Are you having trouble with the patch? The benchmark used is included in it. The benchmark varies by the number of puts and gets, so 500 = 500 puts + 500 gets, so would have been suprised to see sub-linear time.
IE is the primary motivator for this set of changes, as it has the slowest JS system and is most sensitive to memory constraints, so the IE benchmarks are the ones I am focusing on. On Wed, Sep 10, 2008 at 6:47 PM, John Tamplin <[EMAIL PROTECTED]> wrote: > On Wed, Sep 10, 2008 at 6:29 PM, Emily Crutcher <[EMAIL PROTECTED]> wrote: > >> The benchmarks are somewhat interesting, as it turns out that memory >> allocation from one can serious screw up the others. Therefore here I've >> included each variant as separate benchmarks posted on the JsMap >> wiki<http://code.google.com/p/google-web-toolkit-incubator/wiki/JsMap?updated=JsMap&ts=1221085506>page. >> > > I haven't looked at the code yet, but what about non-IE benchmarks? Also, > it looks like these results show linear time, and don't specify exactly what > is being benchmarked -- unless you are including iteration in there, I would > expect any hashmap benchmarks to be sub-linear. > > -- > John A. Tamplin > Software Engineer (GWT), Google > -- "There are only 10 types of people in the world: Those who understand binary, and those who don't" --~--~---------~--~----~------------~-------~--~----~ http://groups.google.com/group/Google-Web-Toolkit-Contributors -~----------~----~----~----~------~----~------~--~---
