It doesn't matter how much pre production testing is done, how many production models used in the field the percentage will always be minimal in comparison to a full porduction run. The odds of identifying a limited glitch or intermittent problem is not very good.
I am not surprised at all, in fact I would say I expected something and it will be remedied in due course.
This is just another example of why I never buy any industrial product anymore during the first few weeks of production. This is inevitable and related to the complexity of the product. In this particular case, there seems to be a number of similar problems (just looked at other forums), so this could be a hardware problem (meaning replacement) or firmware (could be solved with an upgrade).
Anyway, all serious manufacturers (and Nikon is one of them) are aware of those risks and will handle those issues professionally on the longer term (cf D2H, D70...).
Gerald, I wouldn't worry too much, the problem will be handled and I am pretty sure you will get another copy you will be happy with...but I totally understand your frustration, this is not a funny experience to say the least...
I just now checked my images in Capture...I shoot in 100% RAW and I am not seeing banding at least at 100% mag. I see a little noise at ISO 200, but nothing the likes of your image above.
I hope you can get this addressed to your satisfaction.
This is why I have two very expensive bodies...Murphy's Law I also have two cars.... two dogs well, you get the idea.
As well as returning the camera to my dealer, I contacted Nikon about it. Despite sounding surprised and claiming it was the first they'd heard of a banding problem (I wouldn't expect them to say anything else, to be honest), their response after I sent them a sample NEF file was very quick. Within an hour of the file being sent to Nikon Europe I'd been emailed by Nikon UK to the effect that 'there may possibly be a problem with your camera. With this in mind I would suggest that you send the camera to us directly so that we can have an engineer to asses (sic) the problem.'
This sounds to me like the response of a company that has identified and acknowledged a problem and is working hard to locate the cause. Whilst waiting around for my brand new body to be repaired isn't a thrilling prospect (I politely declined and pointed out that my dealer would be supplying me with a new one; the faulty one will be making its way back to Nikon service anyway), it makes perfect sense for Nikon to call in as many of the banding-afflicted samples as possible.
My hunch is that, from a production point of view, it's a problem that will go away pretty quickly.
Just bought my D200 2 days ago and checked all my images. Thanks goodness there's no banding in any of them. Colors and saturation are excellent. Love it and it's comparable to the D2X quality.
As well as returning the camera to my dealer, I contacted Nikon about it. Despite sounding surprised and claiming it was the first they'd heard of a banding problem (I wouldn't expect them to say anything else, to be honest), their response after I sent them a sample NEF file was very quick. Within an hour of the file being sent to Nikon Europe I'd been emailed by Nikon UK to the effect that 'there may possibly be a problem with your camera. With this in mind I would suggest that you send the camera to us directly so that we can have an engineer to asses (sic) the problem.'
This sounds to me like the response of a company that has identified and acknowledged a problem and is working hard to locate the cause. Whilst waiting around for my brand new body to be repaired isn't a thrilling prospect (I politely declined and pointed out that my dealer would be supplying me with a new one; the faulty one will be making its way back to Nikon service anyway), it makes perfect sense for Nikon to call in as many of the banding-afflicted samples as possible.
My hunch is that, from a production point of view, it's a problem that will go away pretty quickly. ...Show more →
This is fantastic news. This is the response I expected from Nikon. The tech I spoke to must have been an exception. I'm really glad to hear it. Gives me my confidence back.
I also expect the problem will go away quickly, and that the camera will be top notch. I think we are VERY early in the production cycle.
Incidentally, I NEVER received the promised response back via e-mail, so never got to send my NEF at all. I may have to follow up on that... I think it still matters because I intend to repurchase the camera after the holidays.
chemprof wrote:
Not exactly what I had in mind when I bought the camera!
Gerald
Thank you for posting this. I immediately looked at the few test images I took ... and lo and behold I got it. It does look like corduroy. What a bummer. I am in the process of returning it to Ritz.
chemprof wrote:
This is 100% crop from the upper right corner (shade, window frame, smooth wall). Note that this is NOT an anomaly. Now that I know what I'm looking for, I can find it in ALL of my images, including out of focus areas in any image. It's most noticable in mid-tone areas.
