Stack the plates. Pick the round. See what actually happens.

ArmorSim is a 3D terminal ballistics simulator built for people who argue about armour for fun. One question drives the whole thing: will it go through?

You build the target — plates, thicknesses, angles, spacing, materials. You build the projectile, or take a real one from the American, Soviet/Russian and German arsenals. Then you fire, and the impact is solved, not animated. The rod grinds itself down. The plate dishes, tears, punches out a plug or opens a ragged hole. Fragments fly with real mass behind them. You orbit the camera around the whole thing in slow motion and look at the damage from any angle you like.

No sparks. No glow. No explosion smoke. Just metal doing what metal does.

Three solvers, one sandbox

NORMAL — the fast one. Analytic terminal ballistics: Alekseevskii–Tate erosion integration, Recht–Ipson residual velocity, a Rosenberg ricochet criterion, Grady fragmentation. Answer in a fraction of a second. Fire, tweak one number, fire again.

ADVANCED — the same core with its guts on the table. Layered and spaced arrays, reactive armour packs, per-layer energy accounting. Ballistic limit, wear fraction and residual velocity reported for every plate the rod passes through, so you can see exactly where the shot ran out of energy.

ULTRA — a genuine continuum solver. An axisymmetric Lagrangian mesh running Johnson–Cook plasticity, a Mie-Grüneisen equation of state, adiabatic shear banding and element erosion. It takes minutes to produce a single impact. The crater shape is not drawn by an artist — it falls out of the physics.

Physics, not hardness points

There is not a single hit point anywhere in ArmorSim. Every outcome comes out of published models driven by real material properties — density, yield strength, shock data from laboratory tests. Rod erosion, ragged hole edges, plug ejection and fragment formation are consequences of those numbers, not animations played back on cue.

It is not a hydrocode, and it does not pretend to be one. An ANSYS-class simulation of this impact runs for hours on a cluster. Here is what ArmorSim does instead of pretending:

• Every material constant carries a source. The material table tags each number MEASURED, DERIVED, FITTED or ASSUMED, and names the paper it came from. It is not hidden — you can open it and check.

• Absences are documented, not filled in. Where the open literature holds no usable data, the material is not quietly faked with a stand-in. It is marked unsupported, and the reason is spelled out.

• ULTRA states its own limits. The continuum mode is axisymmetric. It runs head-on impacts, and it tells you plainly when a scene falls outside what it can honestly solve instead of guessing at it.

• The solvers never contradict each other. ULTRA renders the shape of the damage; the analytic core is the only thing permitted to declare an outcome. You will never get two different answers to the same shot.

What you get

• Free-build scene editor — add, drag, rotate and re-material plates; spaced and layered stacks

• An extensive material database: 14 armour materials and 4 penetrator cores

• Kinetic penetrators drawn from American, Soviet/Russian and German service ammunition

• Reactive armour, ceramic with backing, composite backing plates

• Shaped-charge jets at normal incidence

• Slow-motion playback with a free orbit camera

• Shareable replay files — a full ULTRA run is a couple of megabytes

• Five interface languages: English, Русский, Deutsch, Polski, Татарча

Who this is for

If you have ever paused a documentary to argue about whether that shot should have gone through — this was built for you. No engineering degree required, and no engineering software to learn. Set it up, hit fire, watch it happen.