Django is remarkably faster on Pypy than CPython: http://t.co/QKpHr2Tb
miniblog.
Related Posts
Wow, Pypy now supports numpy and pandas! https://morepypy.blogspot.co.uk/2017/03/pypy27-and-pypy35-v57-two-in-one-release.html (a major reason for using CPython previously)
Source-level coverage metrics are really hard in the presence of a JIT. Here's pypy's solution: https://morepypy.blogspot.co.uk/2013/10/making-coveragepy-faster-under-pypy.html (cf -Og in gcc)
Wow, pypy now has a reverse debugger! https://morepypy.blogspot.co.uk/2016/07/reverse-debugging-for-python.html (also a nice demo of how RPython facilitates experiments)