Why onion skinning is the most underrated tool in frame-by-frame animation

Every frame-by-frame animation software has it. Almost none of them treat it like it matters. Here’s why that’s backwards.

Ask an animator to name their favorite feature in a tool, and you’ll hear about the brush engine, the rigging system, the export options. Nobody says "onion skinning." It’sthe feature that just has to be there, checked off a spec sheet somewhere, rarely the reason anyone picks a tool.

That’s strange, because onion skinning might be the single most load-bearing feature in frame-by-frame animation — the one thing standing between "this motion feels alive" and "this motion feels like a slideshow."

Why onion skinning matters for animation spacing

The idea is old: animators were holding drawings up to a lit glass table decades before anyone had a computer to do it for them, tracing the ghost of the previous pose onto the new page. Digital onion skinning is that same light table, except the ghost updates itself: draw a frame, and the frames around it show up faintly behind your pencil, past poses in one shade, future poses in another.

The reason it matters more than it gets credit for: frame-by-frame animation is fundamentally a spacing problem, not a drawing problem. Any individual pose can be beautifully drawn and still ruin the motion if it’s spaced wrong relative to its neighbors. You can’t judge spacing by memory. You need to see the actual gap between where the foot was and where it’s going, while you’re deciding where to put it now. Without that, you’re animating blind and hoping the arc holds up on playback — and it usually doesn’t, not on the first pass.

Common problems with onion skinning tools

Look closely at how differently tools treat this "obvious" feature and the gap is bigger than it should be. Some implementations only show a flat silhouette — no sense of what was actually drawn, just a shape standing in for it. Some don’t let you control depth (how many frames back or forward you’re seeing) or opacity independently. Some render the ghost as a completely separate pass that lags behind your current strokes, so the light table itself feels laggy while you’re the one trying to move fast.

And an unsolved problem across the industry, even in respected tools: once you’re working with multiple layers per character (a face layer, a body layer, an effects layer), almost nobody has cleanly figured out whether onion skinning should show the whole scene’s ghost or just the layer you’re working on. Both have real uses. Most tools pick one and don’t let you switch.

What good onion skinning should look like

In Formo, the ghost isn’t a stand-in. It’s rendered through your actual hand-drawn strokes, in the same order you painted them, so what you see behind your current frame is what you drew, not a simplified silhouette pretending to be it. You get two ways to read it: a filled tint for fast silhouette reading, or an outline for when you need to see through the ghost to what’s underneath it. And because it runs on every redraw — every drag, every scrub through the timeline — it never lags a beat behind your hand.

None of this is a flashy headline feature. It’s not going to be the first thing anyone screenshots. But it’s the difference between animating with your eyes open and animating from memory, and if you’ve ever redone a walk cycle four times because the spacing quietly drifted somewhere you couldn’t see, you already know which one you’d rather be doing.

See the frame before and after, while you draw