Complete.Org: Mailing Lists: Archives: freeciv-dev: June 2002:
[Freeciv-Dev] Re: CMA and observer (PR#1557)
Home

[Freeciv-Dev] Re: CMA and observer (PR#1557)

[Top] [All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index] [Thread Index]
To: freeciv-dev@xxxxxxxxxxx
Cc: bugs@xxxxxxxxxxxxxxxxxxx
Subject: [Freeciv-Dev] Re: CMA and observer (PR#1557)
From: Per I Mathisen <per@xxxxxxxxxxx>
Date: Thu, 13 Jun 2002 04:16:14 -0700 (PDT)

On Thu, 13 Jun 2002, Raimar Falke wrote:
> > > If you give the two CMAs different goals it is interresting to see
> > > that the clients will change the city forever. I have no answer what
> > > should happen. So in short it isn't supported.
> >
> > Ideally, there should be one CMA per _player_ instead of per _client_ -
> > so when one client changes the CMA the changes should be mirrored in
> > other clients playing the same player...
>
> I think that multiple controlling clients are a nice technical idea
> but I don't see the value. IMHO splitting the control space is a
> better idea. This can be done at spatial (one client per island) or
> type (one client for cities, one for units) or time (even turns, odd
> turns).
>
> Has anybody any real experience with multiple controlling clients?

It is not just a nice technical idea, it is actually quite fun, when it
works. You have to split the units and areas between you somehow (ie
telling the other guy "I'll take this part"), though when we introduce AI
sidekicks, it all becomes a lot more difficult. I love playing this way
(so-called "team melee") in Starcraft, but it definitely requires having
non-annoying people to play with on your side.

What is needed here, I think, is a way to claim a unit or city for a
certain connection and lock out all other connections for the same player.
These units and cities would be off limits to the other connections until
the controlling connection release them or disconnect.

But this isn't trivial and there are plenty more issues like this with
allowconnect. I don't think we should start fixing these issues now, but
instead we should #ifdef DEBUG this entire option and try fix them after
the release.

Opinions?

Yours,
Per

"It is difficult to catch a black cat in a dark room.
Especially if there is no cat there." - Confucius




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