I could use some advice on some strange images coming off my D3. I went to a local dance studio to sample the lighting for an upcoming shoot. While taking the samples, I noticed some of the shots were perfectly white balanced but then some would go deep orange or yellow. I thought the Auto WB was getting fooled so I turned it off. Same results. Then I started getting some half white/half yellow - almost like the look when your shutter is too fast for your flash - but I'm not shooting flash. The line between yellow and white runs horizontal on landscapes and vertical on portraits so I assume that the shutter is involved but don't understand how or why. I can take the exact same shot within a second of each other and get totally different results.
The studio was lit only by plain flourescent light - no daylight - no flash. The only thing I can think of is that I am somehow using a shutter speed that just happens to be at the right sync rate with the "flutter" of the lights - kind of like the banding you get when you try to use a video camera to film a tv screen. Am I right or is there some other explanation?
I took the shots on a D3 - 70-200 2.8 @ 2.8 - 1/200 - iso 1000
#1. "RE: D3 - Banding?" | In response to Reply # 0fourbyfive Registered since 14th Mar 2007Fri 08-Feb-08 12:11 PM
This is probably not going to be a D3-specific issue. There was something posted about this on one of the forums not too long ago. I believe you're correct in your diagnosis about the light cycles. From what you're describing, this has nothing to do with the "banding" issue in the D200.
#3. "RE: D3 - Banding?" | In response to Reply # 2billyboy14 Registered since 11th Nov 2005Fri 08-Feb-08 12:23 PM
The previous poster is correct. The lights flicker...imperceptibly to our eye, but the camera's shutter is picking it up in mid cycle. Some previous posters have had similar prolems in gymnasiums and hockey rinks.
#6. "RE: D3 - Banding?" | In response to Reply # 5AlwaysLearning Registered since 06th Jan 2008Fri 08-Feb-08 02:18 PM
Thanks for the responses. I will check the podcast asap. Here are some of the samples - I am new to this forum so if I am not properly uploading, please let me know. These images were shot in RAW and simply converted to jpg - quality 5 - to show the effect. To keep the file size low enough, I reduced ppi to 100 in Photoshop
I have shot other dance studios under similar lights with my D200 and never had this problem. I am wondering if this is related to the D3's ability to shoot at iso 1000 with a shutter speed faster than I ever could have with the D200. With the D200 I had to settle for lower iso and longer exposures - thereby giving the flicker enough time to "even out". Crazy idea?
Attachment#1 (jpg file)
Attachment#2 (jpg file)
#7. "RE: D3 - Banding?" | In response to Reply # 6Jess Sturgeon Registered since 28th Oct 2005Fri 08-Feb-08 04:45 PM | edited Fri 08-Feb-08 04:45 PM by Jess Sturgeon
Its interesting when your camera is advanced enough to reveal deficiencies in various lighting systems.
This has come up before with the D2X/D2H, but with the great strides in high ISO performance of the D3, it will a much more regular occurance. Strobes and flash can help in some cases, lowering your ISO and good hand-holding technique can help in others where stop-motion shutter speeds are not required, but in some cases, it will still be impossible to get good pictures, even with a D3
#13. "RE: D3 - Banding?" | In response to Reply # 7ajdooley Nikonian since 25th May 2006Sun 10-Feb-08 04:06 PM
How ironic it is that the final product should be degraded by the the newest and best technology. Maybe I just didn't know how good I had it when I was constrained to Tri-X at ASA 400 and Ektachrome at ASA 125! Even if you can get fluorescents all to "fire" at the same time, they produce a discontinuous spectrum that's virtually impossible to color balance other than subjectively anyhow. Of course, that's an advantage to some imagery -- namely astronomical imagery. Since there are "spikes" and "gaps" in the spectral output of many non-tungsten sources, like high pressure sodium, they can filter out the "spikes" and thus create artificially enhanced "dark skies."
I have maintained for years that photography is neither an art nor a science. It is a craft. Like wood working -- where my capabilities seem to be confinded principally to making smaller boards and saw dust -- we must constantly strive to master the whole spectrum (pun intended) to produce the best results. The D3 opens doors, but we have to watch where we go when we pass through them and realize lots of surprises await us on the other side.
Alan Dooley in Waterloo, IL
#8. "RE: D3 - Banding?" | In response to Reply # 6walkerr Nikonian since 05th May 2002Fri 08-Feb-08 05:03 PM
>I have shot other dance studios under similar lights with my
>D200 and never had this problem. I am wondering if this is
>related to the D3's ability to shoot at iso 1000 with a
>shutter speed faster than I ever could have with the D200.
>With the D200 I had to settle for lower iso and longer
>exposures - thereby giving the flicker enough time to
>"even out". Crazy idea?
No, not a crazy idea. It's exactly what we discuss in the podcast. Slow shutter speeds even out the cycling that occurs, but increase the likelihood for motion blur.
Incidentally, changing the PPI setting doesn't reduce the size of images on a display, which is why these are very big. You need to reduce the actual pixel dimensions to decrease the size.
#10. "RE: D3 - Banding?" | In response to Reply # 8AlwaysLearning Registered since 06th Jan 2008Fri 08-Feb-08 06:15 PM
Thanks for the link, Rick. I just listened to the podcast - it describes my problem perfectly! TOo bad I hadn't heard the podcast a day sooner! It is somewhat ironic that I wanted the high iso and shutter speed to catch indoor activities but now I have to deal with this issue which I frankly never considered. Now if I can just get Nikon to refit every arena with better lights, we'll be all set!
Thank you everyone for your input!
#9. "eRE: D3 - Banding?" | In response to Reply # 6sorin Registered since 06th Sep 2007Fri 08-Feb-08 05:04 PM | edited Fri 08-Feb-08 05:17 PM by sorin
this is kinda cool effect.
you know that the shutter is open completely only at speeds under the flash sync value (depending on the camera 1/200-1/250 etc). if the speed is faster then the shutter blades don't have time to be fully opened so they "move" the opening from one side of the sensor to the other, exposing light in a progressive way.
it seems as the shutter opening passes in front of the sensor the fluorescent lights recharging cycle changes so parts of the sensor are processed by the camera with a different WB than others (the original measurement).
to fix this in PS use a gradient. i'm sure it will be easy to implement a fix (a WB gradient adjustment or some other trick) either in capture nx or camera's firmware if there is need.
#12. "RE: D3 - Banding?" | In response to Reply # 11MikeW2ck Basic MemberSun 10-Feb-08 04:28 AM
I don't see what you could do even in post processing with such a difference across a single frame. Also, I could give you much worse examples, so I guess it depends what you're starting with too.
You not only have a color difference but also exposure.