I am writing this post because I am quite pissed off with the support I receive, I bought this package because you told me that it would be possible to apply the motion system as dota2, lol, etc do and this was 2 years ago and I have not yet been able to implement this package in My project.
I send him private messages in order to solve my doubts since they are related to his tool but it seems that he does not feel like answering the messages
Hi
You sent me your PM late on Friday (2 days ago) in my timezone. Most people do not expect support over weekends.
I would also like to clarify that I did not tell you that it would be possible to use the same motion system as dota2, I said, and I quote
I have done experiments in a beta branch with movement that is a lot more similar to Dota however (it is based on the movement in the SC2 engine).
Note that for a game like Dota, you will most likely have to write your own custom movement scripts at some point as there are so many special movement behaviors in that kind of game that using the built-in ones will likely not be enough.
That said, I am sorry that I haven’t managed to provide you with a good solution for your game. If you are not happy with the package I am of course willing to provide you with a refund.
Hello again, I see that you have answered my private message.
I would like to know where I can get this beta that you mention?
thanks.
Hi
That quote was from long ago, I think you have already tested the beta, haven’t you?
It’s the one that you can download here: https://www.arongranberg.com/astar/download
I think yes, I remember that this system moved the stationed agents and that system is not worth me, I would love to be able to explain in detail what the movement system I need consists of, this system I have managed to implement with the unity navigation system The problem is that the routes are very expensive and I would love to be able to adapt that same system with your project, which I am sure is much faster.
I have done some investigation into what Dota 2 actually does (it’s kinda hard to find info on it, they haven’t published anything). It seems that they use a 2 tier approach. One rough pathfinding calculation (virtually identical to a regular grid graph in this package that ignores all dynamic obstacles) and secondly a short range geometric pathfinder that does not operate on a graph, but it does take dynamic obstacles into account. This short range pathfinder is not something that is implemented in this package. The closest thing is the local avoidance in the beta which avoids locked rvo agents sort of like that, but Dota’s version is more advanced. I don’t think anything like it will show up in this package in the near future I’m afraid. Here are some more details if you are interested: https://www.reddit.com/r/DotA2/comments/43ii8k/massive_pathfinding_problem_or_just_me/
1 Like
Hello! Thank you very much for the information, it is very interesting to know that they use 2 systems to calculate the routes.
I am trying to implement aipath with RVOCOntroller and park those agents that are stopped I think it can give me a good solution