Hi Zheng, Thank you for your response. It is really faster than using fields::total_energy(). But am I doing it right ? To calculate the total field energy in every step I do :
double ex = fields->field_energy_in_box(meep::Ex, volume.surroundings()); double ey = fields->field_energy_in_box(meep::Ey, volume.surroundings()); double ez = fields->field_energy_in_box(meep::Ez, volume.surroundings()); double hx = fields->field_energy_in_box(meep::Hx, volume.surroundings()); double hy = fields->field_energy_in_box(meep::Hy, volume.surroundings()); double hz = fields->field_energy_in_box(meep::Hz, volume.surroundings()); double e=ex+ey+ez+hx+hy+hz; Is it optimal ? It is still not as fast as the algorithm in Fortran I mentioned earlier. Thank you, Thomas Jefferson 2008/7/28 Zheng Li <[EMAIL PROTECTED]>: > Jefferson Thomas wrote: >> >> Hi, >> >> I want to calculate the total energy in my simulation for every step. >> I am using the function fields::total_energy() >> The problem is that it runs veeery slowly. It takes several times more >> time than the actual step takes ! >> Is there any other way to calculate the energy faster ? It would be very >> useful. >> I know it is possible because I am using paralelly with meep a >> photonic crystal simulation written in Fortran >> and there the calculation of energy is a side product of the actual >> FDTD calculation so the energy >> data is simply available for every step without any additional >> overhead. Is there something like that in meep ? >> >> Thank you for any help, >> Thomas Jefferson >> >> _______________________________________________ >> meep-discuss mailing list >> [email protected] >> http://ab-initio.mit.edu/cgi-bin/mailman/listinfo/meep-discuss >> >> >> > > Hi, > If you use C++ interface and want to calculate the energy in each step, then > I think you should circumvent fields::total_energy. Due to > the different evaluating time and spacial points between the H and E fields > in the > Yee lattice, when you call the total_energy() function, meep should backup > the fields > , run a half step for H, average the H energy to get the right one and > restore the fields again, > which is obviously unsuitable for your case. You can just record the H and > E energy in each step > with > double fields::field_energy_in_box(component c, > const geometric_volume &where); > and average them by yourself. > > Zheng > 2008-728 > _______________________________________________ meep-discuss mailing list [email protected] http://ab-initio.mit.edu/cgi-bin/mailman/listinfo/meep-discuss
