Author collinwinter
Recipients collinwinter, eric.smith, gawain, gregory.p.smith, haypo, mark.dickinson
Date 2009-09-10.14:20:51
SpamBayes Score 5.00829e-05
Marked as misclassified No
Message-id <1252592454.08.0.130043182744.issue6713@psf.upfronthosting.co.za>
In-reply-to
Content
I ran this patch against Unladen Swallow's slowspitfire template 
benchmark, which does more int->string conversions than any of our other 
benchmarks. When run against Python trunk r74737, I get these results:

slowspitfire:
Min: 0.888772 -> 0.867427: 2.46% faster
Avg: 0.891857 -> 0.872461: 2.22% faster
Significant (t=45.532127, a=0.95)

(./perf.py -r -b slowspitfire ../a/python.exe ../b/python.exe)
This was an idle MacBook Pro, OS X 10.5.8, Apple gcc 4.0.1, 2.4 GHz Core 
Duo.

Other benchmarks benefit, but are only barely statistically significant.
History
Date User Action Args
2009-09-10 14:20:55collinwintersetrecipients: + collinwinter, gregory.p.smith, mark.dickinson, haypo, eric.smith, gawain
2009-09-10 14:20:54collinwintersetmessageid: <1252592454.08.0.130043182744.issue6713@psf.upfronthosting.co.za>
2009-09-10 14:20:52collinwinterlinkissue6713 messages
2009-09-10 14:20:51collinwintercreate