Skip to content

Dynamics

MindFry’s dynamics module handles time-based processes: decay and propagation.

The Decay Engine reduces lineage energy over time.

E(t) = E₀ × e^(-λt)
where:
E₀ = initial energy
λ = decay coefficient (default: 0.001/tick)
t = time since last access

Decay is lazy — computed only when accessed:

fn compute_decay(&self, lineage: &mut Lineage, now: u64) {
let elapsed = now - lineage.accessed_at;
let decay_factor = (-self.lambda * elapsed as f32).exp();
lineage.energy *= decay_factor;
lineage.accessed_at = now;
}

Benefit: Zero CPU cost for dormant memories.

The Synapse Engine propagates energy through bonds.

δ_target = δ_source × strength × polarity × damping
ParameterDefaultDescription
strength0.0-1.0Bond intensity
polarity+1/0/-1Synergy/Neutral/Antagonism
damping0.550% loss per hop

With 50% damping, propagation is limited:

HopEnergy Remaining
150%
225%
312.5%
4+< 7% (negligible)

This prevents runaway activation cascades.


See Persistence for storage layer details.