Page 1 of 1

A couple nagging questions regarding Kprobe/CDSpeed. Please

PostPosted: Mon Jun 28, 2004 5:26 pm
by GizmoC
I use mainly KProbe to test my media (and the quality of my burns).

However, I stumbled upon this thread (Is KProbe unreliable? : ... ge=2&pp=25 )

I realise now that KProbe alone is not the best indicator of media quality. I am supposed to use CDSpeed in conjunction with KProbe.

Sorry if these questions sound naive, but they are nagging me.

- CDSpeed3 comes with 3 tests. The ScanDisc test, the CDquality test and thirdly, the default test when you start the program (its called the Transfer speed test, right?).

- Correct me if I am wrong, CDSpeed's CDquality test is the same as KProbe's BLER test, right? THEY ARE THE SAME THING RIGHT? So, why use Kprobe when CDSpeed provides an all-in-one solution?

- The ScanDisc test tests the PHYSCIAL surface of the disc, right? So.. uhm... generally speaking, it would be pointless running this test on BRAND new media; because brand new media is always unscratched, right?

- What significance is the default CDSpeed test? (ie, the Transfer Speed test)

- Which of these tests is the most "realistic" one? I mean, if I have a 700mb movie stored on my CD, and I want to copy it to my harddisk... which test would most accuately predict the success of my copying (using normal WinXp copying from CD to harddisk?)

- I would appreciate if some of you experianced folks would answer this question:
Someone comes up to you and gives you a CD. He points a gun at your head and asks you, "Is this CD good or bad?". What test(s) would you run to give him the MOST accurate answer? and more specifically, what variables would you alter while doing these tests? (For instance, what is "Accurate Stream" checkbox mean in CDSpeed, and minor variables like that).

Once again, thank you.

PostPosted: Tue Jun 29, 2004 10:40 am
by MediumRare
The thread at CDFreaks refers to DVD-scanning. The counts reported by various LiteOn drives may vary significantly for the same disc (even with the same model). The acid test, then, is if the disc is readalbe. The situation with CDR's is similar, but not as extreme. I've found good correlation between readabilty and reported BLER rates.

- CD Quality test is similar to KProbe BLER- but it combines the reading speed. Also, KProbe runs on LiteOn (and OEM) drives only, while CDSpeed can check C1/C2 on further drives. On LiteOn drives, the error counts shown by KProbe and CDSpeed are pretty well the same.

- Scandisc checks each sector for readability. It has nothing to do with the "physical surface condition". This is the earliest test that was implemented and (for drives capable of reporting C2 errrors) and suffices to gauge the readability of a disc. Yellow blocks are damaged (contain some form of C2), red blocks are unreadble. This test is the basis of the extensive Media Compatibilty thread here. The criterion is very simple: a good burn shows no yellow or red blocks.

- The default reading test tells you something about the performance of drives- even those that do not report C1/C2 (or PI/PIF/POF for DVD's) counts externally. Internally, the drive can still adjust the reading speed when it encounters problem spots. The other components of the default test (e.g. seek time) are performance indicators.

- the "accurate stream" checkbox refers to audio extraction capabilities of the drive, usually in connection with Exact Audio Copy. This is a desirable property for copying audio CDs.

I would normally do a surface scan to check "is this disc readable" because the answer is clear. Personally, I scan all CD's I copy with KProbe because the C1 or (shudder) C2 count gives an indication of reserves that the disc has against deterioration (scratches, UV damage or whatever). These errors are correctable, so they do not render a disc unreadable. However, a disc with high C1 counts is more likely to have problems with aging or mechanical damage than an impeccable one.

KProbe, CD-Doctor or CD-Speed are all viable on LiteOn drives, so your choice is a matter of personal preference. I like the various configuration options KProbe offers (like the logarithmic count scale). Then again, I like the (combined) quality test, especially with the new options in the latest version 3.1.

My reccommendation: pick one method and use it.


PostPosted: Tue Jun 29, 2004 11:28 am
by GizmoC
Thanks for your excellent reply.

Speaking of C1 errors, I hve over 1500 CDs, and some of them are quite old. I value my collection a lot. I've spent years on it. I ran some tests on some old CDs and some CDs reported a huge avarage for C1 errors... like 500 average!!
A few CDs also had some C2 errors.

However, whats strange is that they work perfectly fine. I am able to copy ALL the data from the CD to the HD. What does that mean? Should I re-burn them eventho they work? I am shot on money... very short..

PostPosted: Tue Jun 29, 2004 5:20 pm
by MediumRare
Personally, I wouldn't trust a disc with C2's. Of course the count isn't just a property of the disc- it also depends on the drive used to read it and the speed. So a disc that shows a high error count at maximum speed may look a lost better when scanned at, say, 8x. On the other hand, we normally read our data discs at max. speed, so that test is realistic.

If you value the information on a disc that shows C2 errors, copy it to quality media (e.g. TY, MCC, Ritek). If you're extremely short of cash, test them at intervals and back them up if they start to deteriorate.

If these are audio CD's, back them up sooner rather than later. Data discs have an extra level of error correction information that audio CDs (and video CDs) do not have. Also, it's more difficult to exztact audio data than to copy files.