Integrators are a set of mathematical methods available to calculate the movement of particles. The following guidelines will help to choose a proper integrator, according to the behaviour aimed at by the animator.

• Euler: Also known as Forward Euler. Simplest integrator. Very fast but also very unstable. If no dampening is used, particles get more and more energy over time. For example, bouncing particles will bounce higher and higher each time. Should not be confused with Backward Euler (not implemented) which has the opposite feature, energies decrease over time, even with no dampening. Use this integrator for short simulations or simulations with a lot of dampening where speedy calculations is more important than accuracy.
• Midpoint: Also known as 2nd order Runge-Kutta. Slower than Euler but much more stable. If the acceleration is constant (no drag for example), it is energy conservative (also known as symplectic). It should be noted that in example of the bouncing particles, the particles might bounce higher than they started once in a while, but this is not a trend. This integrator is a generally good integrator for use in most cases.
• RK4: Short for 4th order Runge-Kutta. Similar to Midpoint but slower and in most cases more accurate. It is energy conservative even if the acceleration is not constant. Only needed in complex simulations where Midpoint is found not to be accurate enough.

The initial velocity of particles can be set through different parameters, based on the type of the particle system (see Particle System tab). If the particles system type is Emitter or Hair, then the following parameters give the particle an initial velocity in the direction of...

• Object: ...the emitter objects movement (e.g. Let the object give the particle a starting speed)
• Normal: ...the emitter's surface normals (e.g. Let the surface normal give the particle a starting speed)
• Random: ...a random vector (e.g. Give the starting speed a random variation)
• Tan & Rot: ...a tangential vector along the surface rotated by "Rot"
  • Tan: Let the tangent speed give the particle a starting speed
  • Rot: Rotates the surface tangent

If the particles system type is Reactor, then the following parameters give the particle an initial velocity in the direction of...

• Particle: ...the target particles velocity (e.g. Let the target particle give the particle a starting speed)
• Reactor: ...a vector away from the target particles location at the time of the reaction (e.g. Let the vector away from the target particles location give the particle a starting speed)

These parameters specify how the individual particles are rotated during their travel.

• Dynamic: Only initializes particles to the wanted rotation and angular velocity and let's physics handle the rest, otherwise the angular velocity is predetermined at all times (e.g. Set rotation to dynamic/constant)
• Rotation: Sets the initial rotation of the particle by aligning the x-axis in the direction of... (e.g. Select particle rotation mode)
  • None: ...the x-axis
  • Normal: ...the emitter's surface normal
  • Velocity: ...the particle's initial velocity
  • Random: ...a random vector
• Amount: How much to align the x-axis (e.g. Rotation amount)
• Phase: Rotation around the x-axis (e.g. Initial rotation phase)
• Angular v: The magnitude of angular velocity, the dropdown specifies the axis of angular velocity to be...
  • None: ...a zero vector
  • Spin: ...the particles velocity vector
  • Random: ...a random vector

These parameters specify how external physical factors can enhance or impede particles velocity. Useful when simulating various phenomenon like gravity, air-drag, friction and such.

• AccX, AccY and AccZ: An acceleration in the direction of the global axes. Use this to implement gravity by setting AccZ to a negative value, for example.
• Drag: A force that reduces particle velocity in relation to it's speed and size (useful in order to simulate Air-Drag or Water-Drag).
• Brown: A random force that changes from frame to frame. Simulates Brownian movement which is an effect seen on small particles where forces from individual molecules are unbalanced over time (most simulations shouldn't use it).
• Damp: Reduces particle velocity (deceleration, friction, dampening).