function run(self)
{
self._x+=Math.sin(self._rotation*(Math.PI/180))*self.theSpeed;
self._y+=Math.cos(self._rotation*(Math.PI/180))*self.theSpeed*-1;
}
Planet Gravity
Author
I was trying to figure out the math for a 2d game were you cruise around in space and get pulled around by different planets gravities, similar to an old atari game (forgot the name) or master of orion.
I know the formulas for calculating mass, gravity, gravitational acceleration, and such, but im not sure how to turn it into code, specifically, the players rotation and x/y values. At the moment im calling this function every frame:
And allowing the arrow keys to change _rotation and theSpeed. How exactly does the gravity come into play, specifically, how does it affect the angle that the player is moving in?
Thanks!
Actually to compute precisely the torque generated by the gravitational attraction, you would need to compute an integral:
torque = summation over all mass elements of r(x,y,z).cross( F(x,y,z) )dxdydx
x,y,z being the local coordinates of a mass point.
My guess is that value should be quite small, since the variation of F(x,y,z) over the volume defined by the spaceshift is quite small( it depends of the distance of the element considered to the center of the planet ). You could safely consider that each mass element is at the same distance from the center of a planet and each element of torque is annihilated by another one. The effect could be meaningful if and only if the size( and therefore mass ) of the spaceshift is comparable to the size of the planet. You could then simply compute the forces at two "end points" of the space shift( you approximate its shape by a segment ), and compute the related torques generated by each planet on the spaceshift, and sum them.
Since, your game is two dimensional, there is a direct relationship between the torque variation and the variation of the orientation angle of the spaceshift.
[Edited by - johnstanp on March 9, 2010 3:38:15 AM]
torque = summation over all mass elements of r(x,y,z).cross( F(x,y,z) )dxdydx
x,y,z being the local coordinates of a mass point.
My guess is that value should be quite small, since the variation of F(x,y,z) over the volume defined by the spaceshift is quite small( it depends of the distance of the element considered to the center of the planet ). You could safely consider that each mass element is at the same distance from the center of a planet and each element of torque is annihilated by another one. The effect could be meaningful if and only if the size( and therefore mass ) of the spaceshift is comparable to the size of the planet. You could then simply compute the forces at two "end points" of the space shift( you approximate its shape by a segment ), and compute the related torques generated by each planet on the spaceshift, and sum them.
torque_acc = zerofor each planet compute the attraction force, force_1, at end_1 of spaceshift compute the torque, torque_1, it generates at the center of mass compute the attraction force, force_2, at end_2 of spaceshift compute the torque, torque_2, it generates at the center of mass add torque_1 and torque_2 to the torque accumulator, torque_accSince, your game is two dimensional, there is a direct relationship between the torque variation and the variation of the orientation angle of the spaceshift.
[Edited by - johnstanp on March 9, 2010 3:38:15 AM]
Can't help with the torque, but gravity works something like this: in the beginning you have the player-planet gravitational system (F = (G * m1 *m2) / (r*r)). You know the distance and both masses so you solve it and get a force. If you have more than one planet attracting the player you could just solve both systems and add the forces (ignoring attraction between planets). Knowing that F=ma you can turn that force into acceleration, then you change position by velocity * time and change velocity by acceleration * time (p, v, a, F are all vectors obviously). This is the very basic Euler integration (the one that so many people use and so many people say that it sucks / is unstable. But probably good enough for first approximation and for simple 2D game).
Quote:Original post by lightbringer
Can't help with the torque, but gravity works something like this: in the beginning you have the player-planet gravitational system. You know the distance and both masses so you solve it and get a force. The force is along the vector (planet minus player). Knowing that F=ma you can turn that force into acceleration, then you change position by velocity * time and change velocity by acceleration * time. This is the very basic Euler integration (the one that so many people use and so many people say that it sucks / is unstable. But probably good enough for first approximation and for simple 2D game).
My bad...
I assumed that he knew how to compute a gravitational force, and integrate it...I understood that he wanted to know how the gravitational forces could affect the rotation of his spaceshift(angular momentum). I skipped force and torque ant the integration computing because I thought that he wasn't asking for it.
[Edited by - johnstanp on March 10, 2010 2:57:30 PM]
Quote:Original post by johnstanp
I assumed that he knew how to compute a gravitational force...I understood that he wanted to know how the gravitational forces could affect the rotation of his spaceshift(angular momentum). The method you provide, will only give him effects on the linear velocity of its spaceshift. I skipped it because I thought that he wasn't asking for it.
I skipped torque because I honestly have no clue about it :) But given that Azh321 is trying to add rotation to position, I think he needs to hear about the linear stuff, too.
Quote:Original post by johnstanpYeah, I wasn't sure if that's what he wanted. Azh321, do you want the player's ship to rotate due to gravity, or just get pulled by gravity?
I understood that he wanted to know how the gravitational forces could affect the rotation of his spaceshift(angular momentum).
Author
This seems a lot more complicated than I was expecting, luckily im currently taking calc 3 :) I will have to look over the math when I get back to my desktop.
To answer your question johnstanp, im not sure yet. I really need to reference a similar game to see how it deals with it. Im thinking I do want the ship to rotate to point in the direction that its currently moving towards, so I will need to have a separate variable for the angle and for the actual rotation of the player.
To answer your question johnstanp, im not sure yet. I really need to reference a similar game to see how it deals with it. Im thinking I do want the ship to rotate to point in the direction that its currently moving towards, so I will need to have a separate variable for the angle and for the actual rotation of the player.
def gravitate(planet, body): dx = planet.x - body.x dy = planet.y - body.y dist = sqrt((dx*dx) + (dy*dy)) #normalize dx and dy dx /= dist dy /= dist force = (body.mass * planet.mass)/(dist*dist) dx *= force dy *= force return (dx,dy)Just add dx & dy to your ship's x,y position
Author
Thank you guys very much. Youve all been helpful, I will make a prototype of it in action tonight and post a link for you guys to see.
thanks!
thanks!
Quote:And allowing the arrow keys to change _rotation and theSpeed. How exactly does the gravity come into play, specifically, how does it affect the angle that the player is moving in?The force exerted by each planet will have an effect mostly on the ship's velocity, but, as has been stated before, will have very little effect on it's angular speed, unless you really want to complicate things.
Each force contributes to a change in the ship's acceleration, related by the know equation
accumulated_force = vector2d(0,0)for each planet: force = compute_force_from_planet_to_ship() accumulated_force += forceNote that if you do this, you don't use an angle to represent the ship's velocity anymore, it really is just the velocity vector. You may want to compute an angle out of it to align the sprite of the ship to the direction it is currently flying to, but that would be computed after that vector.
This topic is closed to new replies.
Advertisement
Popular Topics
Advertisement
