The current iteration of the code was written using some features of C++17, and was compiled using CUDA-10.0 and CUDA 12, with a relatively modern version of cmake
Multirank simulations tested with openMPI 4.0.0 (sudo apt-get install libopenmpi-dev, e.g.)
The GUI needs Qt6 (e.g., apt-get install qt6-base-dev) and also openGL.
A flag during the cmake phase can be used to not compile the GUI (particularly useful when you just want to run the command-line code on a cluster), and another flag can be used to build the entire project without any of the GPU / CUDA stuff (particularly useful, e.g., on a mac now that Apple has stopped supporting cuda). See below for those options.
Basic compilation (with CUDA support, and with the graphical user interface also being compiled):
$ cd build/
$ cmake -DENABLE_QT=ON ..
$ make
$ cd build/
$ cmake ..
$ make
$ cd build/
$ cmake -DDISABLE_CUDA=ON ..
$ make
(And of course, the cuda and QT flags are independent, so cmake -DDISABLE_CUDA=ON -DENABLE_QT=ON .. will work as expected).
Please note that on modern macOS, you might have to explicitly specify the location of the mpi cxx compiler, along the lines of
$ cmake -DCMAKE_CXX_COMPILER=/opt/homebrew/bin/mpicxx -DDISABLE_CUDA=ON ..
Specifying the MacOSX15 SDK via the SDKROOT environment variable (and then setting the CMAKE_CXX_COMPILER to clang++) may also work.
By default, the above steps will create two executables, "openQmin.out" and "openQminGUI.out", in the build directory. The GUI executable will... launch the graphical user interface. The non-GUI executable is meant to be run with various command line parameters.
Are you wondering about the mysterious third executable, "customScriptFromGUI.out", that also gets made? We currently have an experimental (but functional) feature by which you can record sequences of actions in the GUI and save them to a new, compilable "customScriptFromGUI.cpp" file. This can be used, for instance, to specify specific initial conditions more easily than by fussing with the command line or writing your own cpp codes. Note that the "customScriptFromGUI.out" executable itself has command line options (such as changing the lattice size), and is suitable to be run as an MPI executable.
To make additional executables on compilation, copy a cpp file into the base directory, and then add the name of the cpp file to the base CMakeList.txt file in the "foreach()" line immediately following the comment that says "list the names of cpp files cooresponding to linked executables you'd like..."
The requirements can be obtained by looking at the info on the following:
CUDA: https://developer.nvidia.com/cuda-downloads
CMAKE: https://cmake.org