BenQ DW1600A Preview - 16x DVD+R Burning

[Ian]

OC-Freak has a short preview of BenQ's new 16x DVD+R/RW drive, the DW1600:

http://www.cdfreaks.com/article/145

Our DW1600A should be showing up this week.. hopefully. I have a feeling its on a slow boat from China.

[aviationwiz]

I posted this there, and I'll post it here too. The times they got at 16x were slower than what you got at 12x on the PX-712...

Benq has something to work on.

[shimman]

the wopc2 kicked in too much, philips engineers need to think about the balance between jitter values & writing error rates. even though the writing quality in terms of pi/po is good, it must be the high jitter values made reading problems @ high speed.

though very interesting it seems 5.5min is about the limit for single layer full size burning.

wouldn't it be wonderful if 712a can do 12x -r buring at least + booktype change...dl support would be nice, too

[Halc]

PIe/POe and jitter rates are not mutually exclusive.

In fact, jitter that the drive sees from a disc, is one of the causes for PIe/POe at the chip level.

The question is, how do they balance rotational speed, WOPC, BLER OPC and TC, when they are not calibrated with each other.

[dolphinius_rex]

I'm not too impressed. Why did they choose to put more credence on a disproven drive/software combo, then a strict transfer rate test? I'd say the writing quality even at that slow speed for 16x, was not that good

[Ian]

I posted this there, and I'll post it here too. The times they got at 16x were slower than what you got at 12x on the PX-712...

It's not faster. They're writing more data than I do (4.4gb vs 4gb). Even then, if you look at the CD Speed tests, the BenQ is only a few seconds faster than the PX-712A.

Keep in mind too that this is an early sample. Things will (hopefully) improve by the time the drive ships in July.

On a related note, BenQ seems to be sending these samples out to a lot of publications, probably in an attempt to head off some of the 12x drives that have started shipping.

[dolphinius_rex]

hehehe, even if the drive slows down to be the same speed as 12x, people will buy the BenQ because it is a "16x" drive

I have to give them credit, BenQ has been doing some incredible marketing lately! Too bad their products aren't that impressive in my opinion.

[Bhairav]

I have to give them credit, BenQ has been doing some incredible marketing lately! Too bad their products aren't that impressive in my opinion.

COmpletely agree, their CDRW drives sucked. Died in 2-3 months, their firmware flashing was SCARY! You never knew what was about to happen

[pranav81]

Well,my brother has a 52X BenQ CD-ROM and I was shocked that it read a severely scratched disc that was not read by Asus 52X CD-Writer.I like that CD-ROM drive.It shows "50X" read speed in Infotool and 52X is stamped on the tray of the drive.


::Pranav::

[shimman]

jitter is caused by incorrect size of lands & pits & pi/po are caused by incorrectly recognized lands & pits. they are not somewhat related because too long & short lands & pits can be incorrectly recognized & ecc code kicks in.

unlike pressed disks, recordable disks use dye to have two different reflective indeies representing land and pit. when laser did not put enough power on dye, the reflective index could be somewhere between lands & pits. this is probably the main cause of pi/po errors; causes of this thing is physical & dye irregularities & spindle servo & laser power fluctuation & viburations

i am suspecting that cd/dvd speed is not reporting actual spindle servo speed but amount of data transfered to the drive. because spindle servo cannot be controlled finely enough the drive probably recalibrate laser power then wait for the troubling spot to be come back on the next rotation. then write the data, but because the troubling area was not written correctly, the burner cannot really continuously write data as if data was written continusly though good enough. this can increase jitter, thus @ higher linear speed, it would be difficult to distinguish between errors & jitter, thus the drive slowed down bit to cope with jitters.

if cdfreaks.com had cat machine or measured jitter with 712a, it would have fairly big values. imho, wopc2 was also tuned for better k's probe pi/po values to impress dvdr fanatics like me

of couse, this is just a speculation

[VEFF]

My DW822A-OC2 has worked flawlessly.
I am very happy with it, and will definitely buy a 1600A, unless another 16X DVD burner eclipses it in terms of features and/or performance.

