2 ms·
Nice work! Did you consider using an RK method with fewer stages but "better" properties? I'm thinking in particular of symplectic methods like velocity Verlet.
by semi-extrinsic 11y ago
Nice work! Did you consider using an RK method with fewer stages but "better" properties? I'm thinking in particular of symplectic methods like velocity Verlet. Depending on the details, it could give you both speedup and improved accuracy.
- flannelhead 11y agoThank you! A comparison between different integrators is in the plans. It would be certainly interesting to try e.g. velocity Verlet and some other symplectic integrators.