This document is both a current-state atlas and a design deck for where the simulation can go next. The strongest axis in the project right now is already heating, cooling, curing, softening, burning, and gas-driven motion. The next step is to make temperature not just a color or stiffness modifier, but a source of internal stress, layered transformation, gas generation, and structural memory.
Two of the directions sketched below have since shipped and are documented in full on The Explosive Lab: gas generation inside a soft matrix, and reactive burst / fragmentation events.
The project is already strongest when multiple representations overlap: deformable MPM matter, SPH liquid, hot buoyant air, and conductive rigid SDF walls.
Best for water-like behavior, splashes, drips, filaments, and codimensional flow through narrow gaps.
Dam Break and Codim ThreadsAlready supports curing, glass-like softening, fracture, fiber anisotropy, burning, embers, porosity, and hot-fluid oddities.
Smoke, buoyancy, wind, thermal diffusion, and drag make heated matter feel alive instead of just color-shifting in place.
These are the most interesting current heat-first material families and what they are trying to show.
| Family | Examples | What it demonstrates | What to expect |
|---|---|---|---|
| Glass-like | Glass [experimental], Tempered Glass [experimental], Hot Glass [experimental] |
Reversible soften / reharden behavior | Should stay rigid when cool, get viscous only in a hotter band, then recover stiffness on cooling |
| Ceramic / kiln | Terracotta, Porcelain, Bloom Clay [new], Kiln Reed [new][experimental], Puff Clay [new][experimental] |
Curing, firing, cracking, shell-like hot fracture | Hardening first, then chunking, petal breakup, or grain-guided splintering |
| Fiber and wood-like | Soft Wood, Hard Wood, Fiber, Rope, Bamboo, Bone |
Anisotropy and heat-weakened structure | Different failure along versus across grain, with more visible heat-driven weakening than before |
| Gas-forming / porous | Bread Dough [new][experimental], Puff Clay [new][experimental], Intumescent Block [new][experimental] |
Expansion, pore growth, softening | Ranges from dough-like puffing to cured clay swelling to hot porous foam, with pressure now able to tear or crumble the body |
| Burning / reactive | Lamp Oil [new], Firecracker Pellet [new][experimental], Wood Log, Paper, Coal, Embers |
Combustion, smoke injection, buoyant hot debris | Ignition, smoke, hot plume behavior, weakening as burn state progresses, and first-pass pressure bursts for reactive pellets |
[new] marks recent additions. [experimental] marks first-pass behaviors that are intentionally visible before they are fully tuned.
The new oven and pot scenes are especially good for heat/cooling experiments because they create stronger thermal asymmetry than the generic sandbox.
Good for observing heated wall glow, escaping plumes, directional draft, and uneven heating on suspended or shelved materials.
Good for testing heat transmission through rigid structure into a second chamber or pot-like space, especially with fluids or foam-like matter.
Best for separating pure flow effects from thermal reactions.
The new hard-ball tool makes repeatable before/after-heating impact tests much easier.
Still useful for simple progression tests: heat while moving, burn while rising, or cure near a source.
The missing pieces are less about adding many unrelated solvers and more about adding the right hidden state fields.
For convincing hot metal, temperature must generate internal stress, not just a change in stiffness.
Pottery becomes much richer once the surface and interior are allowed to mature differently.
Bread-like behavior needs expanding gas inside something that stretches and then sets.
Current fire is already visually useful, but the big missing jump is pressure and staged reaction.
The long-term route to concrete, laminated shells, better bone, smarter bamboo, glazed pottery, puffed bread, and treated metal is a small set of extra internal fields rather than a giant set of hardcoded presets.
If the main goal is that heating and cooling remain the most interesting axis in the sim, these are the highest-value additions.
This is the missing foundation for convincing hot metal, better ceramic cracking, and more realistic glass failure.
This unlocks glaze, crust, case-hardening, shell-core fracture, and a lot of kiln / baking behavior in one move.
This now has a first-pass prototype in Bread Dough [new][experimental] and Puff Clay [new][experimental], and it is the bridge from intumescent behavior to true bread, puffing, bubbling, and burst-driven combustion effects.
This now has a first-pass prototype in Firecracker Pellet [new][experimental]. Once local pressure can rupture a body, firecrackers, popping resins, seeds, and brittle explosive ceramics all become possible.
This is the long-term route to materials whose behavior depends on how they were heated, cooled, coated, baked, or forged.
The current glass model is already more interesting than before, but it still behaves closer to a hot viscoelastic body than a brittle network with strong residual structural memory.
The best next step is probably not another preset, but a better temperature-history model: hot enough to relax some stress, cool enough to reharden, and different outcomes depending on how quickly the body cools.