Hi again.
I'm not sure what the "HOT SPOT" method you refer to actualy is.
I'm assuming you want your ai cars / vehicles to race around a track or at least follow a road / path.
I'm not a user of dark basic any more but a rival product so cant really provide any code examples, sorry.
I've used the waypoint method with great success, giving quite a professional game feel to the ai cars. I can adjust the difficulty levels so they are far greater than me, and i've been playing racing games for many years now. My ai cars can easily outperform me.
I would advise trying the waypoint method first to see what sort of results you get.
My ai cars actually head in the general directon of the waypoint, +- 10 - 15 degrees to the left or right of the waypoint. This allows the cars to drift down the track and to not strictly stick to the ideal line which the waypoints mark out.
This allows the cars to duck and dive for position with each other, and i also give each of the ai cars slghtly different handling and acceleration characteristics. This allows the cars to overtake each other, push each other off the road ect.
A slightly more advanced technique is to use splines or curves to define your best line. These again are waypoints / vertices wich define a curve based on much fewer waypoints or 3d points. This will give you a much smoother line for your ai cars to follow but is a bit harder to code. I would advise the waypoint method first and then move on to splines if you require more accuracy.
But in my experience, splines don't really give you that much of a benefit over waypoints. Maybe smoother movement thats all. The coding is trickier as you have to seek down the spline to find the point your heading for and also need to work out the speed at that spline point too.
An even more advanced approach is to use re enforced learning.
This is basically a way to teach your cars how to drive. By repeatedly driving around the track, saving the lap time, then making a small random adjustment to the ai / physics, then do another lap and see whether that lap time was quicker than the last. If it is, you keep that new change and make another random one .
This way, the car teaches itself how to drive.
Think of a rat in a maze. He randomly searches each corrider learning and remembering dead ends as he goes until he finds the exit. Finding the best path.
This is much more advanced vehicle ai and to be honest isn't necesary at all for game racing ai. But i'm sure if you search the net you'll find lots of tutorials on path finding, vehicle ai ect. There are plenty of AI specific websites out there.
If your racing on a track, then you generally want to follow the fastest / best line.
You could also check out geoff crammonds f1gp games, which use a CCLINE approach. There is some info out there on the net about it, and might lead you on to other areas. This again is a best line to follow approach.
Hope that helps.
A choc a day keeps the doctor away