[Python-Dev] peps: PEP 432 updates in response to initial comments (original) (raw)
Antoine Pitrou solipsis at pitrou.net
Fri Dec 28 22:59:50 CET 2012
- Previous message: [Python-Dev] PYTHONPATH processing change from 2.6 to 2.7 and Mac bundle builder problems
- Next message: [Python-Dev] question about packaging
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
On Fri, 28 Dec 2012 14:54:55 -0500 Terry Reedy <tjreedy at udel.edu> wrote:
4. Running ./pythond from within a PCBuild/pythond interactive window and on a regular disk averages .10 seconds. The slowdown is probably a mixture of disk access and extra debug code, but is not bad. There is no flashing (probably because there already is a window, whereas on Windows IDLE runs code within a windowless pythonw process) and ^C works. This is definitely a better environment for this type of test ;-).
You'd get more meaningful numbers by using a non-debug build (PCBuild/python.exe, I guess). Our debugging additions + the lack of compiler optimizations butcher performance.
It would be extra nice if you had numbers comparing 3.3, 3.2 and 2.7 (under Windows, that is).
Regards
Antoine.
- Previous message: [Python-Dev] PYTHONPATH processing change from 2.6 to 2.7 and Mac bundle builder problems
- Next message: [Python-Dev] question about packaging
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]