.com.unity Forums - AI unable to build big army (>25000 pts)

.com.unity Forums (http://forum.shrapnelgames.com/index.php)

- Campaigns, Scenarios & Maps (http://forum.shrapnelgames.com/forumdisplay.php?f=140)

- - AI unable to build big army (>25000 pts) (http://forum.shrapnelgames.com/showthread.php?t=41472)

Re: AI unable to build big army (>25000 pts)

Quote:

Originally Posted by francoisD (Post 658755)

i have just found out another soution to the problem of big battles:

the ai purchase loop should be redone until the total number of units of the ai is below 500. indeed 500 is a proof that the ai has not enough points.

this is a very very very small modification of the code, however, it raises strongly the duration of the purchase step, because you rely solely on probabilities...

and it might never stop...

Of course - we have all missed the obvious other variables.

1) In preferences, turn training to OFF.
2) Set the OPFOR up to a higher than default setting like say 120
3) leave yours at the default for the year (70 or whatever for Nazis)

Buying a 44 Soviet army for the assault, that moved the force value from the 32K (65) to the 48K region (120) immediately.

It will make your enemy more experienced (elites in this case) - and thus balance your core's experience gain, too.

4) set AIAdjustpercent to say 150, even 200. That wont help the assaulting Russians in that case (they are banging off the unit limit already), but may let it have more to defend with, or in the meeter.

5) before you set up your campaign - set difficulty to hardest, and you will have had less repair points (20%) overall to build such an expensive core with, in the first place.

Cheers
Andy

Re: AI unable to build big army (>25000 pts)

Quote:

Originally Posted by Mobhack (Post 658859)

Stugs in the German army can appear as tanks, assault guns, and even as tank destroyers at various times.

The German AI pick is a very complex piece of code, and date, battle type, points remaining, snow, city, batloc, who the actual opponent is and several other factors in the 2-3K lines of code determine what gets bought.

That fills in a lot of blanks on what is considered during the selection.

Quote:

So simply running a single purchase trial is only a guide. Try 50-100 or so of each battle type at a specific batloc and points value, then analyse the trends.

Andy, I'll take your word for it. I really don't want to got through all that trouble just so I can say, "Yep, you were right" :D

Re: AI unable to build big army (>25000 pts)

andy, you are genious!

indeed with elite troops, russians have with heavy off armies worth more than 31k pts in 1944 (min was 29000)

so now i can have a german core of 12k pts!

i will have to face elite troops in training instead of elite troops in weapons... wonder what this will give...

well, anyway, many thanks!

Re: AI unable to build big army (>25000 pts)

Well this game never ceases to amaze me. Much respect on picklists realised bought diffrently just not to what extent.

Re: AI unable to build big army (>25000 pts)

yes indeed

but i hope andy and don will consider my proposition to inform the player of the pts the ai has versus the pts it should have.
because problems may appear even for smaller battles (13000 pts), at the beginning of the war. because weapons tanks and so on, cost less (except the infantry)

again thanks for the wonderful solution.