[Freeciv-Dev] Re: (PR#4387) Suggestion for deterministic moves

["Christian Knoke" <chrisk@xxxxxxxxx>]

On Sun, Aug 31, 2003 at 11:01:27AM -0700, Gregory Berkolaiko wrote:

> I would like to draw the Design Boarders' attention to the issue of random 
> moves.

As you may have guessed, I like random moves. Being non-deterministic is an
advantage IMHO. After all, it's a game, not chess.

The kind of randomness that we have here is a good one: it has a rather
small effect, and the limits of the effect are known. It gives the game some
flavour.  In contrast, I don't like the randomness of the reputation and
consequences: you can easily loose your game if the senate decides to not
follow you any longer. This is rather like roulette.

> PRO:
> 
> 1. Reduction in the complexity of the current code -> less bugs and 
> crashes.
> 2. Non-ambiguous definition of the shortest route.
> 3. Ability to display correct #turns a path will take.
> 4. Working "dangerous routes" for triremes.
> 5. Better AI: 
> * If AI moves a "stack" of units, say a charge and a bodyguard, it won't
> end up with one of them being allowed to move while the other won't, thus
> splitting the "stack".
> * It becomes quite hard for an AI to plan coordinated attacks when the
> ETA of units is not known. A human player can easily improvise, an AI
> player cannot. For example, if the offensive units get there faster than
> the defensive units, you have a problem. This can happen with all of
> best-case, average-case and worst-case pf methods.

I see that deterministic moves will make the life easier for developers.
OTOH, path finding works quite well now, and AI is doing better and better.
A bad time for such a deep change maybe. The benefits you mention do not
convince me that it is necessary.

I suppose, the design board will be perfectly able to present an alternative
to current random move model.

But as said, I'm not convinced that it is necessary. I'd rather think that
in a real game (AI's and humans) there will be more things that are not
deterministic (from AI's POV), for example this damn old phalanx on a
mountain blocking your way, nullifying your ETA.

> AGAINST:
> 
> 1. Not civ2 compatible  <-  this smells of fundamentalist and anyway civ3 
> have ditched the random moves.

At least this has been one of the design goals so far.

> 2. Upsets game balance  <-  not by much, especially if we use mixed model 
> with threshold 1/2.  And there are other way to restore game balance.

Christian

-- 
Christian Knoke          * * *          http://www.cknoke.de
* Please note new mail address - Neue Mailadresse beachten *
* * * * * * * * *  Ceterum censeo Microsoft esse dividendum.