Complete.Org: Mailing Lists: Archives: freeciv-dev: August 2001:
[Freeciv-Dev] Re: [Patch] Introduction of turns
Home

[Freeciv-Dev] Re: [Patch] Introduction of turns

[Top] [All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index] [Thread Index]
To: "Ross W. Wetmore" <rwetmore@xxxxxxxxxxxx>
Cc: Trent Piepho <xyzzy@xxxxxxxxxxxxx>, freeciv development list <freeciv-dev@xxxxxxxxxxx>
Subject: [Freeciv-Dev] Re: [Patch] Introduction of turns
From: Raimar Falke <hawk@xxxxxxxxxxxxxxxxxxxxxxx>
Date: Tue, 28 Aug 2001 08:45:06 +0200
Reply-to: rf13@xxxxxxxxxxxxxxxxxxxxxx

On Mon, Aug 27, 2001 at 07:25:48PM -0400, Ross W. Wetmore wrote:
> At 03:07 PM 01/08/27 -0700, Trent Piepho wrote:
> >On Mon, 27 Aug 2001, Ross W. Wetmore wrote:
> >> Unless the computation is incredibly arcane, or there are update issues
> >
> >That's pretty much it.  In order to compute the turn from the year,
> >you need to know if and WHEN certain advances took place.  If I say
> >it's year 1950, you can't tell me the absolute turn number with that
> >information.  
> 
> Yes if the conversion depends on various parameters, the client needs those
> to run the algorithm.

As Trent pointed out you don't need parameters but the history. A list
of (year,game.global_advances) pairs. Trust me it is easier to just do
a turn++ on the server side an sent it over the network.

> >It doesn't seem unreasonable to me to add the turn number to packets that
> >contain the year.  The new_year and game_info packets are the only ones that
> >need it.  The year and turn number could even be changed to 16-bit, so the
> >bandwidth stays the same.
> >
> >> But maybe I am missing something that makes the network solution
> "easier" or
> >> more efficient?  Or how this helps to control network traffic and load?
> >
> >It's not like the client needs to query the server for the turn number.  You
> >just stick that data in a packet that contains the year and only gets sent
> >once per turn.
> 
> Updating existing packets to send dual info is fine, especially if this
> is/coincides with the current year update. I got the impression from Raimar's 
> wording he was using two (extra) new packets. 

I'm sorry when you get this impression. You should look at the
patch. It add fields to packets no new packets.

        Raimar

-- 
 email: rf13@xxxxxxxxxxxxxxxxx
  +#if defined(__alpha__) && defined(CONFIG_PCI)
  +       /*
  +        * The meaning of life, the universe, and everything. Plus
  +        * this makes the year come out right.
  +        */
  +       year -= 42;
  +#endif
    -- Patch for 1.3.2 (kernel/time.c) from Marcus Meissner


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