[Jonas B]

Do the test, check the threads at DPR and you'll soon learn that it is not a JPG versus raw thing.

[nosnoop]

Unfortunately, some formerly well-respected contributors 'over there' have taken to shouting down any discussion about VPN,

Ha, I think the reverse is true. Many people, including me, do not see and do not have VPN issues. And if I dare posting my experience in any of the thousands of VPN threads, I would be accused of being a Pentax fanatics, or being in denials! It was those who did not have VPN issues being "shouted down"! Those who did not have VPN would be questioned as to why they posted in the VPN threads, and that their reporting of no VPN would cause Pentax not to take action!

Incredible, isn't it?

[Jonas B]

Incredible, and a quite strange view of it also.

[Alvin]

Jonas, that was an interesting post. I did not look too heavily, but noticed in picture 3 and 6. Won't comment on DPR because I don't have an account there. Does this show up only on the K10D? I haven't noticed something like this in the K100D.

[jfdavis58]

We can avoid the shouting and abrasive behavior of 'over there' if we stick to the facts and pertinent related issues.

Pattern Noise, vertical or horizontal is a well know and reasonably well understood physical phenomena (Google 'Vertical Pattern Noise' and scan for proceedings of IEEE--i.e. technical papers). Most of these authors and all of the reviewers of such proceedings are experts in the field.

There are two solutions: hardware remodeling (preferred), and software re-write(probably necessary for legacy equipment--i.e. that camera in the consumers hand). The software solution is problematic: is there sufficient memory to hold the required code? Is the camera CPU sufficiently fast to execute the code without a noticeable increase in lag time between shot and writing to card/reviewing on LCD?

Currently the assumption is that enough space exists and the CPU is sufficiently fast. (Read the GordonBGood thread at DPR; he appears to have the requisite knowledge of the subject to effect a software solution, independent of Pentax).

That pretty much covers the facts, so what are the issues?

Two issues remain that matter: "Should V/HPN be addressed by the camera manufacturer?" and "does every user require the fix?"

The answer to the first question will be based strictly on market conditions: Do enough users operate their camera in a manner that will cause them to experience V/HPN? In this regard, it's counter productive for either side to 'shout down' the other as such shouting cheapens the argument with useless and meaningless human noise.

Common sense says that if you need a solution to V/HPN then a direct approach to the manufacturer will be more productive than any forum rehash of the issue. Following onward: a loud, persistent, demanding mob outside the front door will gain more than a single dilettante whispering in some minor functionaries ear over cheese and crackers.

The answer to the second question is simple logistics and a method exists for rapid deployment; common sense says that if a fix is developed then it should be made available to everyone.

[Jonas B]

Only the K10D. See it summarized here:
One to Two Stop Reduced Dynamic Range: Pentax SLR Talk Forum: Digital Photography Review

[carpents]

Ha, I think the reverse is true. Many people, including me, do not see and do not have VPN issues. And if I dare posting my experience in any of the thousands of VPN threads, I would be accused of being a Pentax fanatics, or being in denials! It was those who did not have VPN issues being "shouted down"! Those who did not have VPN would be questioned as to why they posted in the VPN threads, and that their reporting of no VPN would cause Pentax not to take action!

Incredible, isn't it?

Sorry, I haven't seen any threads that call people a stupid **** for not seeing banding. I can't say the same for the reverse.

Again, my problem is with the tone, and the shouting down. If you can't have a civil discussion about it, I suspect that you don't have much to say. Someone told me once: "Those with the least knowledge feel the need to speak loudest."

[benjikan]

Hi Benjamin,

Have you tried to reproduce this phenomenon.

Could it be older lenses, different firmware, higher dynamic range, different color light?


Hello;

To be quite frank I have. In fact being that I still have all of my Canon gear, I tested both cameras'. Even though the noise levels at high iso was somewhat cleaner with the Canon, I could not produce VPN. What I did find however was that when under exposing by two to three stops, the two cameras' would produce all kinds of spurious noise. Therefore there was random pattern noise in both cameras'.

