Muddy colors Using an ND Filter

[fike]

I have been experimenting with using very long daylight exposures with a 10 stop B+W ND Filter.  When I use the filter, I get some pretty interesting shots, but I have found that I get a brown cast in the images.  I can improve it slightly, but I want to figure out how to best correct the issue.  

I have a graycard, and I was considering using it to set the middle gray, but I am not sure that will totally address the brown color cast in the highlights.  Do I need to get a color checker and do a test calibration shot?  Can I shoot a test shot once and get a general sense of the correction in daylight and then apply the same adjustments programatically to other scenes?  I really don't want to carry a color checker or graycard around everywhere.

Another interesting note.  10 stops is really dark.  I am going to get a 6 stop filter, because working with 10 stops basically means shooting totally blind.  I think it claims 0.1% light transmittance.  It results in some cool effects though.

[Jonathan Wienke]

I'd take some shots of a ColorChecker and make a custom DNG camera profile for shots using the ND filter. That should take care of any differences in the color cast between the shadows and highlights.

[Dick Roadnight]

Neutral density filters should be ...neutral ... do digi cameras suffer from reciprocity failure? and are they all as bad or which is best?

[Alan Goldhammer]

I have a graycard, and I was considering using it to set the middle gray, but I am not sure that will totally address the brown color cast in the highlights.  Do I need to get a color checker and do a test calibration shot?  Can I shoot a test shot once and get a general sense of the correction in daylight and then apply the same adjustments programatically to other scenes?  I really don't want to carry a color checker or graycard around everywhere.

You can get the new X-Rite Passport which contains the color checker in a solid clam shell case.  It which would permit you to create not only the calibration shot but offer you white balance correction as well.  The changes are subtle but noticable in the histogram (and the resultant image) after you apply the profile.  I have seen some mild color shifts but can't comment as to whether it will solve your particular problem.

[fike]

I was hoping to avoid carrying a color checker.  I also was hoping not to have to buy an expensive little color-card.  

I am not doing DNG conversions, and I don't plan on starting any time soon. (I understand the risks.)  Is there any way to make a ACR base setting that I can use when I us the ND filter?  How do you take that color checker into ACR and make mods to get the colors right.  I could understand if each color in the checker pattern had an equivalent series of RGB #s, but I don't know how you would read and correct that in ACR.

I am trying to make this correction as simplistically as possible.  I use this ND filter once or twice a month.  I don't frequently have substantial issues with color balance, so I am disinclined to implement a major color workflow change like one would for product photography.  

I already do panoramic photography (sometimes with the ND), so I don't need more processing obstacles to getting images out.  

[Paul Sumi]

Singh-Ray says this in their FAQs.  Not quite what you've noticed, but may be similar:

http://www.singh-ray.com/faqmain.html#top

"Q13) I'm getting a color cast with my Neutral Density filter. What's the problem?"

"A: When using a Neutral Density filter (graduated or solid) you are generally causing your camera to make a longer exposure than normal, sometimes much longer. The software in some digital cameras uses response curves that adjust the RGB sensors sensitivity based on the exposure time, which may introduce more red into the image. Sort of a digital form of reciprocity failure. In other instances, blocking visible light and a longer exposure allows infrared light to "build up" on the red sensors leading to a red or purple-ish cast. If your filter looks neutral but you see the cast in your images, try taking some test shots using the filter with a different brand of camera, or with a film camera. If you still see the cast, and you're within the warranty period, contact Singh-Ray to have your filter evaluated and possibly repaired/replaced."

Paul

[Panopeeper]

I am not doing DNG conversions, and I don't plan on starting any time soon

The DNG profile is called so, because it needs a DNG file for its creation. However, the profile you created can be used in ACR directly, with the native raw files.

[cyberean]

I have been experimenting with using very long daylight exposures with a 10 stop B+W ND Filter.  When I use the filter, I get some pretty interesting shots, but I have found that I get a brown cast in the images.  I can improve it slightly, but I want to figure out how to best correct the issue.  

I have a graycard, and I was considering using it to set the middle gray, but I am not sure that will totally address the brown color cast in the highlights.  Do I need to get a color checker and do a test calibration shot?  Can I shoot a test shot once and get a general sense of the correction in daylight and then apply the same adjustments programatically to other scenes?  I really don't want to carry a color checker or graycard around everywhere.

