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

[CDRnovice]

I would like to verify that the data I've just written to my CD-R matches what is on my computer drive. I've been told to use the Nero "verify written data" option but I cannot find it. The Nero help screens say the option is located on the burn status dialog. But the option does not appear at the time I initiate the burn through the "Burn Wizard." The option also does not appear after the burn is completed on the burn status dialog box (even though the Nero help screen says the option appears there). This is strange.

What steps do you take to enable the "verify written data" option in Nero when you burn discs?

Can I verify written data at any time after a burn or only immediately following the burn?

[Inertia]

CDRnovice,

You posted this question to the wrong forum. It should be in the Nero forum.

To verify a disc after a burn, don't use the Wizard. Turn it off. The verify option will be available in the burn dialog when burning in the normal mode. This option within Nero works only after a burn is completed.

If you want to verify written data at any time, use a program like the shareware CDCheck.

Nero CD Speed in the Nero Toolkit is useful for general quality testing of recorded discs. Look under Extra for the ScanDisc and CD Quality Tests. Read the help file for information on the tests.

[Halc]

I just burned a bunch of discs on my wife's Mac G4 tower using LG 4480B (latest firmware) on TDK SpeedX 48x 700 Mb CD-R media (TDK Corp).

They all verified as "no errors" after the burn on the Mac. The verification process verifies the files by reading them off the disc and comparing them to the original files on the HD (not just reading C1/C2 data).

I then tested them with LiteOn 48246S (SS0B) with latest KProbe (1.1.12). The test range was set to "Disc Size".

All of the three discs had unrecoverable C2 errors at the 99% mark of the disc (capacity) and the testing was stopped. The discs did not contain more than 650 Mb of material each (of the total capacity of 700 Mb).

This leads me to believe that the there is something wrong with the way I tested or that Kprobe has problems with testing some particular disc types?

I also ran the test in Nero CD Speed ScanDisc using the "Surface Scan" option (can't use "File scan" as the cd-rom format is Macintosh HFS format).

All blocks on the chart are fine, except the very last block which says "unreadable".

I just had my wife go through all of the discs and copy all of the contents onto a hd. No problems with any of the discs.

Can anybody guess what I am doing wrong with my KProbe testing? It reports uncorrectable C2 errors at the part of the disc that is not even burned with data and stops testing the disc.

regards,
Halc

[dolphinius_rex]

I'm betting it was a multisession disc and the session was left open at the end. I've never found that Nero's verification process was particularily useful under any circumstances. I've seen discs that could not be copied back to HDD pass the verification flawlessly.

[Inertia]

My guess would be that the Macintosh HFS format is involved in the C2 error/unreadable sector reports at the end of the disc.

If there truly were unreadable sectors, the discs would have failed the verification on the Mac. It is possible that the LiteOn has trouble reading the discs created on the LG, but I don't think that is likely.

The tests may be reading something related to the Macintosh HFS format at the end of the discs which can't be properly interpreted by Windows test programs. This may be the reason for the "unreadable C2 errors" reported.

[rdgrimes]

unrecoverable C2 errors

There's no such thing. But putting that aside, KProbe does not report "unreadable sectors". You need CDSpeed for that. When KProbe reports those "errors", it's usually due to what the author calls "PUH slipping". Kprobe does not re-read or slow to read damaged sectors, it just skips over them. It may or may not report an "error" at that point, but it is a different kind of error than the C1/C2 it is measuring.

[Halc]

Thanks for the suggestions. More info:

- Discs were NOT multisession
- HFS is not the problem (I just tested several HFS format discs, no problem)
- Verification was NOT done in Nero, but in Roxio Toast (it was a Mac, remember :)
- LG was not the (only) problem (I have tried several TDK corp discs burned on a PC on the LG and they all measure excellent: below 10 C1 and no C2)


I thought that when the KProbe stops the testing before 100% is reached (due to slipping or whatnot), that it is due to not being able to read the disc/sector correctly. Does this not constitute a CU error (regardless of whether KProbe reports it or not)?

I had my wife burn a 100 Mb disc on the same 700 Mb TDK Corp media at 12x (rather than at 40x) and it tested 100% ok in KProbe.

