R0J0hound's Forum Posts

  • Temp =cosp(low, high, month*2/12)

    Should do it if month is 1-12

  • Very cool tweak!

  • Arrays aren't always needed, but they can be useful if you want to keep track of a bunch of numbers. In this case I used it to keep track of the two end positions of a bunch of line segments. Most of the cool stuff is from math and logic.

    Anyways, I found a simpler way without much of the math stuff that works with colored bubbles. It uses the paster third party plugin. Basically it uses the "destination out" blend mode to slice off parts of the circles. You could do it without the paster plugin if you had a layer per bubble, but that would be harder to work with. Paster lets you draw to a image so you can use it for more complicated layer composition stuff.

    ucb6ec2ce00b10ba0e311e6f3347.dl.dropboxusercontent.com/cd/0/get/Ch7-jWW4O6GKgirUnAxRpBCHsQR9zmGHoqUQ_yYS6ZxL8CL1WYmKlHoA8RgTgEBhfwWmSF5f2Urco3HpX77zRFD1ILLwVSpttIgyvhDGqWhGKe98xdzny2kbwGQgutVe3vM/file

    Edit:

    It may be satisfactory as is, but some more work is needed to make clicking select the correct bubble. AKA we'd want to do out own collision detection.

    The more I fiddle with ideas, the more impressive the game in the op's video is. They clearly put a lot of work into it.

    Edit2:

    I tried adding an effect to make a the bubbles more soft, but eh, it doesn't look great.

    I'm more convinced a the bubble would look better with soft body physics instead of rigid. A simple way could be to make a bubble out of multiple smaller rigid bodies connected with springs and some forces to keep them from collapsing on themselves. Then we'd stretch the bubble over that so it looks distorted. The paster plugin's draw quad action could help with that. The bubble intersection logic would have to be reworked again though.

    It may be enough to just make the collision solver more spongy though. I guess that's the advantage of rolling our own physics, we can tweak stuff like that.

    Here are some older topics about soft body physics for possible reference.

    construct.net/en/forum/construct-2/how-do-i-18/bouncy-jelly-effect-43593

    construct.net/en/forum/construct-2/how-do-i-18/how-do-i-jelly-floor-from-the-85899

  • One idea is to make the collision polygon slightly smaller in the image editor.

    The crop button adds a transparent line of pixels around the image. So that would create a gap if the colored pixels went to the edge of the image before.

    The physics engine also adds a small gap between objects called “slop” that increases stability. We don’t have access to tweak that though.

  • Here's a test. Basically physics with a collision radius smaller than the circle and some code to calculate the edges in between for a visual. Physics was done from scratch but the behavior could be used. For that you'd need two objects, one for the physics and one bigger one for the visual.

    uc92941268f0229c018a1317df3e.dl.dropboxusercontent.com/cd/0/get/Ch424sQbkuZ08pm0aeNfPmZBOMh_IUggjXTO_DE3vYjD8Ow1A6NjxlzimbEY9nYucoUDIpBesqHS_nkuYm86hYOv2ztgtjCtdG9_t086n_vJsdd_SR38_0VFf5L9eGoZ740/file

    The edge detection has issues though. It's probably as far as I want to go for now. I guess the next idea to tackle is coming up with a way to color each bubble individually or make the edges between circles curved.

  • It never hurts to go a bit more realistic. Buoyancy force is proportional to the area of the object underwater.

    So a very rough calculation utilizing the sprite's bounding box could be:

    apply force 0, -k*(sprite.bboxright-sprite.bboxleft)*max(0, sprite.bboxbottom-waterY)

    where waterY is the y of the water level and "k" is some number to tune it. You could probably start by setting k to 1 and go from there. I had to fiddle with it quite a bit to get something i was happy with.

    The only drawback is no torque is applied so I utilized imagepoints to be the centers of the box divided into four, and doing the approximate force for each. It actually gave very pleasing results and long planks would level out. To avoid bouning out of the water I cranked up the damping when in contact with water.

    dropbox.com/scl/fi/zweqaf1pcui7tm4my1aep/approx_buoyancy.capx

    Here's an old topic with a few more ideas:

    construct.net/en/forum/construct-2/how-do-i-18/physics-float-in-water-133996

    One was to use sine which may work depending on what you're looking for.

    The second was to calculate the actual area under the water, and the Center of mass of that area. It's much more realistic. The box2d example bounces a bit much, but turning up the angular damping helps.

  • I don’t think there is any plugin that would help.

    My current idea is to use the physics behavior for the bubbles. Then just apply a force from every bubble to every other bubble to have them go together. They will just be rigid circles though, not soft.

    To make them soft the idea is to draw a larger circle around the bubbles and everywhere those larger bubbles overlap it would become flat. Easier said than done, I have a few ideas but they are involved and I’d have to actually try it out to see if it looks good.

    The circle outline could be done as a bunch of sprites as line segments. Filling these outlines could be done by triangulating the polygon shape and then placing two right triangle sprites to fill each triangle.

    As is construct 2 doesn’t have the drawing features we want. Namely drawing lines and polygons.

