Time Integration

Explicit Euler is the simplest and most intuitive time integration method. It calculates the future state of a system using the derivative (e.g., velocity, acceleration) evaluated at the current known state.

• Mechanism: x(t + Δt) = x(t) + v(t) * Δt
• Pros: Extremely simple to implement and computationally cheap per step.
• Cons: Prone to energy gain and instability, especially with large time steps or stiff systems (e.g., stiff springs). It is only conditionally stable.
• Use Case: Quick prototyping, educational examples, or non-critical simulations where absolute accuracy is secondary.

Implicit Euler is a first-order method renowned for its unconditional stability. It solves for the future state using the derivative evaluated at that future, unknown state.

• Mechanism: x(t + Δt) = x(t) + v(t + Δt) * Δt
• Pros: Unconditionally stable for linear problems, allowing for much larger time steps without simulation blow-up. Excellent for damping out high-frequency oscillations in stiff systems.
• Cons: Computationally expensive per step, as it requires solving a (often non-linear) system of equations. Introduces numerical damping, which can make simulations appear overly "mushy."
• Use Case: Simulating stiff materials like rigid constraints, cloth, or hair in real-time applications where stability is paramount.

Verlet integration is a second-order method popular in molecular dynamics and particle simulations. It does not explicitly store velocity, instead deriving it from position history.

• Mechanism: x(t + Δt) = 2*x(t) - x(t - Δt) + a(t) * Δt²
• Pros: Time-reversible and symplectic, meaning it conserves energy well over long simulations. More accurate and stable than Explicit Euler.
• Cons: Slightly more complex initialization. Velocity must be calculated separately if needed for forces.
• Use Case: Long-term stable orbital mechanics, molecular dynamics, and particle systems where energy conservation is critical.

The Fourth-Order Runge-Kutta (RK4) method is a family of higher-order explicit methods that achieve greater accuracy by taking multiple intermediate derivative samples within a single time step.

• Mechanism: Calculates four weighted derivative estimates (k1, k2, k3, k4) between t and t + Δt.
• Pros: Fourth-order accuracy, meaning error per step is proportional to Δt⁵. Much more accurate than Euler methods for a given step size.
• Cons: Computationally expensive, requiring four derivative evaluations per step. Still conditionally stable.
• Use Case: High-accuracy simulations in scientific computing, orbital trajectory prediction, or any system where precision is more important than raw speed.

Symplectic Euler is a popular compromise method that updates velocity and position in a staggered, semi-implicit fashion. It is the workhorse of many real-time physics engines.

• Mechanism: v(t + Δt) = v(t) + a(t) * Δt then x(t + Δt) = x(t) + v(t + Δt) * Δt
• Pros: More stable than Explicit Euler, simple to implement, and symplectic for separable Hamiltonian systems, leading to good long-term energy behavior.
• Cons: Still only first-order accurate and conditionally stable.
• Use Case: The default integrator in many game physics engines (like Box2D) for simulating rigid body dynamics with good performance and acceptable stability.

The core challenge in time integration is managing stiffness—systems where different components evolve at vastly different rates (e.g., a stiff spring connected to a heavy mass).

• Explicit methods (Euler, RK4) have a stability limit: the time step Δt must be smaller than the system's fastest dynamics, or the simulation explodes.
• Implicit methods (Implicit Euler) are unconditionally stable: they can take large steps without blowing up, but at the cost of accuracy and computational solve steps.
• The Trade-Off: Engineers must choose between fast/accurate but potentially unstable (explicit) and stable but slow/damped (implicit). Adaptive and semi-implicit methods seek a middle ground.