I'm assuming that the errors at the very end of the 700 Mb disc (with c. 650 Mb of data) were due to a bad combination of LG / media and/or burn speed.

I'll do more tests, but it appears that the problem is not with Kprobe (I think).

regards,
Halc

[CDRnovice]

Halc: I have the same question since I burned a Prodisc CD-R that could not complete the KProbe test. So, I burned a second copy of the same music and KProbe did complete a test of that second disc.

I too am wondering whether the fact that KProbe cannot complete a test on a specific CD-R indicates the disc in question is faulty.

[KCK]

Halc and CDRnovice:

Could you tell us at what speed KProbe was run? Do you get similar results while testing at a slower speed, e.g., 8x?

rdgrimes:

We don't really know how KProbe handles reading/slipping errors. Actually Karr Wang asked for suggestions on how to handle PUH slipping:

http://www.cdrlabs.com/phpBB/viewtopic. ... 3383#63383

http://www.cdrlabs.com/phpBB/viewtopic. ... 3386#63386

but apparently his request has been ignored.

[CDRnovice]

Could you tell us at what speed KProbe was run? Do you get similar results while testing at a slower speed, e.g., 8x?

I run all of my KProbe tests with the "Max" speed option checked. I tested the first Prodisc 48x 80 min CD-R and KProbe locked up (froze) at 5% complete. I repeated the test with that same disc -- same exact result. I then burned a second Prodisc 48x CD-R with the same music and the KProbe test breezed through that second disc, 100% complete.

I did not try a slower speed for the first disc since KProbe should be able to complete its test at max speed. I have no idea what causes KProbe to freeze at only 5% complete while testing the first disc.

I wonder whether KProbe's failure to complete its test on the first disc is an indicator of a faulty disc.

[rdgrimes]

I can only say that I have yet to see KProbe fail on a really good disc, even at full speed. Once i establish that I'm dealing with a questionable disc, I switch to CDSpeed. Kprobe is great with C1 measurement, not so great with C2 and unreadable sectors.

[CDRnovice]

KProbe provides very detailed C1 and C2 error information whereas CDSpeed seems to provide a much less detailed report. Based on what you've said, I guess I'm not sure to what degree I can rely upon KProbe's detailed test results. How accurate is KProbe?

[rdgrimes]

You're comparing apples and oranges. Kprobe is designed to read at a constant speed, CDSpeed allows the drive to perform as designed, which means is will slow and re-read. It will continue re-reading until it fails, which gives you the unreadable sector. So you now know that the drive has had to slow and re-read and that the sector is unreadable. (or that it's readable but only after re-reading at a slow speed). Kprobe gives you different information, which is that the drive was unable to read at full speed. For testing purposes, forcing the drive to read at a constant speed eliminates several variables. To summarize, one is a pure testing tool, and the other is a tool to determine the "real-world" readability of a disc and it's data. KProbe is very accurate in what it does, which to display what the drive is reporting, C1 and C2 error counts. A "good" disc will read at full speed and report no C2 errors whether you check it in one or the other. To put it another way, the nature of C2 errors is that they are corrected. However KProbe does not allow a drive to slow or re-read, which is another level of the correction process. My own interpretation of KProbe's behaviour is that if a disc does not complete the read process, it has failed. I use CDSpeed to determine if the disc is in fact readable. However, the fact that is readable does not make it a "good" disc. Hope I've explained this well enough, it's late and I'm fried. You can either use KProbe to scan at a slower speed, (you can even scan just the specific portion of the disc), or you can use CDSpeed to scan it instead.

[KCK]

CDRnovice:

Although my message was composed before rdgrimes' reply, I'll post it as is, since it addresses other aspects.

-----------------------------------------------------------------------

Just as rdgrimes observed, KProbe typically handles good discs very well even at full speed. In such cases, the only concern is that the fraction of sectors sampled by KProbe decreases for higher reading speeds on slower systems; see S8 in the list of suggestions for improvement

http://www.cdrlabs.com/phpBB/viewtopic. ... 4360#64360

This concern, however, is of minor importance because in practice KProbe does provide useful statistical estimates of a disc's quality, even though a single sampling run is not guaranteed to discover "rare events", such as C2 errors on a single sector. To understand this aspect, just read cfitz's contributions to this thread.

