I've just got started with physics. I'm using Java, though language does not matter obviously. Now I though I'd do something like this:
A ball with a certain speed, radius and mass hits a rectangle with a certain mass, width and height. Depending on where the ball hits the rectangle (how high up), and all the properties the ball and the rectangle have that i just mentioned, there will be different outcomes of the situation.
These are the four possible outcomes:
The ball bounces back because the rectangle was too heavy
The rectangle starts to wobble, but then goes back to normal
The rectangle falls to the right
The ball strikes through making the rectangle fall to the left
Please note, I don't expect you to write a program for me. I understand it is a lot to think off. But I have no idea how to start. I would really appreciated some guide lines and links to further reading about this (I was not sure what to google to find info about this.)
And also, I'm doing this to learn, so don't tell me to use an engine or anything like that.
You are trying to build a simple physics simulator. This is a pretty involved problem, and you'll have to learn a certain amount of physics along the way.
I suggest you develop the simulator to handle these situations, roughly in this order:
An object moves through space (constant velocity, no gravity).
An object moves under the influence of a constant force (such as gravity).
An object moves with a constraint (e.g. a pendulum, a rolling square).
An object slides across a surface, with friction (both static and kinetic).
Two objects collide inelastically (they stick).
Two objects collide elastically (they bounce).
Once you have all of these, you will be able to simulate your ball and rectangle.
Related
Is it possible to have a physics object in GameMaker Studio use precise collisions?
Here's some context for my question. I'm making a pirate game where the player sails around a large ocean with a number of islands. I've been using the physics engine to control the movement of the ship, and that is working well. However, the problem arises when trying to introduce collisions between the ship and the various islands. As far as I can tell, the underlying physics fixtures can only be formed into fairly simple shapes. Specifically, the collision shape editor is limited to 12 points, and only convex shapes. This is a problem, because many of my islands are relatively complicated non-convex shapes, and aren't necessarily a single piece. It would be nice to be able to use the island sprite as a precise collision mask, as would be possible for non physics based objects.
Is there a way to do this, or a possible work-around that I'm missing? Here's an example of one of my islands:
I can see two solutions to your problem.
1 - The easiest, but performance-unfriendly.
In the sprite editor, click "Modify mask". There should be a "precise collision checking" box you can tick. This means that your sprite will be checked pixel by pixel for collisions. As you can guess, this is not performance friendly, but will do exactly what you want.
2 - The one I would recommend.
What you could do is just draw the island sprites, either through the background or via a dedicated object, and then create some simple shape objects (rectangle, circle and diamond), that would be invisible, and place them over your islands in the room editor. (Don't forget that you can stretch them).
These simple shaped objects would be the ones to check for collisions.
I used this technique make a hitbox for complex-shaped clouds in one of my games, so I know it works.
I believe that the island you show us can be fairly well covered with a few ovals and a long rectangle.
Bonus : after doing that graphically, you can copy the creation code of the shapes from the room create event to the island create event to repeat for multiple identical islands. Just don't forget the position/angle offset !
By using the Shape options when defining the collision shape, you can have any kind of Convex polygon as your collision shape. Example:
The spot where you choose the Shape option is in red.
After you select that option, you can just click & drag to add/edit a vertex to the polygon. Just bear in mind, it has to be a convex polygon, GameMaker is very strict about that. You can also remove vertices by right clicking on them.
I am working with kinect in openframework using the ofxKinect addon, which is great and plenty fun!
Anyway I am looking for some pointers or a direction when dealing with multiple bodies on the screen. I was thinking of making a rect around each detected body and when the rects intersect something could happen, an effect or anything.
So what I am looking for are ideas or something that could point me to the right direction of detecting multiple bodies when using a kinect.
Right now based on the depth image I get from the kinect I go through each pixel and create a bunch of smaller rectangles with a padding and group them in a larger rectangle bound if they are separate from another rectangle group. This is not ideal as it only deals with the pixel values and is not really seperating bodies from eachother and is not giving me the results I am looking for.
So any ideas would be greatly appreciated!
If you want to use ofxKinect a quick solution would be to threshold on depth and assume bodies and no other objects will be within a depth range. This should make it easy to use the OpenCV's contour finder to isolate the outlines of the bodies and get the bounding rectangles. If the rectangles intersect(and ofRectangle already does the math you), trigger the reaction you need. Also don't forget to do that once if the effect isn't showing already, otherwise you will trigger the effect multiple times per second while the two bodies' bounding rectangles intersect.
You could try something a bit more hardcore and using ofxCv(not just ofxOpenCV) to tap into the HoG functionality. This is slow in itself and not ideal with the depth map, but hopefully you can run in every few seconds just to detect a person and the depth, then keep tracking that movement.
Personally, if you want to track people with the Kinect I recommend using ofxOpenNI as if already provides the scene segmentation feature and even if you don't track the skeletons you can still get useful information like the pixels pertaining to each body and they're centre of mass. I'm guessing Microsoft KinectSDK has a similar feature and there should be an oF addon, but it's windows only.
ofxKinect/libfreenect does not offer any people detection features, so you will need to roll your own.
