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

[CDHero]

kprobe v.1.1.18 released
1.Kprobe now can save speed setting.
2.I also fixed the start address bug.

Thanks all !!!

[CDHero]

kprobe v.1.1.19 released
1.Add feature:Write speed for DVD+R/DVD+RW

[dodecahedron]

hmm...this was a very short-lived version - less than 4 hours!

great work, Karr

[aviationwiz]

Where can I download it?

[cfitz]

Karr's download site says:

Download KProbe full installation version (with ForceASPI v4.7)

<snip>
what's this ForceASPI v4.7 ?

That was a typo. He has updated the site and it now says ForceASPI v1.7.

cfitz

[cfitz]

Where can I download it?

At Karr's new site he just told us about:

http://karr.myweb.hinet.net/

cfitz

[BBB]

Hi,
Maybe these questions are answered before in the thread, but I'm on
a dial-up, so the phone bill will be really huge/ in my country/
that's why I don't have enough time to check all the posts

Could you tell me/or give me a link/ what are all of these:
in kprobe=> MSF;LBA; and what are C1& c2 errors.

And how Kprobe determine who is the manufacturer of the CD,
if there is no info in the ATIP?

Or better you may give me some link of documentation file or something.

Thanks to all

[cfitz]

This goes against my better judgment, but...

MSF = Minutes Seconds Frames (there are 75 frames per second)

LBA = Logical Block Address, expressed in hexadecimal notation in KProbe (1 logical block = 1 frame)

Both are measures of the position within the disc, expressed using different, but equivalent, units. Use them if you only wish to test a portion of the disc. For example, if you want to test only minutes 30 to 50 on a disc, select the MSF radio-button and set Start MSF to 30:00:00 and End MSF to 50:00:00.

In general it is easiest to just leave Disc Size selected so that KProbe automatically checks the entire written portion of the disc.

C1/C2 errors:

http://www.roxio.com/en/support/cdr/cderrors.html
http://www.cdpage.com/dstuff/BobDana296.html#3
http://www.mscience.com/faq13.html
http://www.cdrlabs.com/phpBB/viewtopic. ... 6820#46820
http://www.cdrlabs.com/phpBB/viewtopic.php?t=8194
(a little more specifically, http://www.cdrlabs.com/phpBB/viewtopic. ... 6820#46820 )
http://club.cdfreaks.com/showthread.php ... adid=61943

Determining manufacturer from ATIP:

ATIP start of lead-in codes are assigned to different manufacturers. KProbe has a table built into it containing a list of these assignments so it can look up the codes and display the manufacturers. These are not always reliable, though, since sometimes second- or third-tier manufacturers will borrow or steal ATIP codes.

Click on "ATIP Info and Guide" on the following page for more information about identifying discs by the ATIP code:

http://digitaldolphin.netfirms.com/Links.htm

cfitz

[aviationwiz]

With 1.1.19 with my LTR-48125W, when the speed selected is Max. It reads it at 12x. The disk is an Audio CD burned on my Plextor Premium, the disk is manufactured by Mitsubishi Chemicals.

[dolphinius_rex]

For some reason K-Probe sometimes reads discs at 12x, regardless of the speed you tell it to read at. This also happens on my 48125W.

Karr Wang: Could you PLEASE add a counter for unreadable sectors? I'd like it both in the testing screen, and one in the saved graph as well please Sometimes a single unreadable sector is REALLY hard to see, but still detrimental to the quality of the disc, and I'd much prefer knowing for sure if there is one there or not. Thanks!

[rdgrimes]

KProbe does not measure "unreadable sectors", only CDSpeed does that. KProbe is reporting "read errors", which can occur for a number of reasons. As Karr has posted, PUH slipping is the most common reason. when the drive cannot/does not read a sector, KProbe skipps ahead and continues scanning at the set speed. CDDr differs slightly in that it will re-read the sector to a point, then will likely fail to complete the test.
But in no way should the KProbe "read errors" be interpreted as "unreadable sectors", although they could be. If you see that, it's best to slow the scan speed and try again.

[dolphinius_rex]

Well, I think that it is fair to say that if the drive can't read the sector properly at that speed, and a read error is generated, it isn't wrong to refer to it as an unreadable sector, as long as you take the speed it was read at into account. Yes I understand that reading the disc at a slower speed may result in being able to read these sectors, but I'm just thinking in terms of my reviews. In my opinion, it is easier, and not entirely in-accurate to say unreadable sector with reference to the speed tested at, then try to explain all the possible reasons.

