Ariel, Thanks for the suggestion, after removing an element I found a bug in the implementation... ;-)
I'll try a correction later, so that I can post the benchmark for it... Diogo On Tue, Dec 2, 2008 at 9:09 PM, Ariel Flesler <[EMAIL PROTECTED]> wrote: > > Yeah... I suppose we'll go with the bifurcated then. > > Last request, benchmark a situation where the element isn't included > into the document. > > Thanks :) > > On Tue, Dec 2, 2008 at 7:39 PM, Diogo Baeder <[EMAIL PROTECTED]> > wrote: > > Alright, jqbench updated... now, traversing 10 levels of divs: > > > > Using parentNode: 264ms; > > Using the bifurcated code: 89ms. > > > > Now, with more levels, it starts to be a big difference... what do you > think > > of it? > > > > Diogo > > > > > > On Tue, Dec 2, 2008 at 5:40 PM, Ariel Flesler <[EMAIL PROTECTED]> > wrote: > >> > >> Ok... but this is not something we'll call too often... > >> > >> Did you try with different depths ? that is, nodes that are more and > >> less nested into the dom. > >> > >> That should change the numbers, probably making the bifurcated one a > >> little faster (when deeper). > >> > >> On Tue, Dec 2, 2008 at 3:46 PM, Diogo Baeder <[EMAIL PROTECTED]> > >> wrote: > >> > OK, let's talk about numbers - 1000 calls to the methods, in my > >> > benchmark: > >> > > >> > Using parentNode: 156ms; > >> > Using the bifurcated code: 99ms. > >> > > >> > Diogo > >> > > >> > > >> > > >> > On Tue, Dec 2, 2008 at 4:26 PM, Ariel Flesler <[EMAIL PROTECTED]> > >> > wrote: > >> >> > >> >> If the bifurcated one isn't CLEARLY faster, then +1 for Diego's. > >> >> We don't tend to do that and it's ugly code IMO (the bifurcated one > >> >> that > >> >> is) > >> >> > >> >> On Tue, Dec 2, 2008 at 2:17 PM, John Resig <[EMAIL PROTECTED]> > wrote: > >> >> > > >> >> >> To be honest, John, I haven't tested this implementation using > >> >> >> frames... if > >> >> >> you say yes, it works, then I think we could stick with it... but > >> >> >> wouldn't > >> >> >> it be a little (just a little) faster to get "doc" outside the > >> >> >> method > >> >> >> call, > >> >> >> in an outter scope, to maximize performance? It seems to me that > it > >> >> >> was > >> >> >> one > >> >> >> of the points that lowered 30% the time for processing the call, > in > >> >> >> my > >> >> >> benchmarks... > >> >> > > >> >> > Well, the problem is that if we move doc to an outside scope it'll > >> >> > either A) Cause some code to break (outer documents, etc.) or B) > >> >> > Require us to do a conditional to determine which document to use - > >> >> > at > >> >> > which point we're worse off then we are now. > >> >> > > >> >> > IMO I would go with whichever one is faster - the one that I > proposed > >> >> > or Diego's (since they work equally across all documents). > >> >> > > >> >> > --John > >> >> > > >> >> > > > >> >> > > >> >> > >> >> > >> >> > >> >> -- > >> >> Ariel Flesler > >> >> http://flesler.blogspot.com > >> >> > >> >> > >> > > >> > > >> > > >> > -- > >> > Diogo Baeder > >> > http://www.diogobaeder.com.br > >> > > >> > > > >> > > >> > >> > >> > >> -- > >> Ariel Flesler > >> http://flesler.blogspot.com > >> > >> > > > > > > > > -- > > Diogo Baeder > > http://www.diogobaeder.com.br > > > > > > > > > > > -- > Ariel Flesler > http://flesler.blogspot.com > > > > -- Diogo Baeder http://www.diogobaeder.com.br --~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "jQuery Development" group. To post to this group, send email to [email protected] To unsubscribe from this group, send email to [EMAIL PROTECTED] For more options, visit this group at http://groups.google.com/group/jquery-dev?hl=en -~----------~----~----~----~------~----~------~--~---
