R0J0hound's Forum Posts

  • You’ll need a bad bottleneck to make doing that worthwhile.

  • You could do it like this with some variables. 

    Radius=random(128,256)
Ang=random(360)
X=posX+radius*cos(ang)
Y=posY+radius*sin(ang)

    Alternately you can avoid some of the math with the move at angle action. 

    Create sprite at (x,y)
Sprite: move random(128,256) pixels at angle random(360)

  • Try again. Dropbox changed the format of its links which seem to break on this site.

  • Try Construct 3

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

    Try Now Construct 3 users don't see these ads

  • I'd say file a bug report or just allow for some tolerance in your level design for the jump height. It's supposed to be framerate independent, and was in the times I've seen it tested before. Maybe it's something like the jump sustain that can't really be 100% framerate independent?

    If it's really a nuisance you can try cramming dt in the y velocity somewhere and see if you can tune it to be more consistent. Two possible ideas are below. Just try different values of k starting at k=1 and see tuning that up or down does anything useful.

    -500*k*dt or -500+k*dt

  • That should already be frame rate independent.

  • You can always do particles with sprites instead. While it could be slower you get a lot more control so you potentially can get a better result.

    Anyways the key takeaway is you'd add the player's velocity to the particle's velocity. or you could just ensure the particle speed is faster than the player's speed so they can't step into the fire. 

    https://www.dropbox.com/scl/fi/skk3qe6gdta5bvcjql0pb/custom_flameThrower.capx?rlkey=7ruzs0o9fo5vw0mee7aubp7jf&dl=1

  • It mostly just provides function not available otherwise. To do it with sprites you’d have to slice up an image beforehand. You could even use it to do your own spritefont.

    It’s basically a lower level tool.

  • Pretty much any c3 example of raycasting won’t work with this plug-in. Raycasting isn’t supported.

    Two options would be to either:

    * modify the plugin to access the vertices and triangles of the objects. With that you’d need to then transform the triangles into world space and then do some math for each tri vs a ray.

    * or just skip the plug-in, make an obj loader, transform triangles into world space…etc. basically the same steps.

    This plug-in really doesn’t provide anything that would make ray casting easy to do. I’d recommend doing other things instead of raycasting.

  • Gotcha. Looks like that works by having the mesh points as physics objects and connect them with distance joints. I found a way to make a 2d grid into a circle, but I haven't investigated a way to use that with that example.

    After a few failed attempts to move the 2d grid points into a circle shape I ended up making this tool to do the transform gradually so that points are evenly distributed as possible. I think this could be modified to do different shapes too. 

    https://www.dropbox.com/scl/fi/ju8w4uunxngf9umsbwizs/grid2circle3.capx?rlkey=wogbt7c8abo59cp4tob8uj116&dl=1

    Anyways here I made the soft body simulation directly without the physics behavior. It loads the mesh point positions from what that other tool generated. 

    https://www.dropbox.com/scl/fi/7ywcwxb88mdbbuerw6zxx/circle_softbody.c3p?rlkey=fejgxqht5q0wk1pu188030gjm&dl=1

    After the points are moved into a circle shape the uvs are calculated easily: 

    u = (p.X-ball.BBoxLeft)/ball.Width
v = (p.y-ball.BBoxTop)/ball.Height

    Admittedly it's an incomplete solution. I didn't get around to adding volume preservation or collisions with other objects. Here's an older example that does those things though.

    dropbox.com/s/ray56gskqwpiwfy/jello_c3.c3p

  • The tutorial randomizes the tiles by literally setting them to random values. To make the puzzle always solvable you’d want to start with the solved state and have it do say 100 random moves, but I guess that’s a different problem to solve.

  • Realistically I don’t think they have plans on making it free.

  • Dropbox links seem to be breaking when posted on this forum now.

  • The distort mesh always looks like that. It distorts the sprite image which is always a rectangle.

    What kind of mesh were you hoping for?

    You can move the points around to make any shape you like, and you can make it so the image isn’t distorted if you adjust the uvs too, but I don’t know if you can do that in the editor.

  • Hmm this forum broke the link. Copying it from the code block seems to work.

  • There's a bunch of ways you can tweak it. The motion is done with verlet integration currently which just infers the velocity after moving objects apart. We could change that to deal with the velocity change with each collision instead. That probably would fix the explosions where crowded objects end up moving rather far and get a lot unintended velocity because of it.

    Having bigger objects, slower speeds, and simulating in sub steps (kind of like the physics behavior iterations) are also few ways to make it more stable. Even the physics behavior itself has a threshold where it breaks under pressure.

    Utilizing the velocity to do a continuous collision detection is an idea, but it would be more complex and slower to do.

    Also there's no reason this can't handle different sized objects. It just needs some reworking.