R0J0hound's Forum Posts

  • For something simple like variable that only increases or decreases you can utilize the trigger once condition.

    Variable timer = 10

    Every tick

    — subtract 123*dt from timer

    Timer <= 7.5

    Trigger once

    — set text to “three quarters left”

    Timer <= 1/3

    Trigger once

    — set text to “a third left”

    Timer <= 0

    Trigger once

    — set text to “done”

    Something that counts up would be >= instead.

    Another option could be to still change the variable by whole numbers but use a dt multiplied one to control it.

    Variable timer =0

    Variable countdown = 10

    Every tick

    — add 100*dt to timer

    Repeat int(timer) times

    — subtract 1 from timer

    — subtract 1 from countdown

    — if countdown=5

    — — set text to “halfway”

    — if countdown = 2

    — — set text to “a fifth left”

    — if countdown = 0

    — — set text to “done”

    Anyways those are two ideas that come to mind.

  • One way is to just spawn the objects in random positions and if the object collides with the player move it to a random position.

    Here’s one possible example. Create it on the player, and repeatedly move it to a random position till it’s not touching the player.

    On function “add spear”

    — Create spear at (playerbox.x, playerbox.y)

    — while

    — spear: overlaps playerbox

    — — spear: set position to (random(windowWidth), random(windowheight))

  • One way to do it would be with a second array. So if A is the first one B could be the second. And you’d do this:

    On function “journal”

    — B: set size to (0,2,1)

    — A: for each xy

    — — B: push A.at(A.curX, A.curY, 1) to front

    — — B: set at (0,1) to A.at(A.curX, A.curY, 0)

    — B: sort

    Then B would have the sorted list in the 2nd column.

  • Possibly. The only physics part I touch is enabling and disabling it for a frame.

    Actually, the issue probably has to do when the polygons have an odd number of points. Distort meshes only have an even number of point with the way I’m using it, so it just sets two points to the same spot. Which is probably breaking the physics library’s collision response sometimes.

    Edit:

    That doesn’t seem to have fixed it. It seems to work fine without the physics behavior. Probably need to debug the polygons. Likely some numeric errors that are cropping up.

    This is assuming that the physics behavior, how the collision shape it generated, and how the collision polygon is distorted are working without bugs.

    Edit2:

    Found it can crash too. Seems to be in the physics behavior. Anyways, best I can do is add code to make sure the polygons are convex and ordered clockwise. Maybe on another date. If it still fails then the issue is out of my control. Guess there’s always event based physics if I can fit it in and get the inspiration.

    Edit3:

    Can't fix. My code is generating convex polygons. The mesh distort feature also seems to be giving the correct collision shape. The physics behavior seems to bug out sometimes?

  • So c3 has a feature that can let you distort sprites with a a grid mesh. The cool thing about it is it not only visually distorts the sprites but it affects their collision shapes too, which in turn is transferred to the physics behavior.

    Anyways, I finally got a chance to test it and confirm it lets you do that. Next I was able to use that as a tool to slice up physics object. The basic idea is each sprite is paired with an array making up the points of the polygon. From that with some math I was able to divide two polygons into two utilizing a line. The final step was to take that array of polygon points and setup the distort.

    Here's the example of it working. drag a line to cut the sprite.

    dropbox.com/s/r8lhe8sha0wa33m/poly_split.c3p

    One quirk with the physics behavior that I had to work around was when creating objects you need to disable them for a frame before enabling them again or they will jump. It's been there for ages.

    This only scratches the surface, and was mainly just to show how to slice objects up. I could be adapted to do dynamic breaking by running it in a loop where you'd set the line to be random angles from the collision point.

  • So the overall logic would be when a collision occurs see how hard of a collision it was and if it was hard enough break the object up.

    Seeing how hard the collision was should be a matter of looking at the difference of kinetic energy (KE=0.5*mass*velocity^2) of the object before and after the collision. You can do that by saving the KE to a variable at the end of a tick to reference later.

    Breaking the object up is either done with pre-made pieces or by somehow breaking it up dynamically.

    The pre-made pieces way is the more doable approach. There are two ways to do it though.

    1. Destroy the main object and create all the pieces.

    2. Just have all the pieces joined together with joints which are later unjoined.

    There are pros and cons of either. With 1 you’ll need to be able to place the pieces in the right position and angle, as well as set their linear and angular velocity from the original object.

    With 2 setting up the joints between the pieces can be tedious, and you’ll need a way to look at all the pieces as a group with events. Also joints tend to be springy, and you may need multiple joints between pieces to keep them from rotating.

    Now the fanciest approach would be to dynamically break the objects up as needed. For that we need to be able to change the collision polygon and what is drawn. For c2 you’re out of luck there with vanilla features, but in c3 you can in theory do it with the distort mesh feature. However it wasn’t meant for that sort of thing so some creative use will be needed.

    Anyways, using mesh distort as a tool we will want a way to slice a convex polygon into two separate convex polygons. With multiple cuts we should be able to break the objects up. There are many ways we can take the idea from there such as making the breaks originate from the contact points and use voroni cells to change up how it breaks.

    Anyways that’s a general overview. There are probably lots simplifications that can be done to make it more feasible.

  • You don't need javascript, here's a way to do it with events. It uses the tiledbackground object to draw it, and uses some math to do the raycast calculation.

    dropbox.com/s/s8mz4i34djr89xy/raycast.c3p

    Kudos to construct for giving access to collision points. I didn't use the line of sight raycast feature since that doesn't provide the necessary data to do the texturing.

