4
\$\begingroup\$

I've taken a few photos with a new B+W ND 6-stop filter and found that it leaves them quite warm. I'm able to correct this in ACR using temperature/tint sliders. For the few photos I've taken so far, the same temp/tint settings seem to work for each one.

I've heard that color cast from a given ND filter can vary between cameras and lenses. But presuming I continue to use this filter with the same camera and lens, is it safe to assume my WB correction will be the same every time, for any photo? Or will it be dependent on other conditions like lighting and exposure/aperture?

Note that I ask, because if the color cast varies based on some conditions, then I will probably take no-filter companion shots for any filter shots that are important, so I'll have a reference point for matching WB later. Though, it would be nice to not have to always do this.

EDIT:
To clarify my question, here is a concrete example. Please excuse the horrible photo.

Here is a RAW file shot with no filter. I used auto-WB, and ACR says that "as shot" the temp is 5400 and the tint is +13.

And here is another RAW file, shot with my 6-stop ND filter. Same tripod, same composition, same lighting (different exposure settings though). Also used auto-WB. Resulting in temp of 5700 and tint of +43. It is noticeable warm. If I adjust temp to 4300 and tint to +10, I think it is a pretty close match to the first one.

So, I needed to drop the temp by 1400, and the tint by 33, to correct it. My question: can I make these exact same drops to every shot I ever take with this filter/lens combo, to make it match what would have been the result if I hadn't used the filter? Or will it vary?

\$\endgroup\$
7
  • \$\begingroup\$ Why would you expect WB adjustment to be the same in different lighting conditions? Shouldn't you expect them to be different under sunlight vs artificial lighting, for instance? \$\endgroup\$
    – xiota
    Commented May 29, 2020 at 4:00
  • \$\begingroup\$ @xiota I don't necessarily expect anything, it was just an arbitrary suggestion of a possible conditions. Yours is another. I'm interested in any such conditions, including ones that haven't occurred to me. Although, doesn't artificial light vs sunlight qualify as "different lighting conditions"? \$\endgroup\$
    – The111
    Commented May 29, 2020 at 4:14
  • \$\begingroup\$ Color balance in post-prod will compensate for everything at once: ND filter, actual color temperature of light, color casts due to reflection on colored objects (walls...)... \$\endgroup\$
    – xenoid
    Commented May 29, 2020 at 8:08
  • \$\begingroup\$ Where did you get the ND filter? If you bought it from amazon, it's as likely as not to be a counterfeit. \$\endgroup\$
    – Michael C
    Commented May 29, 2020 at 11:51
  • \$\begingroup\$ How are you setting the white balance (both CT and WB correction) in camera? What are the lighting conditions under which you are shooting? Are the lighting conditions consistent between different shoots or variable? Without this information it is difficult to answer your question authoritatively. \$\endgroup\$
    – Michael C
    Commented May 29, 2020 at 19:40

2 Answers 2

1
\$\begingroup\$

For a given lens/camera, all images taken with or without any filters will need WB correction to account for the varying shooting conditions. Even when no filters are involved, one can't use the same WB settings for an image taken in 2700K light that one would use for the same scene taken under 6400K light and expect the colors of the objects in the scene to look remotely similar.

And here is another RAW file, shot with my 6-stop ND filter. Same tripod, same composition, same lighting (different exposure settings though). Also used auto-WB. Resulting in temp of 5700 and tint of +43. It is noticeable warm. If I adjust temp to 4300 and tint to +10, I think it is a pretty close match to the first one.

Your issue may have absolutely nothing to do with whatever color cast is being introduced by the ND filter.

When you use Auto WB you're at the mercy of the camera's programming or ACR's analytical engine. It may well be that your camera is designed to render long exposures of very dim scenes warmer than shorter exposures of brighter scenes.

Looking at the two example images¹ you linked, it appears the metallic roof on the lean-to next to the barn as well as the white building near the left edge are fully saturated in the filtered shot. The filtered shot is also much brighter overall than the unfiltered one. Any time you have a completely saturated area in the image, the way the camera or a raw conversion application does Auto WB can be thrown off. Exposing slightly dimmer to prevent full saturation anywhere in the frame can significantly shift the color temperature and WB correction the camera or raw conversion application uses to render the preview image of the raw file. So can selecting things such as "ambiance WB" or "highlight WB" in cameras or raw conversion applications that offer such settings.

The only way to check to see what is the effect of the filter is to set color temperature and white balance correction manually to the same settings for both the unfiltered and filtered photos and then compare them.

¹ Of course it is difficult to tell much from the example images when we can't even tell what image format they are in as delivered to us by google drive, much less see any EXIF info. Are these JPEGs you exported before uploading? PNGs? TIFFs? Or did you upload actual ARW files and google is converting them with whatever unknown default settings it uses to interpret what we are seeing?

