Complete.Org: Mailing Lists: Archives: freeciv-dev: November 2001:
[Freeciv-Dev] Re: Reproducable core dump (PR#1051)
Home

[Freeciv-Dev] Re: Reproducable core dump (PR#1051)

[Top] [All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index] [Thread Index]
To: jdorje@xxxxxxxxxxxxxxxxxxxxx
Cc: freeciv-dev@xxxxxxxxxxx
Subject: [Freeciv-Dev] Re: Reproducable core dump (PR#1051)
From: Mike Kaufman <mkaufman@xxxxxxxxxxxxxx>
Date: Fri, 30 Nov 2001 14:37:40 -0600

On Fri, Nov 30, 2001 at 12:21:45PM -0800, jdorje@xxxxxxxxxxxxxxxxxxxxx wrote:
> Here's a new patch.
> 
> Changes:
> 
> - Constants AIR_ASSUMES_UNKNOWN_SAFE and AIR_ASSUMES_FOGGED safe are 
> used.  These are defined at the top of gotohand.c.  They are used by 
> both AI and human players - but note that they're not used by whatever 
> function is used to eventually find the path, only by this function 
> which tests if a path exists.  So they are incomplete.
> 
> - An ai_handicap check is done for both unknown and fogged tiles.  I 
> created a new handicap type H_FOG.  I'm not quite sure if this is the 
> right thing to do, but using H_MAP for the fogged case didn't seem quite 
> right.
> 
> - I made a few tiny changes elsewhere, mostly cleaning up comments IIRC.
> 
> - There are no parameters passed in - all the different cases are 
> handled as above.  Of course, it'd be pretty easy to change 
> AIR_ASSUMES_UNKNOWN_SAFE and AIR_ASSUMES_FOGGED_SAFE laster to use 
> player-specific values.
> 
> I'm not sure if that's all the issues, but it's all I remember.
> 
> I haven't done any testing with the new version - although I did with an 
> older version.  Some testing would be nice, but since the current code 
> doesn't work anyway testing it in the field isn't all that bad.

I realize that it could be nasty to implement, but at least for fogged
tiles, tiles that I just saw a turn or two ago feel safer to make a
beeline through (less chance that an enemy moved there in that time)
Also fogged tiles out in the middle of the ocean are safer than over
land; desert, arctic, mountains, and  tundra, tiles might be safer to 
transverse than grassland or plains would.

For unknown tiles, it's obviously harder to predict, but if I feel that
a blotch of unknown is over ocean, I don't worry too much about it, but
if I've traced out a coastline (with possibly an enemy city or two in sight), 
I automatically assume that the unknown interior is crawling with
enemies, so flying an airplane blindy wouldn't be such a good idea.

-mike

> 
> jason


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