It's another story that KProbe doesn't make it easy to detect discs of questionable quality with reading/slipping errors; see S5, S6 and S7 in the list of suggested improvements.

Anyway, for questionable discs, CD Speed is the only alternative. Note, however, that even for good discs you can't expect perfect agreement between KProbe and CD Speed, since they use different commands for retrieving C2 counts from the drive.

[dolphinius_rex]

That's exactly why I continue to use CD Speed's Scan Disc, along with CD Doctor/K-Probe.

[Halc]

Ok, more data to help me track down this problem.

I had my wife burn another disc on her Mac at 24x on a TDK Speed-X 48x 700Mb rated media (TDK Corp). I have burned several of these with the exact same LG 4480b unit, but on a PC and they all measured fine.

Well, the ones she burned on her Mac, didn't do so well:



That particular disc contains 690 MB of data.

As you can see, the results are really nice up to about 70 minutes or so, after which things go haywire.

She burned two other discs at 16x and they measure almost exactly the same (I won't bore you with more screen shots).

All of the files remain 100% readable and pass all verification tests on a Mac.

However, when tested with Kprobe (see previous posting for specifics) I get a lot of "slipped" information and the results are really nasty for the last few "minutes" of the disc.

Any ideas as to what is going on?

Are these discs just bad after the 70 minute or so mark?

Has anybody else experience anything similar with Kprobe testing`?

regards,
Halcyon

PS I plan to get more of these and burn them on my LiteOn (on PC) to test more. <sigh> when I finally thought I had found a good quality CD-R for my wife's needs....

[rdgrimes]

Those are pretty typical test results. The discs burned on the mac are just low quality.

[Halc]

Even more fuel to the fire:

The EXACT SAME DISC as above, but now tested with CD Doctor using exact same test gear:



I don't know about you, but I'm running out of excuses on behalf of KProbe. To me it looks like KProbe is somehow behaving oddly while CD Doctor gives results that are in line with the actual quality of the disc.

I need to get my Plextor Premium soon to verify this.

As for the "low quality burn on a Mac". I've never heard it myself nor experienced it like this. A burn is a burn is a burn and dictated by the burnind cd-rw and not the OS or software (unless raw burn is used).

And like I stated before. I've burned the same batch of discs on PC and they were almost faultless even if burned at 48x. I'm not so convinced this is a 'bad media' issue anymore.

regards,
Halc

[cfitz]

No fuel there. You tested at 8x read speed in CD Doctor, and 48x reading in KProbe. That accounts for the difference. Reading at high speeds is much harder than reading at low speeds. Test everything at 8x or everything at 48x.

cfitz

[rdgrimes]

Scan the same disc in CDSpeed Quality test and see how it compares. I note that CDDr is skipping large sections near the end.

[KCK]

Halc:

Why are the reading errors not displayed for KProbe? Maybe the "Error value" chosen in the Setup window was too small.

Also it's a bit unusual that the "Disc Manufacturer" field is empty for KProbe. Does KProbe recognize the disc manufacturer in the Disc Info tab?

[Halc]

Open Questions
============

KCK,

I don't know why the "Disc Manufacturer" window is empty. Maybe because I have auto-insert notification turned off? Anyway, I can read the Mfg info by going to "Disc Info" tab and pressing "Refresh". It doesn't read this automatically nor does it include this information in a saved "Write strategy" graph, unless I manually go to "Disc Info" and press "Refresh".

My error value was set to 1. I do not know if this is too small?

cfitz and rdgrimes were of course right. I should think more clearly (and not during night time), before I post :)

Kprobe at 8x
=========

Let's see the Kprobe at 8x (compare to the CD Doctor at 8x):



Looks almost the same compared to the CD Doctor at 8x (sans the c2 spikes), just like cfitz assumed.

CD Doctor at 48x
=============

Also, see the CD Doctor at 48x (compare to earlier image of KProbe at 48x):



Looks like CD Doctor is skipping large sections at the end, just like rdgrimes assumed.


Why measuring get progressively worse towards disc end?
===========================================

Now, the only questions remain (perhaps off topic for a Kprobe thread):

1) Why do the burns verify at 48x 100% ok and the files remain readable on the LG at 48x, but measure unreadable in CD Speed ScanDisc Surface scan (for last part of the disc) and also fail to be tested at a higher speed in KProbe (and assumedly CD Doctor) on a LiteOn at 48x? All this, at the END of the disc.

