How to eliminate the Fuji Velvia “Magenta Cast” 

I love the colors of Fuji Velvia, but I don’t like the magenta color cast that can result during daylight shooting. You will find many discussions related to this if you google “velvia magenta cast”.

I wrote this article specifically to address how to remove Fuji Velvia’s magenta cast when shooting landscapes with a camera on tripod and Neutral Density filters alone.

Before we gat started, let me provide you with an example of the color cast I’ll be dealing with:

It may not be immediately apparent but you can see a distinct magenta color cast in image above. It’s particularly pronounced in the bricks on the bottom left and the sky. We’ll be coming back to this scene quite a bit but before we do, let me rewind and tell you how this experiment came about.

One day while reviewing my shots from the beach, I realized that there were some images that had no magenta color cast (something regular Velvia shooters should be all too familiar with). I tried to remember what I had done in those particular shots, but I wasn’t 100% sure…

I don’t use any color compensation filters, only Neutral Density (ND) filters for blocking light. After noticing that I had dealt with the color cast using ND filters alone, I thought to myself that if I could find out how to eliminate the color cast without color compensation filters, it would help me to create consistent results under any conditions.

I should say that I believe that using color compensation filters can also be helpful, but only if we know the correct filter to use according to specific conditions. This article shows the results of tests I performed to eliminate Fuji Velvia’s magenta cast with only correct metering and the use of ND filters alone. Here’s what I’ll be covering in:

Equipment used

To perform the tests, I used the following camera, film and filters:

• Leica M7
• Leica Super-Elmar-M 21mm f/3.4 ASPH
• Metering: internal camera meter
• Mode: aperture priority
• Fuji Velvia 50 (RVP50) and Fuji Velvia 100 (RVP100)
• Neutral Density (ND) filters: 3-stop and 6-stop
• Graduated Neutral Density (GND) filters: 3-stop and 6-stop*

• The Graduated ND (GND) filter is used to balance the sky (highlights) and the shadow. It blocks the light from the sky so that the sky will not be blown out.

Test 01: Baseline with ND filters

For reference, I noted the metering results from the sky (highlights) and the shadow. Then I took pictures with and without filters.

The order of the images below is:

• No Filter
• GND
• 3-stop ND + GND
• 6-stops ND + GND

For reference, the base metering of the scene was:

Let’s take a look at the images. Please note that for all of the samples provided here, clicking on any of the thumbnail images will open a larger version. You can close and view each image separately or use the arrows to the left or right to cycle through the examples.

A surprising result occurred when I was using the 6-stop ND filter. Look at the final image above and compare it to the other pictures. The 6-stop ND helped remove the magenta cast!

Here is a direct comparison without and ND filter and with a 6-stop ND filter us the the first and last image above:

Images using the 3-stop ND filter still had the magenta cast. As I had no magenta cast at 6 seconds with the 6-stop ND filter, the initial test above seemed to suggest that the longer the exposure time, the better…but actually, that was not true.

As I found out in my next test, I can also eliminate the magenta cast with faster exposure times.

Test 02: Exposure time / aperture test

Coming fresh from my simple baseline filter test, I moved on to testing various exposure times and apertures with my 3-stop and 6-stop ND filters. But first, let’s remind ourselvs how the scene looks without the ND filter applied:

Using the 6-stop ND filter, I tried various exposure times to see if this affected the results. For this, I simply changed the aperture of my lens from f/3.4 to f/16, and adjusted my exposure time accordingly.

Here are the results (click on each image for a large version):

If you are having difficulty seeing the differences between each frame, I would advise you to focus on the bricks at the lower left of each image and the cloud on the top right of the image above.

I then performed the same test using my 3-stop ND filter:

We can see again that the longer the exposure time, the better the result. The final images have no magenta color cast at all. Is this only because of the longer exposure time? Not really…

As we have seen already, we are able to use faster exposure times with the 3-stop ND filter than with the 6-stop ND filter. And here is the interesting thing: there is almost no color cast at the smallest aperture (f/16) on either test group, regardless of the ND filters used or the exposure time!

We can also see that apertures from f/8 to f/16 reduce the color cast significantly. So, the test suggests that small apertures (from f/8 onward) perform very well in reducing the color cast, regardless of the ND filter used.

For shooting on the beach, for example, in order to capture smooth movement of waves, you will need to expose at around 1 sec using a 3-stop ND. If you want to make the sea calm, shoot for 5 sec or longer with the 6-stop ND.

To conclude, let’s once again compare the shots taken with and without using ND filters.

Without ND filter and with the 3-stop ND filter (at f/16):

Without ND filter and with the 6-stops ND filter (at f/16):

From what I can see, the 6-stop ND produced better saturation and completely removed the magenta cast, in comparison to the 3-stop ND.

Test 03: Overexposure tests

This test was to see how overexposing the film would affect the resulting image. If we use Velvia 50, for example, to overexpose 1-stop means we set our ISO to 25. Some suggest that for Velvia, the best approach is to set the ISO to 40 (overexposing about 1/3-stops). Others prefer shooting at box speed (EI 50).

In this test, I used Velvia 100 (RVP 100) with no filter applied:

Clearly, overexposing +2 is not an option! So, we have four good options in this test: 0, +1/3, +2/3, and +1 stop.

From the test above, we can see that the color of the sky is still visible when the film is not overexposed. So, if we shoot without the Graduated ND (GND) filter and still want to bring out the sky’s color, we shouldn’t overexpose my film.

However, If I have a GND filter at hand, then I prefer overexposing the film for less saturated images.

I also found out that overexposing +1 is sometimes too much and causes the sky to still be blown out, even using the GND filter. This causes inconsistent results.

When I correctly apply the GND to balance the sky and shadow, the sky still has +1 overexposed, and this can cause a blown-out sky on a bright day, too. So the options are +1/3 or +2/3. And for safety, I chose +1/3.

Sample photographs

I used the methods described above to create the following same photographs. They include a mixture of filtered and unfiltered images (depending on the scene). To scan, I used the default settings on my Epson Perfection v600 scanner. I made no adjustments, except cropping some of the images.

With this method we can see the actual color results from Velvia 50:

Conclusion

In short, Fuji Velvia’s magenta color cast can be significantly reduced when shooting landscape on Velvia 50 or 100 by using ND filters and small apertures to limit the light. Some quick takeaways from my tests:

• A 6-stop ND filter is better than a 3-stop ND filter for reducing color cast.
• Aperture f/16 is better than f/8 for reducing color cast.
• With a 6-stop ND filter and the lens at f/16, I get no magenta color cast.
• If 6-stop ND results in exposure times which are too long, I can still use 3-stop ND and a small aperture to obtain a decent result.

To simplify things, I use f/8-16, with a 3-stop or 6-stop ND. I also prefer overexposing the film by +1/3 stops (about EI 40 for Velvia 50).

For filters, I recommend Formatt Hitech Firecrest. These filters are made from glass and the good thing is that I don’t need any color correction filters.

Using a neutral (ND) filter is sufficient to remove the Velvia magenta color cast and can produce consistent results in almost any light conditions. I should mention that during sunrise or sunset (golden hour), the magenta cast produces nice images, so I wouldn’t want to remove it then. Besides that, in low light conditions, the ND filter is not really suitable.

With the method described here, I get more consistent results when shooting landscapes using Velvia. So before I sign off, let’s check the before and after results one more time:

Thanks for reading!

~ Yusuf Wiryonoputro