How I Simplify a Reference Before Painting
A reference photo has too much information. Every leaf, every wrinkle, every reflection — a camera captures all of it with equal emphasis. Your job as a painter is not to copy that information. Your job is to design a painting from it.
That means simplifying. And simplifying before you pick up a brush saves you from the most common trap in painting from photos: getting pulled into details before the big structure is working.
Here's the process I use every time I sit down to paint from a reference. It takes about five minutes, and it's the single thing that's improved my painting the most.
The Problem With Painting Directly From a Photo
When you set a reference photo next to your canvas and start painting, your eye does what eyes do — it goes straight to the interesting parts. The catchlight in the eye. The pattern on the fabric. The way the light hits one edge of a leaf.
None of that matters yet.
What matters is the big picture: where are the large shapes? What are the major value relationships? Where is the lightest light and the darkest dark? Is the composition actually working?
If you don't answer those questions first, you end up with a painting that has beautiful details sitting inside a weak structure. We've all done it. The eye is perfect, but the head is the wrong shape. The tree bark looks great, but the overall value pattern is flat.
The fix is to strip the photo down before you start — so you can only see the things that matter in the first hour of painting.
Step 1: See the Big Shapes First
The first thing I do with any reference is shrink it. Not literally make it smaller on screen, but reduce it to such a low resolution that detail disappears entirely.
When you're looking at a photo that's been reduced to roughly 80 by 60 pixels and then scaled back up, you can't see detail anymore. You can only see shapes and values. That's exactly what you need.
This is the digital equivalent of squinting — something every painting instructor tells you to do, and almost nobody does enough. Squinting works because it knocks out the mid-tones and details, leaving you with just the big value masses. A pixelated thumbnail does the same thing, but you can study it without giving yourself a headache.
Open the Thumbnail Sketcher and load your reference. Drag the resolution slider down until you can only see 4-6 major shapes. That's your composition.
What you're looking for at this stage:
Is there a clear pattern of light and dark? If the thumbnail looks like a uniform gray mush, the photo might not have a strong enough value structure to make a good painting — or you'll need to push the values yourself.
Where do your eyes go? In the thumbnail, with all the detail gone, your eyes will go wherever the strongest contrast is. Is that where you want the focal point? If not, you know you'll need to adjust.
Does the design hold up? You can see the abstract composition clearly now. Does it feel balanced? Is there a dominant value? Are the shapes varied and interesting, or are they all the same size?
I spend about a minute here. Sometimes I realize the photo I thought would make a great painting actually has a weak composition — and I'd rather discover that now than two hours into a painting.
Step 2: Block In the Color Masses
Once I know the big shapes are working, I want to see the photo reduced to flat masses of color. Not the infinite gradations of a photograph — just the five to eight major color areas that I'd actually mix on my palette.
This is where block-in comes in. By reducing the photo to a limited number of colors, you get a version that looks much closer to what an actual painting looks like in its early stages — before any blending or detail work.
Open the Block-In Tool and load the same reference. Start with 5 or 6 colors and adjust until the image reads clearly as flat shapes. Toggle grayscale to check that the values still separate.
This step answers a critical question: what are the actual color masses in this scene?
Photos trick you. A "green" hillside might actually contain warm yellows in the sunlit areas and cool blue-greens in the shadows. When the photo is reduced to flat color masses, you can see those distinct zones clearly. That's valuable information when you're mixing paint.
The grayscale toggle is especially useful here. When you switch to grayscale, you can instantly see whether your color masses are also separating in value. If two different colors collapse into the same gray, you'll lose that separation in your painting — unless you consciously push one lighter or darker.
I often screenshot this blocked-in version and keep it next to my easel as I paint. It's a better reference than the original photo for the first pass of a painting, because it shows me what to simplify rather than what to copy.
Step 3: Check the Value Structure
Value does the heavy lifting in a painting. Color gets the compliments, but value is what makes the image read. A painting with accurate values and wrong colors will still look convincing. A painting with beautiful colors and wrong values will look broken.
After I've seen the big shapes and the color masses, I want to check the value structure specifically. How many distinct values are actually in this scene? Are the values in the right relationships?
Open the Value Scale Mapper and load your reference. Hover over key areas to read their value numbers. Then toggle to a 5-step view to see the simplified value map.
Most paintings work best with a dominant value — one value that covers the majority of the canvas. When you look at the stepped value map, you can see immediately whether your scene has that dominance or whether the values are spread too evenly.
A landscape with strong sunlight might be mostly high-key (light values dominating), with a few dark accents. An interior scene might be mostly mid-tone, with one bright window pulling focus. The value map makes these patterns obvious in a way that staring at the full-color photo doesn't.
I pay special attention to the value of my focal area versus the value of the surrounding areas. The focal point almost always needs the strongest value contrast in the painting. If the value map shows that the contrast is actually stronger somewhere else, I know I'll need to make adjustments.
Putting It Together
Here's what this process gives you in about five minutes:
From the thumbnail, you know the composition works and where the eye goes. From the block-in, you know the actual color masses you need to mix. From the value map, you know the value structure and where to place your strongest contrast.
That's a painting plan. Not a vague sense of "I'll just start painting and see what happens" — an actual plan for what goes where, how light or dark each area should be, and what colors you'll need on your palette.
When I start painting with this kind of preparation, I spend the first hour laying in big shapes with confidence. I'm not guessing. I'm not constantly checking the photo. I already know what the simplified version looks like, because I've already seen it.
And that first hour of confident big-shape work is what separates paintings that hold together from paintings that fall apart despite having good detail.
When to Break the Rules
This process is not a rigid formula. Some paintings don't need all three steps.
If you're painting a simple scene with a clear composition — a single figure against a plain background, say — you might skip the thumbnail step. The composition is already obvious.
If you're doing a value study or a grisaille, you might skip the block-in and go straight from thumbnail to value map.
And sometimes, honestly, you just want to paint. You know the scene, you've painted similar subjects before, and you want to respond to the reference intuitively. That's fine. The tools are there for when you need them, not as a mandatory checklist.
But when I'm working on a painting that matters — a commission, a competition piece, a subject I haven't tackled before — I always come back to this process. Five minutes of simplification saves hours of correction.
The goal is never to copy the reference. The goal is to understand it well enough to paint something better.