Elven Academy

Overall this was an outstanding book, and affordable for such an abundance of color images and illustrations. Many concepts were not adequately explained in the book, which is ironic considering how lengthy and pedantic the book is in general. I don’t think I’d recommend this book to a beginner, or someone who gets bored easily.

In fact, I had to turn to ChatGPT to explain things to me, and frankly the AI did a better job than the book. In these cases, Color & Light was more of a summary or starting point. Since I was taking notes and adding from ChatGPT, I recorded these notes and personal takeaways from this book. This is sort of a summary of key points in the book for future personal reference.

***

Everything is subjective. We know hue, color temperature, and value shift based on their surroundings, but don’t forget the size of a shape also changes. We can use all of these things to our advantage.

Munsell Color Theory. This is the Hue, Chroma and Value we are familiar with. Chroma is actually purity of color, not intensity or saturation. There is a distinction.
Value – First and most important.
Chroma – Second importance, but most undervalued by students.
Hue – actually not that important. Overvalued.

Limited Palettes. These can be temperature limited as well as value (i.e. Grisaille or Verdaccio), and hue (Zorn). A temperature limited palette might be. Titanium White, Ivory Black, and Cad Orange.

Value grouping. Since value is most important, our first goal in painting is to simplify and establish major value groups. This is called “value grouping.” This may use a technique called “value clipping”, which is like overexposing or underexposing a photo to blast out a range of tones.

Value keying. This is the process of choosing our value range, i.e. high key, low key, and middle key. Each key has a mood. Mid-key feels restful and harmonious. Low key can feel dark and brooding. Opposite key points of the painting become focal points.

Exponential Drop-off. We will always see value scales as linear, but our eyes have evolved to enhance contrast between light and dark, which means value hits darks on a more exponential scale than linear, at the transitions. So a lit sphere doesn’t actually look evenly lit. It drops more quickly from dark to light at its middle, than we would logically think.

Simple Shapes (Bargue drawing). Simpler value masses make stronger shapes. Adding more and more planes and shapes can result in overworked, broken-up, and overmodeled artworks.

Edges. “Soft”, “firm” and “sharp”. Used to draw attention or de-emphasize to release attention.

Color. We have different theories because light combines additively (RGB model is Red + Green + Blue = white, due to cones in our eye), while paint mixes subtractively (CMY model is cyan, yellow, magenta mix to black). Due to this, it is recommended to limit the number of pigments in a paint mix.

Split-primary palette. One way to resolve the color theory problems is to view RGB as warm, and CMY as cool.

Book example of a palette – Cad Red, Cad Yellow, and Ult. Blue are warm. Bismuth Yellow, Cobalt Teal, and Quin Rose cool.

ChatGPT palette – uses Quin Magenta instead of Rose. Rose and Magenta are both PV19, but Rose is warmer, mixes better oranges but worse purples than Quin Magenta. So Quin Rose is better for flesh tones, but Quin Magenta is better for pure color. This palette uses Pthalo Blue (Greenish) instead of Cobalt Teal, because the teal is weaker tinting strength. Also uses Hansa Yellow Light (PY3) for same reason, higher strength, over Bismuth Yellow (PY184).

Complements. RGB color scheme complements mix to muddy brown, while CMY complements (Red + Cyan, or Green + Magenta) should mix to grey, which is interesting.

Color Temperature. Warm colors tend to advance, feel energetic, and catch the eye. Cool colors tend to recede, feel calm or distant. But temperature is relative. A warm blue still looks cool next to orange, but warm next to violet.

Warmth of Sunlight. Sunlight warms by passing through the earth’s atmosphere, where blue light is reflected and scattered by the earth’s atmosphere. The sun angle in morning and evening causes the sunlight to pass a greater distance through the atmosphere, removing the most blue light. So conversely it is least warm at midday.

Color of Shadow. Shadow color should be the color of ambient light reflecting onto that surface. Typically light areas are warm and shadows are cool in paintings. This works for landscapes with blue skies for example. With cool light, we would render shadows as warm, but only as an unspoken convention to create energy in the painting, per ChatGPT. Sometimes landscape objects do not have sky ambient light affecting them, i.e cliffs facing a bright yellow plain. Here the yellow is the ambient light on the cliff face.

Hue value. Important to keep in mind every hue also has an inherent value.

Ambient Occlusion. Environmental ambient light is the primary driver of colors within the shadows. Every object radiates its own ambient light, and this is why shadows are not black. As objects become closer together, they block more and more ambient light. This ‘ambient occlusion’ is a very subtle but observable effect, darkening the spaces where objects are close or in contact with each other.

Ambient Shadows. A basic rule is “50% to black.” Ambient shadows will be 50% to black against a white background, 75% to black on a mid-grey background, and 100% to black on a pure black background (i.e. the Moon).

50% to black. So shadows will be at least 50% to black, depending on the background. And this value relationship should be consistent through the painting, adjusted for ambient occlusion. Use a printed value scale to judge the relationship. Squinting can also help judge these value relationships.

Conclusion:

The main things I learned from this book were names for concepts I was already practicing intuitively. These would be value grouping, value keying, and ambient occlusion.

I also found useful the idea that shadow color should be the color of the ambient light. It was interesting to confirm that warm brown shadows on snow, for example, which are seen in master paintings, is not reality. It’s a convention.

I learned the difference between Quin Rose and Quin Magenta, and when they should be used. The idea suggested by ChatGPT to use Pthalo Blue instead of Cobalt Turquoise is interesting. Since I do not use high chroma in my paintings, I will probably not go on that adventure.

And then exponential dropoff seems like a key concept to painting realistic fabrics. for example. I am probably trying to paint more logically and not taking this into account. So that’s my review of this book. Hope you learned or reinforced something. Thanks for reading.