I do make sure to scan CD-Rs that give me read errors a couple times to see if it is a consistant problem though. For instance, if I testing a CD-R and brwosing the web, it is quite likely that I will end up with a read error on the test (this is *NOT* something I do when testing media for a review by the way).

[MediumRare]

kprobe v.1.1.18 released
1.Kprobe now can save speed setting.
2.I also fixed the start address bug.

Thank you for fixing that bug, Karr Wang. The appropriate radio button is also activated now when the start or end address is changed (it wasn't previously). The scan I tried earlier now works as expected.

Saving the scan speed is useful too. Could you please add the drive used to the saved data too? My CD-RW drive is secondary slave and I have to choose it manually each time.

G

[rdgrimes]

not entirely in-accurate to say unreadable sector with reference to the speed tested at, then try to explain all the possible reasons.

Not "entirely" inacccurate? That's like being not "entirely" pregnant? There are other reasons for seeing those read errors, besides just an unreadable sector. Calling them unreadable sectors is confusing because of the precident set by CDSpeed. In most cases, they are in fact readable sectors. The most accurate approach is to ignore them, they are reported mostly as an indicator of the skipped sections.

[dolphinius_rex]

rdgrimes:

This is the reason why I switched to testing at both highspeed and lowspeed. The highspeed test performance under regular conditions, and the low speed test to show how the media would react if you were desperate to recover the data from it.

By the way, Scandisc isn't so good at reporting unreadable sectors, it is about as accurate as the read errors on K-Probe, maybe a little less actually, from my experiences at least. I recently tested a disc, and it had several damaged blocks, and one unreadable block. However I was able to copy the full contents of the disc easily to my harddrive, and it didn't even slow down that much when it reached the spot (slowed to 16x). The copy of the disc worked perfectly....for that matter, the original worked perfectly too, but I wanted a backup given the low quality of the previous burn.

[Justin42]

I just got around to putting in my LiteOn 48246S drive into my system, which has a SCSI UltraPlex Wide 40x reader in it (as well as a Plex 12/4/32). I installed KProbe 1.1.19 (honestly, the only reason I cared about getting a LiteOn at this point! and it caused the Ultraplex to freak out -- the lights would blink green/yellow/green/yellow and my Win2K event log showed a ton of SCSI errors (appears like it was causing a ton of SCSI bus resets).

The system would become basically unresponsive and I'd have to CNTL-ALT-DEL to get to task manager, and then restart the system.

I found a copy of 1.1.14 from earlier in the thread and it does a similar thing, but only takes about a minute of bus resetting the SCSI drives before it just goes ahead and seems to work ok.

Is there a way to make KProbe ignore certain drives? I know the Plextors are too old to use with KProbe, so I don't care-- I'd just like to be able to use the newer versions (or even the older ones without waiting for my system to un-lock!

I use Kprobe at work with no problems, so it seems mostly related to the SCSI stuff...

thanks!

[Halc]

Justin,

the behaviour you are seeing, is probably related to CD-RW drive initialisation routines.

My guess is that you have either improperly installed ASPI layer OR another control layer, that takes precedence on the Plextor unit instead of the ASPI layer. Such a layer could be ASAPI (not ASPI) from VOB or another virtual device driver.

I recommend you go through your Device Manager and see if you have CloneCD/Alcohol virtual drives active. If so, disable them for the duration of testing.

Then, check your ASPI layer installation that it it fully operational and works ok.

After that, check for existence of ASAPI.VXD, ASAPI.DLL or ASAPI.SYS in your system directories.

If all those are ok and you still can't get KProbe to function with your Plextor, then check your IDE controller drivers. If you are using an nForce-based motherboard, try both 'Standard PCI IDE Dual Channel controller' (Microsoft standard drivers) and the nVidia drivers provided by your motherboard maker, not by nVidia (this is especially true if you use an Asus board).

Best of luck in debugging.

cheers,
Halc

[Justin42]

Thanks for the long reply! Unfortunately, none of the issues you mentioned seemed to apply in my case...

My guess is that you have either improperly installed ASPI layer OR another control layer, that takes precedence on the Plextor unit instead of the ASPI layer. Such a layer could be ASAPI (not ASPI) from VOB or another virtual device driver

Nope. Never have installed VOB, and no sign of any of the ASAPI files that you mentioned later in the message. I have ASPI 4.60 installed, and ASPICHK shows that all is well.

I recommend you go through your Device Manager and see if you have CloneCD/Alcohol virtual drives active. If so, disable them for the duration of testing.

Again, nope...

If all those are ok and you still can't get KProbe to function with your Plextor, then check your IDE controller drivers.

