99Instances2Go's Forum Posts

  • Geez i forgot to include the link. You solved you probs, buet here is it anyways:

    https://

    drive.google.com/open?id=0B1SSuCVV8v74aVhQUDBSdU9MVEE

  • Find a path with a solid sprite is in fact easy. Just set the origin on the right and 3 pixels outside the collision polygone. With a good cell size and a relative big border, it will avoid other solids. You can add the 'push out of solids' from the 'custum movement' to avoid them getting stuck or (better) use wat R0J0hound wrote. But you can not use the origin to rotate. Dooing this with a gost could work, keeping the face sprite aligned facing to the current waypoint in the path.

    Pathmoving will avoid solids because of the cell size and cell border. NOT beacuse of the fact that the obstacles are solid. To include moving solids you will have to recalculate regions arround those objects a lot. There is an action for that. But do not expect wonders. Do not expect close encounters. Do not expect it to move trough small gates.

    Give me a base capx. And i try.

    Other (probaly better and more elegant) solutions go from using bullet to move an object from waypoint in the path to the next waypoint(i can show you that), to coding your own pathfinding (there are examples in the forum).

  • Is this what you want ?

    Your local variable movement got reset when its topevent runs in its next tick. Always beeing zero, every tick. So, it kept finding paths, not ever coming to move on it.

    Local variables are heaven, but you know that they reset (if you dont make them static) every time theire top event runs.

    I added obstacles in event 12. I recalculated them localy for the battlefield. INCLUDING the battlefield.

    I made the pathfinding logic in events 19 to 21.

    I made the cellsize 30 and border 1. See manual why.

    Hope you get the hang on the pathfinding thing.

  • Maybe you have to set the origin in the middle in the first frame, and the apply it to all frames of the animation ?

  • For the 9 points, just set 9(n) imagepoints , and use the expressions Sprite.ImagePointX(n) & Sprite.ImagePointY(n) as the positions to find a path to. Pick n in a choose(...,n)

    https://www.scirra.com/manual/63/image-points

    Set the obstacles on 'custom' in the pathfinding behaviour. And add obstacles, if you need to, in actions. And do use the 'create obstacle map' at least once somewhere, if you have obstacles.

    The object will stop moving when reaching end of path, if it does not, you probaly let it find a path each tick. Or youre pathfinding has a problem.

    If i may say some things that i personal dislike in your project. They are just personal taste. Just gentle tips:

    Plz, name you sprites. Its difficult to look into a capX where everything is named 'sprite(n)'.

    You can as well after the "direction = 1" condition move the sprite. Dont need the second 'action in between agent'.

    When you enter the battle arena, all those blue solids still count as obstacles when finding a path. Needs more time to calculate, can lead to unecpected behaviour. Since now you have solids set as obstacles in the path behaviour.

    All the blue obstacles are so many collission calculations. You can speed things up by 600% when you bring them in a tilemap.

    Not that my opion counts in anyway or any other way on this forum.

    Plz yell if you still have problems.

  • The first thing is see is that the pathfinding object is a solid it self, while having the pathfinder behaviour set to avoid solids. It can not not reach out itself. Either dont make the pathfinding object a solid, or dont make solids an obstacle.

    Also. I am not sure that the collision polygones as set in the battlefield sprite work as expected when they have to repesend a solid to a pathfinding. Would it not be esayer to just use 4 solid solids ?

    Also. I have no idea where 833,456 is supposed to be, but it do not think it is inside the battle area.

    Maybe later i look deeper in it.

  • Wich plugin does the falling ? Physics ? Platform ? Bullet ?

    How is the player controlling the object ?

    Would be nice if you just post a CapX.

  • Make a new layer. Move sprite2 (the arrow) to that new layer. Set parrallax of that new layer to zero. Move the arrow in the layout to the top. Remove the action in the events. Now you have a HUB layer.

  • This is not meant to be functional, it is only meant to show multi touch.

    https://

    drive.google.com/open?id=0B1SSuCVV8v74ek1JWGI4b3RET3c

  • Since you want each trunk to have an personal JapTreeTrunk count, you need to use instance variables.

    https://www.scirra.com/manual/73/instance-variables

    Now you can pick the right tree based on the trees's personal count.

    https://www.scirra.com/manual/124/system-conditions

    (search "pick")

  • Try Construct 3

    Develop games in your browser. Powerful, performant & highly capable.

    Try Now Construct 3 users don't see these ads

  • Working example:

    https://

    drive.google.com/open?id=0B1SSuCVV8v74NV9Odmg0Wm1wMkE

  • Do you mean that, if you run it in the debugging mode, the variable loopindextest is either =zero or =the end of the loop ?

    Well, yes, it happens in 1 tick. You dont see that loop counting, its is done in less then a blink of an eye. So it or zero, when the condition is not met, or the end of the index when you look at it. It is Schrödinger's cat.

    If it does not loop (from zero to end), then you really need to upload the capx. But, i expect it to loop because you basecaly say that the variable does indeed show the end of the index.

  • Now, but, how did you pick the instances ?

  • You mean this ?

    https://

    drive.google.com/open?id=0B1SSuCVV8v74bGcxa0NjTTdtdkE

  • Easy Way:

    https://

    drive.google.com/open?id=0B1SSuCVV8v74cThiVTROSGxKdDQ

    I wonder how you picked the instance though.