New Toy: DVDScan and Feedback Here:

Alexanderino mentioned a lot of what I've noted, but I still have some questions, suggestions and a problem.
Questions:

• are the timing graphs with the pit and land distributions dynamic, i.e. for the last sample, or are they the average over the disc?
• can a jitter and PI/PIF scan be combined like CD-Speed does for BenQ drives? Or are the sampling points too different?
• how is the running total that "seems" to be the PI total determined? In the attached 8x PI/PIFSum8 graph, the value shown is 5535 but the sum of the PI counts from the scrolled scan data window is 17702. From a 4x scan (not shown), the corresponding values were 2755 and 4722. (The second attachment is jitter. See this post for other data from this disc: MCC04, 165P6S MV7U 16x, HT/OS on)

Suggestions:

• Please set the initial window size for the display so that it just fills the window without scrollbars
• I'd like to be able to save chosen speeds and choice of PI or PISum8 for tests from session to session
• I appreciate the possibility of saving the raw measurements (Ctrl-A, Ctrl-C in the scrolled window). Could you make a menu-entry/button for this like the suggested screenshot option?
• please activate the option to suppress realtime graph !! Karr removed it from KProbe because it caused problems with the sampling. I can see this happening here too: the difference between successive samples increases from ca. 25 LBA to over 500 at the end of an 8x scan. I'm reasonably certain that refreshing the display (which requires more time with increasing data) is responsible for this.

I have a problem with the FE/TE tests. For both discs I've tried, the LBA address freezes after ca. 35-37%. The scan continues to the end of the disc and then appears to restart at a lower speed. The third attachmnet is the FE/TE scan for the ancient disc from my previous post (RicohJpn R02 burned with SOHW-1213S fw TS08, 12x ZCLV. This is my "testbed"- I've scanned this disc more than 30x. ). I've marked the apparent end of the disc at LBA=C8454.

G

I've had a hard time finding information about whether Beta and Assymmetry are the same thing or not. The few sources I have found disagree on this.

If you have some information on the subject, could you please post it or post a link? Thanks!

As far as I know FE/TE tests are only reliable when performed on blank discs, and you seem to be performing your FE/TE tests on burned discs?!

This is a great start on a comprehensive new test suite for our Lite-On drives.
Thank you to all involved in bringing this to us!

What I would like to know is how to start interpreting the jitter results.
From the Ecma-279 standards document:
What value to set the DCJit spec at to measure against the above stated standard?
Does the "110" represent "11%" or what? If so then we should set to "90"?

Or is this scale set to something completly different?
Sorry if this sounds like a dumb question, I just want to understand what I see.

Thanks again for the beginnings of this test suite. Looking forward to future enchancements and improvements.

- assuming the graph units are 0.1%, then verticle "160" = 16%, "110" = 11%, etc.
- if above is true, then DCJit are also in units of "0.1%" (110 = 11%)
- appears changing DCJit only moves the "red" reference line
- I say we set DCJit line to 100 (10%) and make Wind/Tom sweat a bit (joking)


Comment collection from WIND:
- You can change the interval value to shorten the testing time.
I usually change it to 0x8000.
The interval value is not used for PI/PIF scan at this version, next version will be.
USB/Multi drives supporting has problems and Tom is working out.
I think there would be soon a new version for you.
Thank you for all the great testing and feedback.

- You can change the orange-red line by yourself, just modify the jitter spec. value. I think if the jitter value is under 11% measured by 165p6s I think that would be OK. So I asked Tom to set the initial value to 110. Of cause you can change to what ever you want, that's just for reference only.
Sometimes, even the jitter value is slightly higher than 110, the pi/pif results are still good. Everything is for reference only. No big deal.

---

Last edited by bichi; 20-05-2006 at 19:53. Reason: spelling

165P6S - #1
MV7U - All ON
Maxell - Prodisc F02, -R, @ x16

Two jitter scans with different settings
- interval = 8000 and speed = 1
- interval = 8000 and speed = 4

---

Last edited by bichi; 20-05-2006 at 20:30. Reason: edit

165P6S - #2N
MV94 - All ON
Maxell - Prodisc F02, -R, @ x16

Two jitter scans with different settings
- different physical drive, same drive model, but with MV94
- same disk as above
- interval = 8000 and speed = 1
- interval = 8000 and speed = 4

- 8000 interval setting, per Wind, does take much less time
- illustrates small result deivations with different FW and hardware
- apologies for graphics fuzzieness, using PS Elements, 641 horz and forced to save in JPG, to stay under image size limit

---

Last edited by bichi; 20-05-2006 at 20:50. Reason: edit

This is a jitter test of a Verbatim DVD+R DL MKM 003
I'm guessing that DVDscan v0.01 Beta is not yet supporting DL medias?

165P6S - #2N
MV8V BS=ROM - All On
MKM 001 +R DL
DVDScan v0.01 beta

Pchilson:
- noticed the same as you
- manually stopped test when "mild" clicking (head seek/thrashing) started on layer 2

---

Last edited by bichi; 20-05-2006 at 21:57. Reason: edit

Another comparison of jitter at 4x and 8x. Once again, clearly the 8x result is very erratic and 4x is a better speed to test at.

