I have written applications both in Qt and PyQt and found out that in the real world, Pythons performance is more than apt for creating a UI.
Even more so, the PyQt application had a NumPy backend and again, it performed flawlessly. The Qt application had a C++ backend that did similar computations as the Numpy backend, but with a much more painful programming process.
So in that line of thinking, I am a big proponent of having a dynamic language as frontend for low level libraries. Both Qt and the Numpy core are plenty fast for what they need to do and having a dynamic language like Python as a frontend makes programming very convenient. The best of both worlds, really.
I have never experienced Python being the bottleneck of my algorithms using these libraries.
Comments
I have written applications both in Qt and PyQt and found out that in the real world, Pythons performance is more than apt for creating a UI.
Even more so, the PyQt application had a NumPy backend and again, it performed flawlessly. The Qt application had a C++ backend that did similar computations as the Numpy backend, but with a much more painful programming process.
So in that line of thinking, I am a big proponent of having a dynamic language as frontend for low level libraries. Both Qt and the Numpy core are plenty fast for what they need to do and having a dynamic language like Python as a frontend makes programming very convenient. The best of both worlds, really.
I have never experienced Python being the bottleneck of my algorithms using these libraries.