[jeffkrol]

yup, thought i wish pentax wouldnt take another 5 weeks to tell me they couldnt reproduce the problem.

interesting info you got there on nikon. i'll go tho the store tomorrow and probably buy a Km so i dont miss a camera for the holidays.

thanks

no problems. Intermittent problems are always a bear...
Like the engine noise that's not there when you take it to the garage....

I assume Pentax didn't go out in their mountain air and let the camera freeze before testing it
They should have just replaced it the first time.

UPDATE: Appears to be bad ADC/clock circuit.
http://books.google.com/books?id=OZVeoomxxhYC&pg=PA17&lpg=PA17&dq=types+of+f...esult#PPA24,M1
See pg 23-24........

---

Last edited by jeffkrol; 12-15-2008 at 05:46 PM.

[jeffkrol]

Here is one clear example:

Send yours in as well..... The more sent back, the quicker it's "fixed".......

[raider]

I remember someone mentioned that the images which show vertical banding can be resolved by using a software. Can't remember the name though. Anyone?

[jeffkrol]

I remember someone mentioned that the images which show vertical banding can be resolved by using a software. Can't remember the name though. Anyone?

Silkypics can take it out apparently.
SILKYPIX:Use example of "Geometric NR"

[Ray Pulley]

A little bit condescending with your closing sarcasm.

My "helpful" advice is that you get over it.

Getting back to the REAL ISSUE, it is reasonable to assume that the K20D should function perfectly at all the temperatures this would normally encounter in the real world.

No it is not. It is reasonable to assume that it will work as advertised under the specified operating conditions, and that goes for all brands.

Ray

[Ray Pulley]

i don't know what your referring to about the 80's, is it the expression "dude" i used or something else. english is not my first language.

Yes "Dude" was a popular English saying in the 1980's. Note the smiley after my comment.

i repeat myself though, your posts are not very interesting, if you don't want to help in there, go have some fun elsewhere.

I will post when and where I choose, just as you apparently do. Your permission is not necessary.

fyi, this problem here is very particular (see the fft i posted). enough that pentax engineers definitely need to pay attention. and they're probably actually doing so for some time, but not talking about it, because they haven't sorted it out and don't want to do bad advertisement.

Yep, I am sure that you know EXACTLY what the Pentax engineers are thinking and doing, and furthermore, they will certainly fix this problem in the defective K20D design due to your posts here...

Ray

[pentaxmz]

I think you can use this tool

PhotoME - Exif, IPTC & ICC Metadata Editor

i use exiftool in command line under linux

on any images the data is in the exif information

Thanks for the tip on PhotoME. Very NICE program. Odd that PhotoME offers all this information when not even Pentax's Photo Browser/Laboratory displays this information.

Thanks again.

Unfortunately, I did not have enough daylight time to take enough photographs (the sun sets at 4:14pm this far north), however, my initial tests do not exhibit the vertical banding problem.

The one issue I had noticed is that my focus seems too soft. Apparently, there is a way to correct this?

Unless my Tamron SP AF 24-135mm F3.5-5.6 AD AL (190D) Gold Anniversary isn't that good. None of my lenses were designed for DSLRs but I haven't seen any vignetting yet so this may not be a problem.

I took photos of the same image at every available ISO (I omitted 100 ISO because there was too much shake (1/8 sec exposure).

I also noted the camera's internal temperature in each photo never went below +15 degrees Celsius; that is despite it being about -12 c. I wasn't outside long enough but I promise to give the camera a real workout this weekend.

Info from PhotoME:

Pentax Model: K20D
Manufacture Date?: 2008-02-09
Camera has been serviced?: Serviced bit not set (Value: 2?)
Production ID?: 3
Internal Serial Number?: 42480

Take care!

Attached Images

 

[pentaxmz]

Apparently it occurres in the Nikon d200 as well.......
Maybe searching the Nikon sites will reveal the cause.
guess I'm a bit surprised in the fact it's in a 20d w/ a CMOS sensor. Leads me to believe it's a problem not related to the sensor but more to do w/ a bad board/ ADC/ power supply....whatever

Nikon D200 Reviewed. Noise Performance
Discussion Forums @ Nikonians - D200 and banding
CCD Noise Sources

Great academic article but it is not the same issue that some people are seeing with the Pentax K20D.

