On Thu, Sep 29, 2011 at 1:05 AM, Juan Pablo Carbajal
<[email protected]> wrote:
> Hi all,
>
> I have implemented a function that calculates one iteration of the
> velocity Verlet algorithm (with velocity tracking)[1] as an m-file and
> in C++. This algorithm is basic for the computation of molecular
> dynamics and belong to the class of symplectic integrators [2], in
> contrast to the standard Runge-Kutta methods.
>
> The C++ code is here http://agora.panocha.org.mx/snippet/LDw1/
> The m-file is here http://agora.panocha.org.mx/snippet/gG1h/
>
> Using tic toc, the C++ code runs ~16 times faster.
>
> However, either code can be improved (the two codes are not exactly
> equivalent, C++ takes the interaction function in any format, while
> m-file only handles). Do you have any suggestions?
>
> I will be adding this code to the Mechanics package.
>
> Thanks
>
> Note:
> The global error of the algorithm of order two in the time-step for
> velocities and order four for positions. However, if the system
> integrated is Hamiltonian the energy of the approximation oscillates
> around the constant energy of the exactly solved system, with a global
> error bound of order two. The same goes for all other conserved
> quantities of the system like linear or angular momentum, that are
> always preserved or nearly preserved in a symplectic integrator.
>
> [1] http://en.wikipedia.org/wiki/Verlet_integration#Velocity_Verlet
> [2] http://en.wikipedia.org/wiki/Symplectic_integrator
>
> --
> M. Sc. Juan Pablo Carbajal
> -----
> PhD Student
> University of Zürich
> http://ailab.ifi.uzh.ch/carbajal/
>

BTW, the function assumes the action-reaction principle is not
necessarily valid. The interaction function should return F(x1,x2) AND
F(x2,x1) as row vectors.



-- 
M. Sc. Juan Pablo Carbajal
-----
PhD Student
University of Zürich
http://ailab.ifi.uzh.ch/carbajal/

------------------------------------------------------------------------------
All the data continuously generated in your IT infrastructure contains a
definitive record of customers, application performance, security
threats, fraudulent activity and more. Splunk takes this data and makes
sense of it. Business sense. IT sense. Common sense.
http://p.sf.net/sfu/splunk-d2dcopy1
_______________________________________________
Octave-dev mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/octave-dev

Reply via email to