Thanks for playing and the thoughtful feedback! Honestly I got stuck on that problem, I figured if you could replay old races people would just go farm parts on race 1 while getting super OP then storm through the rest, but on the other hand watching a whole race and then rolling the exact same new car sucks! I like your idea that scrapping should do something, maybe like every 2-3 scrapped cars gives you a part or something. My other thought was maybe the motivation for sticking with later races was you get more premium parts with better mutation chances, so you can choose to farm more worse parts of fewer, higher odds parts. It’s fun to try to figure out how to solve it!
Viewing post in Turbo Mutant Speedy Beasties jam comments
I had those same thoughts! It's difficult to decide the best way to do it. If it's too easy to farm parts it's not as satisfying. But maybe just increasing the likelihood of a mutation (maybe guarantee at least 1 mutation?) would make things go a lot faster, and would still require replaying races, which I think is ideal. Either way, nice work!
Question: how did you implement the car behavior during races? Did you use a navmesh? or use Path3Ds? Sometimes the cars realistic performed overtakes, which was really interesting, so I was curious what you used to make them follow the track
Getting the behavior down was quite a process! As far as getting the cars around the track I set up a waypoint system with area3ds, and the cars stored the waypoint number they were on. So when they hit a waypoint the waypoint system fed them the next position to go to.
For navigating to the waypoint the car checks the angle to the waypoint and sets the steering angle accordingly. The cars got stuck on walls a lot so I had to just manually make sure the waypoints were far from the walls and the line between two waypoints never crossed a wall.
Then for overtaking I have 3 ray casts, one forward and then 45 degrees out to left and right. The car checks its desired direction by eliminating any rays that intersect walls or cars, then averages the remaining rays together (eg right is blocked, left and forward average to slight left). Then I averaged that direction with the direction to the waypoint to come up with where the car should steer to.
Basically a Frankenstein of every method I found online, but I was so pleased when it all came together!