[Halc]

jitter is caused by incorrect size of lands & pits & pi/po are caused by incorrectly recognized lands & pits.

That is only half the answer.

DVD measurements are done with two possible ways:

1) non-rotational microscopy based optical imaging (low level, mostly true disc characteristics)
2) rotational dvd transport like CATS/other Pulsetec variants (indirect measures mostly)

Jitter as per DVD standard (DC tilt peak normalized to channel avg) is both specified and professionally measured using rotational devices only.

It is not a true measure of land/pit change structure modulation, but a measure of how a rotational device sees many low level characteristics of the disc and how they affect the signal that the drive reads.

In short: jitter is a measure of the disc and the drive as a combo (not separately). It's a synergistic measure.

So, jitter when measured with a rotational tranport is always a combination of the disc and drive. It is not a true measure of disc characteristics, but rather an indirect one.

This is different from CD-R where one is more accustomed to seeing references and indirect measures to pit/land length variations.

This is not the way jitter is measured on DVD.

somewhere between lands & pits. this is probably the main cause of pi/po errors; causes of this thing is physical & dye irregularities & spindle servo & laser power fluctuation & viburations

Again, only half the answer. Other characteristics, like tracking, refletivity and true disc characteristics like birifringence, thickness, etc play into this. Yes, it's arguable that they induce jitter, but it's important to note that they do so without affecting the pit geometry optical modulation.

So, it's not all down to jitter (as you have defined it). Even if it was, it wouldn't be that easy to make a dvd burner that gives very good jitter figures, because at least three parameters opf the drive must be optimized together (tracking/focus offset and slicer margins), to find lowest level of jitter.

increase jitter, thus @ higher linear speed, it would be difficult to distinguish between errors & jitter, thus the drive slowed down bit to cope with jitters.

Distinguish between errors and jitter? Eh, I lost you there...

Let me repeat:

1) jitter is not defined/measured as pit/land deformations on dvd specs

2) jitter is not the only thing that causes causal PIE/POE results

3) jitter and error rates are linked, but they don't map 1:1 and they change from one drive to another

if cdfreaks.com had cat machine or measured jitter with 712a, it would have fairly big values.

And you base this on what exactly?

I'm not saying it's not true, but I can't see the logical path here. Care to fill me in?

Again, please understand that jitter is a measure of the drive + disc. What the drive sees on chip level after it has read the disc.

Even the beloved Pulsetec is calibrated on industry averages. While the DVD specification gives detailed outline on how to measure jitter and what are the peak levels, it is important to understan that:

A) consumer drives do not measure true peak tilt dc jitter

B) there are several ways to get to certain amount of jitter, when jitter is measured on Pulsetec. However, these several ways are more than likely to give different jitter readings on consumer drives, because the lower level tolerances of the consumer drives cause them to see different amounts of jitter.

imho, wopc2 was also tuned for better k's probe pi/po values to impress dvdr fanatics like me

For me, the jury is still out on BenQ DW1600A quality. I'm gonna need a BenQ myself, to get a Nexperia chipset into the measurement mix. On a theoretical level, what BenQ does with TC, WOPC and BLER PC is very good. Their CATS scans on previous models are very good (both causal and indirect measures).

I don't really think they would be stupid enough to try and calibrate their burners to any other standard than the industry average (i.e. Pulsetec) and own conformance/compatibility tests.

Calibrating them to give good measures on some LiteOn drives would be very foolhardy as 99% of the buying population has no idea what kProbe is and LiteOn isn't the market leader in dvd-burners (or not necessarily even in dvd-roms, not to mention rack hifi dvd players), on all of which the burned discs should ideally work without a hitch.

Now, don't get me wrong. I think you make good points, but I don't agree on all of the interpretations you make about the dvd specifications or dvd standard measurement techniques.

Further, I strongly object to general/growing tendency of using kProbe scans as some sort of measure other than showing how much errors does the tester drive see, when reading a particular disc.

