|Marc Chagall has flattened the space somewhat in this fantasy, but he still provides enough clues for us to tell where things are
Depending on the feeling you want to convey, you can manipulate the variables of form to show the viewer how much depth you mean to suggest in your painting.
The effectiveness of your illusion depends largely on understanding what we are used to seeing. In the painting below, for example, Chloe Yingst uses the scale and placement of the poles to convince us that these few shapes represent miles of space. Her composition, simple as it is, gives us plenty of information about where things are in relation to each other. We will be giving emphasis to composition in today's exercise.
Our brains know that the poles are all the same size, but our eyes are used to seeing them as smaller in the distance. The illusion of space comes from depicting what our eyes are accustomed to seeing. Imagine if the poles were painted all the same size.
In class everyone selected a photo. Those will work well. There are a couple more at the end of this post, or look for something you like that clearly depicts depth.
Begin by identifying the major shapes in your scene. These are the shapes that need to be separated in order to understand where things are. Make a simple pencil drawing outlining the shapes. No need to describe them. We don't need to know what they are. At this point it only matters where they are.
Keep it very simple. The number of shapes should not exceed 10 or 12. If there are lots more objects than that in your scene, consider combining some. Adjacent shapes of similar value can often be combined to good effect. In the study below, there is a definite foreground, middle ground and background. Notice how the various dark buildings in the background have been encouraged to run together. Instead of several shapes, there is now only one. The individual identities of the buildings were not an essential feature of the space. They could be implied rather than specified. The group of buildings did need to be separated from the taxi, however, to better understand how much space is being depicted. We want to know that there is some room between the middle ground and background. This was easier to achieve once the background had been simplified. If those buildings were separated from each other they would have had much more in common with the taxi, making the space more ambiguous.
Read that last paragraph one more time. That point about combining shapes is important.
Once the major shapes are identified and drawn , take a moment to check for unfortunate convergences. Increase overlap wherever necessary to make it easy to tell what's in front of what. It's much easier to move the shapes around when they are only pencil lines.
As you begin layering the lights, middle values and the darks, consider when in the sequence you need to get shapes to separate from each other. It may be that the early stages of the process don't require that you keep the individual identity of each shape intact.
In Trevor Chamberlain's sketch, below, look at the sunlit buildings in the background.
When the pale, warm washes were being applied it was not necessary to make sure the viewer could tell how many buildings there would ultimately be. The washes could run together and still get the definition they need later.
The separation between foreground and background also happens at a relatively late stage. The dark arch does its job more easily if it can be applied right on top of those pale washes, rather than by painting very carefully along an edge.
When did the windows come along, by the way?
OK, here are a couple more images. Have fun!