My system is an ancient BX-based motherboard, so it's using whatever is built into Windows 2000. The LiteOn seems to work fine, it's the SCSI's drive's initialization that seems to hang the system so bad. (and I am using the latest Adaptec drivers for my 29160) Both Plextor drives are SCSI; but only the 40WIDE is seeming to act up on startup (or maybe the CDRW just knows to stay quiet and not flash its lights like it's having a seizure!)

Best of luck in debugging.

Thanks for your help... it seems very weird that 1.1.14, while not perfect by any strech, is much better than 1.1.19. I am going to maybe try to dig around to find some older versions than 14 to check out.

[rdgrimes]

By the way, Scandisc isn't so good at reporting unreadable sectors, it is about as accurate as the read errors on K-Probe

You cannot compare results from CDSpeed with results from Kprobe, they test in a completely different way....apples and oranges. While a C2 reported in CDSpeed is similar to one reported in Kprobe, the drive is controlled in a completely different way in Kprobe, so results will differ.

[dolphinius_rex]

By the way, Scandisc isn't so good at reporting unreadable sectors, it is about as accurate as the read errors on K-Probe

You cannot compare results from CDSpeed with results from Kprobe, they test in a completely different way....apples and oranges. While a C2 reported in CDSpeed is similar to one reported in Kprobe, the drive is controlled in a completely different way in Kprobe, so results will differ.

I wasn't trying to allude that they were testing the same thing, or in the same way, I was just trying to say that both results are not deffinitive as they can be changed by reading the CD at a slightly (or greatly) reduced speed.

[Halc]

Any update from Mr Wang on C1/C2 classification?

It would be very nice to get information on how C1 and C2 flags are raised by LiteOn hardware and what are reported by KProbe?

So, to repeate the earlier questions:

Are the following classifications correct for Kprobe?

C1 error = E11 + E21 + E31
C2 error = E12 + E22 + E32

Or is it something else? This does not require any new features in Kprobe, just a little word or two from Mr Wang :)

Further, would it be possible to get (when we have the above information) to have comments on the following feature wishes:

1) A count of uncorrectable C2 level errors (let's call them CU)
Reason: now when we get C2 errors, we cannot know if they are correctable or not even in the drive that performs the test

2) A standard BLER count, instead/in addition to C1/second
Reason: AFAIK, BLER count is an industry standard measure to classify audio discs into various quality categories based on the rough BLER count (I know, these are not absolute figures)

regards,
Halcyon

PS OffTopic, isn't it interesting that even though E42 should be correctable by many CIRC implementations, even PlexTools and pro analysers flag E32 as an uncorrectable error? If truly all readers can correct E32 and E42, then shouldn't at least E32 be classified as a correctable C2 and not CU?

[rdgrimes]

Was just using this version for the first time, and I couldn't believe it when I heard the drive downshift speed during the test! It was set to max, got to about 80% and downshifted to 32x speed. I tried the same disc again, and this time it downshifted to 24x at about 40%. Can anybody else duplicate this odd behaviour? The disc in question has no particular error issues, no C2. I suspect this might be a glitch with the new live speed indicator.

[rdgrimes]

I went back through the stack of CDR's I was testing and found the one that does this. It's the only disc out of 13 that does this. Here's the proof:



Look at the speed indication and the "max" setting.
This disc consistantly does this, usually around 90-95%. I don't see any errors that would account for this, besides KProbe isn't supposed to downshift for errors.
CDSpeed reads the disc at full speed with no errors, CDDr also has no issues with the disc.
This is pretty weird.
On some scans, the discs completes at full speed.

[MediumRare]

I've observed this downshifting on a number of disks, but only after a spike in the error count (sometimes manifested as C2). With the new version, I saw this on an audio CD (original, not CDR)- the scan continued at 32x . This is why I really appreciate the speed indicator- I no longer have to rely on my ears for this information.

I think this happened on earlier versions too. I'll try and find an earlier post here in which I mentioned it. No time right now- have to get to work.

[edit]
I found the post (back on page 9!!). Comment after "case 3". cfitz found similar behaviour. I think these observations went under because "the great sampling rate discussion" started at the same time.

rdgrimes- did you try scanning this disk on your 52327S too?
[/edit]

G

[rdgrimes]

Scanned it a number of times, and it stopped doing this. As a testing tool, Kprobe is much less valuable if it doesn't maintain a set speed, I hope this gets fixed. I've certainly never seen this in earlier versions, even with some nearly unreadable discs. I guess I'm back to 1.1.14.