Jitter Questions

[VideoRoy]

Since I purchased the dw1655 to help with the scanning of my archives I have been paying more attention to jitter. I rarely scanned in my PX-716A since it is my video production rig and my DRU-710A did not measure jitter.

From all the reading I have done a good rule of thumb is 8% or less based on the DVD spec but it also depends on how your other error rates look. I noticed that the PX-716A routinely produces discs with Avg <8% jitter (but just under 8%) but spikes to 11% at times. The dw1655 produces discs with Avg <6%.

So my questions:

1. From practical experience of the users here, how much it too much jitter to hinder play back or just reading a disc on a computer?

2. Can PC drives compensate better for jitter that DVD players? (my gut says yes).

I create 2 kinds of discs. First DVD movies of course and the second in raw footage archives in UDF format. I am thinking I do not need to worry as much about jitter on the data discs.

I do not want to overthink this but just want some help with some good guidelines. The forum here has helped me so much with drive selection and technical questions I really appreciate it.

Here is a typical scan with my preferred media, burned in the PX-716A and scanned with the dw1655.

Thanks.

[VideoRoy]

One more produced on my Pioneer DVD recorder. It was my understanding that Pioneer are some of the best drives for jitter control.

[Ian]

Ideally, when testing with a BenQ drive, you don't want the jitter levels to go above 12%.

[VideoRoy]

Thanks Ian.

It looks like I am in pretty good shape in all my drive / media combinations that I have tried.

[dolphinius_rex]

Ideally, when testing with a BenQ drive, you don't want the jitter levels to go above 12%.

Even if it goes above 12%, MOST drives will still have no problem reading it, but that's when you get into the "potential compatability problems" zone.

For my own use, I don't mind discs <13%, but when I'm reviewing a drive/disc combination, anything at 12% or more receives a fail.

[VideoRoy]

Even if it goes above 12%, MOST drives will still have no problem reading it, but that's when you get into the "potential compatability problems" zone.

For my own use, I don't mind discs <13%, but when I'm reviewing a drive/disc combination, anything at 12% or more receives a fail.

Good feedback, thanks!

Not trying it generalize too much, but is the 12% rule the same for all drives / discs? For instance if I do a test in my PX716A should I expect the jitter to be similar to the DW1655? Unlike PIE/PIF which I have seen can be somewhat different based on the error correction code in the drive.

For my archive project I am failing anything that spikes >=12% in the DW1655 scans but I am also looking at the overall errors on the other parameters.

*EDIT* - BTW I am scanning everything at 4x CLV in the DW1655 to be comparable to my other drives.

[dolphinius_rex]

Even if it goes above 12%, MOST drives will still have no problem reading it, but that's when you get into the "potential compatability problems" zone.

For my own use, I don't mind discs <13%, but when I'm reviewing a drive/disc combination, anything at 12% or more receives a fail.

Good feedback, thanks!

Not trying it generalize too much, but is the 12% rule the same for all drives / discs? For instance if I do a test in my PX716A should I expect the jitter to be similar to the DW1655? Unlike PIE/PIF which I have seen can be somewhat different based on the error correction code in the drive.

For my archive project I am failing anything that spikes >=12% in the DW1655 scans but I am also looking at the overall errors on the other parameters.

*EDIT* - BTW I am scanning everything at 4x CLV in the DW1655 to be comparable to my other drives.

No, the Jitter rule doesn't apply to Plextor drives. Why? Because Plextor drives don't make any sense when testing for jitter. BenQ's drives actually are relatively similar to CATS jitter testing, whereas Plextors doesn't make a heck of a lot of sense from what I've seen. I pretty much ignore it on my Plextor scans (But the BETA testing on the other hand, now THAT I pay close attention to!).

Regarding BenQ testing, please confirm are you testing at 4x CAV or 4x CLV? BenQ drives can NOT test at CLV properly, and that's why Nero CD/DVD Speed is the only one to support it (big mistake in my opinion!). It seems to grate against the drive in some way, and causes unreliable results. There's been some testing of this on CDFreaks.

Personally, I still think BenQ's should be used at 8x CAV. Comparing the scans with other drives is fine, but don't expect that keeping the reading speeds the same will make the scans more comparable. You're still comparing apples to oranges basically.

