Great to hear - thanks for the review!

Hello, The Navmesh sample project is in 3D but you can follow the same steps for 3D. In addition, both the A* Pathfinding Project and PolyNav support 2D. This is more related to the underlying pathfinding implementation rather than the Movement Pack - the behavior tree itself doesn't care if the perspective is in 2D or 3D. Feel free to reach out at opsive.com/forum if you still have questions. Justin

Hello, With your Movement Pack invoice number you can download the A* or PolyNav integration from the downloads page. Please get in contact at support@opsive.com if you are having trouble downloading the integration. In some sense it doesn't matter how many lines of code is written and what matters more is the design/functionality. I believe that you got the 150 number from an abstract base class. This class is used as the base for most of the NavMesh Movement Pack tasks but each task contains its own logic (and thus more lines of code) for its respective functionality. Justin

Thank you - I'm glad that you're enjoying the Movement Pack!

Thanks for the review! The Movement Pack tasks are self contained in that you can just add them to your tree and are good to go assuming you have the underlying pathfinding implementation setup. There is a sample scene included with each pathfinding implementation that provides a good example for how that particular task is used. There is a sample scene for the NavMesh and A* integration that you can compare against. The A* and PolyNav integrations are separate so they can be updated independently of the Movement Pack packages. Of course, if you have any questions definitely reach out and I'd be happy to help! Justin

> It's not integrated with the A*pathing like it advertises. It is - you can find the integration on this page: https://opsive.com/downloads/?pid=1092 Updated for edit: > Both the wander and flee task constantly pick locations outside of the pathfinding graph. Have you tried the most recent integration? I spoke to the A* developer and recommended a method which prevents this from happening. > You replied to me on the forum asking if I knew of a better way to sample a random position. I vaguely remember this case and I asked if you knew of a generic way to determine if the location was valid in A*. It was easy to get for one graph type but not a generic graph. I do not know intricacies of the A* Pathfinding API so I thought that you might while I wait for an answer from the A* developer. In any case that should be fixed now with the latest integration. If it's not please let me know via the forum and I'll continue to work with the A* developer. Justin