A* pathfinding project pro and Playmaker

#1

Hello there,

First of all let me tell you something about me : I have absolutely 0 knowledge in coding.
As I learnt alone I may have a strange vocabulary or a hard time understand basic stuff.
Same goes for the english language… lol

So I’ve been posting stuff on the playmaker forum and I came here for a copy paste (maybe here people know the answer eheh)

Here is the link
(http://hutonggames.com/playmakerforum/index.php?topic=15268.0)

So I’ve fiddled for hours with A* pathfinding project pro and Playmaker here are the results :

I can make the local avoidance to work, the pathfinding is working very well too.

However to make the game I want to I lack of 2 Playmaker’s actions

1- The Scan, to make me able to SCAN at runtime (this way if I tell my units to HOLD a position, by changing their layer/tags to obstacle, I have to scan the scene for the Astar path stuff to check the new tiles that the agents cannot go thro, basically I could make a wall of units)

2- I’ll try to explain, when an agent is doing nothing, you suddently order him to go to A, it’s instant, the agent will instantly go to A but on his way to A if you order him to go to B the agent will still go to A while losing speed and then after a little while it will finally go to B. (this is not related to the “repath” that I set at 0.1 for the try),
when I right click I see the new path instantly on my screen, the problem is really something related to the agent behavior.
As I need to be able to micro like in starcraft (Instantly go to A then B then …You got the idea, intense micro management) I really need this.

Maybe we can do something about it with the set/get property ? I’ve checked so many things but I could have missed it (I hope to).

If there is no solution I think I’ll be ready to pay to have few playmaker actions but the best would be to sell a playmaker’s actions pack in the assets store, many people are using playmaker now, and i’ve seen more than one people asking for it.

If someone have a solution it would be wonderful

Thanks for reading,
Have a nice Day!

#2

Hi

You can scan the graph using a single line of code

AstarPath.active.Scan()

You can also update the graph using the GraphUpdateScene component, but for your use case I don’t think it will be easy to do it without any code at all.

  1. That looks like the agent has an acceleration that is too low.
    Currently the acceleration is not exposed in the inspector, but based on the max speed of the character. I am planning on exposing it though.
    However you can easily change it by opening the AIPath.cs script, find the line that looks like

    float acceleration = maxSpeed / 0.4f;

and change it to say

float acceleration = maxSpeed / 0.05f;

That means that previously the agent would reach its top speed in 0.4 seconds, now it will do the same in 0.05 seconds.

#3

Thanks for the help Aron. NewTo, I will turn these into playmaker actions for you.

Aron, I am working on a playmaker action set for Astar version 4. Still in the early stages. (I made the playmaker actions for VRTK, textmesh pro, unitys new post processing package, etc). Its a big project for Astar, but hopefully I can handle it. I might have to annoy you from time to time on the forum with newb Astar questions however :smile:

#4

I misread that, seems that the acceleration variable is not public.

#5

Hello there,

The playmaker action’s made by Tcmeric that we are currently trying make me able to scan at runtime and toy with “float acceleration” value and it’s exactly what I needed for some litle and quick agents (fast micro-management).

Thank you Aron for your answer.