Possible explanations:

1) CD burned at 48x is seriously worse quality at the end of the burn than at the beginning. I don't think so, because the level of C1 errors actually goes down towards the end of the disc, if tested at 8x (and not 48x, which stresses the drive more than the disc quality). I believe the 8x test more than the 48x, just like professional cd quality testers do. It is a better sign of cd burn quality than the 48x test.

2) CD read at 48x is seriously dependent on the quality of the reading/testing drive, especially towards the end of a disc. This I think is supported by the fact that the LiteOn LTR-48246S cannot read the last part of the disc properly, whereas an LG 4480B can (at the same speed)

Is there a difference between a Mac and a PC burn?
=======================================

Question 2:

2) Why did my earlier burns on the same media, but on PC measure faultlessly using same burn gear/speed and same test equipment/speed?

Possible explanations:

1) The PC was better at burning with high speeds. This is not supported by my wife's Mac burns at 24x and 16x which measure just as good/bad as 48x burns.

2) I wasn't sure if my PC burns were not as big (not taking almost the whole of 700Mb). Fortunately I found one of the 78 min burns I had done on my PC using the same LG4480b at 48x with the same TDK Speed-x 700 Mb 48x media (TDK Corp)



Look, no C2 at all and no rising of the C1 towards the end, even though the burn is almost to the max capacity of the disc. However, all my wife's burned 700 Mb discs (with the same burner) have huge amounts of C2 errors towards the end of the disc, although they verify 100% ok on her Mac. I've tested 6 of these burns already.

I'm puzzled.

What I have learned from this testing myself is that:

Read tests at 48x are not very useful, as it appears that the quality test reliability at that speed from drive to drive (between the LiteOn that fails and LG that does not fail) varies quite a lot. If we are testing the DISC quality and not the drive quality, we should not test at 48x+ speeds. Especially we should be weary of the test data towards the end of the disc, which can be (IMHO, based on above tests on the same disc at 8x and 48x) extremely drive dependent.

regards,
Halcyon

PS My aplogies if I have taken this thread off topic too far. Feel free to moderate me away if need be.

[dolphinius_rex]

Halc: I agree to a point with what you are saying regarding testing at high speeds. That is why I generally test at 32x instead of 40x or 48x.

You have to remember, that the C2 error reporting accuracy of the LG drives is lower then that of the LiteON, on the other hand, the error correction ability is a little higher I believe, so these two things put together can make for some odd looking comparisions.

High speed testing is best used for seeing how the disc will fair in real world situations, and testing the disc at 8x shows how it will fair if you are desperate to recover the data off of it. A disc that performs VERY well on a high speed test is a disc you want to keep, where as a disc that performs very poorly on a high speed test, but still performs very well on the low speed test is one that you want to keep an eye on for quality slippage.

I generally, I would suggest testing your media at full speed or 32x or something in that area, and if the disc performs poorly, then test it again at 8x. If the disc passes the highspeed test, GREAT! if it fails that, but passes the low speed test, then good, if it fail the low speed test as well, reburn the disc and be happy that you bothered testing in the first place!

[CDRnovice]

Thanks to rdgrimes and KCK for their very helpful responses. You definitely know a lot more about this stuff than do I but I'm starting to better understand this CD-R frontier, in great part due to this forum's contributors. This technology seems to be more stable than it was a few years ago when I first looked into CD-R technology. I have much more confidence in it now and that's why I've started burning CD-Rs. I've only burned a few discs so my next little project will be to buy some of those TY discs everybody raves about and see how they perform with my 52MAXX drive, KProbe, and CDSpeed.

[Ian]

This might have been asked before, I just can't find it in the 18 pages of discussion.

Anyway, has anyone seen C2 errors at the end of disc recorded using TAO? I'm using the latest KProbe and a Lite-On LTR-48125W (firmware VS06). I've tried running the tests at 24x, 32x and 40x and they keep popping up. They aren't there with WSES.



I'm going to try to upgrade the firmware on the drive. If that still doesn't help, I'll try a LTR-52246S.