>Mike >It was iso 100 >the histogram looks to be a good curve but still a slight >bit of grain in the darker portions >Regards >Paul > >Focus on what's right in your world instead of what's wrong >
Try the same shot at ISO 400. Banding seldom occurs at ISO 100. Mostly you would have to zoom 200 percent to see it anyway. I've seen several D200s that exhibit no banding at ISO 400 with extremely bright objects in the frame. I've also seen one that showed banding at ISO 400 with a dynamic range well within the cameras ability. It was one of the first units released. After being sent back it now shows no banding. Mine will occasionly show banding at ISO 400 but only if you blow the exposure and exceed the camera's dynamic range. You still have to zoom to 200 percent to see it. I'm well satisfied with mine.
Thom Hogan at www.bythom.com suggests in his D200 review if you force any DSLR far enough it shows banding, and his 20D can be induced to show more banding than his D200. Thom does not say if 6,7, 8, 9 or 10 stops under or over exposure are needed to induce banding but he confirms he has never encountered it in an actual shooting situation.
I go further - I have yet to see an example of type 1 where a reasonable attempt has been made to control DR by the use of fill flash, reflectors or similar means.
Photography is a bit like archery. A technically better camera, lens or arrow may not hit the target as often as it could if the photographer or archer does not practice enough.
Try it with a D70. I've never been able to get it to show any kind of banding.
Actually, I have, since the banding hype got so much attention with the D200. Shots straight into the sun with flare or moire show very faint dark banding (as opposed to the D200 bright banding) through the entire image, which is visible at 100% if you pay attention.
FWIW - I read a fairly good explanation of banding at Imaging-Resource.com. Their qualitative findings on banding agree with Thom Hogan, et al, but they also state the conditions they found banding to occur. From their article, "It is very ISO-dependent, invisible (in the cameras we've seen at least) at ISO 100, becoming more prominent up to about ISO 400, and then diminishing again at higher ISO levels" and "It is most commonly triggered by a moderately-sized area of significant overexposure (more than 1.5-2 stops beyond the level that produces luminance values of 255) Small catchlights don't seem to trigger it. More extreme overexposures (more than about 5 stops beyond being blown out) don't seem to trigger it either."
I plan on testing the dynamic range of my D200 when it comes in and also to see if I can induce banding. If my sample performs as well as the camera in the review I won't worry about it. I don't believe this would even be visible in an 8x10 print.
"It is very ISO-dependent, invisible (in the cameras we've seen at least) at ISO 100, becoming more prominent up to about ISO 400, and then diminishing again at higher ISO levels" and "It is most commonly triggered by a moderately-sized area of significant overexposure (more than 1.5-2 stops beyond the level that produces luminance values of 255) Small catchlights don't seem to trigger it. More extreme overexposures (more than about 5 stops beyond being blown out) don't seem to trigger it either."
I think there is one condtion missing from their statement. At least for my D200 which has only shown minor type 1 banding.
In my D200 the above conditions must also have a very sharp contrast change, black to white edge, to show type one banding. I don't think my D200 would band on the electric arc picture either. It has never banded on sunsets or stadium lights. I got a bit of banding shooting a dark walled room with a sun lit window. Just a bit at the window edge where it rapidly changed from light to dark.
I found type one banding in a night shot! A building had spotlights and brightwork on the roof. There were type 1 banding lines in the shadow areas near the spots. I shot on a tripod at several ISOs. I think I see some banding at ISO 100 but it is definitely visible at 100% at ISO 400.
When printed to 8x10, the banding is not visible. At 24x30 it probably would be.
PS: There was obviously no way to correct the DR with flash or reflectors. The building was 3 blocks away. I would also call the shot under exposed
Sorry for the delay I have been on shift It was saftey training for emergency responders and it was being back fed @ 6300 volts from 110 Here is a picture of a glove with a pin hole The hot dog in the thumb cooked real fast
Focus on what's right in your world instead of what's wrong