Benq 1640 scan for reference, followed by 4x and 8x Jitter tests on Liteon 160P6S. Disc is a Spin-X brand Prodisc R02 burned at 6x in Liteon 160P6S PV8Q SB HT OHT all on.

Wind:

Another small issue to address with the program - When starting a PI/PIF test (and probably other tests, don't know) without spinning up the disc beforehand, the initial errors reported are very elevated/exaggerated. I have seen this from a 'cold' start of a disc tested from the beginning, as well as when I started a test from the middle of the disc. It's not that big of a deal since it is easy to spin up the disc beforehand, but it would be good to have a 'pre-spinup' of the disc before starting a test, such as how Nero CD-DVD Speed does.

No clicking or thrashing here.

Tested media is a Memorex branded RITEK G05
Burned on LG GSA-H10N @ 8x
pic 1: DVDScan Jitter test on Lite-On SHM-165H6S.
pic 2 CD/DVD Speed quality test on BenQ DW1655.

Lower jitter while in the 6x zone and a sudden rise as the drive changes to 8x.
Steady line with a very slight increase across the disc.
Both drives are essentially reporting the same pattern of jitter.

This is very good to see IMO, as it gives more validity to the jitter levels that Benqs have been reporting. I haven't doubted them, but alot of people just shrugged off abrupt increases in jitter reported by Benqs, believing they were 'biased' in Benq-burned discs and that other drives weren't actually burning with sharp increases in jitter. One interesting thing about your two scans though, with all of the other comparisons I've seen so far and all of my comparisons so far, the Liteon reported higher overall jitter levels. In your case it's the opposite (not significantly different though), although the overall scans follow a very similar trend.

From my own comparisons so far, the Liteon tends to show an even more exaggerated jump in jitter levels than my Benq drives. Below is a good example. In particular, I've seen these abrupt increases in jitter in NEC and Liteon burns (as well as some other drives, but those are the two that come to mind immediately). Most burns I've seen from Benq, Plextors (older versions anyway), and most Pioneer burns don't seem to burn with abrupt jumps in jitter. One scenario where I often see abrupt jumps is with Z-CLV burns, although it can happen even with CAV (one example is some 16x Liteon burns, where there is an increase in the jitter in the first 1/3 of the burn where the burn has a momentary slowdown).

disc below is a fake TYG02, burned at 6x in NEC 2510, the burn is a Z-CLV and jitter increase corresponds to the shift in burn speed from 4x to 6x. The two drives show a similar 'story', except the Liteon shows a far larger jump in jitter.

I'm on board with you scoobiedoobie.


Memorex branded CMCMAG AE1
Burned on LG GSA-4166B @ 8x
pic 1: DVDScan Jitter test on Lite-On SHM-165H6S.
pic 2: CD/DVD Speed quality test on BenQ DW1655.
pic 3: CD/DVD Speed quality test on Lite-On SHM-165H6S.

The Lite-On drive shows a definite drop in jitter at the Zone change not quite as dramatic or noticable on the BenQ but you can see it if you look carefully.
Both drives show a decreasing jitter pattern across the disc.
Both drives react to the PIF clump at the 4.1 GB mark.
Again both drives are reporting essentially the same jitter pattern.

---

Last edited by pchilson; 21-05-2006 at 04:48.

It's very interesting to see that jitter measurements on the LiteOn drives show the same abrupt jitter jumps as those we frequently see in BenQ scans of "foreign" burns made on LiteOn, LG, NEC, and (some) Plextor drives.

We shouldn't conclude too much from just a couple of scans, but getting this kind of second opinion is just what I hoped would happen when we got the possibility of performing jitter scans on LiteOn drives.

The second opinion is not the opinion I hoped for however, because I would rather have BenQ drives misreporting jitter than have at least four other brands of drives burn discs with sudden jitter jumps.

The Pit and Land distribution represent the status of each time measurement.Step by step, I think Tom would consider it.This is the measurement times only.I think Tom will check these item by item. There seems a lot to be improved, Tom might have to work very hard in these days.

Yes, you are right on one condition: using standard measurement tool under standard environment.

It's late and I'm getting tired, but I really don't understand your answer to this question, Wind!?

While I agree that it would obviously be better if these jumps in jitter were invalid from a burn quality standpoint, I will admit that I was getting tired of all of the suggestions that Benq's were 'biased' and that the jitter jumps were invalid. The claims had nothing to back them up, and most of the claims to invalidate the jitter levels were basically based on a refusal to believe (aka, denial).

With the tests I've done so far, if anything the Benq was quite a bit more forgiving in the jumps in jitter levels vs. the Liteon, the Liteons seem to be exaggerating the jumps to a much greater degree. Another way of observing this, it seems that the Liteons show a greater sensitivity overall regarding jitter, they show a greater level of variation overall vs. the Benqs, although they both show similar trends.

Wow, bichi, you must got a very good disc or a very good drive.

This is a good suggestion!

Spot on scoobiedoobie, the "claims" were not based on any irrefutable evidence but rather on the observers own "bias".

Again, I'd like to point out that it's all read-relative.