Another interesting note.  10 stops is really dark.  I am going to get a 6 stop filter, because working with 10 stops basically means shooting totally blind.  I think it claims 0.1% light transmittance.  It results in some cool effects though.

per B+W's product sheet, the higher valued ND filters (those beyond 3-stops) will introduce
warmer tones to an image, given the filter's higher red transmission levels beyond 660nm.
for their 6-stop filter B+W suggest the use of a UV/IR 486 (blocking) filter as a remedy.
no such suggestion is made for their 10-stop filter, which is even warmer ...

[fike]

okay, I went back to some of my samples, looking for example.  Brown was a bad way to describe the tonality that I am worrying about.  Here it is.


F/8 @ 2 sec ISO 200

[fike]

per B+W's product sheet, the higher valued ND filters (those beyond 3-stops) will introduce
warmer tones to an image, given the filter's higher red transmission levels beyond 660nm.
for their 6-stop filter B+W suggest the use of a UV/IR 486 (blocking) filter as a remedy.
no such suggestion is made for their 10-stop filter, which is even warmer ...

I am not enthusiastic about stacking filters like that.  I would rather figure out a way to correct digitally.  That may give me an idea of where to start though.  I don't really mind making changes individually every time I use the filter.  I don't use it THAT often.  I just need to have some ideas about where to start to get a more neutral output.

[Wayne Fox]

I am not enthusiastic about stacking filters like that.  I would rather figure out a way to correct digitally.  That may give me an idea of where to start though.  I don't really mind making changes individually every time I use the filter.  I don't use it THAT often.  I just need to have some ideas about where to start to get a more neutral output.

Just curious what camera you are using?  I've heard some cameras/sensors handle this better than others, and was thinking about trying a 10 stop ND myself.

[new_haven]

Bright blue skies create blue casts in certain situations, so I suspect that a long exposure would only add to the problem. Anyway, it's easy to correct in PS with straight curves in LAB mode (left), or the white balance tool in ACR (middle). The original is on the right.

[Dick Roadnight]

Singh-Ray says this in their FAQs...   The software in some digital cameras uses response curves that adjust the RGB sensors sensitivity based on the exposure time, which may introduce more red into the image. Sort of a digital form of reciprocity failure. In other instances, blocking visible light and a longer exposure allows infrared light to "build up" on the red sensors leading to a red or purple-ish cast.

So, like I surmised above, according to this, it is digital reciprocity failure, and it camera/sensor/software error not compensating for it. ...so which cameras are best for long exposures?
The P65+ is a top pro sensor that can do long exposures - anyone had cast problems with it?

[fike]

Bright blue skies create blue casts in certain situations, so I suspect that a long exposure would only add to the problem. Anyway, it's easy to correct in PS with straight curves in LAB mode (left), or the white balance tool in ACR (middle). The original is on the right.

Those conversions are similar to what I was able to do. I still wasn't particularly happy with the shots.  Color still seems off to me.  I am having a hard time describing my feelings about shots with this filter.  It almost looks like slide film, and not in a good way.

[Wayne Fox]

So, like I surmised above, according to this, it is digital reciprocity failure, and it camera/sensor/software error not compensating for it. ...so which cameras are best for long exposures?
The P65+ is a top pro sensor that can do long exposures - anyone had cast problems with it?

Actually the p65+ tops out at 60 seconds.  So I guess it depends on your definition of long.  As a comparison the p45+ I believe handled exposures as long as 1 hour.  I have done full moonlight exposures with a p45+ (8 to 10 minutes) that showed no issues with color cast or reciprocity ... indeed they appeared as normal daylight lit images as one would expect.  As to what problems are exhibited if you exceed the 60 seconds I do not know.  Perhaps they are similar to those from the OP.

I have done many 5 to 10 second exposures with the p65+, but then I've never had any camera where this was a real problem ... not like the example being discussed.  I guess I'm not aware from the OP how "long" his exposures are.  After all if I use a 10 stop filter in a normal full sun situation, even at f/32 that puts my shutter speed at 30 seconds.  I would think most current cameras can handle that long of an exposure without a real issue - I've done this with my Canon cameras and p45 on many occasions.

To me this implies the filter might be the major contributing factor ... after all what may be theoretically neutral might not perform as such on a digital camera, especially if the filter was designed when film was predominant.  I've read this is the case with Kodak Grey cards ... neutral to film but not completely neutral to digital.