fdtd software

I think there are so many FDTD codes because they are so easy to write Truthfully, I think the reason is simply that no one can bring out a FEM code that competes with HFSS - they have been devolping that product since the dawn of time, and it will take bill gates himself to fund the development of a competitor.

In my opinion the best FDTD is CST MWS. Their PBA is the perfect way to solve the problems with curved structures, and their CAD is very, very nice.

I think that MoM is better for large problems or when distances between objects gets large, because the field is not discretized. With FEM/FDTD/FIT a finite (but small) error is made each cell, so if you have millions of air cells next to each other you get significant numerical dispersion. Numerical dispersion is inherent, but can be minimized, but the bigger your structure gets the worse it becomes.
With the advent of solution techniques like the MLFMM, VERY large structures can be solved with MoM in a reasonable amount of memory. The only commercial code that I am aware of that has MLFMM built in is FEKO. At our company we recently used it to simulate car structures at cellphone frequencies - we could solve a full sized car at 1800MHz well inside 2 GB RAM and in only a few hours!
If you are solving small structures with very complicated materials and feeds, then the FDTD or FEM algorithms come into their own. If you are doing UWB, then what choice other than FDTD do you really have?!?
Each technique has its place and that is why each one has a big name busy driving it.

Why commercial? There is a reason that those other packages are free... It costs far more in designers time to use free software than it does to just buy stuff that you know works.

FDTD is a general purpose numerical method and is very powerful. FEM generally is more resource memory expensive.

Regards

hi guys,i think that FDTD IS one of most em simulator such that fidelity under zeland and XFDTD ARE USED THAT SIMULATION

Although many researchers have improve the perfance of the FDTD by ADI or CN-FDTD, but the best FDTD method always the Yee's conventional method. Unless the Grid is an great modification, the simulation of EM will not take advance. I think.

Dear guys,
I think the best simulators are:
1. IE3D 10+, for Micrsotrip circuits and antennas, its a MoM code,
2. HFSS agilent 4+, for 3d structures, its a FEM code,
3. Fiedelity 4+, for 3D stuctures, it is a FDTD code.

I used IE3D for more than a thousand simulations, its acuracy for microstrip antenna and circuits with arbitrary shape and dielectric type and thickness, is unquestionable.

the other two, I used them to simulate Multisegment Rectangular Dielectric Resonator Antennas, they are in very good agreement, but HFSS needs about 10 times more time! they are also in good agreement with experiments.

anyway, if someone really needs this softwares, be my guest and contact me.

urs, mamadreza.

hi,

can anyone share C/C++ codes of CN-FDTD? I need it badly...

BR // Obhro

obhro23@hotmail.com