How Does Fujifilm Film Simulations For Olympus E-M1 Mark II Files Sound?

Olympus E-M1 Mark II RAW image, made in full direct sunlight, used as the basis for the newly
created Fujifilm film simulation presets to be applied. (click to enlarge)

I'm a bit schizophrenic when it comes to my photo gear.  I love my Fujifilm gear and I also love photographing with my Olympus micro M4/3 gear.  For different reasons, of course.  I think there may be many like me in owning and using these two particular brands. Recently, I wrote a post on what I thought the fundamental differences were between the two camera systems.  You can find that post here if you are interested.

The two main reasons I love photographing with my Fujifilm X-T2 is a) the straightforwardness and simplicity of how the camera is designed and engineered with its knobs and dials and b) the colors and "look" of my Fujifilm image files when I use the film simulations that are built into the camera. The one caveat to using the film simulations is that one has to utilize JPEG files in order to benefit from Fujifilm's engineers' work in creating simulations of their films.  I've tried applying Adobe's versions of Fujifilm film simulations in Lightroom's Calibration menu, but I don't think they are quite as good as those in the camera itself.  They aren't bad but they aren't the originals either.  However, for many of my images, the JPEGs can suffice nicely and I'll use them when circumstances allow.

When I photograph with my Olympus E-M1 Mark II, I kind of miss the Fujifilm color.  Being that I photograph regularly with both cameras, it would be nice if the color palates from both cameras could match.  Hmmm.  That got me to thinking....  A couple of weeks ago I asked myself, "I wonder if I could replicate the Fujifilm film simulations, by creating custom Lightroom presets, so I could then apply them to my E-M1 Mark II RAW files?"  I thought about it for a while, then set out to see if I could.  This post is a narrative of my logic and my journey.  

If you are interested in how I made the presets to replicate the Fujifilm film simulations, read on.  But beware, this is a long post outlining a detailed process.  If you want to see how close I came to replicating the Fujifilm film simulations in the real world, then come back as I have not yet had time to go out with both cameras, side by side, photograph the same subjects, then compare the final images for rendition. Between weather and other pressing priorities, I've not been able to do that as of yet.  Also, adding that to this post would make it dreadfully too long and complicated and I don't think anyone would be interested in reading it.  I plan to do the comparison within a week, hopefully, and write about it in a near future post.  I will be as interested as you to see how close I came and if my method was successful, partially successful or a failure.  Also, others may have done just this but I haven't looked nor read about others embarking on this same journey and I don't want to any other work to influence what I intended to to.

This journey is not one of strict science, but one of thinking things through and using logic to figure out how to create the film simulation "look" to be applied to the Mark II RAW files.  I work with a calibrated computer monitor and the X-Rite ColorChecker Passport, which I deemed essential for some semblance of color, contrast and saturation accuracy as well as a real chance for success.  Your mileage may vary if you were to try to replicate my efforts.  My goal was to create a methodology that would "standardize" the entire process so everything possible was equalized and was reproducible.  One more thing—the presets I developed apply to he E-M1 Mark II only.  They may not be valid for any other Olympus camera model, etc.

Here is what I did to make the film simulations to be applied to my Olympus E-M1 Mark II RAW files:

A) Before I started, I reset all of the X-T2's film simulations to default settings.  I wanted to create the film simulations at default settings, just as Fujifilm designed them. RAW + Fine JPEG was the file selection (my normal setting) and ISO was set to 200, the lowest for best quality. White balance was set on Auto. (White balance isn't absolutely critical at this stage as the subject I would photograph was the X-Rite Colorchecker Passport, which has a series of neutral gray patches on which I would fine tune the white balance.  Couple that with the excellent auto white balance in the Fuji and Olympus cameras and the final white balance was very close to dead on.  Dynamic Range was set to 100.

B) I then chose a summer day with mostly clear skies full sunshine.  I placed the sun squarely at my back about 1/2 hour before noon so the angle of the sun had very little influence on color temperature of the daylight (i.e., not early in the morning or late in the evening when the light outdoors is very warm so as to minimize any necessary changes in white balance). No clouds obscured the sun while making any of the test photos.  All exposures were made under exactly the same ambient conditions.

C) Holding the X-Rite Colorchecker Passport out in front of me at eye height and arm's length (as illustrated in the image at the top of this post), I determined a "good" exposure on the histogram on the X-T2 to ensure there were no clipped highlights or shadows.  In other words, a good "average" exposure that any of us would make that makes the image look good right out of the camera.

I then made an exposure with each of the film simulations, both color and Acros with color filtration and without.  I did not make test exposures for the other black and white setting as I only use the Acros settings.

C) I then stood in exactly the same position with the Olympus E-M1 Mark II, set to auto white balance, ISO 200 but instead of recording JPEG files, I set the camera to record RAW files and made a series of images with slightly different exposure settings.  Why a series?  I wanted to be able to choose one of the exposures that exactly matched the luminance level of the Fujifilm exposures I already made.  If I could make everything equal in the field, it would remove any variables in color matching later.

