[Freeciv-Dev] Re: (PR#6815) Balancing Battle Response Time for Different Client CPU/Network connection

["Raimar Falke" <i-freeciv-lists@xxxxxxxxxxxxx>]

<URL: http://rt.freeciv.org/Ticket/Display.html?id=6815 >

On Tue, Nov 11, 2003 at 03:26:34AM -0800, Christian Knoke wrote:
> 
> <URL: http://rt.freeciv.org/Ticket/Display.html?id=6815 >
> 
> On Mon, Nov 10, 2003 at 02:22:35PM -0800, Paul Zastoupil wrote:
> > 
> > <URL: http://rt.freeciv.org/Ticket/Display.html?id=6815 >
> > 
> > On Mon, Nov 10, 2003 at 01:17:49AM -0800, Guest wrote:
> > > 
> > > <URL: http://rt.freeciv.org/Ticket/Display.html?id=6815 >
> > > 
> > > I play Freeciv 1.14.0 on a moderately old Pentium-233MMX Win 95 
> > > system.  This works great for the economic and diplomacy aspects, but I 
> > > find myself at a distinct disadvantage come wartime.  For 
> > > those 'counter-spy' type actions where reaction time is everything, 
> > > those folks with 2Ghz+ systems may have a factor of 10 on me for any 
> > > CPU related processing delays.  I'd like to suggest that it would be 
> > > possible to, upon connection to the freeciv server, do a Micro 
> > > benchmark to determine client connection speed and/or transaction 
> > > processing speed, and then scale the game mechanics appropriately to 
> > > avoid over favoritism for the gamers with the fastest computers (and 
> > > noting that this would need to rebench whenever they disconnect and 
> > > reconnect).
> > 
> > Since I work for a gaming company I know these sorts of things are full
> > of nightmares.  Even worse, since the players have the freeciv source
> > in their hands, why wouldn't I alter my client to always say I'm on a
> > 486 speed machine?
> 
> Yes. I bear an idea for long, though, that it might be possible to let the
> server put in a delay for quicker connections, so that everybody virtually
> has the same ping time. 

> If you use ICMP packets for that, it should be hard to betray the
> measure.

No. While I may don't know how to delay ICMP packets I know how to
disable them completely.

> What do you think of that?

If would be more fair to rate limit the commands to n
commands/time. But I don't know if this is a good idea.

        Raimar

-- 
 email: rf13@xxxxxxxxxxxxxxxxx
 "Make it idiot-proof and someone will make a better idiot."