What this does in effect is to trace out a polygon path that approximates the curve. Intuitively Euler’s method requires small step sizes to be accurate. It is simple to understand but is neither the most accurate nor the quickest method available.Baraff [1] details exactly why Euler’s method can fail and how to reduce the error by including more derivatives from the Taylor series. He also includes a more accurate ODE called the Midpoint method.Returning to the particle example, at any point in time the acceleration can be obtained by summing the forces to produce a net force acting on the particle at any one time, then dividing by the mass of the particle (Eq. 2.1). Equation 2.2 shows acceleration to be the derivative of velocity with respect to time. It is therefore possible to use the ODE solution to numerically integrate acceleration to obtain the velocity, as at any point in time the derivative of velocity (acceleration) is known. That just leaves the initial value of the velocity; this is simply the initial velocity i.e. the velocity at time t0. The same method can be used to go from velocity to position. Therefore, given a particle with mass, position, velocity and acceleration, it is possible by numerical integration using an ODE to obtain new values for velocity and position at future times.