 GM Stormcrow wrote:
Whilst also being a better solution (I'm not quite sure what lesser/quicker/simpler approximations without terrain cost of movement etc actually bring to the gameplay end of Illy...?), there's also the commercial reality that work performed on the more difficult 'intractable' end of the underlying problem qualify for R&D HMRC Tax Credits for Illy Corp 
SC |
You still misunderstand my post, Stormcrow!
What I say is to start by introducing, as you put it "terrain cost of movement", but WITHOUT the hazzle of optimal paths. Just make it straight-line movement, in lines defined by the players. My suggestion was to restrict it further to only allow 8 directions. Actually, the choice between any combination of directions would not be necessary if you require the player to set a waypoint - then you only need to check if the waypoint is on a gridline from the starting point.
Example 1:
I want to move from (0,0) to (60,34), having researched "axis movement" and one waypoint.
Setting the waypoint anywhere but (0,34) or (60,0) would give me an error message, using any of those two points would give the calculated travel time and a "confirm" button.
Example 2:
I want to move from (0,0) to (60,34), having researched "axis and diagonal movement" and one waypoint.
I would now have between 8 and 12 options depending on whether the diagonals work, but the computational effort would still be limited to checking if the move is allowed and adding the movement costs of the involved squares.
Example 3:
I want to move from (0,0) to (60,34), having researched "axis and diagonal movement" and two waypoints.
Choosing the first waypoint would be checked only by ascertaining if it is on an allowed line going out from the origin, while checking the second point would be as per example 2. Calculation of distance is still only summing terrain costs for traversed hexes.
Example 4:
I want to move from (0,0) to (60,34), having researched "Thundercats wonderful shortest path avoiding all dangers".
I just select the target square, looks at the travel time, checks accept and off the caravan goes, calculations proceeding in a less than 2n!/(n!)^2 complexity that I will not notice except as a smooth and safe movement of my caravan.
Straight lines, rather than restricted grid directions might be simple enough to calculate costs for to not motivate the rather artificial limitation of using only grid directions. But I think player defined waypoints is actually quite as good as automatic shortest distances. (Obviously, that opinion is disputed.)
Tweet
Pathfinding - how hard is it?
Topic Options
