[Freeciv-Dev] Re: CMA passes back control without reason (PR#1505)

[Raimar Falke <rf13@xxxxxxxxxxxxxxxxx>]

On Mon, Jun 03, 2002 at 06:00:41PM +0200, Reinier Post wrote:
> > > > If you want a server without attributes just remove the attributes
> > > > capability from the server.
> > > 
> > > Ah, I didn't realise you had already implemented this.  Wonderful!
> > > I may do ome playtesting to see if it really makes a difference.
> > 
> > I will probably don't work because the attributes capability is a
> > mandatory on the clients (with good reasons, see earlier
> > threads).
> 
> I don't understand, does this mean only the GTK+ client will work
> from now on?

No. But only the GTK+ client has the CMA and so understand the
attributes. The infrastructure of the attributes is in common and
client and so will work even if the attributes aren't used.

> > Lets take another look:
> >  - if you use a client and don't want attributes, just don't use CMA.
> 
> Then it seems to me the capability is not mandatory?

See old thread.

> >  - if you host a server over a modem, just ask your players to not use
> >  CMA.
> >  - if you talk about civserver: civserver isn't connected over modem
> >  (or is it?!)
> 
> No, users connect to it with a modem.
>  
> > So where is the problem?
> 
> The problem is that I don't know how to turn CMA off and get rid of
> the overhead that way.  Not much of a problem as far as I can see.

The CMA core will only use attribute space if you say "control
city". I'm not sure about the CMA frontend (coded by Per) and its use
of the attributes.

> > > > There are ways to reduce the traffic (both normal and attributes). It
> > > > is just a "feature" that isn't requested often.
> > > 
> > > If it's optional, it's a great feature.
> > 
> > It isn't there is you don't use CMA.
> 
> As long as attributes are only used by CMA that is fine.

This will change. But you don't have to use the other agents.

> It is mostly the CMA, rather than the attributes storing itself, that
> may cause overhead

Correct.

>, but perhaps I'm wrong, I have to test this.

        Raimar

-- 
 email: rf13@xxxxxxxxxxxxxxxxx
 "We've all heard that a million monkeys banging on a million typewriters
  will eventually reproduce the entire works of Shakespeare.
  Now, thanks to the Internet, we know this is not true."