We Keep Coding

How to create a 3D mesh from a heightmap represented as a float array

I am attempting to understand how a 3D mesh has been constructed from a heightmap stored as a one dimensional float array. The only examples I have seen previously have made use of a 2D float array, and I am struggling to wrap my head around the math involved here. Any insight into it would be appreciated. I have commented the code which I do not quite understand yet for your convenience.
Source of code: https://github.com/SebLague/Hydraulic-Erosion
public void ContructMesh () {
Vector3[] verts = new Vector3[mapSize * mapSize];
int[] triangles = new int[(mapSize - 1) * (mapSize - 1) * 6];
int t = 0;
//Note that default mapSize is 255
for (int i = 0; i < mapSize * mapSize; i++) {
//Following code is not properly understood
int x = i % mapSize;
int y = i / mapSize;
int meshMapIndex = y * mapSize + x;
Vector2 percent = new Vector2 (x / (mapSize - 1f), y / (mapSize - 1f));
Vector3 pos = new Vector3 (percent.x * 2 - 1, 0, percent.y * 2 - 1) * scale;
pos += Vector3.up * map[meshMapIndex] * elevationScale; //Elevation scale is 20 by default
verts[meshMapIndex] = pos;
//End of misunderstood code
if (x != mapSize - 1 && y != mapSize - 1) {
t = (y * (mapSize - 1) + x) * 3 * 2;
triangles[t + 0] = meshMapIndex + mapSize;
triangles[t + 1] = meshMapIndex + mapSize + 1;
triangles[t + 2] = meshMapIndex;
triangles[t + 3] = meshMapIndex + mapSize + 1;
triangles[t + 4] = meshMapIndex + 1;
triangles[t + 5] = meshMapIndex;
t += 6;
}
}
Mesh mesh = new Mesh();
mesh.indexFormat = UnityEngine.Rendering.IndexFormat.UInt32;
mesh.vertices = verts;
mesh.triangles = triangles;
mesh.RecalculateNormals ();

What specifically do you not understand?
int x = i % mapSize; // Get the x location of the current point
int y = i / mapSize; // Get the y location of the current point
// This should be equal to i, IDK why this is even calculated
int meshMapIndex = y * mapSize + x;
// How far along either direction are we?
Vector2 percent = new Vector2 (x / (mapSize - 1f), y / (mapSize - 1f));
// Make a new vector that scales the X and Y coordinates up.
// The Y coordinate is set to the Z element in this vector
// Presumably because whatever you use to render uses the Y element as "up"
// And the X-Z plane is the horizontal plane
// Also normalize X and Z to lie between -1*scale and 1*scale
Vector3 pos = new Vector3 (percent.x * 2 - 1, 0, percent.y * 2 - 1) * scale;
// Add the value at the current index, times the scale, as the Y element of pos
pos += Vector3.up * map[meshMapIndex] * elevationScale; //Elevation scale is 20 by default
// The X-Z values of pos give you the location of the vertex in the horizontal plane
// The Y value of pos gives you the height
// save the newly calculated pos in verts
verts[meshMapIndex] = pos;

Calculation island error with perlin noise

I'm developing a small project, which is a grid of 100 x 100 hexagons.
In the script below, I paint my hexagons with the perlin noise, but the format I want to island does not go away.
I'll leave my code and 2 examples as my map stays and how I wish it to stay.
My island
My Island
As i need
Im Need
int getColor(float x, float z)
{
xTO = (int)x / terrainWidth - 30;
zTO = (int)z / terrainHeight - 30;
float v = Mathf.PerlinNoise((xTO + x + seed) * freq, (zTO + z) * freq);
// v += 0.001f;
float form = formWorld(x, z);
if (v < 0.25f)
{
//water
return 0;
}
else if (v < 0.5f)
{
//sand
return 1;
}
else if (v < 0.75f)
{
//grass
return 2;
}
else
{
//Trees / Forest
MakeNewTree(new Vector3(xx, 0, z * 7.5f));
return 2;
}
}

