# GOL-021: Two Rings

Eight near neighbors and sixteen further neighbors follow independent rules. Let them decide together with separate birth and survival priorities, or alternate one ring per Step.

[Open simulation](https://100gol.com/experiments/021) · [Directory](https://100gol.com/experiments)

## What to explore

Make the difference between two scales acting on one frozen board and two scales
alternating across evolving boards directly observable. Both rings must matter,
with disagreement and the exact next phase visible rather than hidden in code.

## Rules, update order and replay

Binary square lattice, 80/120/160 square; toroidal by default, optionally bounded
with off-field inputs dead. Inner ring is Chebyshev distance exactly one (8
positions). Outer ring is distance exactly two (16 positions). Neither includes
the center; the outer ring excludes all inner positions. Minimum engine dimension
is five so wrapped positions remain distinct. Each ring has independent exact
birth and survival masks. Empty masks intentionally disable the corresponding
path. All cells read frozen current occupancy and commit synchronously.

Simultaneous mode evaluates both rules against the original current cell state
and current ring counts. Agreement stands. On disagreement, birth priority
selects the winning ring at a dead position and survival priority selects the
winner at a living position. Priority is authoritative in both directions: the
winner may permit or deny life. This first iteration offers inner/outer priority,
not extra AND/OR combination operators.

Sequential mode applies only the active ring for one complete Step, then flips
inner to outer or outer to inner. The outer Step reads the board committed by the
preceding inner Step. Reset starts inner. Mode changes preserve the field and
Step number, clear prior evidence, and restart at inner. Other rule edits pause,
preserve phase and occupancy, and clear evidence. Inactive precedence settings
remain editable and preserved. Painting preserves phase/Step and clears evidence.
Topology changes pause and preserve the field and phase, clearing evidence.

Replay state includes board, next sequential ring, Step number, last applied
phase, each prior ring count/proposal, prior occupancy and decision source.
Configuration, geometry and topology accompany the state. No evolution RNG or
scan-order priority. Reset recreates the selected starting field from its seed
and clears phase/evidence. Clear starts an empty field at Step zero, inner next.

## Reading the simulation

Canvas 2D with binary occupancy; living inner decisions cyan, living outer
decisions violet, living agreement lime. Dead decisions stay background. Inspector
outlines inner inputs solid cyan and outer inputs dashed violet, reports both
current counts/proposals, agreement/conflict, birth/survival priority, next phase,
exact next result and previous-Step evidence. Written labels preserve all meaning
in simplified Color mode. Keyboard and pointer inspection pause painting.

## Reproduce a starting configuration

[Download example JSON](https://100gol.com/api/experiments/021/configuration). Import it using Share & reproduce in the simulation controls. It starts paused at Step zero. Files contain rules and initial conditions, not evolved boards or manual edits. See [format, limitations and automated access](https://100gol.com/agents).