I have a limited area (screen) populated with a few moving objects (3-20 of them, so it's not like 10.000 :). Those objects should be moving with a constant speed and into random direction. But, there are a few limitation to it:
objects shouldn't exit the area - so if it's close to the edge, it should move away from it
objects shouldn't bump onto each other - so when one is close to another one it should move away (but not get too close to different one).
On the image below I have marked the allowed moves in this situation - for example object D shouldn't move straight up, as it would bring it to the "wall".
What I would like to have is a way to move them (one by one). Is there any simple way to achieve it, without too much calculations?
The density of objects in the area would be rather low.
There are a number of ways you might programmatically enforce your desired behavior, given that you have such a small number of objects. However, I'm going to suggest something slightly different.
What if you ran the whole thing as a physics simulation? For instance, you could set up a Box2D world with no gravity, no friction, and perfectly elastic collisions. You could model your enclosed region and populate it with objects that are proportionally larger than their on-screen counterparts so that the on-screen versions never get too close to each other (because the underlying objects in the physics simulation will collide and change direction before that can happen), and assign each object a random initial position and velocity.
Then all you have to do is step the physics simulation, and map its current state into your UI. All the tricky stuff is handled for you, and the result will probably be more believable/realistic than what you would get by trying to come up with your own movement algorithm (or if you wanted it to appear more random and less believable, you could also just periodically apply a random impulse to a random object to keep things changing unpredictably).
You can use the hitTest: method of UIView
UIView* touchedView=[self.superview hitTest:currentOrigin withEvent:nil];
In This method you have to pass the current origin of the ball and in second argument you can pass nil.
that method will return the view with which the ball is hited.
If there is any hit view you just change the direction of the ball.
for border you can set the condition for the frame of the ball if the ball go out of the boundary just change the direction of the ball.
Examples of what i'm going to need:
I'm using cocos2d to draw a CCTMXTiledMap, on those tiles i'll have to draw the LOS triangle.
How would i test if the player is within that triangle, taking obstacles into account?
How would i draw the line of sight area like in the examples above?
BTW, i wasn't sure if this should have been posted here or on gamedev, don't be mad.
You may wish to look at point-in-polygon algorithms such as the ray casting algorithm described here.
You can break up the triangle to account for obstacles, or just make a more complex polygon. You should be able to find an implementation to suit your needs online.
You may also want to take a look at this article for some inspiration. You can maintain a tree like structure, a root triangle (or fulcrum) that can be used to determine whether a point is in general line of sight, with the children (triangles) taking obstacles into account. That way you can quickly eliminate more complex checks.
In the image below the dark blue dots are quickly eliminated from further checking as they do not fall within the root viewing triangle.
For a university term project, I'm working on a graphical roguelike (I'm aware of the contradiction in terms :P) that uses an isometric display. What I'm trying to figure out is, since drawing all the tiles is stupidly expensive and unnecessary, I'm wanting to figure out a relatively fast algorithm to determine which tiles should be drawn to fit within an NxMpx window, given that the tile graphics are XxYpx.
I'm not doing smooth scrolling for this, so that's not an issue. I'm also not worried about being perfect - a little unnecessary draw is fine, I just don't want to draw a huge amount of unnecessary tiles that won't show up in-game.
You need to think about two concepts: Screen space and world space. These are very important in 3d engines, but they apply to all but the very simplest games. In the isometric engine your world-space is a 2d array of tiles.
So you are looking first of all at a way to covert between these two co-ordinate spaces. Once you've done that, it'll be obvious that screen space maps onto the world as a rectange that's turned at 45 degrees. You can determine a formula for that, but it's only important if you are trying to determine which part of the world space is visible so you only simulate monsters in that area (an efficiency necessary on 8 bit consoles, probably not on a modern PC!). When it comes to actual rendering you don't really need to determine this visible region of the world, because it's implicit in the way you render the tiles onto the screen:-
You work out which tile location is at the top-left of the screen, call this O (for origin) - that's going to be a fixed offset from your point of interest, usually the player, that you want to keep in the centre of the screen.
Once you have that you paint the tiles in the top row of the screen, stepping +1X and -1Y in world space for each tile (if you are looking north-east).
Then you paint the row below. That is offset minus one half a tile width in screen space and starts at O -1X in world space.
You repeat steps 2 and 3, modifying your starting position in world space by -1 in X and Y from what you used in the previous stage 2 until you reach the top of the screen.
Other tips:-
Obviously you don't draw any tiles that are outside the map. You might also, depending on game design, not draw any tiles outside a particular room the player is in.
There's not just floor tiles to draw, theres also players, monsters, scenary, etc. The rule is you draw everything in a paricular world location in the same pass. That way objects closest to the 'camera' will obscure stuff behind them (which is why you start drawing at the top of the screen).
Also, you don't just have floor tiles in most iso engines, you would also have furniture and wall segments. You might designate areas of the map as belonging to a particular room, when the focus is on that room (because the player is in it for example), you don't draw the wall segments for the side of the room closest to the camera.
Anyway, that's enough to be getting on with, hope it's helpful and your project goes well.