[pentaxmz]

no problems. Intermittent problems are always a bear...
Like the engine noise that's not there when you take it to the garage....

I assume Pentax didn't go out in their mountain air and let the camera freeze before testing it
They should have just replaced it the first time.

UPDATE: Appears to be bad ADC/clock circuit.
The Image Processing Handbook - Google Book Search
See pg 23-24........

That's it! Good find!

[pentaxmz]

Geometric noise problem (vertical banding) might be related to when the camera was manufactured.

Problem is evident:

Manufacture Date : 2007:12:22
Production Code : 2.1
Internal Serial Number : 14901

=========================

Problem is not evident:

Manufacture Date: 2008-02-09
Production Code: 3
Internal Serial Number: 42480


Please fill in your values under one of the above headings. This information was obtained from EXIF data using the free program: PhotoME from: PhotoME - Exif, IPTC & ICC Metadata Editor

[raider]

Thanks. That's it! Silkypix.

So if silkpix can remove the vertical banding, is it still a big issue that the k20D exhibit banding in only certain circumstances or the banding effects cannot be totally removed via software?

[phler2]

as said, it's an issue, because, first it adds a manual work on your images that you shouldn't have to do
second it's like putting eyeglasses to the device to correct its perception issues. the thing is just suppose to work
third, if some k20d don't have the problem then why should mine have it

according to jeffkroll

UPDATE: Appears to be bad ADC/clock circuit.
http://books.google.com/books?id=OZV...esult#PPA24,M1
See pg 23-24.......

so it does seem to come from the clock, which randomly interferes in a quite destructive way. hence the device has to be replace, because i doubt they can change only parts of it in that case.

some people had the problem and it took them 3 sent back to pentax before they finally change the camera... (as read on the dpreview thread on the subject)

@ray, sure you can post here or anywhere, you don't need my permission. does it makes you feel smart to be condescending?
no i don't know what pentax engineer think, but i do suppose they do their job rightfully as engineers who care. in that perspective. if i where working on that camera, i would try to enhance the design so the clock interferes less often with the AD converter.

[Cuchulainn]

I've been seeing it. Here are my camera details:

Manufacture Date : 2008:02:19
Production Code : 3
Internal Serial Number : 60267

Going to wait till after christmas to get it fixed - I went by Jessops last week and they agreed that it is a problem and that they would help me get it fixed by Pentax.

[pentaxmz]

I've been seeing it. Here are my camera details:

Manufacture Date : 2008:02:19
Production Code : 3
Internal Serial Number : 60267

Going to wait till after christmas to get it fixed - I went by Jessops last week and they agreed that it is a problem and that they would help me get it fixed by Pentax.

Wow, that concerns me because my camera is just a little bit older than yours.

Does the banding appear all the time or in specific conditions. I have only taken about 20 photos thus far and I have not encountered the banding yet.

What do I have to do to encourage the banding problem to appear (if it exists in my camera)?

This is very preliminary but it seems as though the problem may not be due to a specific production run or older components.

Perhaps the issue:
- component(s) prone to failure (and I'm not affected because I haven't
taken enough photos yet) or or simply a random failure (some do/will,
some do not/will not).

Did the K20D ALWAYS exhibit this problem or did it appear eventually?

My picture count is about: 300 (this info is in the EXIF) what is yours?

- component(s) are temperature sensitive. I think Phillip suggested this is not
the case. What is the temperature recorded for the photos that exhibit this
problem? The lowest internal temperature in my photos, thus far, is 15 c.

- component(s) exhibit a failure only during specific settings. My first test photos
were all taken in Av mode at all ISOs. Perhaps I should take other settings?
What are the specific settings used when you see this problem?

Anything else?

[jeffkrol]

second it's like putting eyeglasses to the device to correct its perception issues.

Not quite sure that this applies. If it's just a rounding error caused by a clock very little detail if any will be lost due to recalculating the correct values. From the Silkypics example not only is it apparently easily corrected, detail does not seem to be smeared as in other forms of NR. Just a minor visible observation. Again, it shouldn't be there but it is much more handlable then the k10 VPN, which was way more random and hard to deal with without destroying some detail