take the G05 disaster for example those discs showed good results but soon turned to poop, and how did we know? well not only did they have readback problems later on but the 1st clue for me atleast was that the scans showed a fast level of degradation after only a matter of weeks, way before it affected the TRT.

Which is actually the very reason why I won't stop scanning anytime soon: I use scanning MAINLY to check for degradation (+ finding the best drive/media/firmare combination, + checking each new batch for quality variation). I don't care that much that my burns are top-notch or not (in terms of PIE/PIF) to start with (even if it's fun to achieve beautiful scans ), as long as they are stable. With random re-scanning of batches on a regular basis, one may catch any problem before data loss occurs. Yep. Definitly.

As you put it scanning gives more information, but the important point that I wanted to stress is that about the actual readability of a disc, a scan gives only (important) clues, not definitive answers. A TRT gives the answer to the basic question: is the disc entirely (and easily) readable. Personally if the scan is great but the reading curve is poor, I consider there is something wrong with the disc (assuming the drive used usually reads the same MID nicely, of course). Also I've seen some pretty good scans showing unreadable sectors in a TRT or a surface test.

I'm not in the least trying to diminish the importance of the PIE/PIF testing, I rely heavily on these tests myself. I just want to turn people's attention on the fact that TRTs are often disregarded as if they were "redundant" to scanning, though, just like you say, they give different information about the disc/burn.

Excuse my ignorance. Given that both tets only work to show clues to a possable problem with reading would it not be an idea then to to do a scan disc? Wouldn't doing all three together give a much clearer idea or is that just me being a bit too simple?

Ignorance never needs excuses.

Scandisc (in CDSpeed) is a simplified interface, that basically give the same information than TRT and PIE/PIF scanning.

When set to "C1/C2 - PI/PO", you only get a different view, much less detailed, of a PIE/PIF scanning. PI failures over the theoritical limit will be shown in yellow and PO failures in red (if I'm not mistaken).

When set to "read test", you only get a simplified view of a transfer rate test: if the drive has to resort to slowdown to read a sector (how many I don't know), the sector will be yellow, and if the sector is unredable, it will be in red. Thus the differences with a TRT will be:
1. In a TRT, if the drive can't read a sector, the test will stop.
2. With scandisc set to "read test", you won't notice all of the slowdowns, only the ones that the sofwtare judges as "relevant".

This may not be 100% accurate, but I think it's basically the way it works.

Of all my drives only the Benq reads @ 16 but has yet to read a disc (TRT) at max speed. Always fails when it reach max. Of all my other drives i have yet to see a non perfect TRT from an expected ok disc. I have also yet to see a bad burn/unreadable in standalone (except from the Benq) of any decent disc. Basically i use QT since it gives me more precious info than the speedier TRT. I scan to compare batches/id's and for later disclife information. TRT isn't useful in those ways - actually i can tell the TRT in advance when it's a decent disc

Maybe you're psychic?

Seriously though, if you only perform quality scans and not transfer rate tests, scandiscs or other sorts of verification, you are in effect assuming that there's nothing seriously wrong with your disc.

As others have stated above, quality scans will only show you some low-level information about the behaviour of your discs, but they won't show you the most important one - whether your disc is actually readable!

Nonsense.

I realise that. My goal in this case was to get a feel for the worthiness of media and burner combinations. The biggest eye-openers for me were some Memorex hub printable 8x DVD-R's (CMC MAG. AE1) that were burnt in my Pioneer A06P. When TRT'ed on my Pioneer A10XL the curves look terrible, but at least the data is all there. But the same discs burnt on the A10XL look beautiful. (Not that I'm a fan of CMC MAG. AE1's, but I wanted some hub printables on Boxing Day, and that's the best that I could do.)

What I haven't done yet, however, is to TRT the AE1's that were burnt in the A06P back in the A06P itself. I've told my wife's niece that she can have my old writer, so I'll check them when I install it. FWIW, other discs that were burnt in the A06P have "normal-looking" TRT curves when tested in the A10XL.


Colin

rdgrimes, if you're trying to pick a fight with me, you're gonna have to do better than that.

Since you're not actually providing any useful argument, I'm not going to bother refuting it.

On what grounds?
I've seen more than one disc with a good PI/PIF scan having unreadable sectors nevertheless..

I agree with rdgrimes and i don't see him picking any fight just stating a simple point. You guys just carry on testing cheap discs with TRT and see what it gets you. All TRT i ever made with any drive (and decent disc) but the Benq look like the one below if not a little better Had i a large site like this and was testing discs i might consider a TRT just to cover as much as possible but personally i don't consider it relevant. (And yep fellow dane, i burn a few discs now and then and i know how the discs i use behave in the drives i use. I don't experiment with cheap stuff and that saves me a lot of testing)

I only saw it once. TYG02 in my Benq, tested on the Benq?? That's why i (almost) only use it as reader - never know what to expect from it

Apples and oranges. A TRT with a read failure is not the same thing as "unreadable sectors", it simply indicated that THAT drive was trying to read it too fast or has an incompetant error correction scheme. Granted, it may not be a great disc, but it also may be a pretty good disc. I seriously doubt that the same disc has "good PI/PIF scans" if scanned in a competant drive that reports errors correctly. But if you can post such a disc, please do. It is certainly the exception, and pretty rare. I can tell you that in several thousands of discs scanned and tested, I've yet to see such a disc. (again referring to unreadable). But even if 1% of all burns can reproduce that effect, it does not mean that scans are flawed. It probably means that you need a different drive for doing TRT.

