Re: [OFFICIAL] Some FT background stuff (guidelines for writers) - LONG POST!

On Wed, 11 Feb 1998 08:55:44 -0800, Tom Granvold writes:
>John Leary <realjtl@sj.bigger.net>
>>	Drone fighters can be trusted to always make the same choice
>> given the (basically) same conditions.   As a result the drones
>> win round one big time, and become progressively less effective
>> as the countermeasures come into play.   The human fighters 
>> would soon acquire a special fighter that carried ECM or 
>> Weasel sets to force the 'incorrect' choice in the drones 
>> program.   
>
>    The drone fighters could easily be programmed to avoid this.
>The simple way would to have a number of choices for any given
>situation and randomly pick between them.  A different method would
>be to give the drone a history so that if it used the same tactic
>a couple of time, it would do something different this time.  I
>think that this would not be a problem.
>

One thing I haven't seen mentioned here that seems obvious to me would
be to have the fighter be tele-operated for the carrier.  The fighter
could have limited on board smarts, like for evasive action and
targeting and such, but could get arger movements and tactics and such
from a human on the carrier.

This wouldn't work in all genre's though.  Somewhere where fighters
operate multiple light seconds away from their carrier, the time lag
for the communications would be too large.  But in something like a B5
based world, where ships fight in the 10's of kilometer range, the
delay for a signel would be almost zero.

This would allow you to use less expensive hardware, and not tend to
loose the expensive pilots.

-=- Matthew L. Seidl		email: seidl@cs.colorado.edu		
   =-=
=-= Graduate Student			Project . . . What Project?	
   -=-
-=- http://www.cs.colorado.edu/~seidl/Home.html 	 -Morrow Quotes 
   =-=
=-= http://www.cs.colorado.edu/~seidl/lawsuit				
   -=-