[VideoRoy]

Regarding BenQ testing, please confirm are you testing at 4x CAV or 4x CLV? BenQ drives can NOT test at CLV properly, and that's why Nero CD/DVD Speed is the only one to support it (big mistake in my opinion!). It seems to grate against the drive in some way, and causes unreliable results. There's been some testing of this on CDFreaks.

Personally, I still think BenQ's should be used at 8x CAV. Comparing the scans with other drives is fine, but don't expect that keeping the reading speeds the same will make the scans more comparable. You're still comparing apples to oranges basically.

Yes, I just started testing at CLV and actually it *seems* to be working fairly well. If I may be getting false readings I will go back to CAV or what ever the default was before Nero CDSpeed started supporting CLV. I am pretty sure it was CAV.

You have helped me understand that I cannot really compare drives against each other in the past and I appreciate that. I was really asking is 12% a standard assuming you have equipment that can test accurately.

So if I had drive "A" and drive "B" that could test jitter properly should I expect to get the same jitter results scale assuming everything else is equal?

Here is a sample scan since I have been using CLV.

[Francksoy]

Hi there

[Buck] pointed me to this very interesting thread...

No, the Jitter rule doesn't apply to Plextor drives. Why? Because Plextor drives don't make any sense when testing for jitter. BenQ's drives actually are relatively similar to CATS jitter testing, whereas Plextors doesn't make a heck of a lot of sense from what I've seen.

How did you compare the CATS jitter reporting with the Plextor and Benq ones?

I'm asking since the article at Cdfreaks "CATS vs. Homemade scans" found the opposite, that the CATS and Plextor jitter plots were very similar, thought the Philips one (a rebadged Benq 1620) was very different...

Now I'm confused!

Also if you apply this:

when I'm reviewing a drive/disc combination, anything at 12% or more receives a fail.

, and you use a Benq for the jitter plot, it means that all 8X Nec Z-CLV -R burns will fail in your tests...? Benq drives report extremely high jitter in the 6X area of Nec 8X Z-CLV -R burns... whatever the MID.

So do you imply that NEC drives have a major flaw in their Z-CLV strategies, that only the jitter plot from Benq scans reveal?

[VideoRoy]

Interesting scan Francksoy. I assume CDSpeed gave you a red indicator for the maximum Jitter in the results?

I burn most of my video projects on my PX-716A and the PIE/PIF are always excellent but the Jitter seems to run on the high side hitting the 12% mark once in a while. I burn most of the video projects at 4x on Verbatim 8x and everything looks great except the Jitter.

This is why I was trying to understand this topic better. If I take an image of a disc I burned on the PX-716A and burn it on the DW1655 I get much lower Jitter with all other variables being equal.

I rarely scan with the PX-716A because it is fairly slow and I use that particular rig for production only.

[Francksoy]

Interesting scan Francksoy. I assume CDSpeed gave you a red indicator for the maximum Jitter in the results?

No, it doesn't... only PIF levels have "official" maximums in CDSpeed, and according colors/background colors. PIE and jitter figures are entirely left to the user's appreciation...

Er... I just notice that I posted a @12X scan.
Sorry about that. Before someone asks, the jitter levels reported for the 6X writing area are the same in @8X scans of my NEC burns.

[VideoRoy]

Interesting scan Francksoy. I assume CDSpeed gave you a red indicator for the maximum Jitter in the results?

No, it doesn't... only PIF levels have "official" maximums in CDSpeed, and according colors/background colors. PIE and jitter figures are entirely left to the user's appreciation...

This is not true at least with CDSpeed 4.50 and the DW1655. I do not have a scan handy with the results page but I will find one this evening. The Maxium Jitter number will be GREEN even at 11.9% but if I hit 12.1 % the number turns RED as it does on PIF that is out of range. I picked those 2 numbers because I happen to hit discs with both of them yesterday.

I say this is true at least with CDSpeed 4.50 because I have not tried an older version with my DW1655 and before that I did not have a drive that showed Jitter in CDSpeed.

Not sure if the varies by drive or not.

[Francksoy]

Thanks for the heads-up. I've been using 4.50 for a week or so now, but as I didn't happen to come across a disc with high jitter, didn't notice this. Interesting.

[VideoRoy]

Thanks for the heads-up. I've been using 4.50 for a week or so now, but as I didn't happen to come across a disc with high jitter, didn't notice this. Interesting.

