Hopefully a simple question

[Mike Bonner]

Finally am getting back on my feet(health stuff, long story) and want to try
and learn some new things.

I'm looking at making a very simple overhead view rpg, and my thought was to
use tilesets and set the imagesource of the chars to the desired tile based
on letter. (meaing w for wall, f for floor b for building etc)  The viewport
will be a small 9x9 area with the current 'map' held in a variable.

Before I go too far down this path, does anyone have any
tips/pointers/warnings or better direction for me to go?  My thinking was
track the current position of the character against the map file, then on
move, update the char postion (which will always be centered in the
viewport) and use that to get the chars in the surounding area, lock the
screen, update the field contents with the areas text representation and set
all the image ids, then unlock.

Would a datagrid be a better method for this? With a row template of 9 tile
size images viewing 9 rows, and use fillindata to choose what imagadata to
set each row tile to.. I was thinking this might work pretty well as I will
only have to worry about the charoffset for each data row, and the datagrid
will handle which rows are displayed.

Other methods? At this point i'm not even really worried about collisions or
anything of that nature.  And  I'm still determining if the map should bind
edge to edge (making around the world trips possible)

Any and all hints would be welcome.
Thanks!