EM photonics is proud to announce the release of our newest product, FastFDTD. This free toolkit enables accelerated electromagnetic simulations on your desktop.
FastFDTD is a free 2D and 3D accelerated FDTD solver based on GPU (graphics card) technology. The FastFDTD toolkit contains all files and documentation necessary to accelerate your FDTD computations using a simple input file format. No proprietary hardware is required to utilize the FastFDTD toolkit. Rather, individuals are free to use any supported graphics card of their choosing (the installer will verify that your graphics card is supported). The 2D and 3D solvers include a variety of sources and materials, with more constantly being added (see the Feature list [PDF] for more details).
Modern graphics cards are very powerful custom processors designed for the computationally intense rendering of 3D video games. However, these cards can be used to perform additional computations, such as those required by scientific computing algorithms. By understanding the computational pipelines present in graphics cards, we have been able to utilize these platforms to perform our own computations, at speeds 30-50x faster than standard software implementations.
Customers who currently use MATLAB, C/C++, FORTRAN, or other such programming languages for electromagnetic simulations will find that FastFDTD fits easily into their current workflows. Users simply export the data that already exists in their code to our simple file format. Customers who use a third-party simulation platform should contact their software vendor to inquire about built-in support for FastFDTD.
Using FastFDTD is a simple three-step process.
(1) Enter the simulation parameters using an XML input file. The input file is intuitive and thoroughly described with our Javadoc-style documentation. Further, the toolkit comes with numerous examples to get you started.
(2) Specify the computational region using a binary input file. Simply specify the material types at each node in the mesh. Again, this file format is described in detail in the documentation.
(3) Pass the input files to the FastFDTD executable. Your simulation will begin and the detector results will be produced automatically.
(1) Enter the simulation parameters using an XML input file. The input file is intuitive and thoroughly described with our Javadoc-style documentation. Further, the toolkit comes with numerous examples to get you started.
(2) Specify the computational region using a binary input file. Simply specify the material types at each node in the mesh. Again, this file format is described in detail in the documentation.
(3) Pass the input files to the FastFDTD executable. Your simulation will begin and the detector results will be produced automatically.
Using FastFDTD has been extensively tested on the NVIDIA GeForce 6 and 7 series graphics cards. Although it may work on other platforms, such as the NVIDIA Quadro line, it has not been fully tested on these platforms. Feel free to use the FastFDTD forums to request support for a specific graphics card or to report other graphics cards on which FastFDTD successfully executes.
