Flag. Adjust sliders. Dots stick to previously stuck dots. 1st stuck dots are blue, newly stuck dots are redder. Tips are most exposed to moving dots so grow fast. Max step size is the max distance dots move (min is 1) before dot changes direction randomly each step. Click background to hide variables.
Sticky dot model of directional dendrites. Models tree-like (dendritic) growth when dots' direction is biased in one direction (towards right). Tips have more growth potential, because tips are most exposed. But this exposure also represents higher pressure difference or electric field potential, so models growth in fields. My scratch version uses five colors to mark quintiles: first 20% of dots stuck are blue, last 20% red, with mid colors for in-between. This helps see tips, which have greater likelihood to grow more, and branching (dendritic) shape. Bluer colors show areas deep within fjords, areas physically shielded from further growth. In the model, fjords are shielded because particle trajectories are less likely to reach those inner areas. Blue has lower potential, red higher potential. If modeling electric field, the blue has lower potential since shielded by other particles. If modeling pressure, blue experiences less pressure difference, red experiences more pressure difference so more growth potential. Slight variation of my earlier Diffusion Limited Aggregation. Dots start at random positions, move randomly, stick to stationary particles. Starts with one stationary "seed" dot. Moving dots (white) stop when touching colored (so no need to limit to only 1 moving dot at a time, as in the days when programs had any touching dots stop).
