Tools to measure DVD+/-R(W) burn / read quality?

[Halc]

On the CD-R side we have tools like Nero CD Speed, Kprobe and PlexTools Pro to measure the cd-r(w) burn/read quality on a particular drive.

Are there any similar tools for DVD-/+R(W) burn/read quality measurement?

I know that KProbe does some, but AFAIK that feature is still highly theoretical as no LiteOn drive shipping at this time supports that feature.

Are there any other utilites?

regards,
Halcyon

[cfitz]

I know that KProbe does some, but AFAIK that feature is still highly theoretical as no LiteOn drive shipping at this time supports that feature.

Not so. You can test PI/PO read errors with KProbe using Lite-On DVD-ROM drives right now. What you can't do yet is do the Tilt Angle write test. Although even this may be possible now if you have the new true Lite-On LDW-401S burner. Maybe dhc014 can report on the Tilt Angle tests.

Here is an example:



(see http://www.cdrlabs.com/phpBB/viewtopic. ... 0199#60199 )

cfitz

[Ian]

cfitz, do you have any info on what good PI/PO results should be?

[dhc014]

Not shipping? I think it works on Ltie-ON DVD-ROM's, and besides, the LDW-401S is shipping...

This tilt analysis test takes FOREVER:
[Images Removed Due to time lapse]

I don't know if these tests properly correlate since I'm not even an expert on C1/C2 testing... When C1/C2 testing burned CD-R's, it likes to make large spikes so that you can't see the rest because it's small...

Here's the performance write test, nothing too exciting. I think that this is where LO needs to do some improvements in stabilizing the speed:
[Images Removed Due to time lapse]

There's so much to mess with with this drive, I've made millions of screenshots...

If you guys have any suggestions for tests, just let me know. I'm just happy that I've burned 5 working DVD+R's at 4x so far. As you might imagine, some of these tests take a very long time...

[cfitz]

cfitz, do you have any info on what good PI/PO results should be?

Unfortunately no, since I don't own a writer and haven't gained any experience with writing DVDs. I can tell you that the only specification on PI/PO errors is that the sum of PI errors in any eight consecutive ECC blocks must be less than or equal to 280. And, of course, you don't want any uncorrectable PI or PO errors.

To add to the uncertainty, I don't know what Karr is plotting on the PI/PO axes. Presumably he is plotting the running 8-block sums, but I suppose there is also a possibility that he is plotting the count per block, so we would need to sum them up in blocks of 8 to compare to the 280 error standard. We will have to ask him if he can clear this up.

A few people have posted PI/PO plots over at CDFreaks. In general they seem to show fewer errors than what I have shown here. That probably isn't too surprising, since this cheap-o disc came free inside a box of Cheerios.

cfitz

[cfitz]

Hey Dave, thanks for the Tilt Angle test results. Now to figure out what they mean.

I don't believe it is anything we average users really need to be concerned about, because my understanding is that the tilt angle settings are set in the firmware to give best results, and we can't control them outside of this one testing utility anyway. So, the results are somewhat academic. But they are still interesting. And maybe someday there will be a Vari-Tilt feature like the Vari-Rec of Plextor Premium that lets people play with these low-level parameters. Of course, if they did add such a "feature", it would probably work about the same as Vari-Rec: mostly just moves the setting from its optimal position and makes the burn quality worse.

cfitz

[Halc]

Ok, I stand corrected on LiteON drives :) Thanks!

Any other software? Free or not free?

AudioDev Cats SA300 is still a wee bit too expensive for me...

On a related note: does anybody know of any good / authoritative / accurate sources on DVD-error correction, disc tolerance specifications and measurements practices?

regards,
Halcyon

[cfitz]

I'm not sure how much you will find readily available. I think that some or maybe even a lot of that type of information is considered to be trade secrets, and they don't just hand it out to people without a license. You can get an idea of some of the things the -R camp tests for in these documents:

http://www.dvdfllc.co.jp/forms/forms.htm

cfitz

[Halc]

Thanks cfitz!

dhc014,

do you have a LiteOn CDRW drive also?

Could you test a problematic disc in both LiteOn CDRW drive and the new DVDRW drive?

I'm eager to see if the DVDRW drive is more accurate than the LiteOn DVD-ROMs in reading CD-R discs?

I'm still waiting for the dual format LiteOn DVDRW writer myself, but I'm not sure if I need to keep my cdrw drive as well, if the dvdrw writer is no good for measuring CD-R discs.

cheers,
Halcyon

[dhc014]

I do, but it's disconnected at the moment. I'll do a test for ya on in a sec and update this post...

This drive likes to find a HUGE spike in all discs, usually right in the middle... kind of a pain.

OK, DYOC (Draw your own conclusion) time: http://dhc014.rpc1.org/ [link removed due to passage of time]

[Halc]

dhc014,

thank you ever so much!

Looks like I'll have to keep my LiteON CDRW. The c1/c2 testing inaccuracy seems to be carried over from LiteOn DVD-ROMs and I would not trust those results.

On a related note, I think I'll wait for the next generation LiteOn DVDRW writer :)

regards,
Halcyon

[cfitz]

You're welcome Halycon, and that was a good question to ask about the C1/C2 scanning capabilities of the drive.

Dave, thanks for running those C1/C2 tests. I appreciated them as well. I guess we have to stick with CD drives for testing CDs, and DVD drives for testing DVDs.

By the way, your test result pages just keep getting fancier and fancier!

cfitz

[aviationwiz]

Not shipping? I think it works on Ltie-ON DVD-ROM's, and besides, the LDW-401S is shipping...