It wasn't very long ago that NO drives would read recordable discs at 16x, but now many do and many of those have modded firmware. Take that same failed disc and run a TRT at 8x and you may see very different results. Just a thought. Or, do several TRT on the same disc and you may well get different results each time, just like scanning.

In a scan, there's no such thing as "unreadable sectors". POF may be interpreted as such, but such a disc may still be readable at lower speed or in a different reader. So, POF really should not be interpreted as "unreadable". You can call it crappy, but unreadable is something else.

You can make the opposite argument, that some discs with high PI/PIF still read well in TRT or any other "test". But that also does not mean they are good burns or good discs. (Certainly, it's not a disc that I would keep.)

I don't buy any cheap discs at all. The only DVD media I have right now are Verbatim (MCC/MKM), Plextor (Taiyo Yuden) and TDK ScratchProof.

Do you really think you are immune from getting a bad burn just because you only use quality media?

Many (most?) scanning drives will not tell you that a sector is unreadable when it performs quality scanning. LiteOn drives don't, and AFAIK neither do NEC drives. I think I have seen BenQ drives reporting Parity Outer Failures (POF) which is almost the same as an unreadable sector, and I know for sure that Plextor drives report POFs if you use PlexTools for scanning (but not PxScan).

So in general a quality scan will not show you if there are unreadable sectors, but with some drives you will get that information. Maybe this is what rdgrimes meant with his "nonsense" remark, but since I'm not a mind reader I can't say for sure.

If the reason you do quality scans is that you want to see a pretty picture with low numbers, then it is irrelevant for you to check if your disc can actually be read.

If OTOH you scan discs in order to predict readability, then you are ignoring the most important and direct measurement by not performing either a Read Transfer test, a ScanDisc, or some other type of actual verification.

You are assuming that you don't get coasters because you burn quality media, so that's the reason you don't perform any other tests except quality scans. But if you are already assuming you get good burns, then why perform a quality scan at all? You see my point?

When you perform any type of pass/fail tests, there are two types of flaws that can happen: False positives and false negatives.

You are correctly stating that a Read Transfer test can give you a false negative (falsely state that disc is unreadable, even though it is not).

But you fail to mention that you cannot get false positives with a Read Transfer test! Any disc that passes a Read Transfer test was, without question, readable at that speed in that drive at that time, and that tells you something!

Quality Scans OTOH can usually give both false negatives and false positives, depending on the drive used to perform the quality scan.

Agreed.

Technically this is also correct, but any completely unreadable sector on a disc will either always give you a POF at that location on the disc (if your drive and software supports reporting POF), or it will cause the scanning application to fail at that point.

And any disc that gives you a POF (or CU for CD media) will be either unreadable or very close to unreadable, so it is a warning that must not be ignored!

Provided that the PIE/PIF scanning drive is actually sane (as opposed to e.g. my crazy NEC 4551 drive), then yes I would agree with this.

No, it's not. Not even close. Scanning drives are locked at speed and cannot perform a re-read. POF means nothing except an error was not corrected at that speed with one reading pass. Even at 1x scanning, a POF may not mean "unreadable". (although the odds increase a lot) There's plenty of examples of discs with POF in the hundreds, even thousands, that still can be read.

There's only one way to demonstrate a disc is unreadable, and that's to copy the burned data to HD. Even that only confirms that it's readable or unreadable in that drive.

Um, huh?

A disc with high PI/PIF that reads at 16x in a TRT is certainly a "false positive".

@ Drage - If you read above you'd see my reasons for testing. I have a few pc's and a few extra drives so it doesn't matter much to me to put the disc in a drive and do a test. Again, the same straight line in a TRT doesn't interest me. Reread my above reasons for scanning.

And yes, i do backup same valuable work on more discs to improve lifespan of my work. The scanning will only be for future reference or to conclude a certain disctype/id is more or less preferable than another. Yet again i only backup on TY and i only change disc type/brand rarely so whatever i do it's pretty predictable. And yes i have several 100's of dvds with audio/wave backup and some 100 cdrs so off course i'm interested in keeping track of my stuff's condition

I re-iterate my statement above: And any disc that gives you a POF (or CU for CD media) will be either unreadable or very close to unreadable, so it is a warning that must not be ignored!

Not all drives are read-speed locked when you perform a PIE/PIF scan, as an example Plextor PX-712 drives are not. But that's not really the point.

The point is that if you are performing quality scans in order to weed out the bad discs, you'd be crazy not to throw away a disc that gives you even a single POF, because that is worse than any amount of PIE and PIF!

I have a couple of audio cds that have deteriorated so much, that I literally had to spend hours getting a perfect copy of the disc, with hundreds of re-reads for some sectors. The discs had massive amounts of C1, C2 and CU when scanned.

Using one of your definitions, those discs are still readable. Even if the discs had been unreadable in all my drives, the discs would still be readable by your definition if there existed a single drive in the world that could read these discs!

Using another of your definitions, discs that have only a single C2 would probably be considered bad, or DVD discs having more than 5 PIF in a single ECC block, or whichever limit you are applying when scanning your discs.

I think your definitions are confusing people, even if you can concstruct an argument that you are in fact "correct" in your definitions.
That is not what I meant by false positive / false negative.
What I meant is that you cannot have a successful Read Transfer test on a disc that isn't readable in that drive!

Your (personal or official) limits for PIE/PIF values allowed in a scan may cause you to throw out a disc that shows perfect Read Transfers, but those PIE/PIF values are no more and no less than your own criteria for keeping/discarding that disc. Drives may succeed or fail to read that disc regardless of whether it has passed or failed your PIE/PIF criteria; the probability of a disc passing or failing to read in a drive has a correlation with the PIE/PIF numbers, however. How strong that correlation is, is one of the contended points in this forum.