FreePainter Studio Notes · Pilot issue
Look closer.
Paint with intent.
Controlled medium studies and practical visual decisions for turning a photo or an idea into a painting—not a filter with brush texture.
- 25
- original studies
- 1
- variable at a time
Studio index
Choose the decision you need to make
Each note ends with a reusable brief, a failure check, or a selection rule.
The Best Photo for a House Portrait: A Five-Gate Source Test
We redrew one fictional house from four source-photo conditions. The useful reference kept the roofline, façade rhythm, defining details, and setting legible without forcing the painter to invent hidden geometry.
Grid Method Drawing From a Photo: A Ratio and Cell-Count Test
We held one fictional portrait and every landmark fixed, then transferred it to a proportional canvas, a mismatched canvas, and two grid densities. The grid only protects proportion when the crop ratio and cell map agree.
How to Simplify a Photo for Painting: A Four-Pass Test
We held one still life, crop, and light fixed, then reduced it to five values, seven large masses, and one selective-detail map. The goal is not a blurry photo; it is a reference with a clear order of importance.
How to Paint Water Reflections From a Photo: A Three-Pass Test
We rebuilt one fictional lakeside reference in three passes. The reflection became believable only after we aligned its large vertical masses, compressed its value range, and let horizontal water marks interrupt the image.
How to Combine Reference Photos for a Painting: A Four-Gate Test
We assembled one fictional still life from three reference sources, then fixed viewpoint, scale, light, and color in order. The reliable composite has one lead photo and several evidence photos—not three competing realities.
How to Crop a Reference Photo for Painting: A Fixed-Frame Test
We held one fictional harbor scene and a 4:5 canvas still, then compared a centered crop, a subject-only crop, and a story crop. The useful frame preserves the relationship that makes the scene worth painting.
How to Simplify a Photo Background for a Portrait Painting
We held one fictional portrait still while copying, blurring, erasing, and deliberately editing its background. The useful version keeps one story cue, groups values, quiets competing edges, and integrates the silhouette.
Focal Point in Painting: A Three-Pass Competition Test
The same landscape, crop, and subject show why a focal point works when value, color, edges, and direction agree—and why one bright accent is often not enough.
How to Photograph Artwork Without Glare: A Three-Capture Test
A direct-flash, angled-camera, and controlled two-light comparison shows why geometry, glare, and color reference must be solved before editing.
Can You Paint from a Low-Resolution Photo? A Four-Source Test
One fictional portrait, two downsampled copies, and one full-size blurred copy show why usable identity evidence matters more than an enlarged pixel count.
How Large Can You Print a Digital Painting? A Pixel-to-Paper Test
We held one 3000 × 4500px painterly file still and mapped it to 300, 240, and 150 PPI outputs. The useful limit comes from pixel dimensions, target inches, crop ratio, and viewing distance—not the DPI label alone.
How to Change Reference Photo Colors Without Losing the Painting
We held one fictional still life and value map still, then changed hue, value, and saturation separately. The reliable recolor keeps the light-dark roles first, limits the palette second, and spends its strongest color at the focal point.
Atmospheric Perspective in Painting: A Three-Zone Landscape Test
We held one fictional landscape and composition still, then separated contrast, color, and detail to see what actually creates distance. The reliable rule is a coordinated recession—not a blue wash over every far hill.
Hard and Soft Edges in Painting: A Four-Pass Portrait Test
We held one fictional portrait, crop, values, color, and light still, then changed only the edge plan. The useful rule is selective, not stylistic: keep the identity anchors found and let supporting boundaries soften or disappear.
What Color Underpainting Should You Use? A Four-Start Test
We held one portrait, crop, drawing, and side light still, then compared neutral gray, warm earth, complementary green, and local-color starts. Choose the layer that solves your first painting problem.
How Many Values Should a Painting Study Use? A 2–3–5–9 Test
We reduced one still-life source to two, three, five, and nine fixed grayscale bands. The useful answer is not ‘more’: start with three, then add only the boundaries the painting needs.
The Best Reference Photo for a Portrait Painting: A Four-Test Audit
We held one fictional sitter, expression, clothing, and background stable, then changed camera distance, crop, and light. Four checks reveal which photo can carry a likeness into paint.
Crop, Resize, Export: A Social Image Workflow That Protects the Painting
Keep one untouched master, compose multiple crops, resize only at the end, and export format-specific derivatives without turning painterly texture into crunchy noise.
Character Reference Sheet for Artists: What Every View Must Prove
A painter-focused reference-sheet method for preserving proportion, costume placement, materials, palette, face structure, and gesture across new artwork.
Why AI Paintings Look Like Filters—and the Four Structural Fixes
A painted portrait is not a photograph with identical swirls everywhere. Four before-and-after pairs show how value grouping, simplification, edge hierarchy, and directional marks rebuild the image.
Oil, Watercolor, Gouache, or Pastel? A Pet Portrait Style Selector
We kept one golden retriever’s identity, pose, crop, collar, and light stable across four treatments. Here is what each medium preserves—and the source photo each one needs.
Eight Paint Marks Worth Naming: A Visual Glossary for Better Image Prompts
Impasto, dry brush, wash, bloom, scumble, flat opaque shape, pastel veil, and contour are not synonyms for “painterly.” Each mark changes a different part of the image.
Which Photos Survive Watercolor? A Five-Point Source Image Test
A clear pear, a backlit cyclist, a dense garden, and a gray cat reveal the same lesson: watercolor rewards separation and punishes visual sameness. Use this scorecard before uploading.