Getting better at miniature painting

The first year of this hobby has a reliable shape. You learn to prime, to thin your paint, to basecoat cleanly and to shade with a wash, and the models go from embarrassing to perfectly presentable. Then progress stops, and the usual response is to go looking for a technique — wet blending, glazing, non-metallic metal — on the assumption that the gap between your models and the ones you admire is a skill you have not learned.

It usually is not. The gap is nearly always contrast, and the reason it persists is that contrast is a decision made before the brush touches the model rather than a technique applied with it. A model painted with perfect wet blending in a narrow value range still looks flat from arm’s length, which is the distance a tabletop model is actually seen from.

What stalls

Contrast, not techniqueThe most common misdiagnosis in the hobby.

Viewing distance

Arm’s length, not macro photographAlmost all online reference is photographed far closer than anyone will ever look.

What improves fastest

Finishing habitsEdges, basing and a consistent varnish, none of which is a painting skill.

Realistic timescale

Years, in visible stepsThe steps are large and infrequent rather than gradual.

Each of these is worth more than the one after it, and most painters attempt them in roughly the reverse order:

1. 1

   Step 1: Widen the value range before anything else

   Take the darkest recess much darker and the brightest edge much brighter than feels comfortable. Photograph the model in greyscale — if the shapes are not readable without colour, no amount of blending will rescue it. This single change does more than every technique below it combined.

   You will know it worked when: Models that read from arm’s length.

2. 2

   Step 2: Make the edges deliberate

   A clean boundary between two colours reads as competence more reliably than anything happening inside them. Tidying a wobbly edge with the adjacent colour is not cheating; it is how the boundary gets made.

   You will know it worked when: The model looks intentional rather than approximate.

3. 3

   Step 3: Finish the model, including the parts nobody looks at

   Base, rim, undersides. A well-based model with average painting outranks a beautifully painted one on bare plastic in every judgement anybody makes, formal or not — and the base is a fifth of the visible object.

   You will know it worked when: A finished thing rather than a painted one.

4. 4

   Step 4: Vary the surface

   Metal, cloth, leather and skin should not have the same sheen. A satin varnish over everything with gloss on the eyes and metals, or matt on the cloth, separates materials at almost no cost in effort.

   You will know it worked when: Materials that read as different substances.

5. 5

   Step 5: Control where the eye goes

   Put your highest contrast and your brightest colour on the face or the weapon and let the rest recede. A model where everything is equally bright has no focal point and reads as busy.

   You will know it worked when: A model that leads the eye rather than shouting evenly.

6. 6

   Step 6: Then learn the blending techniques

   Wet blending, glazing and layering are genuinely worth having, and they improve a model that already has contrast, clean edges and a plan. Applied first, they are a lot of effort spent inside a value range too narrow to show them.

   You will know it worked when: Technique doing work rather than decorating a flat model.

• Finish models — The single strongest predictor of improvement. A shelf of half-painted models teaches basecoating repeatedly and nothing else, because every skill after basecoating happens at the end.
• Paint the same model twice, a year apart — The only reliable way to see progress in a hobby where your taste improves at the same rate as your hands. Keep the first one.
• Photograph in greyscale — Two seconds on a phone, and it exposes a flat value range that the eye forgives on the desk under a bright lamp.
• Look at models at the distance they will be seen — Arm’s length for tabletop, and further for a unit on a table. Detail invisible at that distance is detail that cost you an evening for nothing.
• Learn one thing per model — A named goal — this one is for edge highlights — rather than a general intention to do better. Improvement in this hobby is a series of specific attempts.
• Get critique from people who paint better — A club night or a forum thread asking what to fix rather than what people think. The useful answer is almost always about contrast or edges, which is the point of this guide.

Two honest notes to set against the online reference. Almost everything you see is photographed under studio lighting at a magnification no observer will ever use, frequently by somebody painting professionally, and often on a display model that took twenty or thirty hours. Comparing a two-hour tabletop model with that is comparing two different objects made for different purposes.

And the improvement curve is stepped rather than smooth. Long stretches of no visible progress are followed by a model that is obviously better than the last one, usually because something finally clicked rather than because that week’s practice was different. That pattern is normal, it is what nearly every painter reports, and mistaking the flat sections for a ceiling is what stops people.