FFX Encounter Time Script

This script is designed to estimate the amount of time spent in encounters for a speedrun of Final Fantasy X.

Output

The script outputs two CSV files which contain estimates of encounter counts on each screen and estimates of time cost of encounters on each screen.

The outputs contain a record for each seed under each combination of the following options:

• Forcing an encounter at the end of one screen
• Lord Ochu on way into Kilika
• Lord Ochu on way out of Kilika
• Luck Sphere in / out of kilika
• Mi'ihen Agency Simulation Encounter
• Old Road Gate Simulation Encounter
• Extra Sandragora (Intentional Sandragora Skip failue)

Note: The script doesn't currently consider both Lord Ochu in and Lord Ochu out as a valid combination.

This makes for 1728 records for each seed at present.

Functionality

The script currently accounts for the following time costs:

• Time to load into and flee from each encounter
• Time cost from outspeeds based on the probability of outspeed occuring
• Time cost from ambush guaranteed outspeeds
• Time cost / gain from Baaj Sahagins ambush / pre-Empt
• Time cost / gain from Geos pre-empt / ambush
• Time to find a Ghost

The script doesn't currently account for the following:

• Time cost / gain from Forced Piranhas ambush / pre-empt
• Time cost / gain from Tros ambush / pre-empt
• Variable load in times of different enemies. E.g. Some flying enemies seem to cause a longer delay before allowing actions to be taken to Flee.
• Risk of ruin due to ambush game over

Assumptions and Limitations

The script makes the following assumptions:

• Movement on each screen is equivalent to the optimal path, yielding the fewest random encounter (RNG0) rolls
• A random encounter has a base time of 10 seconds to load into the encounter and flee. An extra 10 seconds is added to Maze Larva encounters to account for summoning and attacking with Ixion
• Routing is with Terra skip

The script has the following limitations:

• The script doesn't account for worse movement enabling a reduced number of encounters throughout the run. This is partially mitigated by scenarios for forced encounters at the end of each screen.
• The script only cares about what is theoretically the fastest seed and gives no care for how dangerous the encounters could be. A consequence of this is that a seed could have low encounter count and few outspeeds but multiple Thunder Plains ambushes and thus would be higher risk.