    Also since events are so slow I tried utilizing javascript instead:

    dropbox.com/s/b6hy5e5f3xdldmz/raycast_js.c3p

    It's much faster, even though it does the same thing. But man was that unpleasant to get working. Once i got past the syntax errors, i had to deal with all the silent errors that made it not work. Anyways, it's a shame js is needed to make things faster.

  • Well there’s two different approaches to do that kind of 3D with textured walls with vanilla features.

    One is the raycast method. That’s what that link and template do. Logic wise it’s a raycast per vertical line and depending on the distance each ray is to a wall you set the visual height of that slice. You can use the tiledbackground object to draw slices of a texture to do that.

    Floors and ceiling would need to done via an effect.

    The other method is to use distort meshes with a sprite. Rufus’ example does this. Basically you need a sprite per face and you set the xyz of the corners to make a cube. Then you can use some math to rotate and move the points from a camera point of view.

    Construct makes this a bit more complicated though. You can’t set the zelevation of a mesh point to negative value, and the xy positions are relative to the sprite’s bounding box. Probably can get how to work around that from rufus’ example.

    Otherwise 3D shapes aren’t usable because you can’t rotate them. But there is a third party plugin you could use that does do that.

    Sorry I’m not really good at writing step by step solution stuff.

  • Try Construct 3

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

    Try Now Construct 3 users don't see these ads

  • Glad you’re having fun with it. My interest with mode7 sailed a while ago unfortunately.

    Being slow and not working are different things. Paster was made to fill a gap missing in construct’s feature set and performed well enough for my purposes. I have no other solutions for split screen.

  • Most GitHub libraries aren’t so nice to utilize I’ve found. This one for example looks like it wants you to build the js files. Which isn’t ideal.

    Thankfully they have a demo. You can look at the page source to get the built js files and see how it’s used.

    That’s where I’d start. Usually you can strip a lot of from the examples, as construct provides some of the features.

    The demo didn’t work on my phone so maybe something is amiss with it.

    So in general to use a js library in construct you’d:

    1. Include the js files. C3 scripting I think has a way to do that. In c2 you have to load it manually.

    2. Follow the library’s docs to run some js to use it.

    3. Debugging. Some libraries are just broken, or are obsolete. Or you run into a conflicts with construct.

    I may have a look later as I seem to have recommended this library before. Will see if I find time.

    Edit:

    Tried fiddling with that library a bit but lost interest. It’s a bit lower level and is built around the demos player controls. It should be possible to just have it play a file but I ran out of time trying to adapt their example.

  • The example, the effect and construct in general aren’t made for split screen.

    Typical solutions for split screen would be two games that communicate to each other somehow, or drawing to a paster object. Although I suspect neither would work well.

    The third option would be to start from scratch. But at that point construct just gets in the way.

    I don’t have much time to make examples lately.

    Cheers

  • I think the main issue has to do with the picking. The way it works is things are picked from things already picked. So you pick one cell and then try to pick another and it can’t.

    One way to rearrange the logic is to just pick the neighbor cells one by one and adding to a local variable, and then picking the cell itself and setting its neighbors instance variables then.

    function checkNeighbors(x, y)

    {

    Local number count=0

    If Pick cell at (x+32, y) and cell.frame=1

    Then add 1 to count

    If Pick cell at (x-32, y) and cell.frame=1

    Then add 1 to count

    ... so on for the other directions

    If pick cell at (x,y)

    Then set cell.neighbors to count

    }

  • I’d say just make the request and be descriptive of what it should do.

    Like as a start:

    A way to make paths in the editor as well as during runtime.

    The paths are made up of line and curves connected by movable positions.

    A way to save and load the paths during edittime and runtime would be useful too.

    A way to move along the paths. Could existing behaviors be used with it.

    Something like that at least. I don’t think I covered everything you’re after.

    They can change anything they like since they control the engine.

    Users and third party plugins are were things can be less feasible and require more optimal creativity to work around limits.

  • I’ve tried looking into svg files to get path data and didn’t find it useful. But if you have a program that lets you just draw paths and save it as svg then maybe.

    You can make plugins and behaviors that share information. Shadow caster and light come to mind. But I can’t really think of other official ones. There’s the solid behavior and other motion behaviors but that’s mainly because solid is an engine feature.

    I get the impression that they want most all plugins to be self contained and have all interaction between them to be done with events. Understandable from a maintenance perspective.

    That said, being able to pass more around in events than booleans, numbers and text would be useful.

    Or closer to what construct already does you could request those plugins to provide jsons of the list of points to save and load. At the very least one value at a time would let you do it manually.

    The first part of moving between xy positions with lines and curves is probably the simplest.

    Anyways, I’m not sure my input was super helpful. It’s probably doable but I never make requests so I’m not sure how interested they’d be in in doing it.

  • The idea is you start at the enemy and move by steps toward the player and stop when a wall is hit.

    You can do all the steps with a loop.

    Here’s one such example. The max width of the laser will be 300. 

    Every tick
— set laser position to enemy position 
— set laser angle towards player
—set laser width to 0

Repeat 300 times
— set laser width to laser.width+1
— Laser overlaps wall
— — stop loop

    Alternatively you can use the ray cast feature which will give you the position where a ray hits an object. The just use the distance expression between that point and the player.