Official K-Probe Discussion (Tool for Scanning C1C2/PIPO)

[CDRecorder]

Surprisingly, Samsung Combo showed the read speed as 48x (its native read speed) for all 3 media types where Liteon only showed 12X as shown above. Btw, Samsung can only write the advertised speed on all 3 media where Liteon can do 52x on both 80Min media and 32x on 99Min media respectively.

It is normal for the Lite-On to show "12x" as the read speed for unwritten media; it will show the correct read speed for burned discs. It is also normal for different drive models to select different speeds as the maximum for the same types of discs.

Question: In a nutshell, what is the average / acceptable C1/C2 ratio per burn?
And how should one judge whether his burn quality is good or bad by looking at these numbers?

There's no particular rule for how good is average or acceptable. One important thing to remember is that different burners show different results for the same discs, and different scanning speeds also make a difference in how many errors are detected. I personally think that a C1 max of 25 (note: this is the max value, not the total number) is good, and less is better. If you are seeing any C2 errors, though, something is probably wrong. I currently scan discs in my LTR-48125W at 40x; other drives may produce different results.

[rdgrimes]

There is no acceptable level of C2, they are bad. C1 levels should generally be under 50 or so max, but less is better. Most drives will produce levels under 10 max with the right media and burn speed.

[dolphinius_rex]

for myself, my standards are:

C1 Maximum: ~<20.000
C1 Average: ~<2.000

C2 Maximum: ~<0.000
C2 Average: ~<0.000

and yes, I realize the impossibilities of getting less then 0 C2 errors, but I can hope!

[Alfred]

Hi,

I used kprobe 1.23 on my brand new Liteon LDW 401s ES0G and have problems with the disk size detection in PI/PO test. kprobe does not detect the correct disk size when using a DVD+R and tries to test sectors which are beyond the end of the disk, reporting media read or servo errors.
In most cases the lba end address was left on its default 22ffff, once it was even increased to 2305xx.

This problem does not occur when using the DVD-R disk which I copied to the DVD+R using Nero 6.0.0.11.

I have no other dvd burner or Liteon DVD rom drive, so I can not say if this problem depends on kprobes handling of DVD+R media, my burner or nero.

[dhc014]

Hi,

I used kprobe 1.23 on my brand new Liteon LDW 401s ES0G and have problems with the disk size detection in PI/PO test. kprobe does not detect the correct disk size when using a DVD+R and tries to test sectors which are beyond the end of the disk, reporting media read or servo errors.
In most cases the lba end address was left on its default 22ffff, once it was even increased to 2305xx.

This problem does not occur when using the DVD-R disk which I copied to the DVD+R using Nero 6.0.0.11.

I have no other dvd burner or Liteon DVD rom drive, so I can not say if this problem depends on kprobes handling of DVD+R media, my burner or nero.

That may be due to the bug noted here which Lite-ON has supposedly fixed already for their next firmware release.

[Alfred]

Hi,

I used kprobe 1.23 on my brand new Liteon LDW 401s ES0G and have problems with the disk size detection in PI/PO test. kprobe does not detect the correct disk size when using a DVD+R and tries to test sectors which are beyond the end of the disk, reporting media read or servo errors.
In most cases the lba end address was left on its default 22ffff, once it was even increased to 2305xx.

This problem does not occur when using the DVD-R disk which I copied to the DVD+R using Nero 6.0.0.11.

I have no other dvd burner or Liteon DVD rom drive, so I can not say if this problem depends on kprobes handling of DVD+R media, my burner or nero.

That may be due to the bug noted here which Lite-ON has supposedly fixed already for their next firmware release.

Thanks. Dvdinfopro shows the same wrong disk size information, so this is not a kprobe issue. This bug also occurs with DVD+RW media.

[MediumRare]

Does anyone know how the "Disc Identification" in the "Disk Info" panel of KProbe is determined? I burned 2 disks with identical content on media from the same batch and the IDs are completely different: 000DD220h and 0004AEDDh.

[edit 21 Sept]
The disk ID apparently depends on more than the disk content. I just burned the same compilation 3x on the same rewritable and got 3 different ID's: 000C8A4Eh, 0009B602h, 000F182Fh.
The ID is not time dependent- a given disk always shows the same ID.
[/edit]

G

[MediumRare]

I found an MMC-command that reads the ID- this is 51h "READ DISC INFORMATION". The MMC specs do not have additional information regarding the ID, however.