So what is considered acceptable Jitter on your NEC burned discs?

It appears to me that Jitter is a function of the drive and not the media. I am familar with Jitter in electronic components and design did not know much about it on optical media.

[Francksoy]

Just tried again, jitter line is still pink, and stays pink over 12%. Would you post a scan showing these color patterns you mention?

[Francksoy]

So what is considered acceptable Jitter on your NEC burned discs?

I can't reply to this question. This topic is controversial and I can't make my mind up.

It appears to me that Jitter is a function of the drive and not the media.

It's not clock jitter per se, but disc (physical) jitter that is reported, and this is direclty linked to the recording quality. Disc jitter is caused by physical deviation (from the "perfect" shape and spacing) of the pits and lands burned in the dye, and other causes like mechanical properties of the disc, the precise alignement of the reading laser... I'm not an expert in this area so I can't give you better explanations.

Overall jitter levels are both dependant on: the disc burning quality and mechanical properties, and the drive's reading abilities. Just like anything that is reported, actually. PIE/PIF errors reported also depend both on the disc and the drive.

[VideoRoy]

Thanks for your help. This is pretty interesting and I appreciate the experience and knowledge of others.

I cannot attribute high Jitter with any problems I have had on discs but I will be more aware of it as a possible factor going forward.

[[buck]]

Just tried again, jitter line is still pink, and stays pink over 12%. Would you post a scan showing these color patterns you mention?

I'm pretty sure VideoRoy is referring to the jitter entry in that results box that pops up after a scan is completed, in CD Speed.

[VideoRoy]

Sorry did not see that second post. Thanks Buck.

Just tried again, jitter line is still pink, and stays pink over 12%. Would you post a scan showing these color patterns you mention?

I'm pretty sure VideoRoy is referring to the jitter entry in that results box that pops up after a scan is completed, in CD Speed.

Yes that is correct. Sorry for the confusion.

Actually I wish there were some of the color coded indicators on the graphical page as well.

[Francksoy]

OK I got it - didn't notice it because I never look at the stats, except when interested in the number of samples actually taken into account during scans...

[VideoRoy]

I guess this does leave me with one final question. How was the 12% derived for CDSpeed? All the specs I have seen say 8% for DVD-ROM / -R and 9% for -RW as maximums.

[Francksoy]

Specs have to be somewhat adapted for each drive. They all report differently! This 8% spec is for ECMA-compliant scanners, like CATS. It's been observed that Benq drives report higher levels, and unofficial specs were adapted accordingly. At least that's my best guess.

But don't take my word for it, because despite being quite a die-hard scanner, the topic of jitter reporting is still confusing to me (thus my initial questions to Dolphinius_Rex). I understand pretty well what is jitter, but how it's actually calculated and reported by end-user drives, this is still extremely blurry to me.

[dolphinius_rex]

Hey Franksoy,

Sorry I've been MIA lately, I was in the USA on vacation for a while

I did the comparison with CATS VERY similarily to how I did it with the LiteON SHW-1635S in my recent review:
http://www.cdrlabs.com/reviews/index.ph ... e=Features
(scroll down about half way)

I have one CATS test currently, and will hopefully be getting some more professional end tests for comparisons in the next month or so.

As for my opinion on the whole NEC thing.... I have no idea what to say really. But I *CAN* say I have no trust or belief in the Plextor drive series as being reliable jitter testers. I'm not even entirely sure how to read percentage scores off of them!

As for why 12% is considered the maximum by myself and others.... well, 8% is hard to come across (even among PRESSED DVDs). 12% seems a better number, and even really picky players with poor jitter tolerances can handle up to just under 12% or so.

[VideoRoy]

As for why 12% is considered the maximum by myself and others.... well, 8% is hard to come across (even among PRESSED DVDs). 12% seems a better number, and even really picky players with poor jitter tolerances can handle up to just under 12% or so.

Thanks for the good information as usual

For this particular project I have been rejecting dics that spiked to >12% but then I started to wonder if I really need to worry about spikes. The averages seem to be in a pretty good range. Last night I also went back and rescaned a couple immediately after they spike to 12% or more and the 2nd & 3rd times they did not show the spike.

So I am chalking the spikes up to a test anomaly or even possible other PC activity although I do not usually run anything else while scanning and I have the PC tweaked with very few running programs / services.

Great discussion!