Hello Bruno, Jay, (to mark this email as "responded")

Jay brought this up in today's ET call:

http://lists.einsteintoolkit.org/pipermail/users/2020-November/007670.html

and Peter, Zach, I responded:

--8<--
Jay Kalinani: TOV simulation using new Spritz code. When adding a dipolar
magnetic field to the star, there is some confusion about the behavior of
cctk_time. There's a desire to not store RK intermediate values of
cctk_time to cctk_time itself.
Zach suggested that a local variable could be defined that simply outputs
cctk_iteration*dt if desired.
Roland had some comments about scheduling such a routine.
--8<--

basically: "this is as designed in case the RHS in has an explicit
dependency on time".

Yours,
Roland

> Hi Erik,
>   those should be internal steps of the Runge-Kutta integrator when going
> from t to t+dt. Why is the cctk_time changed in those intermediate steps?
> Shouldn't cctk_time be set only at each iteration in time?
>
> Thanks,
> Bruno
>
>
> Il giorno mer 11 nov 2020 alle ore 15:11 Erik Schnetter <
> schnetter@cct.lsu.edu> ha scritto: 
>
> > Jay
> >
> > This intermediate step comes from the time integrator. The Einstein
> > Toolkit usually uses Runge-Kutta type time integrators that perform
> > multiple sub-steps for each time step. These time integrators are
> > provided by the "Methods of Lines" thorn ("MoL"). The respective
> > documentation should have further pointers, and the respective
> > parameter settings should also tell you more. (See also e.g.
> > <https://en.wikipedia.org/wiki/Runge–Kutta_methods>).
> >
> > -erik
> >
> > On Tue, Nov 10, 2020 at 12:20 PM Jay Vijay Kalinani
> > <jayvijay.kalinani@phd.unipd.it> wrote: 
> > >
> > > Dear all,
> > >
> > > I am trying to perform a magnetised TOV simulation using Einstein 
> > Toolkit and Spritz GRMHD code. In particular, currently I have a subroutine
> > that adds a dipolar magnetic field to the TOV star after its initial data
> > has been generated via TOVSolver thorn. 
> > > Within this subroutine, I have added the following lines that print the 
> > 'cctk_time' value as: 
> > >
> > > character(len=200) warnline
> > > WRITE( warnline, '( "cctk_time: ", g16.7)' ) cctk_time
> > > call CCTK_INFO(warnline)
> > >
> > > This subroutine is scheduled in HydroBase_Con2Prim after Con2Prim step. 
> > And I get the following lines in the .out file as: 
> > > .........
> > >
> > > INFO (TOVSolver): Integrated TOV equation
> > >
> > > INFO (TOVSolver): Information about the TOVs used:
> > >
> > > INFO (): TOV    radius    mass  bary_mass mass(g) cent.rho rho(cgi) 
> >   K   K(cgi)    Gamma 
> > >
> > > INFO ():   1   8.12502  1.40016  1.50618 2.78e+33  0.00128 7.92e+14 
> > 100 1.45e+05        2 
> > >
> > > INFO (TOVSolver): Not using old matter initial data
> > >
> > > INFO (TOVSolver): Done interpolation.
> > >
> > > INFO (Spritz_DipoleAndAtmo): cctk_time:     0.000000
> > >
> > > INFO (Spritz_DipoleAndAtmo): Adding a poloidal magnetic field to the TOV 
> > star on reflevel 0 
> > >
> > > INFO (Spritz_DipoleAndAtmo): Adding atmo to the TOV star on reflevel = 
> > '0' 
> > >
> > > INFO (Spritz_DipoleAndAtmo): cctk_time:     0.000000
> > >
> > > 
> > ----------------------------------------------------------------------------------------------------------------------------------- 
> > >
> > > Iteration      Time |              ADMBASE::alp | 
> > HYDROBASE::rho |          HYDROBASE::press |            SPRITZ::B_norm 
> > >
> > >                     |      minimum      maximum |      minimum 
> > maximum |      minimum      maximum |      minimum      maximum 
> > >
> > > 
> > ----------------------------------------------------------------------------------------------------------------------------------- 
> > >
> > >         0     0.000 |    0.6698612    0.9603815 | 5.827798e-11 
> > 0.0012800 |    0.0000000    0.0001638 | 8.399814e-09    0.0000399 
> > >
> > > INFO (Spritz_DipoleAndAtmo): cctk_time:    0.3906250E-01
> > >
> > > INFO (Spritz_DipoleAndAtmo): cctk_time:    0.3906250E-01
> > >
> > > INFO (Spritz_DipoleAndAtmo): cctk_time:    0.7812500E-01
> > >
> > > INFO (Spritz_DipoleAndAtmo): cctk_time:    0.7812500E-01
> > >
> > >         1     0.078 |    0.6698612    0.9603833 | 5.827814e-11 
> > 0.0012799 |    0.0000000    0.0001638 | 8.294987e-09    0.0000395 
> > >
> > > INFO (Spritz_DipoleAndAtmo): cctk_time:    0.1171875
> > >
> > > INFO (Spritz_DipoleAndAtmo): cctk_time:    0.1171875
> > >
> > > INFO (Spritz_DipoleAndAtmo): cctk_time:    0.1562500
> > >
> > > INFO (Spritz_DipoleAndAtmo): cctk_time:    0.1562500
> > >
> > >         2     0.156 |    0.6698612    0.9603850 | 5.827829e-11 
> > 0.0012797 |    0.0000000    0.0001638 | 8.294994e-09    0.0000390 
> > >
> > > ..........
> > >
> > >
> > >
> > >
> > > The simulation is performed in uniform grid with dx=dy=dz=0.3125, 
> > time::dtfac = 0.25 and HydroBase::timelevels = 2. 
> > > As you may see from the output lines, in between iteration 0 which is at 
> > cctk_time=0.0 and iteration 1 which is at cctk_time=0.078125, there is an
> > intermediate time step with 'cctk_time=0.0390625' when the subroutine is
> > called. 
> > >
> > > I was hoping to understand better why this intermediate time step exists 
> > between two consecutive iterations, and if there is any documentation I
> > could look at. 
> > > Any help would be much appreciated.
> > >
> > > Thank you very much.
> > > Sincerely,
> > > Jay Kalinani
> > > _______________________________________________
> > > Users mailing list
> > > Users@einsteintoolkit.org
> > > http://lists.einsteintoolkit.org/mailman/listinfo/users 
> >
> >
> >
> > --
> > Erik Schnetter <schnetter@cct.lsu.edu>
> > http://www.perimeterinstitute.ca/personal/eschnetter/
> > _______________________________________________
> > Users mailing list
> > Users@einsteintoolkit.org
> > http://lists.einsteintoolkit.org/mailman/listinfo/users
> > 
>
>


--
My email is as private as my paper mail. I therefore support encrypting
and signing email messages. Get my PGP key from http://pgp.mit.edu .