I've been guilty of the same universality thinking myself, but have hopefully seen and accepted the error of my ways :)

regards,
halcyon

[shimman]

halc, my post must have bugged a lot for you writing that much with 5 specific quotes i was impressed with your depth in knowledge

i guess i am not a good writer since you misunderstood what i intended to write. :00ps:

it is true that there are other things that can affect pi/po & jitter, but other things you mentioned are from disks themselves (if not most). it is the burner's ability to cope with those things. i think that's why many are trying to find which drives are better dvd burners.

i think that's why k's probe is popular as an indication to tell how good the burning process is. the values probably off from what engineers see, but high values of pi/po errors reported from k's probe probably indicates low quality burning process.

if most dvd burners/readers have similar reading abilities, i don't think end users need worry about whether burnt disks are rigorously adhering to the dvd standard. i don't mind if my burnt disks can be read by most dvd burners/readers without any problem even if those disks are not adhering to the dvd standard rigorously

16x benq burner's k's probe looked impressive, but the reading transfer test was disappointing. then one more thing came in my mind; that was jitter; i think jitter is the culprit of slowing down of reading speed when pi/po errors are low. since errors are correctable, drives don't need to retry reading problematic area, but i think bad geometry of pits & lands can cause too much to cope for most dvd readers to read @ high speed.

this suspected high jitter values are caused by wopc2 in my opinion.

even though 99% of population out there might not know or care about pi/po/jitter results; there were few cases when mfgs cheated benchmarks to impress fanatics like me and others. some infamous examples are ati/nvidia's anisotropic filter cheating & msi's cpu heat modulated auto overclocing & higher than rated fsb


i think i wrote too much, sorry

[jase]

I thought the burn was fairly impressive considering the way it was written -- essentially that disc is equivalent to a disc where the burnproof technology has kicked in repeatedly. Potentially not good.

Interesting that the drive *defaults* to 16x on discs it doesn't recognise.

[Halc]

Shinman,

nah, you didn't bug me. I'm just using this forum as a method to my madness: namely trying to understand measurement issues.

When there are arguments which I don't share on all points, I try to present counterarguments. Usually I learn this way.

So, it wasn't directed at you as a person and I really appreciate you for writing and commenting on the benq review.

I don't pretend to be an expert on this issue either, but I want to learn more and it's much easier if we discuss these things together.

I know I'm often too terse and can come off as aggressive in my writings, but that's just the surface , believe me :) I'm trying to be as accurate in my expression as possible and to formulate exact questions. Often I fail to consider how other people will see my writings. I really hope I could write in Finnish, I'd have much less misunderstandings :)

So, apologies, if you felt I was attacking you.

Back to topic of BenQ and measurements:

it is true that there are other things that can affect pi/po & jitter, but other things you mentioned are from disks themselves (if not most).

Probably most are from the discs themselves. However, as I stated elsewhere, there are at least three different reader characteristics that greatly influence how much jitter a reader sees on a particular disc.

There are no absolute reference standards on how to calibrate these three characteristics.

This is the reason why some drives see more jitter than others when all drives scan just one particular disc.

it is the burner's ability to cope with those things. i think that's why many are trying to find which drives are better dvd burners.

Perfectly understandable (to try and find what is a good burner).

However, our current kProbe scans, do not enable us to really decide whether the disc was good, the burner was good or the reader was good (or some combintion of these).

When we have a lousy test result from an MCC disc burned with BenQ and scanned on LiteOn 812, we can say that MCC media burned at that speed with BenQ (using current firmware) is not very compatible with LiteOn 812 (using the read speed/firmware of the LiteOn tested).

It doesn't really prove us something generalisable/universal about the burn quality, disc quality or the burner quality itself.

Sometimes pretty safe conclusions can be drawn, if kprobe measurements are done and supplanted with other measurements from various other types of drives (non-LiteOn).

i think that's why k's probe is popular as an indication to tell how good the burning process is. the values probably off from what engineers see, but high values of pi/po errors reported from k's probe probably indicates low quality burning process.

