I have a preliminary version of quick entry performance test checked in. I tested it on my machine, and some Tinderboxes now have numbers as well. In short, quick entry is 47% faster the first time and 63% faster subsequent times compared to creating an event by double clicking. This translates to something like 1.14 s first time and .90 seconds subsequent times for quick entry on the slowest reference platform.
One major reason why it is faster is because we don't select the event and display it in the detail view, like we do when creating an event by double clicking. The profile also shows that this case will get faster as we make new event creation and triage faster. Other cases may contribute improvements as well. So it seems to me this test would be unnecessary, and just slow the Tinderbox cycle time even further without adding much new information. I therefore plan to disable this test and take it off from the 0.7 primary performance case list unless I hear objections. -- Heikki Toivonen
signature.asc
Description: OpenPGP digital signature
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ Open Source Applications Foundation "chandler-dev" mailing list http://lists.osafoundation.org/mailman/listinfo/chandler-dev