    There’s a few plugins that can help do that:

    Canvas- has drawline and polygon drawing

    Polygon- has polygon

    Paster- has draw quad

    Custom draw- has draw quad

    C3 has only the built in canvas plugin. I’m pretty sure you can draw at least filled polygons with it.

  • Here’s a few ideas.

    The first is to to make the L shape out of a single object so it’s easy to move and rotate. Then when it lands you create the smaller squares on top of the shape, and then destroy the shape. You could utilize image points to mark where to create the smaller sprites.

    Another idea is to have another hidden object to act as the rotation center and use the pin behavior to attach the squares. Delays of when the behavior runs may or may not be an issue.

    A third idea is a more manual method. First you create the squares in the shape you want, and probably use an instance variable to help with picking just them and then rotate them with this equation:

    Set position to (self.x-centerx)*cos(90)-(self.y-centery)*sin(90)+centerx, (self.x-centerx)*sin(90)+(self.y-centery)*cos(90)+centery

    Where centerx, centery is the position to rotate around and 90 is how much to rotate.

  • Try Construct 3

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

    Try Now Construct 3 users don't see these ads

  • The -180 to 180 range is caused by the math function used to calculate the angle. Namely atan. Radians don’t effect that.

    Probably a good rule of thumb is any angle that is calculated will be in the -180 to 180 (or normalized angle) range, and any angle you set will just keep that value or make it in the 0-360 range. So even though you can set the angle of motion, it’s something the behavior can change so it’s calculated.

    Usually when comparing angles it’s best to just use the angle comparison system conditions. With them it sees things like 270 and -90 as the same.

  • Not easily I think, then again it’s probably not too easy no matter what you use.

    What I see are distorted spheres overlapping each other. I don’t know how you’d do that without 3D.

    You could try just drawing outlines with a bunch of sprites and make it flat between circles. That would likely require a lot of calculations.

    You can’t really draw filled polygons so coloring the bubbles isn’t really feasible.

    The motion could be done with physics pulling the bubbles toward each other. The soft body nature looks to be secondary.

    In short I don’t think construct has the tools to help do it as is. You’ll probably have to do it with a custom Plugin or something.

  • So the issues are the textBox isn't long enough and the text isn't being centered vertically?

    The width should be simple enough. For fixed width fonts like spritefonts it would be len(text)*characterWidth. For normal text objects there is a textWidth expression, but it only gives the width of the previous frame (booo!). You could also do word warping. I think this is handled by the object, but I haven't checked.

    Vertical centering is just a matter of moving things up and down. So if you know you have two lines you can leave things as they are, and if you have only one line you can move the line down by half a the height of a letter to center it. You may also be able to utilize the vertical alignment property.

    There is an issue if you use lines of dialog with newlines, you'd have to look at the width of each line and use the longest one for the width. Here's it in pseudocode: 

    var width = 0
var height = 0
repeat tokencount(dialog, newline) times
-- set width to max(width, len(tokenat(dialog, loopindex, newline)*characterWidth)
-- add characterHeight to height

    I'll probably stop there, hopefully some of that will help.

    Edit:

    Here's an example you can pick apart:

    uc2eb33046c9bb86a6bc36486d20.dl.dropboxusercontent.com/cd/0/get/Ch-aFddCmrOJxdG1BhKzGuiCQY0pPV3Liu0ECEFCCak3zE2d4Z3NumX1aKa9qvF-dTRGx3KPanNhZXYuHRHMhwb5Xm4WUDKM4BBrOeUaH4f_lYKptTm83-0Y0nYgEDZSiXU/file

  • The math for motion can be done per axis which helps simplify things.

    We can just pick one axis as an example. You can do other axis in the same way.

    x is the position, and vx is the velocity.

    Constant speed:

    x = x + vx*dt

    acceleration:

    vx = vx + 100*dt //100 is the acceration

    x = x + vx*dt

    So for a mario style example you could do x with constant speed and y with acceration. Then to jump you'd set vy to -200 or something. 

    On space pressed
-- set vy -200

    You could also just apply the vy the while the button is pressed. Just use a timer too so you'd only set the velocity for so long.

    xyz stuff is mostly a matter of mapping the third axis onto xy.

  • —

    Sorry, I won't be porting them to C3. C3 does seem to have it's own canvas plugin though. Maybe you could open the example in C2 as a reference and see if you can utilize the C3 canvas plugin to do the same thing.

    I haven't really used C3 to give a better idea.

  • Does these files work with C3? I downloaded them but couldn't install it to my C3 Project.

    Thanks

    Naw. They only work with C2.

  • It's always been fuzzy to me, so I had a look at the runtime source in C2. C3 is probably very similar, since if they changed it it would be a breaking change.

    It seems that the it picks the default instance by going through the layouts in order they are in the project, and then the layers in order and the instance with the lowest uid becomes the default.

    I didn't look too much further but another thing to test is the order of the layers. Are they defined by their order in the editor or the order they were added? I'm guessing just their order. I have no idea if global layers mess things up. I think the layout order is the most useful thing to know.

    Anyways, from that it seems that a nice strategy to control default instances is to create a new layout and add one instance of each object. Then drag that layout to the top of the layout list. Be sure to set the initial layout in the project properties.

    Thanks for the interesting question, it revealed something useful I didn't know before.