A85: LaserSquad-Z

To: assembly-85@lists.ticalc.org
Subject: A85: LaserSquad-Z
From: Johannes <porkkanamies@yahoo.com>
Date: Mon, 17 May 1999 23:54:59 -0700 (PDT)
Delivered-To: assembly-85-outgoing@towerguard.unix.edu.sollentuna.se
Delivered-To: assembly-85@lists.ticalc.org
Reply-To: assembly-85@lists.ticalc.org

Hello everyone, anyone!

Hmmm, I wanted to write something for the list since
it seems to have been in coma for awhile. Don't read
this unless you have some time...heh. Maybe some of
you have heard(well there can't be too many, I guess)
I've been programming the good ol' _LaserSquad_ for
TI-85. It is a turn-based combat-strategy game. Two
players each having up to 8 crew members are fighting
in a certain area. I've made a 35x40map for the first
episode.(map engine handles map sizes 256x256, yeah
too much, but there's no need to restrict that I
suppose...) . The map engine is quite ready now
('map-engine +menu-routines(not standard85!) +shop
+episode1 +most of the data +80-90% of the game
GFXs'=3.1KBs at the moment), and the menu system and
the shop should fully work in next week. Well, some of
you now think:"What is he talking about here?!?!" Hmm,
I've ran into somekind of a problems(again? heh)...
Has anyone of you ever worked with
Pathfinding-algorithms or strategy game AIs? I've got
a pathfinding algorithm I think I'm going to use, but
I'd  like to know if you'd have any suggestions about
how to use the map data with it. The map looks like
this:

.db 1,0,0,0,3,27,27,27,0,...

...so that each byte represents a tile #id. The tiles
1-10  are walkable tiles. I've decided to use 32
different tiles, so each map byte uses 5
bits(0d-31d=0b-11111b). I thought that maybe I could
do the pathfinding with the remaining 3 bits. But now
I'm not sure anymore, because I'd like the enemy unit
to not to walk in the Players Line-of-Sight area, if
it has not enough Action Points to shoot. So I'd need
a shadowmap of the each player units' Line-of-Sight? 

Ok then... Another thing is that I want to represent
here is the Pathfinding algorithm I'm going to use
with the game, and I'd like to hear some comments
about it. Since I'm a mainly a graphician here's a
visualization:

..........
.....t....
..........
..xxxxx...
......x...
..s...x...
..........
..........

s-start
t-target
-------------

..........
.....t....
..........
..xxxxx...
.111..x...
.1s1..x...
.111......
..........

I start making the way from the start...

55555.....
44445t....
33345.....
22fffff...
211123f...
21s123f5..
21112345..
22222345..

...and when I reach the 't'-I just look smallest
number to find my shortest route(starting from t this
time).

....p.....
...p.t....
..p.......
.pxxxxx...
..p...x...
..s...x...
..........
..........

I will make a routine that checks the surrounding 8
tiles, then I use it from the start until it has
checked it's way to the target. To speed up process
I'll always check the distance to the 't' and check
only the nearest tiles(if the path is not found it'll
start checking the next nearest, and so on).

Ok, this is quite basic pathfinding routine, so I know
some of you could help me with this... luckily I have
still 20Ks(28K-3K-Usgard=+20Ks) to fill with
Pathfinding-, Line-of-Sight and AI-routines, hehee!
Oh, and yes, eventhough the game looks complicated I'm
going to release it next August(the two player version
will be out sooner, I guess). Sorry if I bored
anyone...

/Johannes Rajala of Icarus Productions


--
e-Mail:
carrot@sgic.fi

Personal WWW: (not related to calcs... too much) 
http://carrot.findhere.com/

IP homepage:
http://icarus.ticalc.org/
--
_____________________________________________________________
Do You Yahoo!?
Free instant messaging and more at http://messenger.yahoo.com