It is much less apparent when printing to hard copy. My concern is that much of this is often pixel peeping at 100 percent on screen and not translated to real world support. I.e. print in magazines, PLV's, posters, brochures etc.

Best Wishes
Ben

[dmdctusa]

The best means I have found of verifying VPN is by using Noise Ninja's ability to view channels. In roscot's truck image, I think I can see some banding. When I isolate the blue channel I can see what "may be banding". However, I am inclined to belive that what we are seeing is reflective elements in the picture that we do no normally see when we physically view the scene. Perhaps the K10's sensor is overly senstive in the blue channel but that may also be necessry to give the highest DMax at lower ISO's. You pays your money and you takes your chances.

Anyway, I took the image and did the Noise Ninja profile using the blue channel as a guide. I then ran the noise reduciton with the default adjustments with no sharpening. Any degradation you see is problably due to the JPEG resave after noise reduction is applied. The first is the original image and the second is the filtered. I have seen no noise in my images at 1600 that Noise Ninja (and I imagine PS and other noise plugin's) could not handle providing I used a reasonable exposure.

[mutley]

Hi Benjamin,

Have you tried to reproduce this phenomenon.

Could it be older lenses, different firmware, higher dynamic range, different color light?

The answer to this is "Not as a requirement to get this issue"

Hello;

My concern is that much of this is often pixel peeping at 100 percent on screen and not translated to real world support. I.e. print in magazines, PLV's, posters, brochures etc.

Best Wishes
Ben

The example I posted was not a 100% crop on screen, and no pixel peeping needed to see it. And, that was a shot I just took yesterday when I posted it. Telling Pentax it must be just pixel peeping might mean the issue for them has now ended. Hopefully your guys read some of the threads that are linked here in this thread. Especially GordonBGood's more technical explanations so they won't have to attempt to understand what the problem is to begin with, and what we may be doing "wrong" to get it. It's not a fault of ours.

It is a RAW & jpeg problem. Actually it appears more in RAW because its not reduced to 8 bit.

You mentioned that you would use any tool to get the picture you envision. Including shooting at high ISO if needed. If you combine a situation where you want the noise of high ISO in your photo, plus have various mid-shadow areas in the shot but the model is of course at the right level - you are setting yourself up for some vertical streaky patterns in the noise in those mid shadow areas. Looking at the whole photo, not 100%, you might get distracted by these patterns and so will others. Depending on if you can see the noise in your prints in these shadow areas (and they're not just too dark), there's no reason you would miss those patterns in those prints.

There's enough info out there describing exactly what is happening so you need not invest much time explaining this issue to Pentax for us. If they just do a little intelligent reading and are smart enough to look past the belligerent posts screaming that it is just everyone's bad overworked photography creating this - and read the actual causes - they can get to work on it and we won't have to keep explaining it as if it is elusive.

Larry

[jfdavis58]

Larry,

I'm actually trying to understand 'just this problem'--the one right here in this thread. I see the lines on two different calibrated monitors. What I need to know is what is the foreground object, what exactly, did you do to the image file from the moment it was written to the card until it appears here in the forum.

The posturing and hand waving and howling don't provide those answers. Nor do any of the links and references.

This image has been reduced to fit the forum format-what else?

These are the same questions Pentax will ask.

[mutley]

Larry,

I'm actually trying to understand 'just this problem'--the one right here in this thread. I see the lines on two different calibrated monitors.

This image has been reduced to fit the forum format-what else?

These are the same questions Pentax will ask.

JPG out of the camera on Bright, -2 saturation, -2 sharpness, -2 contrast.

In photoshop CS2:

Slight increase in contrast (+5),

Reduced,

Noiseware Pro NR slightly applied - slightly more for chroma noise,

Sharpened with smart-sharpen .3 pixel at 150%,

Posted. Blue & white out of focus no-texture baseball hat in foreground.

Larry

[*isteve]

There have been many people on different forums expressing their concerns regarding VPN (Vertical Pattern Noise) in the Pentax K10D. Although I haven't experienced this anomaly, those who have done so feel that Pentax have not addressed these concerns. Can a Firmware Update resolve this issue? Is it exclusive to Pentax or does it come down to understanding the limits of the digital domain. I tend to believe that if ones exposure is correct within reason, that this issue does not crop up.

