Skin Tone Shootout – Cast your Vote!

by planetmatt11 Comments

Why is skin tone such an important deal? Shouldn't we be looking at trees and bricks for aliasing? Eh, not so much.

At the end of the day, your audience – especially with a narrative film – is going to relate with humans. And human faces tend to show emotion which furthers your story. And that's the essence of why we pay our A-list actors so much, right? So shouldn't our cameras help us in capturing this detail?


The skin is translucent and very rich in luminance and color. Film stock has often been touted as rendering this translucence extremely well while digital technology tends to ‘smear' or make people's faces look ‘plasticy' – rather flat in tone, missing subtle flesh richness. In DSLR still photography, huge pains with sensor and color technology, coupled with RAW editing ability have allowed photographers to get back to achieving superb skin tone and detail – which was first missed with early implementations of compressed, low megapixel jpg stills.

Regarding video, it seems we are working where early DSLR still photography began: low resolution (1080P is around two megapixels), highly compressed images with limited color depth (8bit).

So, let's look at a few still images taken from a shootout video Adam Roberts created that will be shown later after the poll ends. A few very popular cameras were used for this shootout. Adam explains the set up was a single 1K tungsten fresnel bounced off a polly board (camera left) with a silver fill reflector (camera right). All Cameras were set to ISO 800, 3200k, 25P, color matched in FCPX. I'll be the first to say this is not completely scientific and the color grade may not be to your taste (or mine). But I do think this is a good test.


So which do you prefer? A, B, C or D and why? Remember, look at color, detail and subtle skin translucent rendering.

After the poll, I will discuss the power of color grading and the difference of still photography and motion photography in regards to how we perceive skin tone detail. Please view full screen and diligently study the images, that would be beneficial.

Click for full size

Click for full size

Click for full size

Click for full size

[poll id=”34″]

Matt Battershell is a professional graphic designer, web developer and photographer who studied filmmaking since first getting his hands on a Media100 workstation in 5th grade. Applying design principles to typography, motion design, editing, color grading and cinematography has been a rewarding experience as he continues to use HDSLR's and other tools to hone his skill. Understanding technology is a passion, so whether he is building a custom hackintosh to run Davinci Resolve faster than a MacPro or learning about CMOS sensor designs, there is always room for is inner nerd to geek out. View some of Matt's photography.

(cover photo credit: snap from the video)


  1. I’ll admit that human skin tones are something that I have a hard time judging. However, I could throw out A for having really saturated reds, especially in the cheeks and tip of the nose. I felt the same way about B, but the shift was to yellow and made your talent look a bit jaundiced. With those two out of the way, I could pay attention to the details in C and D. While I feel that C may have truer color representations in her lipstick and in her hair, it looks to me as if the model has some sort of bronzer on (I’m a redhead too, and I’ve never been able to get that skin tone). D may have undersaturated some external colors, but I believe that the skin is best depicted through the D camera.

  2. I much prefer D. To me it shows the model most natural looking and for sure the most pleasing of the shots. I am no expert, just know what I like.

  3. C and D are from the same camera, I think the BMC. I would also say D is representative of the most natural looking skin tones.

  4. What happens if you correct the colour balance on the BMC in c/d so the wall is white?

  5. The wall looks like a blueish cyan so I assume everything else would have a yellowish orange tint, which is on the -I line of a vector scope so it would exaggerate her skin tones.

    In fact you can see just that happening in A and B; the background there looks more grey and she looks more orange.

  6. I was puzzled with the choice of the tungsten source as the sensors on the current crop of cameras are more friendly to daylight balance. That said:
    A: A bit to magenta. B: The transition from her cheek seems a bit harsh for my taste. C: Favorite. Enough saturation that it looks wonderful. D: Feels like it’s a bit flat…almost ungraded. In each, it’s difficult to make a great choice as we’re not on set and don’t have the benefit of seeing what the model looks like. If I wanted a dull flat sort of look I might go for D in a sort of overcast day sort of way. If I wanted a bit warmer feel I might choose C. It’s all in the eye of the beholder.

Leave a Comment