Complete.Org: Mailing Lists: Archives: freeciv-dev: April 2003:
[Freeciv-Dev] Re: Cheating
Home

[Freeciv-Dev] Re: Cheating

[Top] [All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index] [Thread Index]
To: freeciv-dev <freeciv-dev@xxxxxxxxxxx>
Subject: [Freeciv-Dev] Re: Cheating
From: "Per I. Mathisen" <per@xxxxxxxxxxx>
Date: Fri, 25 Apr 2003 13:19:57 +0000 (GMT)

On Fri, 25 Apr 2003, Rafal Bursig wrote:
> Primo: Some people may want have this functionality (see enemy unit
> stack).
> Then this can be configure via server options.

There is no point in allowing such configurability if the client doesn't
support it. Going to the effort of making the necessary GUI changes on all
clients for this if it isn't going to be the default ( == nobody will use
it )... simply won't happen.

> I don't look in entire code here and I don't know if server send
> diplomat/spy/sub units position to client (IMHO this will be wrong if
> client have info about they positions all the time).

Stealth works. However, diplomats and spies don't have it.

> We should have rule that "Your Ally's unit always protect your enemy's
> unit when your ally and your enemy are ally and have units on the same
> tile".

The current rule is that you can attack enemy units outside cities no
matter whose units they are stacked with.

If you change this rule, then you get the opposite problem: What if the
first ( == only shown ) unit is an enemy unit, and there are friendly
units in the stack? The client won't know, and will mistakenly think the
stack can be attacked.

This problem also arises because of layering (or lack thereof). If there
for example is an air unit in the stack which you cannot see, and you try
to attack with a ground unit, you will fail. The client has no way of
realizing why not.

These things may not be a big deal for human players, who presumably can
be given informative messages by the server saying why he can't attack.
However, this becomes a big deal with client AIs. They will need some kind
of report-back system that they can understand, and such a system would
make it much harder to build a client-AI that maintains little state
information.

So I think showing the (non-transported, non-stealthed) contents of a
stack is the way to go.

  - Per



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