AI Unit Autonomy

[xian]

I have been wondering whether some of the problems with the AI in scenarios is due to the design decision NOT to give AI units any form of autonomy. Every major action needs to be scripted.

This also means that AI units can't decide how to setup at the start of a mission, thereby limiting re playability.

I'm certainly finding the AI far less of a challenge than in CM1.

[nik mond]

Well not exactly scripted. You are limited to positioning, movement location, pathing delays, and posturing. There is more to it. But essentially the rest is up to the Tac AI which is built in. The Tac AI I must say is really smart. Impressive. And there is the Autonomy. But it still requires the user AI input to get it there.

Compared to CMSF, if there are no user AI plots at all, the TaC AI will actually start to take over earlier, and do a basic advance. Thats what I perceived in playing with the editor.

I kinda wish we had some scripts to use in conjuction with the AI we have now. But it works as it is very well.

[George MC]

This also means that AI units can't decide how to setup at the start of a mission, thereby limiting re playability.

I'm certainly finding the AI far less of a challenge than in CM1.

Not correct bud The scenario designer can give the AI differant set-up options. Jts depends if the scenario designer has done that. If you wnat to see how that is done you can check out Busting The Bocage and play it in scenario author mode - you should see the AI set-up in differant ways. If I recall Busting The Bocage has three differant AI Plans for both sides. Or check out Huzzar - that has three differant US AI set-ups and attack plans. Could be the scenarios you have played the desinger for their reasons has not varied the set-up only how the AI reacts?

[xian]

Thanks for clarification. Really glad to know that I was wrong.