Friday, July 28, 2017

Relative color temperature on skin tones

Mathieu asks:
I have been struggling over the last months trying to understand how to handle the green parts (cool notes) of the flesh in a portrait.

In the Fundamentals of Painting by Mogilevtsev I could not understand the following text: "The light consists of three parts: the highlight and halftone are cold, and the light space between them is warm." I thought only the shadows are warm and all the parts of the light will always be cold. So what does he calls "the light space" between them? 

Gurney—On page 22, Mr. Mogilevtsev implies that the "light space" is the area between the highlight and the halftone (the halftone is the area just before the light side turns to shadow). What he calls the "light space" might also be called the "lights" or just "the light side." 

Within that light side there can be subtle variations in color temperature. 

He is indeed painting the halftones cool, specifically greenish. The halftones are an important area to observe closely for their value, for the abruptness or softness of the edge as it turns to shadow, and for their relative color temperature. 

He makes the point that relative warm and cool tones can give life to a portrait and I would agree with that. His states that the rules he's talking about refer to painting a portrait indoors. Traditionally an indoor portrait would be lit by a relatively cool north-facing skylight. In that case, the lights are generally cool compared to the shadow, because of the coolness of that blue skylight relative to the bounced light of a wood floor or warm-colored rugs, etc.

Charles Hawthorne
However, I would be skeptical of any fixed rules about cool/warm relationships, such as saying "outdoors in sunlight the light is always warm and the shadows are always cool." If there is very warm light reflected back into the shadows, or a secondary light that is very warm, the shadows can be warmer than the light side. 

Or the color of light in the shadow can vary according to the direction the planes are facing, such as a person standing at the beach, with blue sky above, warm sand below to one side, and blue water below in another direction. It all depends.

On the portrait of the woman, I don't understand the logic behind the green parts of the flesh colors. Where to put those greens? Are they at the edges of the planes right before they turn? 

Gurney—Yes, he seems to be placing the greens at the turning of the form. This is something that old masters often did. It may or may not look convincing, depending on how it is handled. Sometimes this cool effect in halftones is the result of the way you glaze color over a grisaille or "dead color" underpainting.

Let's step back for a minute to remember that the appearance of any flesh tone color, whether in light or shadow, is a combination of: 
1) the color of the surface (local color)
2) the color of the light 
3) plus additional factors as subsurface scattering. 

So, warm local color plus warm light equals a very warm color note. 

I try to consider first the local color as it varies across the form. The color across the mask of the face can vary a lot, as any makeup or prosthetic specialist will attest. It's often redder in the cheeks and nose, darker around the eyes, lighter and yellower on the forehead, and bluer or greener in the neck or chin, plus there are effects caused by makeup and sunburn. 

The reflectivity of the skin varies too, and that factor can influence your color and value choices.

Then I consider the sum total of the colors of light shining on each plane. It might help to place a white plaster head near the model in the same light in order to study those influences.

The color you mix for any given plane will be a combination of all those factors. 



Mathieu continues: When speaking about "cool" or "cold" colors in the light areas why do I always feel that the yellowish and reddish color of these parts is warm? On the above portrait by Rubens the light areas doesn't seem "cold" to me.

Gurney—You're right that most skin tones are on the "warm" or orange side of the spectrum, but we're speaking of a relative thing here.

Sometimes it can be hard to judge relative light color when looking at a living model. That's why painting from a white plaster cast can be helpful for understanding both form and light. By removing the effect of local color, you can see what's going on with the relative temperature of the light.

Bottom line: be skeptical of fixed rules, be guided by your observation, and always compare, compare.
-----
Previous Post: Color Zones of the Face

7 comments:

artsworth said...

That is a brilliant answer and typical of this blog. A big thank you James,for taking the time out to pass on your knowledge. Regards,Will

broker12 said...

When I run into an obvious green, I often use terra verte mixed with my flesh tone. It is a weak color, and therefore, easy to manipulate. Sometimes, I'll use a dab of viridian mixed in with my flesh tones, but viridian can quickly overwhelm the mixture. In some instances, plain old yellow ocher and ivory black or raw umber will do. I often see subtle green shades, and for some reason it is too easy to get their value too high and they end up looking too bright next to the other flesh tones. This is an informative and educational post. Thanks.

James Gurney said...

Thanks, Broker and Artsworth. Good tip about the terra verte. I like that using in landscape, too.

Don Rogers said...

In Chris Legapi's videos, he explains that it is a 'law of color' that when you change value you should/must change temperature. Therefore in the halftone, we must understand it's temperature and value too; then, when we mix to the next lighter value, it should change value and temperature simultaneously. Adding white to a color will basically make it appear lighter in value and cooler in temperature due to the white pigment. Will that value change also qualify as a temperature change? I too, remain in that area of confusion but we must keep practicing and working on it all. Thanks for the tip to work from a white cast to identify the color of your lights, halftones and shadows.

Don Ketchek said...

While books can certainly be helpful, always beware when anyone spouts "rules." What James Gurney wrote is far better advice than found in most books - and, alas, too short a text to fill a book or even a page:

"Bottom line: be skeptical of fixed rules, be guided by your observation, and always compare, compare."

Color is relative to what other colors are near and adjacent. So comparing is a must. And in some cases, the "rules" have become simplified and distorted over time. Whereas today, many artists quote the rule: "warm light, cool shadows" and "cool light, warm shadows" - it is far better (when using this guideline or rule) to think: warm light, COOLER shadows, and, cool light, WARMER shadows. In other words, with warm light, the shadows need not be a cool color, they are just COOLER than the color in the light, and that color may still be considered warm in comparison to other colors.

Anon Nimus said...

I think of it this way. The light most facing part of the form will be the highest value and chroma of the local color, and as the form gradually turns away from the light towards the shadow terminator it loses both value and chroma. I mostly use raw umber to lower the value and chroma of skin tones. Low chroma colors or near neutrals can appear cooler or warmer depending on the surrounding higher chroma colors due to simultaneous contrast. Highlights are not part of the form but reflection on the form of the light source, which can either be warm (sunlight, incandescent light, candle) or cool (northlight, overcast light). Reflected lights in the shadows can also be warm or cool depending on the secondary light source or color of the object reflecting on the shadow side, and same principle as in the light side, they lose value and chroma as the form turns away from the reflected light towards the shadow terminator.

Mathieu N. said...

Hello James, I just discovered this article now, thank you very much for having developed the answers to my question point by point through examples. I see things more clearly now.

All the best,

Mathieu.