Complete.Org: Mailing Lists: Archives: freeciv: February 2001:
[Freeciv] Re: AI cheating
Home

[Freeciv] Re: AI cheating

[Top] [All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index] [Thread Index]
To: Karl-Ingo Friese <kif@xxxxxxxxxxxxxxxxxxxxxxxxxx>
Cc: freeciv@xxxxxxxxxxx
Subject: [Freeciv] Re: AI cheating
From: andrew@xxxxxxxxxxxxxxxxxxxxxxx (Andrew Pimlott)
Date: Tue, 20 Feb 2001 12:14:55 -0500

On Tue, Feb 20, 2001 at 02:18:49PM +0100, Karl-Ingo Friese wrote:
> the situation you described does not look like a cheat to me.
> For the warrior moved two squares, one at the END of the current
> turn and one at the BEGINNING of the next turn. This is often
> used by human players as well, they set a goto command during
> the last 5 seconds of a turn to make a fast move.

Ok, the problem is, I have never played against human players, and
had no idea this was a "feature".  I assume there is no way to tell
whether a unit has moved this turn?  In that case, one must always
assume that any enemy unit may be able to move twice before you can
react (unless you actually saw him move already this turn).

I suppose this makes sense when playing with humans, but it is
unfair with the AI, because a human has no chance to "race" the AI.
In other words, the AI can always use this tactic, and the the human
never can.

The ideal I think would be to make the make the AI move in "real
time", ie, concurrent with the humans, at a configurable rate (eg,
one unit move per second).  Lacking that, I think it would be more
fair if the AI could only move before or after the human.  I think I
could achieve that by just commenting out the second ai_start_turn()
(which is preceeded by the cryptic comment "How totally stupid."),
but I'm trying to understand the game sequence better.

Andrew



[Prev in Thread] Current Thread [Next in Thread]