Running a Simulation

Once you have a fully described model, it’s time to create and execute a simulation.

In order to execute a model a CUDASimulation must be created by passing it your completed ModelDescription. The CUDASimulation then creates it’s own copy of the ModelDescription, so further changes will not affect the simulation.

The simulation can then be executed by calling simulate():

C++Python

// Fully declare a model
flamegpu::ModelDescription model("example model");
...

// Create a simulation object from the model
flamegpu::CUDASimulation simulation(model);

// Configure the simulation
...

// Run the simulation
simulation.simulate();

This is the most simple case, however normally a simulation needs to be configured, to select simulation option, override the initial state and specify data to be collected.

This chapter has been broken up into several sections detailing these features:

• Configuring Execution
• Overriding the Initial State
• Collecting Data

Additionally, a visualisation can be configured, however that is covered in it’s own chapter.