Gerald
Hi, John! I'll let this thread play out until I get a full JPEG sample. Plus, I can borrow Gerald's crop for a quick assessment. From it, I detect a statistical imbalance between alternating pairs of pixel columns to the amount of almost 3%. This will be increaingly noticeable as the contrast curve is stretched. That column-pairs are involved implies that alternating Bayer cells (2x2 sensor sites in the RGGB pattern) are out of balance (correctable in firmware).
The algorithm I deploy in the following, rebalanced crop of Gerald's does appear to improve matters greatly:
Canon DSLR I have tested (D30, D60, 20D) also exhibit a systematic imbalance, although different in its pattern: the two G measures within a given Bayer 2x2 is inconsistent by about 1%. The effect is seen as a grid or mesh pattern, but often masked by JPEG compression. Noone is perfect, eh?
Based on my quick results, I need to apply a secondary correction to fix some residual banding. Anyone with that full-sized D200 JPEG...?
I no longer have the camera, because I sent it back, but II still have the file. I'm not sure my e-mail can handle it, but perhaps we can try for me to send it to you? This is quite interesting. Thanks for your analysis! Let me know how to send you the file.
Feel free to attempt an e-mail delivery: my server can take it! As you said initially, the low-ISO version will be most suitable, since high-ISO smoothing isn't desired here. Similarly, if you have a HighQuality JPEG, rather than LowQuality, that will limit JPEG-induced smoothing.
My question to those who have had banding is, it seems that this banding does not happen in all your images, I wonder if it has to do with your in-camera settings. Are any of your settings set to "auto", say for contrast or sharpening? This might explain why it is only happening of a sampling of your shots.
rico wrote:
Hi, John! I'll let this thread play out until I get a full JPEG sample. Plus, I can borrow Gerald's crop for a quick assessment. From it, I detect a statistical imbalance between alternating pairs of pixel columns to the amount of almost 3%. This will be increaingly noticeable as the contrast curve is stretched. That column-pairs are involved implies that alternating Bayer cells (2x2 sensor sites in the RGGB pattern) are out of balance (correctable in firmware).
The algorithm I deploy in the following, rebalanced crop of Gerald's does appear to improve matters greatly:
Canon DSLR I have tested (D30, D60, 20D) also exhibit a systematic imbalance, although different in its pattern: the two G measures within a given Bayer 2x2 is inconsistent by about 1%. The effect is seen as a grid or mesh pattern, but often masked by JPEG compression. Noone is perfect, eh? :)
Based on my quick results, I need to apply a secondary correction to fix some residual banding. Anyone with that full-sized D200 JPEG...?...Show more →
It appears to me that all that is being accomplished here, is that you're softening the image---which has a secondary effect of visually deemphasizing the banding artifact.
KURTZ wrote:
My question to those who have had banding is, it seems that this banding does not happen in all your images, I wonder if it has to do with your in-camera settings. Are any of your settings set to "auto", say for contrast or sharpening? This might explain why it is only happening of a sampling of your shots.
Just a thought...
-kurtz
Did you read the whole thread? IMO, those that are seeing this in only some of their images are either not looking in the right areas of their images, have too much detail to see it well, or not looking close enough. After figuring out the pattern of the problem, I can find it in ALL of the images taken with the D200 I just sent back. Low ISO, high ISO, NR, no NR, etc.
It appears to me that all that is being accomplished here, is that you're softening the image---which has a secondary effect of visually deemphasizing the banding artifact.
Scott
I agree that there's detail loss. That's one of the reasons I knew not to keep the D200 I had. I figured any processing would eliminate the apparently high level of detail in the images. Let's see what Rico comes up with with in the next pass, and let's continue to look for an answer from Nikon as well.
genghis45 wrote:
It appears to me that all that is being accomplished here, is that you're softening the image---which has a secondary effect of visually deemphasizing the banding artifact.
I cannot see a decrease in sharpness / contrast. There is a psycho-visual effect that attributes "sharpness" to the banding since the banding is relatively sharp. This is the same effect as adding sharp noise to an image to increase the perception of sharpness. I suspect the algorithm he is using does nothing more than correct the imbalance in the bayerization of the image.
I agree with Tony that banding generates an impression of detail and, when removed, makes the image appear flat. Film-based images enjoy the same effect - until grain lands in the sky! Indeed, the algorithm merely affects alternating Bayer columns: methods like Gaussian blur and sharpening are unsuitable for this D200 problem.
I also have two cars.... two dogs 
well, you get the idea.
. Tough crowd!!!