That was about that.  I returned to my computer and started the process of creating the presets.

D) When importing each of the images into Lightroom CC the only adjustment I made on the X-T2's files was to set the white balance by using the third block from the right on the bottom row of the ColorChecker Passport (see image at the top of the post).  That brings all 18 colors I would use for replication into their proper relative color rendition.  The change in white balance was minimal.  Fujifilm's auto white balance, as I mentioned, is excellent.  If the white balance had been horribly off, JPEG files may not have been able to compensate in such as way that the other colors were properly preserved, i.e., too much shift for the JPEG files to handle.  In this experiment, white balance change was minute.

E) Upon importing the images from the Olympus camera, the RAW files, I first changed the Profile within the Camera Calibration tab (the very bottom right adjustment window in Lightroom's Develop module) to one I had previously made from the ColorChecker Passport instead of using one of the Adobe generated profiles such as Adobe Standard, Natural, Vivid, Muted, etc.  For the past several years I have used the ColorChecker Passport profiles for each of my cameras as it replicates colors exactly (according to X-Rite). I used this profile as I needed to have a method for which I could white balance the Olympus RAW file that was considered "an industry standard" for color reproduction as well as establish a standardized baseline for all the other colors.  After the profile was changed, I then white balanced the RAW files in the same manner as the Fujifilm files.  To be honest, the adjustments were minimal as well as Olympus does a very good job at auto white balance, too.

In sum, my intention was to create a reproducible standard which anyone could replicate by utilizing the X-Rite Colorchecker Passport, the Colorchecker Passport profile in Lightroom and white balancing on that specific neutral gray box.  This seemed to me the easiest and most direct way to accomplish that.

F) The final step before starting the process of creating the presets was to select which Olympus file was the exact exposure as the X-T2 files.  I did that by using the same neutral (and middle) gray box on the ColorChecker to measure red, green and blue density.  In this case, the density of each of the Fujifilm files read 60, 60, 60.  I then measured the density of the same box on each of Olympus files and found one that also read 60, 60, 60.  I then knew that exposures from both cameras were identical—and they visually looked identical.  I wanted to eliminate any potential color shifts by lightening or darkening an image file to compensate for a bad exposure.

G) I then needed to create a table with 24 blocks to record the RGB color values for each of the colors, plus the gray scale blocks, I planned to record.  In a document creation program, I created table (see below) and printed out a copy for each film simulation.  I also printed out one for the Olympus file.  I labeled each with the name of one film simulation.  The intention was to measure each color block as well as the neutral gray blocks from white to black, record the red, green and blue color channels on each color block so later I could adjust settings in Lightroom to change the hue, saturation and luminance of each color—as best I could!  Remember, this is an experiment by an amateur.

Here are the original JPEG files, each of which has a different Fujifilm Film Simulation, all taken under the exact same settings, lighting, etc.  The colors may not reproduce exactly in these examples as the color space upon export has been changed to sRGB (for web use) instead of the original Adobe RGB as recorded in the camera.  There are some differnces in color reproduction in the two different spaces.  Additionally, I don't know what Google does to the files between the time I upload them and the time you see them.  I'm sure there is some sort of processing of files for publication.  The last image is the Olympus RAW file also made under the exact same settings, lighting and, again, with a sRGB color space so its colors will appear different from my original.  When looking at the original files in Lightroom, for many of the colors, the differences are subtle.  I found it interesting to study the differences.

Provia Film Simulation at Default Settings (click to enlarge)

Velvia Film Simulation at Default Settings (click to enlarge)

Astia Film Simulation at Default Settings (click to enlarge)

Classic Chrome Film Simulation at Default Settings (click to enlarge)

Pro Neg Hi Film Simulation at Default Settings (click to enlarge)

Pro Neg Standard Film Simulation at Default Settings (click to enlarge)

Acros (No Filtration) Film Simulation at Default Settings (click to enlarge)

Acros Yellow Filtration Film Simulation at Default Settings (click to enlarge)

Acros Red Filtration Film Simulation at Default Settings (click to enlarge)

Acros Green Filtration Film Simulation at Default Settings (click to enlarge)

Olympus RAW file with ColorChecker Passport Calibration Profile applied in Lightroom CC (click to enlarge)

 My procedure was to measure the RGB values in each of the color blocks in each of the Fujifilm film simulation JPEG files, then alter the contrast, hue, saturation and luminance of each color (as best I could) in the Olympus file to match each of the film simulations.  Sounds easy, doesn't it?  Ha!

I started by taking a RGB reading of the color blocks in the Olympus file, then each of the color blocks in the various film simulations.  Here are examples:

(click to enlarge)

(click to enlarge)

Now is where it took several hours to replicate all of the film simulations!  

