Hi, all,
I am using Nedelec elements to solve the Maxwell equations, which is a complex-valued problem. When search the previous works I found that Ross Kynch was using fe.system_to_block_index() to splitt the system into real/imaginary parts, while QinCe was using fe.system_to_base_index().
Any differences between these two methods? Does either one outperform the other?
Thanks in advance.
Jianan Zhang