Real watercolor doesn't sit still where you put it — pigment rides on a thin film of water and keeps moving until the paper dries. Each cell of the paper tracks a wetness w and a pigment density C per colour channel. Pigment only diffuses where the paper is wet:
∂C/∂t = D · w(x,y) · ∇²C
∂w/∂t = −k_dry (evaporation; diffusion freezes once w → 0)
Colour comes from simple Beer–Lambert transmission through the pigment film — light passes through and out the absorbing layer once, no scattering term — which is a different, simpler law from the Kubelka–Munk reflectance mixing used in the other painting simulator on this site:
colour_channel = paper_channel · e^(−C_channel)
Two effects unique to wet-media painting come out of the same fields:
- Granulation — the cold-press paper has a fixed micro-texture h(x,y). Pigment that settles as water recedes pools in the texture's valleys, so visible density is modulated as C·(1 + g·granulation·(h−0.5)). Payne's Grey and Burnt Umber granulate strongly (their pigment particles are coarse and heavy); Phthalo Blue and Quin Rose are staining dyes that granulate almost not at all — pick either family and compare the same brushstroke.
- Blooms / backruns — where a drying cell's wetness falls below a wetter neighbour's, capillary flow pulls pigment toward the wetter side, depositing an extra ring of density there. That's the classic "cauliflower" edge watercolorists both fight and exploit.
Water dilution sets how much wetness (and how little concentrated pigment) each dab deposits — high water gives loose, far-spreading washes; low water gives crisp, controlled marks. Wet whole sheet pre-floods the paper so any pigment already down keeps diffusing and blooming, the way a wet-on-wet underpainting behaves.