On Sat, Aug 28, 2010 at 14:08:48 -0600, Zooko O'Whielacronx wrote: > I just realized when glancing at this benchmark result: > > http://wiki.darcs.net/Benchmarks/TahoeLAFSOrg > > that darcs 2.4.98.1+62 takes 3 to 4 times as long as darcs 2.4.4 to > run "annotate" on the Tahoe-LAFS repository.
Are these the numbers you're referring to? 2.4 : 17.8s (5.8s) [ignore] 2.4.4 op : 18.0s (6.3s) 2.4.98.1 op : 30.7s (20.9s) The extreme variance on the 2.4.98.1 case makes me think that you may just have had an unfortunate outlier or two in your data, in which case false alarm. Any idea how to account for these sorts of outliers? You can look for them by doing some creative grepping on your ~/.darcs-benchmark directory. I'm hoping in the future, the criterion library will allow us to automatically detect such outliers (using its statistical black magic) so we don't get unnecessary scares like this. Thanks! -- Eric Kow <http://www.nltg.brighton.ac.uk/home/Eric.Kow> For a faster response, try +44 (0)1273 64 2905 or xmpp:ko...@jabber.fr (Jabber or Google Talk only) _______________________________________________ darcs-users mailing list darcs-users@darcs.net http://lists.osuosl.org/mailman/listinfo/darcs-users
