Collection of TMInterface AngelScript scripts.
A collection of ‘modes’ that work incrementally from a start time to an end time, instead of the bruteforce way where it indefinitely picks random times to modify.
This is the SD mode.
There used to be a classic sub-mode for this mode, but it was removed (in case you were wondering).
The normal sub-mode will automatically determine direction, and you just need to set a certain lookahead time. The default lookahead time is 120ms, which should always work given a decent enough setup.
This is the wall-hugging mode.
The classic sub-mode is what the python version more or less did.
The normal sub-mode is an experiment, that tries to automatically determine how far to look ahead based on how long it takes to reach the wall.
This is an input simplifier/smoother. Note: you most likely want to set the evaluation timeframe to 0.0-0.0 (Eval Min = 0 and Eval Max = 0), to simplify all inputs.
Note: this section may be a bit outdated…
For each time in this range, the plugin will run the script as if the simulation just started, with the exact time parameters slightly changed.
This is the timerange where the script is allowed to modify inputs and check the state of the game. This is also where the script can eventually decide to advance to the next iteration by selecting a steering value, that will be printed to the external console.
Let’s say this is a timerange representing a certain simulation from the start to the end of a replay (time goes left to right).
—————-|—-|——–|-
The first vertical bar would be the start of the starting timerange, the second would be the end of the starting timerange, the third would be the end of the evaluation timerange.
The plugin would then go to the starting timerange and pick a time that falls within it. For technical reasons the plugin will always go from the last time to the first.
For each time in the starting timerange, it will then run the script as if we were doing a simulation without a starting timerange. The time that was picked will be the start of the evaluation timerange, and the end of the evaluation timerange will be the “Maximum evaluation time” setting.
This process is repeated until either we run out of starting timerange times to test, or if the simulation is cancelled by pressing Escape with the external console in focus.
execute_commands.SD Entry to a separate plugin.Note: this is me experimenting with AngelScript TMInterface plugins before their public release.