If you want your image to look more like the second one, the best option is going to be adding a circular gradient which offsets your Perlin Noise.
The easiest way to do this is to measure the distance from the center and combine that with the perlin noise.
Here's some untested code.
int getColor(float x, float z)
{
xTO = (int)x / terrainWidth - 30;
zTO = (int)z / terrainHeight - 30;
float v = Mathf.PerlinNoise((xTO + x + seed) * freq, (zTO + z) * freq);
// v += 0.001f;
v -= CircleOffset(x,z)/2; //Change the two to make the island bigger.
float form = formWorld(x, z);
if (v < 0.25f)
{
//water
return 0;
}
else if (v < 0.5f)
{
//sand
return 1;
}
else if (v < 0.75f)
{
//grass
return 2;
}
else
{
//Trees / Forest
MakeNewTree(new Vector3(xx, 0, z * 7.5f));
return 2;
}
}
float CircleOffset(float x, float y)
{
Vector2 center = new Vector2(terrainWidth/2,terrainHeight/2);
float distance = Mathf.Sqrt((center.x - x)*(center.x - x) + (center.y - y) * (center.y - y));
return distance/terrainWidth;
}
Hope this helps!

Point-Zoom on Mandelbrot Set in C# - It works, except when the mouse has moved

I'm able to point zoom on the Mandelbrot set, as long as the mouse doesn't move after zooming has begun. I've tried calculating a normalized delta (new coordinate - old coordinate)*(oldzoom), but what happens is the image appears to jump around to a new location. I've seen this issue before. I'm struggling more here because I have to somehow convert this mouse position delta back to the -2,2 coordinate space of the Mandelbrot set.
Here's my code. What's important is the GetZoomPoint method, and then the lines of code that define x0 and y0. Also, I use the Range class to scale values from one range to another. I WAS using deltaTrans (thats the thing I was talking about earlier where I normalize the mouse delta with the old scale).
using OpenTK.Graphics.OpenGL;
using SpriteSheetMaker;
using System;
using System.Collections.Generic;
using System.Drawing;
using System.Linq;
using System.Text;
using System.Threading.Tasks;
namespace Fractal.Fractal
{
public class Mandelbrot : BaseTexture
{
private static Transform GlobalTransform = SpriteSheetMaker.Global.Transform;
private static Vector3 GlobalScale = GlobalTransform.Scale;
private static Vector3 GlobalTrans = GlobalTransform.Translation;
private static Vector3 LastWindowPoint = null;
private static Vector3 ZoomFactor = Vector3.ONE * 1.2f;
private static Vector3 Displacement = Vector3.ZERO;
private static int WindowSize = 100;
public static Vector3 GetZoomPoint()
{
var zP = OpenGLHelpers.LastZoomPoint.Clone();
if (LastWindowPoint == null)
{
LastWindowPoint = zP.Clone();
}
var delta = zP - LastWindowPoint;
var oldZoom = GlobalScale / ZoomFactor;
var deltaTrans = delta.XY * oldZoom.XY;
var factor = ZoomFactor.Clone();
Range xR = new Range(0, WindowSize);
Range yR = new Range(0, WindowSize);
Range complexRange = new Range(-2, 2);
// Calculate displacement of zooming position.
var dx = (zP.X - Displacement.X) * (factor.X - 1f);
var dy = (zP.Y - Displacement.Y) * (factor.Y - 1f);
// Compensate for displacement.
Displacement.X -= dx;
Displacement.Y -= dy;
zP -= Displacement;
var x = complexRange.ScaleValue(zP.X, xR);
var y = complexRange.ScaleValue(zP.Y, yR);
var rtn = new Vector3(x, y);
LastWindowPoint = zP.Clone();
return rtn;
}
public static Mandelbrot Generate()
{
var size = new Size(WindowSize, WindowSize);
var radius = new Size(size.Width / 2, size.Height / 2);
Bitmap bmp = new Bitmap(size.Width, size.Height);
LockBitmap.LockBitmapUnsafe lbm = new LockBitmap.LockBitmapUnsafe(bmp);
lbm.LockBits();
var pt = Mandelbrot.GetZoomPoint();
Parallel.For(0, size.Width, i =>
{
// float x0 = complexRangeX.ScaleValue(i, xRange);
float x0 = ((i - radius.Width) / GlobalScale.X) + pt.X;
Parallel.For(0, size.Height, j =>
{
// float y0 = complexRangeY.ScaleValue(j, yRange);
float y0 = ((j - radius.Height) / GlobalScale.Y) + pt.Y;
float value = 0f;
float x = 0.0f;
float y = 0.0f;
int iteration = 0;
int max_iteration = 100;
while (x * x + y * y <= 4.0 && iteration < max_iteration)
{
float xtemp = x * x - y * y + x0;
y = 2.0f * x * y + y0;
x = xtemp;
iteration += 1;
if (iteration == max_iteration)
{
value = 255;
break;
}
else
{
value = iteration * 50f % 255f;
}
}
int v = (int)value;
lbm.SetPixel(i, j, new ColorLibrary.HSL(v / 255f, 1.0, 0.5).ToDotNetColor());
});
});
lbm.UnlockBits();
var tex = new BaseTextureImage(bmp);
var rtn = new Mandelbrot(tex);
return rtn;
}
public override void Draw()
{
base._draw();
}
private Mandelbrot(BaseTextureImage graphic)
{
var topLeft = new Vector3(0, 1);
var bottomLeft = new Vector3(0, 0);
var bottomRight = new Vector3(1, 0);
var topRight = new Vector3(1, 1);
this.Vertices = new List<Vector3>()
{
topLeft,bottomLeft,bottomRight,topRight
};
this.Size.X = WindowSize;
this.Size.Y = WindowSize;
this.Texture2D = graphic;
}
}
}