The ID is apparently generated prior to burning the CD. All pressed CDs I've checked have an ID 0h. However, there seems to be a memory effect in KProbe or the drive- the last none-0 value from prior checks is shown for these discs. Not all CDR's have a none-0 ID, though- various older data discs (WinOnCD 3) have 0-IDs.

So the question now is, how is this ID generated? It's not present in the Nero status log, for example.

G

[cfitz]

Before you get too excited about doing error scans in the 166S, you should read THIS thread. Using a ROM drive for scanning is usefull only for comparing different burns with the same burner and media.

(from http://www.cdrlabs.com/phpBB/viewtopic.php?p=81844#81844)

Why do you make that statement? It is true that all KProbe/CD Doctor/WSES scans should be viewed only as measures of relative performance, but that performance can be measured relative to any one parameter you change while holding all others constant. Usually the one parameter we would vary would be one of the following: reader, writer, read speed, write speed or media. Why do you say that when using a DVD-ROM to test DVDs, the only parameter that can be varied is media (or, if I take the more restrictive interpretation of your statement, different instances of a particular media)?

By the way, I do agree that DVD-ROM drives should not be used with KProbe, etc. for CD-R C1/C2 testing because the results don't seem to be valid at all in that case.

cfitz

[rdgrimes]

From what I can see, the 166S handles different recordables differently. (The read speed is constant at 8x no matter what you tell Kprobe to the contrary). On some recordables, it reads higher errors than the burner, on others it's lower. The differences can be striking, particularly in PI. In short, it demonstrates considerable "preference" for certain media types, and a "dislike" for others. This is the case with C1/C2 as well. It does, however, appear to be fairly consistant in it's behaviour. My point is that it demonstrates completely different "preferences" than the burner itself, so using it to compare different media burned in the same burner will result in erroneous assumptions, because it may just not "like" one or the other of the tested media types, or the burn speed, or the read speed. (even though the burner "likes" them both just fine). I feel that this thinking is equally valid when comparing scans in any drive that were burned in another drive, it's just not a great idea.

Sorry, I think if you want believable PI/PO scans, you have to use the burner for testing.

I would also add that there are always examples of media that produces scans that are unmistakably BAD regardless of how they're tested, and other examples of ummistakably GOOD. But what I see are a lot of people trying to make judgments based on scans that are not in those 2 categories, where the variables are too great to draw any conclusion.

EDIT: One qualification; it's fine to use a ROM drive for scanning if you will be using the burned discs primarily IN the ROM drive. In other words, if you will mainly be reading the discs in a 166S, then scanning them in the 166 isn't a bad idea. But only to determine how well they read in that particular drive.

[cfitz]

I think you may be making the mistake of assuming a priori that the DVD burner is correct, and thus any differences in the scans must be attributable to DVD-ROM "preferences" (and those preferences, if I interpret your meaning correctly, have some sort of negative connotation). However, your data can used to argue equally well that it is the burner that has "preferences".

In my opinion, neither view is provably correct. In fact I would posit that both are incorrect and the better way to interpret the results is that both drives have their own "preferences". Or rather, I would argue that they are just different readers with different characteristics, and lacking some other evidence, trying to conclude one is more correct than the other is folly. All we can really say is that testing with a DVD-ROM shows how that DVD-ROM reads the discs, testing with a DVD burner shows how that DVD burner reads the discs, and neither test necessarily shows how other drives such as Toshiba, Pioneer, or set-top boxes will read the discs.

Remember that scan results are a function of many variables, including the reader, and not a function solely of the physical characteristics of the burned disc. I know you have said this yourself before, therefore I know you understand it, but you seem to be forgetting it right now.

cfitz

[rdgrimes]

Whether the burner is "more correct" or not is not relevant if you are trying to measure some kind of burn quality or media quality. The burner is all you've got to measure with that does not inject a second set of varaibles into the mix. I'm not forgetting anything, just pointing out that adding more variables is not the way to get accuracy. If you read with the burner, you're dealing with fewer variables, and the evidence suggests that it does make a difference in the results. I agree that the ulitmate test is the drive that will be responsible for reading the discs, but unless the DVD-ROM is that drive, what it has to say about the disc is not relevant.

