This week, I've been working on how to use color to make a 3D head look like it is round on a 2D surface. My current process starts with making a drawing from observation or a photograph. The purpose of the drawing is to identify the major shapes in the subject and reduce the image 10-15 shapes.
Next, I transfer the drawing to the painting surface and start painting a major shapes in the background first. When I finish that shape, I paint the shape directly next to it. Color changes based on its position on the head, its proximity to the light source, and the colors directly beside it. Working methodically from the background to each neighboring shape preserves the color relationships in the painting.
For me this means I spend most of my time mixing on the palette, searching for the right color because I need to slowly build the form. I get stuck when I'm not entirely sure what to do to fix my color. Does it need to be lighter or darker? More intense or less intense? Warmer or cooler?
I've also started noticing the edges between two different colors. Some of my edges have a clearly defined line and other edges appear to blur together. Hard and soft edges could be used to define the light and shadow areas on the head.
But when I paint, I'm only vaguely aware of when to use a hard or soft edge. It was only this week I learned that hard edges are usually created because of strong differences in value; soft edges between two color are painted with a middle value between the light and shadow.
This week I am experimenting with blocking in the large groups of shadows and lights in the head first. Instead of stitching the individual shapes together like a patchwork quilt, I am working with large shapes and building up layers of opacity and details. I focus on seeing the entire picture and I delay adding hard edges until I am confident with the larger value relationships.
See Noel Robbins' article on "Notes on Perceptual Painting: Perceptual Painting Theory and Practice" for more a checklist of how to paint using color masses and keeping soft edges open longer.
See Alex Tzavaras' video, "One of the Biggest Mistakes People Make When Portrait Painting" on why you need to reduce the portrait to 3 or 4 major tonal values. He also discusses how to use Photoshop to help blur and reduce details into a few value.
I still start with a drawing to place the head in relationship to the edges of canvas, but I draw directly on the canvas and focus on the angles between key points on the head: the bridge between the nose, the distance between the bottom or the eyes to the tip of the nose, the position of the mouth in relationship to the bottom of the chin, the ear in relation to the eyes, nose, and corner of the mouth.
I also reduced my palette to just four colors: titanium white, raw umber, ultramarine blue, and burnt sienna. I want to focus on achieving accurate values through color mixing; having a limited palette simplifies the number of options I have to mix a color. It's less about finding the exact shade of brown; it's more about how this specific brown relates to the colors next to it.
Other New Ideas I Learned This Week
Make Multiple Variations of the Same Thumbnail
In the Learn to See class with Gabriel Lipper we worked on identifying images we liked and designing pictures with thumbnail sketches. I'm already a big fan of thumbnails since the 20 for 20 Art Challenge earlier this year. Gabe suggested printing out multiple copies of your thumbnail images and painting directly on top of them with black or white acrylic paint to explore value.
Research Filipino Historical Photos Online
I discovered historical photos on the Filipino American National Historical Society website and on Flickr.com, primarily posted by John Tewell. Many of the Tewell's photos are reproduced from library collections and his own personal collection of real photographic postcards. An interesting starting point for future paintings.
Use Photoshop to Scale and Print Images on Your Printer
Ali McKay offers a very useful Photoshop tutorial in her Fresh Paint membership community on how to resize images to fit your canvas and tile the images so they can be printed using a printer. This will help with scaling up smaller photos and converting them to paintings.