I refactored my code, and also figured out a solution to this problem. 2 big wins in one. Ok, so I found a solution on CodeProject written in C# which I was readily able to adapt to my project. I'm not sure why I didn't realize this when I posted the question, but what I needed to solve this issue was to create a 'window' of zoom and not think in terms of a 'point zoom'. Yes, even if I am trying to zoom directly into a point, that point is just the center of some sort of a window.
Here is the method I have, which expects start and end mousedown coordinates (screen space), and converts the mandelbrot set window size accordingly.
public void ApplyZoom(double x0, double y0, double x1, double y1)
{
if (x1 == x0 && y0 == y1)
{
//This was just a click, no movement occurred
return;
}
/*
* XMin, YMin and XMax, YMax are the current extent of the set
* mx0,my0 and mx1,my1 are the part we selected
* do the math to draw the selected rectangle
* */
double scaleX, scaleY;
scaleX = (XMax - XMin) / (float)BitmapSize;
scaleY = (YMax - YMin) / (float)BitmapSize;
XMax = (float)x1 * scaleX + XMin;
YMax = (float)y1 * scaleY + YMin;
XMin = (float)x0 * scaleX + XMin;
YMin = (float)y0 * scaleY + YMin;
this.Refresh(); // force mandelbrot to redraw
}
Basically, whats happening is we calculate the ratio between the mandelbrot window size versus the screen size we are drawing to. Then, using that scale, we basically convert our mousedown coordinates to mandelbrot set coordinates (x1*scaleX, etc) and manipulate the current Min and Max coordinates with them, using the Min values as the pivot point.
Here's the link to the CodeProject I used as a reference: CodeProject link

GDI+ curve "overflowing"

I'm currently using GDI+ to draw a line graph, and using Graphics.DrawCurve to smooth out the line. The problem is that the curve doesn't always match the points I feed it, and that makes the curve grow out of the graph frame in some points, as seen below(red is Graphics.DrawLines, green is Graphics.DrawCurve).
How would I go about solving this?

