by OldProf » Wed Dec 14, 2011 10:54 am
Griphos: AI trains sometimes behave like supernumerary actors (people with little or no theatrical experience placed on stage anyway for crowd scenes, etc.): they don't always perform as instructed or consistently. I tested the final version of this scenario at least four times before uploading it and had no problems with the AI traffic, but that doesn't mean that the same will hold true for other players. When you run the scenario again, please try saving it just after you reach the Fred's Spot marker. If the AI goes haywire again, try to make a note of the number of the train that went astray. Don't bother with the coordinates: no one human can write them out before they vanish, but once I know the number of the errant loco, I can check the scenario again myself.
By the way, I'm firmly convinced that the so-called "dispatcher" in RW actually learns by experience, in the sense that I have run a scenario that I'm still editing and encountered an error, such as a derailment, failure to find path, or collision. Without changing anything, I simply run that scenario again and the problem does not reoccur. Maybe that term Artificial Intelligence is more significant than I think?
Thanks for the comments, folks. Please do let me know if you encounter difficulties with my scenarios. I can't promise to fix everything, but I'll do my best.
Tom Pallen
Tom Pallen (Old Prof)
{Win 10 Home 64-bit; Intel Core i7 6700 @ 3.40GHz; 16.0GB Single-Channel @ 1063 MHz (15-15-15-364); 2047MB NVIDIA GeForce GTX 960}