The dynamic pathfinding layer is a combination of navigation and control logic. It's not like in navmesh. In the video, two layers play with each other. Without a player. This is done as an example, but the developer can add a third, fourth, etc. layer, can include agent reactions to the player, et…

What is your opinion? Let's run after monsters in a shooter.

Dynamic emulation of the pursuit of living beings by the player. Agents try to hide behind suitable shelters. Objects that are not suitable for shelter are considered obstacles. 

Third person view.​

Here is the emulation of living creatures running away from the player. I note that the behavior is quite interesting to customize and this gives many options.

Another stage of development of fast pathfinding has been completed. Dynamic multi-targets are now available. Any normal number of dynamic targets is possible. Any scenarios for the distribution of agents and targets that you wish are also possible.

Continued work on a fast pathfinding solution. Now the movement of agents is more interesting. Agent behavior will continue to improve, I hope. 

I would like to point out some of the main advantages of my solution.

• Speed, performance and low consumption of CPU resources.
• Ability to use complex an…

Now development has been moved to Unity. Further, it is planned to optimize and improve the code. Another of the plans is to implement automatic distribution of congestion on the map.

Some people told me it was so fast because it was a raylib.

For porting 3d space was used as it is a very highly comp…