Dev builds farming sim breeding system using continuous traits and biased mutation

A developer has detailed the breeding mechanics built for a small farming simulation game, replacing traditional Mendelian genetics with a continuous, five-dimensional gene system. Each plant strain carries five numerical traits — hue, yield, speed, value, and hardiness — instead of discrete dominant and recessive alleles, giving players a far wider range of possible outcomes. Crossbreeding works by averaging parent genes and applying asymmetric random mutation, where the jitter window leans positive so selective breeding reliably improves traits over generations. A secondary rare large-jump mutation acts as an occasional lottery, introducing unexpected variation without undermining the sense of player progress. The entire genome is encoded into a shareable string, requiring no backend server to store or transfer strain data between players.