driema/euler-fluid-cpp

Euler fluid simulated with CPP and SFML

Euler fluid simulation

Inspiration

This project is based upon an article by Mike Ash.

For more info see: https://mikeash.com/pyblog/fluid-simulation-for-dummies.html

What is it?

This Project is a visualisation of the famous Navier Stokes Equations.

Simplified this project is a grid based simulation that emulates and visualizes how fluids act, as well as how their behaviour changes if you change the viscosity or diffusion of a certain liquid.

If you have any improvements or ideas, please let me know.

Also feel free to report any bugs that may occur.

How to use

Pressing left mouse will add density.

Dragging the mouse will apply force towards the moving direction.

Keyboard options:

• c = switch color mode (default, hsb, velocity)

Building the project

To build the project simply execute make build in the directory of the makefile.

Make sure to also install the library SFML which is needed to build the project.

To install SFML under Linux: sudo apt-get install libsfml-dev

euler-fluid-cpp's Issues

Sim.cpp Compiler Error C2280

Seem to be getting an odd error on line 6 of Sim.cpp

Sim::Sim() : options(Options()), container(Container(0.2f, 0, 0.0000001f)), win(sf::RenderWindow(sf::VideoMode(SIZE* SCALE, SIZE* SCALE), "Euler fluid simulation - Github: https://github.com/driema/euler-fluid-cpp", sf::Style::Titlebar | sf::Style::Close)) {}

Severity Code Description Project File Line Suppression State
Error C2280 'sf::RenderWindow::RenderWindow(const sf::RenderWindow &)': attempting to reference a deleted function Euler C:\VSProjects\Euler\sim.cpp 6

Any solutions?

About Fluids and rigid bodies

Hello.

First; I want to congratulate you for your contribution. It is great!!

Second; I am considering a project where I want to simulate volcanic lava running down the slope of a mountain (in 2D).

I don't know if it is possible to do it with this approach. The first question I have is whether it is possible to simulate the effect of the lava going around obstacles it may encounter on its way, such as houses (rectangles). Do you have an opinion on how to implement this?

In other words, I wonder how this "fluid" will work interacting with rigid objects/bodies. I guess, as it is not a particle system per se there are no collisions to detect. :-D

Thanks!

DJuego

Refactor the pointers in the container header file to arrays

class Container {
private:
	Physics physics;

	int size;

	float dt;
	float diff;
	float visc;
	
	float* px;
	float* py;

	float* x;
	float* y;

	float* previousDensity;
	float* density;

	void InitArr(float* arr, int size);
public:
	Container();
	Container(float dt, float diff, float visc);
	~Container();

	void AddDensity(float x, float y, float amount);
	void AddVelocity(float x, float y, float px, float py);
	void Step();
	void Render(sf::RenderWindow& win);
	void FadeDensity(int size);
};

to

extern const int SIZE;
extern const int SCALE;
class Container {
private:
	Physics physics;

	int size;

	float dt;
	float diff;
	float visc;
	
	float px[SIZE*SIZE];
	float py[SIZE*SIZE];

	float x[SIZE*SIZE];
	float y[SIZE*SIZE];

	float previousDensity[SIZE*SIZE];
	float density[SIZE*SIZE];

	void InitArr(float* arr, int size);
public:
	Container();
	Container(float dt, float diff, float visc);
	~Container();

	void AddDensity(float x, float y, float amount);
	void AddVelocity(float x, float y, float px, float py);
	void Step();
	void Render(sf::RenderWindow& win);
	void FadeDensity(int size);
};