This tilt analysis test takes FOREVER:




I don't know if these tests properly correlate since I'm not even an expert on C1/C2 testing... When C1/C2 testing burned CD-R's, it likes to make large spikes so that you can't see the rest because it's small...

Here's the performance write test, nothing too exciting. I think that this is where LO needs to do some improvements in stabilizing the speed:


There's so much to mess with with this drive, I've made millions of screenshots...

If you guys have any suggestions for tests, just let me know. I'm just happy that I've burned 5 working DVD+R's at 4x so far. As you might imagine, some of these tests take a very long time...

The PI/PO tests look good. The PI is much higher than what I generally get (average wise) The PO is very very low (average wise). Looks good to me judging by that.

[dhc014]

Sorry, I don't even know that those tests were performed properly, I'd hope so though.

[MediumRare]

Thanks for the tests Dave, really interesting info there.

This drive likes to find a HUGE spike in all discs, usually right in the middle... kind of a pain.

Do you suppose it's trying to do a transition between DVD layers on these CDRs?

By the way, your test result pages just keep getting fancier and fancier!

I like it and had a gander at the html. This is a really neat way of doing a/b comparisons (it's a bit buggy with Opera 7- but that's an Opera problem)!

G

[dhc014]

Some more PI/PO tests on burned DVD+R discs and a DVD-R disc. I'm not sure why the DVD+R tests ended with "UNRECOVERED READ ERRORS"...


DVD+R disc written at 4x on LDW-401S
[Image Removed]

DVD+R disc written at 2.4x on Optorite DD0203
[Image Removed]

DVD-R disc written at 4x on Sony DRU-510UL
[Image Removed]

[Halc]

Does CD Speed surface scan (v.2.02 or later) work with checking DVD-R/+R discs?

Or is it purely a C1/C2 test for CD-media?

regards,
Halc

[cfitz]

Even on version 1.02g (and probably earlier) it will run on a DVD-ROM (I don't have a recordable disc to check). Unfortunately, I don't know what it is actually measuring, which is probably the real question you are asking...

I have a suspicion it is only checking whether each sector is readable or unreadable, although it does show the yellow "damaged" box in the legend.

cfitz

[dhc014]

Does CD Speed surface scan (v.2.02 or later) work with checking DVD-R/+R discs?

Or is it purely a C1/C2 test for CD-media?

regards,
Halc

C1/C2 is only a CD test. AFAIK, PI/PO is the DVD equivalent.

CDSpeed wont let you start the test:

[cfitz]

That's "CD Quality Test..." Try "ScanDisc..."->"Suface Scan"

cfitz

[dhc014]

?

[cfitz]

That's what I get too. Like I said, my guess is that it is really just measuring readable/unreadable. In that case, as long as there are no PO failures all sectors will show up green.

cfitz

[Halc]

I asked Erik Deppe (the author of CDSpeed?) about this, but he's yet to reply.

regards,
Halc

[Halc]

I found out about a utility called DVD Info Pro:

http://www.dvdinfopro.com/index.php

It has supposedly some kind of error test function (havent got any idea as to what exactly it measures and how).

Please feel free to try it out and comment there, if you have more info. I don't even have my dvd-rom hooked up before I get my PCI IDE controller.

On a related note, the dvd specs define an upper limit recommendation of 280 inner parity errored rows in eight consecutive rows of ECC blocks. I don't know if MediaTek chipset reports PI errors per block or by row.

If it reports only by block, then I'm not sure it's possible to exactly calculate how many rows had an error in eight consecutive blocks. Of course we can use the average of 208 rows per ECC block and assuming relatively random distribution, this could give us a rough estimate.

In addition, DVD+R specification seems to require an additional maximum of 4 uncorrectable errors per one ECC block after the PI stage errors have been corrected for.

Now, it would be really nice to see units that the KProbe measures, I think it would help us to decide which burned discs (if any indeed) are within that limit on a particular LiteOn dvd reader.

regards,
Halcyon

PS Two interesting references on DVD error correction that I found. Please tell me if you know of more (print or web):

AN INTRODUCTION TO THE PHYSICAL BASICS OF DISC AND DRIVE - parts I and II
http://www.opticaldisc-systems.com/2002 ... SICS76.htm

http://www.opticaldisc-systems.com/2002NovDec/DVD80.htm

Test bench: DVD blanks (C'T Magazine)
http://www.thewholewideweb.com/forum/to ... PIC_ID=167

[cfitz]

I found out about a utility called DVD Info Pro:

http://www.dvdinfopro.com/index.php

It has supposedly some kind of error test function (havent got any idea as to what exactly it measures and how).

I got the same results with it as with CD Speed's scan disc: all green blocks:



Note that DVDInfoPro is adware, although the ad bits aren't too intrusive and obnoxious.

My guess is that both DVDInfoPro and CD Speed are doing the same test, and that test simply marks each sector as readable/unreadble depending on whether or not there are any PO failures.

On a related note, the dvd specs define an upper limit recommendation of 280 inner parity errored rows in eight consecutive rows of ECC blocks. I don't know if MediaTek chipset reports PI errors per block or by row.

If it reports only by block, then I'm not sure it's possible to exactly calculate how many rows had an error in eight consecutive blocks. Of course we can use the average of 208 rows per ECC block and assuming relatively random distribution, this could give us a rough estimate.

In addition, DVD+R specification seems to require an additional maximum of 4 uncorrectable errors per one ECC block after the PI stage errors have been corrected for.

Now, it would be really nice to see units that the KProbe measures

I had mentioned the max 280 PI errors per 8 consecutive ECC blocks criterion for -R before, but didn't know about the extra requirement for +R. And I did ask about the exact units Karr uses, but didn't get an answer. This may be like the seemingly unresolvable question of exactly what C1/C2 units are being used.

cfitz