[cfitz]

The burner (in its read capacity) is as much another variable as a separate DVD-ROM is. Thus, you have not eliminated the reader variable by using the burner to read, you have just selected the burner to serve as the particular instance of the reader variable. You still have the same number of variables and are not increasing accuracy in any way.

Of course, you aren't decreasing accuracy by using the burner as the reader either. You are just using a different reader, and thus getting different results, as you note. But the fact that one drive shows different results than the other doesn't tell us anything other than that the drives read the media differently. It certainly doesn't tell us one is more accurate than the other.

As for relevance, the exact same argument can be made with regard to the burner. If the end drive that will read the discs is neither the burner nor the DVD-ROM, then the results measured with the burner are no more (or less) relevant to the final results than those measured with the DVD-ROM.

Of course, that doesn't mean I believe such testing is useless and neither has any relevance at all. The truth is just the opposite. I do think that KProbe testing has value and does give some measure of goodness. However, one must always keep in mind that any and all results measured with KProbe are relative results only. They give a feel for better or worse given a certain set of conditions, and given enough experience may serve as a base from which to make reasonable, although not guaranteed, extrapolations of performance, but they don't give an absolute measure of burn quality.

cfitz

[Halc]

And you are telling me this now that I've ordered a JLMS-166 from another continent :)

Seriously, I think you both have valid arguments, but I'm more geared towards the argument that neither the original burner or dvd-rom drive discussed is necessarily a more accurate or a better indicator of any sort of performance in other drives. And that's what matters, isn't it.

Otherwise, if they both report consistent results from media to media (not compared to each other, but compared to themselves), then they likely just have different media preferences and which one is more useful/accurate/trustworthy has not been proven with an outside calibrated source.

After my CD-R test is finished (should be getting AudioDev CATS scans this week), I'll start my DVD+R/DVD-R test preparation and I'll also be looking into Kprobe/JLMS-166 PI/PO testing usefulness during that test.

cheers,
Halcyon

[cfitz]

I think you have the right idea, Halc.

As for your upcoming C1/C2 comparison report and proposed PI/PO study, I am looking forward to both.

cfitz

[dolphinius_rex]

I'm not going to be put off buying a LiteON 166s DVD-ROM either! It may not tell me what others will experience... but it will tell me what *I* will experience

Since I don't have my own DVD Burner I'm stuck relying on other people to burn DVDs for me... but I have been getting stuck with crappy media! So I need to find a better way to find out which media s going to work then just buying a 25pk here and there and having all the failures re-burned until I need to buy more DVD-Rs!

oh, and for the record, I've come to the conclusion that Leda 4x DVD-Rs *SUCK!* I'm not sure if they are really made by LeadData or not, but they deffinatly suck!

[rdgrimes]

Many of us can remember the exact same arguments (ad nauseum) about the DVD-ROM drives when we started testing C1/C2. As far as I can see, the facts are the same in this case.
I do not consider the burner's reading preferences to be a "variable" but a "constant". In view of the wildly varrying scan results between the burner and ROM drives, and the apparent lack of any predictability in those variations, I think one should be extremely cautious about drawing any conclusions about burn quality or media quality based on ROM-drive scans.
That is not to say that you cannot gain ANY information from ROM-drive scans, but only in the context of A-B comparisons of a given media.
In other words, the information you will get is relevant only to that particular reading drive. If you consider the ROM drive to be some sort of "norm" or "standard" (benchmark), then the scans are of some value. I see nothing to indicate that the ROM drives should be considered in such a way, in fact I see evidence to the contrary.
In the case of C1/C2, it appears that it was decided that the burner should be preferred as the scanning drive simply because there are no better choices. This is still true here.
There is work being done on testing tools for other burners, let's hope it is forthcoming.

[cfitz]

Many of us can remember the exact same arguments (ad nauseum) about the DVD-ROM drives when we started testing C1/C2. As far as I can see, the facts are the same in this case.