The simplest solution is to set a tension:
The green curve is drawn with the default tension, the blue one set a tension of 0.1f:
private void panel1_Paint(object sender, PaintEventArgs e)
{
e.Graphics.SmoothingMode = SmoothingMode.AntiAlias;
e.Graphics.DrawLines(Pens.Red, points.ToArray());
e.Graphics.DrawCurve(Pens.Green, points.ToArray());
e.Graphics.DrawCurve(Pens.Blue, points.ToArray(), 0.1f);
}
You will need to test what is the best compromise, 0.2f is still ok, 0.3f is already overdrawing quite a bit..
For a really good solution you will need to use DrawBeziers. This will let you draw curves that can go through the points without any overdrawing and with full control of the radius of the curves; but to to so you will need to 'find', i.e. calculate good control points, which is anything but trivial..:
This result is by no means perfect but already complicated enough.. I have displayed the curve points and their respective control points in the same color. For each point there is an incoming and an outgoing control point. For a smooth curve they need to have the same tangents/gradients in their curve points.
I use a few helper functions to calculate a few things about the segments:
A list of gradients
A list of signs of the gradients
A list of segment lengths
Lists of horizontal and of vertical gaps between points
The main function calculates the array of bezier points, that is the curve points and between each pair the previous left and the next right control points.
In the Paint event it is used like this:
List<PointF> bezz = getBezz(points);
using (Pen pen = new Pen(Color.Black, 2f))
e.Graphics.DrawBeziers(pen, bezz.ToArray());
Here are the functions I used:
List<float> getGradients(List<PointF> p)
{
List<float> grads = new List<float>();
for (int i = 0; i < p.Count - 1; i++)
{
float dx = p[i + 1].X - p[i].X;
float dy = p[i + 1].Y - p[i].Y;
if (dx == 0) grads.Add(dy == 0 ? 0 : dy > 0 ?
float.PositiveInfinity : float.NegativeInfinity);
else grads.Add(dy / dx);
}
return grads;
}
List<float> getLengths(List<PointF> p)
{
List<float> lengs = new List<float>();
for (int i = 0; i < p.Count - 1; i++)
{
float dx = p[i + 1].X - p[i].X;
float dy = p[i + 1].Y - p[i].Y;
lengs.Add((float)Math.Sqrt(dy * dy + dx * dx));
}
return lengs;
}
List<float> getGaps(List<PointF> p, bool horizontal)
{
List<float> gaps = new List<float>();
for (int i = 0; i < p.Count - 1; i++)
{
float dx = p[i + 1].X - p[i].X;
float dy = p[i + 1].Y - p[i].Y;
gaps.Add(horizontal ? dx : dy);
}
return gaps;
}
List<int> getSigns(List<float> g)
{
return g.Select(x => x > 0 ? 1 : x == 0 ? 0 : -1).ToList();
}
And finally the main function; here I make a distinction: Extreme points ( minima & maxima) should have their control points on the same height as the points themselves. This will prevent vertical overflowing. They are easy to find: The signs of their gradients will always altenate.
Other points need to have the same gradient for incoming and outcoming control points. I use the average between the segments' gradients. (Maybe a weighed average would be better..) And I weigh their distance according to the segment lengths..
List<PointF> getBezz(List<PointF> points)
{
List<PointF> bezz = new List<PointF>();
int pMax = points.Count;
List<float> hGaps = getGaps(points, true);
List<float> vGaps = getGaps(points, false);
List<float> grads = getGradients(points);
List<float> lengs = getLengths(points);
List<int> signs = getSigns(grads);
PointF[] bezzA = new PointF[pMax * 3 - 2];
// curve points
for (int i = 0; i < pMax; i++) bezzA[i * 3] = points[i];
// left control points
for (int i = 1; i < pMax; i++)
{
float x = points[i].X - hGaps[i - 1] / 2f;
float y = points[i].Y;
if (i < pMax - 1 && signs[i - 1] == signs[i])
{
float m = (grads[i-1] + grads[i]) / 2f;
y = points[i].Y - hGaps[i-1] / 2f * m * vGaps[i-1] / lengs[i-1];
}
bezzA[i * 3 - 1] = new PointF(x, y);
}
// right control points
for (int i = 0; i < pMax - 1; i++)
{
float x = points[i].X + hGaps[i] / 2f;
float y = points[i].Y;
if (i > 0 && signs[i-1] == signs[i])
{
float m = (grads[i-1] + grads[i]) / 2f;
y = points[i].Y + hGaps[i] / 2f * m * vGaps[i] / lengs[i];
}
bezzA[i * 3 + 1] = new PointF(x, y);
}
return bezzA.ToList();
}
Note that I didn't code for the case of points with the same x-coordinate. So this is ok for 'functional graphs' but not for, say figures, like e.g. stars..

