According to the error message Prime expects initial conditions for alpha and beta, even though there is no ODE for these functions
According the second approach and the m^2.
I am not sure if this is as it should be (a bug?) but we get the correct result if the initial value is just 0 or Zero (capital Z. no lower case).
I tend to call it a bug. As the derivative (D, tan) is dimensionless, the solution must be of the same dimension as x and the same applies to the initial value.
Limiting the vector x to values between 0m and 1m was necessary because its last value was very slightly (about 10^-16) larger than 1m and so the function y would fail.
@Werner_E wrote:
I am not sure if this is as it should be (a bug?) but we get the correct result if the initial value is just 0 or Zero (capital Z. no lower case).
I tend to call it a bug. As the derivative (D, tan) is dimensionless, the solution must be of the same dimension as x and the same applies to the initial value.
It is informative to try different numerical integrators.
Of course, one could always roll one's own integrator. Here's one I prepared earlier that runs (ran) under Mathcad Prime Express 7.
(I've got a programmed version somewhere of the Runge-Kutta and Adams-Moulton predictor-corrector families that I wrote for M11, but I can't find it at the moment)
