(Comp.sys.handhelds)
Item: 1481 by grue at batserver.cs.uq.oz.au
Author: Paul Dale
  Subj: Reversi for the HP48SX (Machine code version)
  Date: Thu Dec 20 1990 07:53 

        A new improved version of my old reversi program!  The main features
of this version of the proram are:

* Features a true machine code move generator (including ALL searching and
        position evalulation).

* And a machine code move checker.

These improvements mean that the program runs much faster than
before and it responds to invalid moves much faster as well (especially
when the invalid move is a pass).

The user interface remains basically unchanged.  The biggest variation in
this department is the renumber of the rows of the board to conform to
normal reversi ordering (row 1 is now at the top!!).

The game is started by pressing the PLAY softkey.  The board will be drawn
and the player asked if they desire to go first (this question should be
answered Y or N).  To enter a move, you should first type the column number
1 (left most) to 8 (right most) followed by the column number 1 (top) to
8 (bottom).  Your move will be made if it is legal.

The other commands available are:

key | action
----+-----------------------------------------------
 P  | Pass move (only allowed if no moves are legal)
 Q  | Quit game
 O  | Turn calculator off
 U  | Take back last move (once only)

The input code will turn the calculator off if no key is pressed for a
minute and pressing on will continue the game as if nothing had happened.

If anybody finds problems with this program or can think of improvements
I'd like to hear about them.  I don't want to hear from people who think
that the display of moves is too slow; I made the making of moves slow on
purpose (So that a player can see what effect a move really had while
playing.  A move that just appears on the display is quite disconcerting
to the player).

This version of the program doesn't play as well as the previous version
(which is a bit of a pity since the older version played a good novice
game).  This is mainly due to the very primative evaluation function
which is used (it simply counts the number of disks each side has and
maximizes the difference).  I'm currently working on both improving the
evaluation function and making the move generator search more than one
play ahead.  Both of these modifications will improve the play (I hope).
The updated move selector should only require the replacement of the
MVGEN routine (neat huh?).


The usual disclaimers apply.  (i.e. the stuff about using undocumented
features of the machine and any damage caused isn't my problem or fault).
I'm also not responsible for this program causing memory losts.  I'll just
say that in my calculator they don't cause them, but yours may be different.

The program is also copyright myself 1990, and it cannot be used for any
commercial purpose without my permission.  Any NON-PROFIT usage is permitted
without my prior approval.


Enjoy...(full listing for HP48 is included at the end of this file).
If enough interest is shown, I'll post the assembly source for the
various routines.  (this source is commented quite well, but the comments
haven't always kept up to date with the code [ it got modified quite a
lot during its evolution] :-(


                                                        Pauli
seeya

Paul Dale               | Internet/CSnet:           
grue@batserver.cs.uq.oz.au
Dept of Computer Science| Bitnet:      
grue%batserver.cs.uq.oz.au@uunet.uu.net
Uni of Qld              | JANET:          
grue%batserver.cs.uq.oz.au@uk.ac.ukc
Australia, 4072         | EAN:                         
grue@batserver.cs.uq.oz
                        | UUCP:          
uunet!munnari!batserver.cs.uq.oz!grue
f4e6g4Qh4++             | JUNET:                    
grue@batserver.cs.uq.oz.au