Maybe you just want to look at the "overshooting the bounds" problem as not a problem with the overshoot, but with the bounds. In which case, you can determine the actual bounds of a curve using the System.Drawing.Drawing2D.GraphicsPath object:
GraphicsPath gp = new GraphicsPath();
gp.AddCurve(listOfPoints);
RectangleF bounds = gp.GetBounds();
You can draw that GraphicsPath directly:
graphics.DrawPath(Pens.Black, gp);
As far as solving the bounds problem, the line necessarily overshoots the vertex on some axis. It's easier to see this fact when the lines are aligned to the bounds.
Given these points:
In order for them to be curved, they must exceed their bounds in some way:
If you never want to exceed their vertical bounds, you could simply ensure that the bezier handles have the same Y value as the vertex, but they will overshoot on the X:
Or vice-versa:
You could deliberately undershoot just enough to avoid the way curves can overshoot. This can be done by swapping the bezier handles, which would maybe be at the line-centers, with the vertices:

Attraction Force between 2D balls

I have a simulation with multiple circles moving in 2D space, with elastic collisions between them.
I'd like to add an attraction force between particles, so that particles move towards other particles depending on mass, etc. How would I go about this?
My collision management function looks like this:
void manageCollision(Particle particleA, Particle particleB)
{
float distanceX = particleA.Position.X - particleB.Position.X;
float distanceY = particleA.Position.Y - particleB.Position.Y;
double collisionAngle = Math.Atan2(distanceY, distanceX);
double pA_magnitude = Math.Sqrt(particleA.Velocity.X * particleA.Velocity.X + particleA.Velocity.Y * particleA.Velocity.Y);
double pB_magnitude = Math.Sqrt(particleB.Velocity.X * particleB.Velocity.X + particleB.Velocity.Y * particleB.Velocity.Y);
double pA_direction = Math.Atan2(particleA.Velocity.Y, particleA.Velocity.X);
double pB_direction = Math.Atan2(particleB.Velocity.Y, particleB.Velocity.X);
double pA_newVelocityX = pA_magnitude * Math.Cos(pA_direction - collisionAngle);
double pA_newVelocityY = pA_magnitude * Math.Sin(pA_direction - collisionAngle);
double pB_newVelocityX = pB_magnitude * Math.Cos(pB_direction - collisionAngle);
double pB_newVelocityY = pB_magnitude * Math.Sin(pB_direction - collisionAngle);
double pA_finalVelocityX = ((particleA.Mass - particleB.Mass) * pA_newVelocityX + (particleB.Mass + particleB.Mass) * pB_newVelocityX) / (particleA.Mass + particleB.Mass);
double pB_finalVelocityX = ((particleA.Mass + particleA.Mass) * pA_newVelocityX + (particleB.Mass - particleA.Mass) * pB_newVelocityX) / (particleA.Mass + particleB.Mass);
double pA_finalVelocityY = pA_newVelocityY;
double pB_finalVelocityY = pB_newVelocityY;
particleA.Velocity = new Vector2((float)(Math.Cos(collisionAngle) * pA_finalVelocityX + Math.Cos(collisionAngle + Math.PI / 2) * pA_finalVelocityY), (float)(Math.Sin(collisionAngle) * pA_finalVelocityX + Math.Sin(collisionAngle + Math.PI / 2) * pA_finalVelocityY));
particleB.Velocity = new Vector2((float)(Math.Cos(collisionAngle) * pB_finalVelocityX + Math.Cos(collisionAngle + Math.PI / 2) * pB_finalVelocityY), (float)(Math.Sin(collisionAngle) * pB_finalVelocityX + Math.Sin(collisionAngle + Math.PI / 2) * pB_finalVelocityY));
}
Each ball or particle spawns with a random mass and radius.
The function is called within an update type of method, like this:
Vector2 globalGravity = new Vector2(0f, gravityScale / 6000);
for (int i = 0; i < particles.Count(); i++)
{
particles[i].Update((float)updateTimer.Interval, globalGravity);
