If you haven’t looked at our programming tools in a while, we now have improved runtime performance, an integrated debugger (breakpoints, variable evaluation) for graphical and Python languages as well as a smoother user interface.
List of features in this new release can be found here.
Much of the performance work we’ve done has been with regards to memory management. Python allows dynamic allocation of memory and has a garbage collector to free unused objects on the heap. Improved memory performance results in faster programs in general as less memory is used and the GC is required to run less often.
The execution speed of tight loops isn’t something we’ve explicitly addressed yet. Scenarios where you want encoder motors to reach particular targets are all handled natively in our framework so tight polling loops aren’t required for there. We may provide specific PID libraries (which run at native speeds internally) in the future.
Note, that by default, the Python debugger is enabled when building and downloading code which does come with a small speed performance cost. So this should be disabled when comparing benchmarks.