Ben

Camera Performance Anxiety!
Yes its possible to create visible VPN at high ISO without underexposing or pushing a shot. Some folks are accutely sensitive to it even at ISO400 but most people have to have it pointed out they them with little arrows, which gives you some idea how bad it really is most of the time. Some people see it in every shot because they are convinced its there somewhere in the shadows.
Compared to the high ISO performance of the A100 which is there in EVERY shot, or the amp noise on the D80 which is a real bummer in low light night scenes, I would probably make the best of it and get on with my life. I never had a perfect camera yet, so this is just something to bear in mind and work around. I shoot at ISO1000 or 1250 more than 1600 specifically to avoid it and rarely see it to any major degree. I have shot a lot of high ISO shots, and one or two were spoiled by it. A lot more were spoiled for other reasons. Some I could recover in PP to a large extent.
One or two Pentax service personnel have denied being able to see it. That does not constitute an official Pentax statement. How much business Pentax lost because of this issue I dont know. Probably quite a bit. The lather generated around it seems to me disproportionate to the degree of the issue, which was much worse on my DS. Like I said, the perfect camera does not exist so why this has generated such furore I dont know. Over-egged expectations? There are a small number of folk who seem to decide its worth producing a new thread on the issue every week and then flogging it to death on the understanding that such behaviour will force Pentax into providing a solution and in the meantime they seem to be terribly anxious about being in possession of a defective camera which is undoubtedly spoiling their fun (and a lot of other peoples who would never have noticed the problem otherwise).
This is not only deathly boring but probably counter productive. They fixed a more serious long exposure issue, so I think they would have addressed this if they could. However, if the issue proves tricky to fix - which it may well be - the noise created on the forum will NOT influence Pentax to divert resources from new models etc. plus it would be virtual corporate suicide to issue a statement admitting such a problem if there was nothing they could reasonably do about it. What would it achieve except to generate even more negative publicity?
For now, I'm just regarding it as a general characteristic of the camera thats less that optimal, as is the JPEG softness and other minor issues. Pentax do not have infinite resources and this was an entirely new model, and hey, guess what, a few bugs appeared. The *istD and DS also had some known issues which were never fixed but gradual refinement led to the K100D, which has very few. MS Windows has been out for nearly 15 years and it still has bugs! The K10D replacement will undoubtedly extract better performance from the NuCore processor once Pentax's engineers get a bit more practice.
Sure, I hope they fix it too, but another 100 threads on this issue will not make it happen any faster and in the meantime I'm getting some very nice results at ISO1000 that are better than anything I could get out of my DS.

[Richard Day]

Camera Performance Anxiety!
snip .....
Pentax do not have infinite resources and this was an entirely new model, and hey, guess what, a few bugs appeared. The *istD and DS also had some known issues which were never fixed but gradual refinement led to the K100D, which has very few. MS Windows has been out for nearly 15 years and it still has bugs! The K10D replacement will undoubtedly extract better performance from the NuCore processor once Pentax's engineers get a bit more practice.
Sure, I hope they fix it too, but another 100 threads on this issue will not make it happen any faster and in the meantime I'm getting some very nice results at ISO1000 that are better than anything I could get out of my DS.

I strongly endorse Steve's comments. Enough said on the subject!

Let's move on.

[Harald]

I strongly endorse Steve's comments. Enough said on the subject!

Let's move on.

Agreed.... if this "serious" phenomenon has to be pointed out with arrows, well.... I saw those spots, but thought they are some kind of reflections, or flare.

From monster car picture I couldnt find out any bands.... only band-like forms in shadows were windows (left upper corner).

Just came back home yesterday from two week trip to Japan. I photographed several hunreds of pictures in dim lighted dojos (kendo) and surprice, surprice, coudnt find from my pictures anything like this thing called VPN.

(thought VPN is "Virtual Private Network", but working at IT-industry makes me think oddly :-)