They are not the same in my experience. Every C1/C2 scan I ever performed with a 166S showed completely unrealistic, extremely low error rates. This was a systematic problem, since it was always dramatically lower. Furthermore, the low rates were not confirmed by comparing to other tools and in fact were refuted. A disc that was known to have problems as shown by other testing tools such as CD Speed and by actual use (drives slowing down while trying to read it, etc.) would show great quality when tested for C1/C2 using the DVD-ROM.

Contrast this to the situation with DVD testing. You yourself have stated and shown that sometimes the DVD-ROM gives lower rates and sometimes the DVD burner gives lower rates. It is not systematic. And you have correlated higher and lower error rates shown on both drives with other tests/indicators of disc burn quality:

One interesting note is that the 401 slows way down when reading the Ricoh bundled DVD+R in CDSpeed, for the last 25% of the disc, the 166 does not slow at all. That would certainly support the notion that the 401 is not as +R friendly as the 166.

(from http://club.cdfreaks.com/showthread.php ... adid=76007 )

This quotation describes tests where the 401 showed higher error levels than the 166S, and thus one might expect the 401 to slow down when encountering these higher error levels while the 166S would not since it was not struggling with higher errors. From this I conclude that the reported PI/PO error levels are indicative of what is actually happening within the drive, which, as I noted, was not the case for C1/C2 testing.

I do not consider the burner's reading preferences to be a "variable" but a "constant".

It is no more "constant" than constantly using the DVD-ROM for testing. How can you now say that in this special case the reader used for testing is not a variable in testing when you have stated many times in the past that the drive used for reading is a variable that does influence the outcome of testing? And I do not understand why you continue to imply, without any evidence to support it, that somehow the burner's "preferences" are more valid than the DVD-ROM's. Neither is more valid the other. They are just different.

In view of the wildly varrying scan results between the burner and ROM drives, and the apparent lack of any predictability in those variations, I think one should be extremely cautious about drawing any conclusions about burn quality or media quality based on ROM-drive scans.

Again, you are making the implicit assumption that the variability is due to the DVD-ROM, only the DVD-ROM, and not at all the DVD burner, even though there is no evidence, empirical or theoretical, to support that notion. I can just as easily and with as little support say that one should be extremely cautious about drawing any conclusions about burn quality or media quality based on DVD burner drive scans because they differ unpredictably from DVD-ROM scans.

Your entire argument is based on the a priori assumption that the DVD burner scans are right, and thus any scans that differ from them are suspect. But there is absolutely no evidence to support this assumption.

I see nothing to indicate that the ROM drives should be considered in such a way, in fact I see evidence to the contrary.

Then please, present that evidence.

In the case of C1/C2, it appears that it was decided that the burner should be preferred as the scanning drive simply because there are no better choices. This is still true here.

As I explained above, that is not the case. The decision was not arbitray in the case of C1/C2 scans. There was actual evidence to support the contention that the DVD-ROM drives have a systematic problem with performing C1/C2 scans.

cfitz

[rdgrimes]

This quotation describes tests where the 401 showed higher error levels than the 166S, and thus one might expect the 401 to slow down when encountering these higher error levels while the 166S would not since it was not struggling with higher errors. From this I conclude that the reported PI/PO error levels are indicative of what is actually happening within the drive, which, as I noted, was not the case for C1/C2 testing.

I later discovered that the 166S is locked at 8x read speed, the 401 is not locked.
My issue with the ROM is that it reports widely varying error rates when compared to the 401. It's higher on one media, lower on another. So the difference between the 2 drives cannot be characterized, it's just unpredictable at best.

[cfitz]

I later discovered that the 166S is locked at 8x read speed, the 401 is not locked.

The ability to more flexibly set testing speed is a real advantage of the 401.

My issue with the ROM is that it reports widely varying error rates when compared to the 401. It's higher on one media, lower on another. So the difference between the 2 drives cannot be characterized, it's just unpredictable at best.

And the 401 reports widely varying error rates when compared to the 166S. It is simply a matter of perspective, and we have no evidence to support the argument that either perspective is right or wrong.

I agree that the differences are unpredictable. But I disagree that the unpredictability can be used to conclude that one drive or the other is more or less accurate. All the differences mean is that the two drives are different. Again, we have yet to find any evidence that either is right or wrong. I suspect that both are right for showing how they individually read the media.