Again, this is not proven and in fact, a lot of examples show that it is impossible to safely draw conclusions like this.

KProbe can tell you that a certain burn is either problematic or non-problematic (or something in between) using the drive in which you tested it.

It doesn't prove anything universal about the burner quality or even necessarily the media quality in general.

It doesn't even necessarily tell anything very useful about the _general_ quality of the particular burn.

It just tells, whether LiteOn drive liked it or not.

There are cases, where a certain burned disc reads really bad on a LiteOn, but works/measures really well on other drives.

What does it then tell about the burn? About the burner? About the media? About the readers?

It's very hard to draw any reliable conclusions or to put the blame simply on the media, burner or even the reader.

Simple kProbe measurements are not enough for this type of analysis.

if most dvd burners/readers have similar reading abilities, i don't think end users need worry about whether burnt disks are rigorously adhering to the dvd standard.

And herein lies the problem: those dvd burners/readers have quite differing reading capabilities as I've explained above.

Even such thing as jitter tolerance can be almost twice as high on some drives when compared to lesser ones. In addition, drives may see more jitter due to differing reasons.

This can have a huge effec on how the drive is capable of reading various discs.

i don't mind if my burnt disks can be read by most dvd burners/readers without any problem even if those disks are not adhering to the dvd standard rigorously

This quote brings up another interesting point I must make:

There is no single universal truthful measurement tool/method that proves compliance to dvd standards rigorously.

Not even Pulsetec (i.e. CATS/Datarius/etc).

Pulsetec is built on industry averages, using tight tolerances and with particular attention paid to calibration.

But there is no "metric standard" of dvd measurement towards this all dvd measurement equipment should aim for and based on the results of which all burns should be qualified.

All measurement drives are always calibrated to try and get as good performance as possible, but there are many ways to defined good performance, so different testers have different ways of achieving this.

Yes, Pulsetec based units are probably the best we have (currently), but they are still based on averages.

Or to put it in other ways: if I only had single measurement that I had to rely on, of course it would be a Pulsetec based analyzer.

However, I'd rather have multiple different transports doing the measurements, because this is more likely to reflect real world compliance / readability / compatibility.

There exists no way that I know of, to measure universal adherence to dvd standards (such as PIE, PIF, dc jitter), because those results are always a combination of a disc and the reader.

16x benq burner's k's probe looked impressive, but the reading transfer test was disappointing. then one more thing came in my mind; that was jitter; i think jitter is the culprit of slowing down of reading speed when pi/po errors are low. since errors are correctable, drives don't need to retry reading problematic area, but i think bad geometry of pits & lands can cause too much to cope for most dvd readers to read @ high speed.

Actually I believe in a somewhat different explanation.

As long as overall DC jitter remains sufficiently low, in most cases so does the causal error rate (of course there are exceptions).

BUT, when transfer rate graphs fail, this could be down due to focus/tracking errors, which force the reading drive to slow down.

After slowing down, focus and/or tracking is regained, and the low jitter shows in low causal measures even on equipment that doesn't have a high jitter tolerance.

But the above is a belief, I cannot prove it. You could be right as well.

even though 99% of population out there might not know or care about pi/po/jitter results; there were few cases when mfgs cheated benchmarks to impress fanatics like me and others. some infamous examples are ati/nvidia's anisotropic filter cheating & msi's cpu heat modulated auto overclocing & higher than rated fsb

Indeed. This is a good point.

I hope that manufacturers of dvd burners are not stupid enough to start optimizing for readability of the burns in LiteOn drives (in order to get good kprobe scores).

While this kind of "optimization" may be acceptable in visualization equipment, I don't think it has any place in data storage equipment, where reliability should be amongs the key factors (and not false benchmark performance).

So, let's hope that kProbe does not become the wrongly accepted standard of dvd burner "quality measurement" as it surely cannot cope that role by itself.

regards,
halcyon