The ode integrator functions have tolerance arguments that you can use.
Yeah, but just keep the tolerance low enough so that you don’t have to worry about this stuff :D.
@maxbiostat working through doing some custom C++ code that uses integrals and such in a model over here: Problem implementing Zipf (power law) distribution -- external C++ . You might be able to use that here.
It’d probably be faster than the ode integrator for this. This boost integral code is the same stuff that’s being used in the yet unfinished 1d integrator stuff.
Edited for clarity… Twice!