Even if we can prove that one drive or the other more closely matches some industry reference standard, that wouldn't necessarily mean the other drive is wrong. The hypothetical drive that matches the standard would have an advantage for testing because its results could be correlated with a standard, but the other drive could still report completely accurate PI/PO counts for its mechanism.

cfitz

[rdgrimes]

And the 401 reports widely varying error rates when compared to the 166S. It is simply a matter of perspective, and we have no evidence to support the argument that either perspective is right or wrong.

All of which is the exact same argument that went around over C1/C2. Bottom line is you gotta choose one to believe, you can't believe both, and I choose the burner.
I choose it for a number of reasons, not the least of which is the fact that Kprobe is designed to work with burners, not the ROM.

[dolphinius_rex]

I believe there is testing in the works to see how accurate the LiteON DVD Writer and LiteON DVD-ROMs really are! So how about we sit back and relax while we wait to see what drive to trust (if any!).

[cfitz]

All of which is the exact same argument that went around over C1/C2.

No it isn't. I just outlined in my previous posts why C1/C2 testing on the DVD-ROM drives was and is different.

Bottom line is you gotta choose one to believe, you can't believe both, and I choose the burner.

Wrong again. You can believe both if both are correct, and they certainly could both be correct (or incorrect, for that matter). I explained that also. Again, why do you now refuse to acknowledge that the reader one uses for testing influences the results, but does not necessarily invalidate them because the results, by their very nature, must include the effects of the reader. You've said this before yourself. Why the change now?

I choose it for a number of reasons, not the least of which is the fact that Kprobe is designed to work with burners, not the ROM.

How do you know this? Did Karr specifically tell you this? We know that he wrote it in part to test the writing strategies for DVD burners, hence the tilt angle test. But that isn't the only facility he built into it. And even if it were, it wouldn't necessarily imply that he wrote it to test those writing strategies using the burners. For all we know the DVD-ROMs are more "accurate" readers and he wrote it to work with them for the read testing.

I don't get it, rdgrimes. You are smarter than this, and you have lots of experience with testing media. Why are you being so closed minded about this and refusing to acknowledge the obvious truths, truths that you have acknowledged in the past when discussing CD burning? Do you have an agenda you are trying to promote? Are only people who buy Lite-On burners, such as yourself, entitled to test and present DVD media burning test results? In the end all of you have to say is that one has to have faith that the burner is correct. I'm sorry, but that is not good enough.

If you want to decide, without any evidence one way or the other, that your 401S is the only standard for measuring PI/PO errors, that is your business. Test away and be happy in the privacy of your own home. But don't publicly make bold, sweeping pronouncements that only the 401S (and potentially upcoming 411S) can be used for PI/PO testing and that DVD-ROMs can not. You have absolutely no evidence to support such a contention, and you are doing a disservice to the community by propagating such nonsense. Saying something is so often enough does not make it so, but in the world of the Internet unfounded rumors such as this do, unfortunately, take a life of their own. If some evidence eventually does surface that proves that only burners can safely be used for PI/PO testing, then by all means at that time shout it from all the rooftops and I will be right with you cheering you on. But until that time, please refrain from posting unfounded speculation as proven fact.

cfitz

[cfitz]

So how about we sit back and relax while we wait to see what drive to trust (if any!).

I would love to, dolphinius_rex. In fact, that is what I have been advocating. But rdgrimes has already concluded, without waiting for any evidence that would tell us which drives (if any) to trust, that the 401S is the drive to trust and has publicly stated as much. That is misinformation and I can't let it stand unchallenged. The last paragraph of my last post explains in further detail.

cfitz

[MediumRare]

All of which is the exact same argument that went around over C1/C2. Bottom line is you gotta choose one to believe, you can't believe both, and I choose the burner.

It's not the same thing as C1/C2 from a DVD-drive or burner. Karr Wang mentioned the difference: cooked vs. raw mode. I spent some time searching for the comment, and it was in the CDFreaks forum. See this post by cfitz.

I think I'm with cfitz on this subject. Pick a drive and do your differential testing. It would of course be very interesting to see how strongly the results for one and the same disk vary with other models of drives, e.g. LTD163.

For what it's worth, consumer DVD-drives are better readers (more error tolerant) than the "reference" drive used by AudioDev. That's what the German magazine c't found in a DVD-burner test in its current issue (see this thread).

G