Vector2 position = particles[i].Position;
Vector2 velocity = particles[i].Velocity;
collisionWallCheck(ref position, ref velocity, particles[i].Radius);
particles[i].Position = position;
particles[i].Velocity = velocity;
Particle pA = particles[i];
for (int k = i + 1; k < particles.Count(); k++)
{
Particle pB = particles[k];
Vector2 delta = pA.Position - pB.Position;
float dist = delta.Length();
if (dist < particles[i].Radius + particles[k].Radius && !particles[i].Colliding && !particles[k].Colliding)
{
particles[i].Colliding = true;
particles[k].Colliding = true;
manageCollision(particles[i], particles[k]);
particles[i].initColorTable(); // Upon collision, change the color
particles[k].initColorTable();
totalCollisions++;
}
else
{
particles[i].Colliding = false;
particles[k].Colliding = false;
}
}
}
I'm storing the initial position, velocity and masses of each ball.
What I apparently need to do, and don't know how to implement, is:
Calculate the magnitude and direction of the gravitational force.
Knowing the force, you can calculate the acceleration of each body.
Knowing the acceleration you can calculate the new velocity.
Knowing the velocity you can calculate the new position.
I'm shaky with the equations for it essentially, and I'd like to start off by making an attraction force between just two balls.
Using Steven's suggestion, this is the new integrated code.
void updateTimer_Tick(object sender, EventArgs e)
{
const double G = 6.67398 * 0.00000000001;
for (int i = 0; i < particles.Count(); i++)
{
double sumX = 0;
double sumY = 0;
Particle pA = particles[i];
for (int k = i + 1; k < particles.Count(); k++)
{
Particle pB = particles[k];
Vector2 delta = pA.Position - pB.Position;
float dist = delta.Length();
if (dist < particles[i].Radius + particles[k].Radius && !particles[i].Colliding && !particles[k].Colliding)
{
particles[i].Colliding = true;
particles[k].Colliding = true;
manageCollision(particles[i], particles[k]);
particles[i].initColorTable();
particles[k].initColorTable();
totalCollisions++;
particles[i].Colliding = false;
particles[k].Colliding = false;
}
else
{
double distanceX = particles[i].Position.X - particles[k].Position.X;
double distanceY = particles[i].Position.Y - particles[k].Position.Y;
double r = Math.Sqrt(Math.Pow(distanceX, 2) + Math.Pow(distanceY, 2));
double force = G * particles[i].Mass * particles[k].Mass / (r * r);
double theta = Math.Tan(distanceY / distanceX);
sumX += force * Math.Cos(theta);
sumY += force * Math.Sin(theta);
particles[i].Colliding = false;
particles[k].Colliding = false;
}
}
double netForce = Math.Sqrt(Math.Pow(sumX, 2) + Math.Pow(sumY, 2));
double a = netForce / particles[i].Mass;
double aTheta = Math.Tan(sumY / sumX);
// Here we get accelerations for X and Y. You can probably figure out velocities from here.
double aX = a * Math.Cos(aTheta);
double aY = a * Math.Sin(aTheta);
Vector2 accel = new Vector2((float)aX, (float)aY);
particles[i].Update((float)updateTimer.Interval, accel);
//particles[i].Update((float)updateTimer.Interval, globalGravity);
Vector2 position = particles[i].Position;
Vector2 velocity = particles[i].Velocity;
collisionWallCheck(ref position, ref velocity, particles[i].Radius);
particles[i].Position = position;
particles[i].Velocity = velocity + accel;
}
Draw();
}
The Update function for the particles is simple, and before it used a global gravity Vector which was 0,0.
public void Update(float timeStep, Vector2 gravity)
{
velocity = velocity + timeStep * gravity;
position = position + timeStep * velocity;
}
I'm now unsure how to deal with the cases of 0.