1) The first thing I did was make additional "virtual copies" of the Olympus RAW file as I would need one for each of the film simulations I planned to create—10 in all.  I then selected a film simulation to emulate.  In our example above, the selection is the Velvia film simulation.  My goal then is to alter the color, contrast, hue and saturation of each of the colors in the Olympus RAW file to match the corresponding color and gray scale blocks in the Velvia film simulation file.  Good luck!  LOL

2) The very first adjustments I made were to equalize the contrast in the target (Olympus) image to match the contrast in the Velvia image.  One thing that I have learned over my many years of color printing and editing is that altering contrast can also alter luminance and saturation of individual colors (luminance is even more important).  If you later match the overall exposure of an image and contrast during the process, you may have to rematch each of the colors. Save yourself some work and frustration in this exercise and start with matching exposure, contrast, then work on hue, saturation and luminance of each color.

In the example above, I have already matched exposures.  You can see that middle gray (the bottom row, third block from the right both equal 60, 60, 60).  Now I want to equalize the Velvia image's brightest white block to the Olympus' brightest white block. The Velvia image's whitest block has a luminance of 98, 98, 98 (bottom left block). The Olympus corresponding block is 95, 95, 95.  The Velvia file is brighter white.  Then look at the Velvia black block (bottom right block).  The Velvia's luminance is 12, 12, 12.  The Olympus' is 17, 17, 17.  The Velvia's black is quite a bit darker. This tells me the Velvia file is more contrasty, which I would expect.  In Lightroom I altered the Curve in the Develop Module so, in each block, the values matched.  This takes a bit of time and is not quite straightforward as I would have thought.  The Velvia's progression from white to black is not linear.  Once the six bottom row blocks match, the contrast of the two images are now equal.  Next step.

3) I then started working on matching the various color block values for red, green, blue, cyan, magenta, yellow, orange, aqua, magenta and purple.  By altering the RGB values in the Calibration tab (bottom tab in Lightroom's Develop module) and also working with all of the colors the Hue, Saturation, Luminance (HSL) Tab, I was able to pretty closely match each of the film simulations. They aren't perfect, but they are pretty close.  I started on the primary colors—red, green and blue, then worked on the secondary colors—cyan, magenta and yellow.  Once I had those pretty close, the rest lined up pretty well.  This took quite a bit of time because every time you change one parameter in one color, it affects that same parameter in other colors.  There was lots of going back and tweaking.

4) I went through this same procedure for each of the 10 film simulations.  Believe me, it took a few hours to accomplish.  For the most part I was able to closely (but not exactly) match almost all of the colors.  The differences, I believe, are visually insignificant and probably won't be detected.  Interestingly, I found the blues the most difficult to match. The blues generated in the Fujifilm files are just different than those in the Olympus files and I've gone back and tweaked several times, but can't seem to get it as right as I would like it. The blues in the color chart are very close, but the blue sky above the color chart looks different.  Confounding.

5) After completing each film simulation, I created a preset and gave it a name such as "Olympus RAW to Fujifilm Velvia."  I can now apply any of the ten presets to any of my Olympus E-M1 Mark II RAW files and, hopefully, get a reasonable match.  Who knows until I can actually get out with both cameras and make identical images using JPEG film simulations in the X-T2 and RAW images with the Mark II.  Since I made these presets in full direct sunlight, I will be interested in how they all translate on overcast days, in shade, and at different times of the day.  I doubt I will carry out experiments for all of these conditions soon.

As a note, I wanted to personalize one of the presets of film simulations I was creating.  I have come to really like the Classic Chrome film simulation and set my X-T2 for that simulation often. It is most likely because I used to shoot loads of Kodachrome-X, Kodachrome 25, Kodachrome 64 and Kodachrome 200.  However, I don't like the way the Fujifilm Classic Chrome simulation reproduces the sky.  It is too cyan for my liking.  I exercised my prerogative to keep the blues bluer than the film simulation provides.  I just don't like cyan skies.  That is just me.  I do love the colors, contrast, hue and saturation of the rest of the colors of the Classic Chrome simulation, however.

I hope this little non-scientific experiment is successful.  Check back in about a week or so and I should have had time to make some "real world" (this phrase is way overused these days!) images to test my work.  After I do, I'll provide some examples and commentary.

This little exercise is a reinforcement to me as to why I love photography as much as I do. Photography satisfies both my left brain/right brain needs. It allows me to be creative and produces some beautiful works of art but it also allows me to exploit technology, experiment, get obsessed with photography's technical aspects and enjoy the gear as much as the images I make.  It is all good!

Thanks for looking. Enjoy! 

Dennis A. Mook 

All content on this blog is © 2013-2017 Dennis A. Mook. All Rights Reserved. Feel free to point to this blog from your website with full attribution. Permission may be granted for commercial use. Please contact Mr. Mook to discuss permission to reproduce the blog posts and/or images.