SFRailsback
commented
11 years ago Ah- I see that this happens with some input sets and not others, and where it does occur the fish are growing incredibly fast. That's likely the reason so I'll worry about that instead.
Did some measurements; code gets slower as it proceeds & number of fish gets lower. From the start of a simulation, execution time per model step (steady flows, so two steps per day) increased from 1 second up to 8 seconds over the first 80-90 steps (40-45 days). The rate of increase decreased over time. Then it leveled off and stayed at 7-9 seconds through steps 80-240. It should be gradually decreasing over that period as the number of fish decreases. (A few redds were created but those shouldn't slow it down much.)