Start by calculating the force of gravity acting on each object. This is given by
F = Gm1m2/r*r
where m1 and m2 are the masses of two objects, G is the gravitational constant, and r is the distance between the two objects.
Now, r is a vector, so you may want to split this up into separate components - Fx and Fy. You can do this as follows:
Fx = F * cos(theta)
Fy = F * sin(theta)
For each mass, calculate the force of gravity acting on it and every other object. Sum the vectors to get the net force of gravity. (Note - that link is available for your interest, but takes a long time to get to the point). At this point you will have a net force on each object, from which you can calculate acceleration. Here's the code to get to this point:
const double G = 6.67398 * 0.00000000001;
for (int i = 0; i < particles.Count(); i++)
{
double sumX = 0;
double sumY = 0;
for (int j = 0; j < particles.Count(); j++)
{
// Don't add attraction to self
if (i == j)
continue;
double distanceX = particles[i].Position.X - particles[j].Position.X;
double distanceY = particles[i].Position.Y - particles[j].Position.Y;
double r = Math.Sqrt(Math.Pow(distanceX, 2) + Math.Pow(distanceY, 2));
double force = G * particles[i].Mass * particles[j].Mass / (r * r);
double theta = Math.Tan(distanceY / distanceX);
sumX += force * Math.Cos(theta);
sumY += force * Math.Sin(theta);
}
double netForce = Math.Sqrt(Math.Pow(sumX, 2) + Math.Pow(sumY, 2));
double a = netForce / particles[i].Mass;
double aTheta = Math.Tan(sumY / sumX);
// Here we get accelerations for X and Y. You can probably figure out velocities from here.
double aX = a * Math.Cos(aTheta);
double aY = a * Math.Sin(aTheta);
}
NOTES
This doesn't take stuff like 0-values into account - you'll have to clean up this code to deal with special cases before it will run without crashing.
Don't update any positions until you've calculated all the forces, or else you'll be off for later elements in the list.
Another thing worth noting: This algorithm is O(n^2), so if you have more than a few bodies it's going to take a lot of crunching. That's unfortunately just the way it is; if you find a fast way to calculate gravitational attraction for a large number of bodies, you should probably call NASA.
Depending on your coordinate system, you may find the y-vectors getting reversed. This is because Euclidean geometry thinks of positive values of y as "going up" whereas programmers tend to measure y in positive units "going down" from the top of the screen. This can play havoc with your angles and things.

Knowing the position all the balls and their masses, you can calculate the vector of the force felt between any two bodies. Find the vector from ball 'A' to all other balls - 'A' to ball 'B', 'A' to 'C', 'A' to 'D', etc. Then, simple add all of A's vectors up to get a final vector of force acting on A. Repeat for B -> A, B -> C, etc to find B's vector. Do this for all, calculate the new velocity, and adjust the positions for the amount of time between steps.