\$\endgroup\$
8
  • 1
    \$\begingroup\$ Thanks Michael. To clarify, those are ARW files I uploaded to Drive. Google isn't converting them at all, you can download them using the "down arrow" button in the upper right. Google is rendering an on-screen preview (which could admittedly be thought of as a temporary conversion), which I think is the source of your confusion (and I agree, we have no idea how they're processing the file to do this). Download those files and you'll see EXIF and other data. Sorry for the confusion. I almost zipped them before upload, but decided against it, thinking that might make more work for you. \$\endgroup\$
    – The111
    Commented May 31, 2020 at 22:11
  • 1
    \$\begingroup\$ If I make both WBs the same (e.g. ACR's "daylight" 5500K and +10 tint), the warmth is still evident in the with-filter version. As a sidenote I've found many other people saying the same thing about B+W filters, that they are known to leave warm result. I actually ordered a Breakthrough filter, a new brand I'd not heard much about, but they seem to have taken dozens of reviewers by storm over the past few years with the most color neutral results ever. If I confirm this, I'll probably just return my B+W and forget about all of this. One example review. \$\endgroup\$
    – The111
    Commented May 31, 2020 at 22:17
  • 1
    \$\begingroup\$ I've not personally used any Breakthrough filters. I have noticed they get great reviews from sources that are known to give good reviews to just about anyone who will send them free stuff to review. I've also noticed that Breakthrough filters get much more "Meh" reviews from those who are known to not accept freebies in exchange for gushing reviews. I must admit most of my research into Breakthrough filters has been with regards to their polarizers, though. Their X2 was supposed to be the cat's meow according to all the "freebie" reviews... \$\endgroup\$
    – Michael C
    Commented Jun 1, 2020 at 1:19
  • 1
    \$\begingroup\$ ... Then the X4 came out and those same shils were talking about how it corrected the failings of the X2! WTF?!?!?! \$\endgroup\$
    – Michael C
    Commented Jun 1, 2020 at 1:19
  • \$\begingroup\$ yeah my guard is definitely up. But I ordered with free returns, so it's worth my own testing. Also got a ColorChecker to make my tests even better. \$\endgroup\$
    – The111
    Commented Jun 1, 2020 at 2:59
0
\$\begingroup\$

The filters should be replaced. I bought some B+W ones too at some point and found them unusable due to their horrible color-cast. Got rid of them and bought many Hoya filters that are perfectly neutral. I recommend you safe yourself the headache and do the same or anything good brand like IRIX, Zeiss or Lee (Although I did not try them myself but they are highly regarded, so are the B+W but I have no idea why!).

Color cast simply occurs because a filter does not uniformly affects light of all wave-lengths. As you say, it can be compensated in software but it does not depend on anything but the poor quality of the filter itself. You may see a slight difference with certain camera and lens just because the lens themselves sometimes affect colors and the interaction between all these will cause different shifts.

In summary, the cast you see from a poor quality filter is mostly due to the filter but how that cast interact with any optical imperfections in the lens or filter-stack in from of the sensor means that you can get slightly different color shifts with the same filter when used on different lenses and camera combinations.

\$\endgroup\$
5
  • \$\begingroup\$ Genuine B+W filters don't have much, if any, of a color cast. But there are probably more counterfeit B+W filters on the worldwide market than genuine ones. That's the price they pay for being #1, similar to SanDisk in the memory card arena. \$\endgroup\$
    – Michael C
    Commented May 29, 2020 at 19:36
  • 2
    \$\begingroup\$ As far as I understand it, there's no such thing as a heavy ND filter that doesn't introduce some level of color cast. nofilmschool.com/2016/02/… Looking at the images in that article, I feel like B+W looks closest to the control image (but the author claims Breakthrough wins). \$\endgroup\$
    – The111
    Commented May 30, 2020 at 3:14
  • \$\begingroup\$ @The111 Without knowing how white balance was set in the test referenced there, the results are totally meaningless. \$\endgroup\$
    – Michael C
    Commented May 31, 2020 at 6:56
  • \$\begingroup\$ My comment above referenced a review claiming Breakthrough beat B+W on color neutrality, while I thought the images in that article suggested otherwise. I've since read dozens other such articles with lots of image samples, and in most of those other ones it was clear that Breakthrough was the winner. I've ordered a Breakthrough to do my own comparison vs my B+W, and will return the loser. The most comprehensive shootout I found: improvephotography.com/40253/… \$\endgroup\$
    – The111
    Commented May 31, 2020 at 22:21
  • \$\begingroup\$ @MichaelC that link above (the nofilmschool one) actually is referencing this fstoppers test. They explain their test setup in further detail there. They used a reference card and LR WB tool. As stated above, I think the pics in that article make B+W look best (even though the author votes for Breatkthrough), but every other such shootout I've found, I agree the Breakthrough looks like the clear winner. \$\endgroup\$
    – The111
    Commented May 31, 2020 at 23:07

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.