Quote:
Originally Posted by FASTBOAT TOUGH
If anything, we need to find the AI Military Intelligence Agency (AIMIA)  code AND KILL IT. 
Yeah, AIMIA was on purpose there's a hidden message in there. 
Regards,
Pat
 |
Somedays I really do dispair.
It's one thing coming from a newbie to the game...... quite another coming from a veteran player
There is a DEAD CERTAIN way to test "theories" like this that take maybe less time than to type a long post about it
Set up a game like this
Why that way you might ask.
This allows you to pick for the AI "attacker" and see what the AI would buy if the AI was the "defender".
This allows you to test theoies like "When I buy Helos or aircraft it amazes me how the AI seems to know I got them."
The answer to that question is in the game. You just have to look for it
So..........
you will need to run this through a half dozen times because like everything else the picklists have randoms built in so it's not the same every time
You want to be able to purchase for the AI so you can test this "how the AI seems to know I got them." theory.
So..... buy the same thing the first few times or alternately whatever floats your boat
To keep it manageable set Ukraine ( or whoever ) to 5000 points and set player 2 to XXX which allows it to buy to it's max for that battle type
Set player 2 air sorties to 15
Buy nothing but gunships for the P2 AI. Quit deploy for P2 and then you will be brought to the deploy screen for P1. Auto deploy ( it's just easier but clicking on manual deploy works too)
Once you see the units for P1 on the map go to the HQ memu........ check how many
DEDICATED manpads were bought.... ignore the ones that might be bought with an infantry formation as that is even more randomized and the AI does NOT check what weapons are in an unit or formation before buying it
Do that a few times and count how many Inf sams are bought ( and SPAA / SAMS if you like )
Next do the same thing only this time buy nothing but MBT's for P2
Do that a few times and count how many are bought each time for both types of attack and tell me if you see a significant difference ( you won't ). Keep in mind that the larger the sample the less chance of getting repeatable random events. You want the average of as large a sample as you are willing to take