MathCastle
Differential EquationsHardNumerical Methods: Euler to Runge–Kutta

Numerical Methods: Euler to Runge–Kutta · Hard problem

The classical RK4 method (global error ≈Ch4\approx Ch^4) is run on 0≤t≤40 \le t \le 4 with step h=0.2h = 0.2, and the error at t=4t = 4 is 1.6×10−31.6 \times 10^{-3}. The error must be brought down to 1.6×10−71.6 \times 10^{-7} using equal steps. What is the smallest number of steps across [0,4][0, 4] that achieves this?