unfortunately, I fixed the bug in my JS and now it's the same in both browsers. Apparently, the problem was that there were some infinite (or NaN?) numbers produced. Firefox's canvas draws these, but Chromium's does not, and trashes whole path.
On Aug 20, 1:47 pm, Kasper Lund <[email protected]> wrote: > 2009/8/20 Dušan Maliarik <[email protected]>: > > > I'll do that and report back, thanks > > Perfect. Thanks a lot, Dušan! > > > > > On Thu, Aug 20, 2009 at 12:58, Mads Sig Ager <[email protected]> wrote: > > >> Since we know of no bugs in the floating point number processing it > >> would be great if you could narrow down the problem and produce a test > >> case where the output differ. Without that, we have very little to go > >> on. > > >> Thanks, -- Mads > > >> On Thu, Aug 20, 2009 at 12:36 PM, skrat<[email protected]> wrote: > > >> > Yes exactly, Chromium is much faster but it produces faulty results. > >> > There is bug in that geometry code, as you can see in the FF image, > >> > but regardless of that, same code should produce same results across > >> > browsers (IE not a browser) > > >> > On Aug 20, 12:28 pm, Kasper Lund <[email protected]> wrote: > >> >> On Thu, Aug 20, 2009 at 12:07 PM, skrat<[email protected]> > >> >> wrote: > >> >> > Before posting any code, I would like to know if there are any known > >> >> > differences in floating point number processing, eventually Math.* > >> >> > functions. That would make it clear why are the drawings different. > > >> >> I'm not aware of any intentional differences. I take it you're less > >> >> happy with the drawings produced by Chromium than the Firefox > >> >> equivalents? > > >> >> Cheers, > >> >> Kasper > > > -- > > Best regards, Dušan Maliarik // cell: +31681556139 --~--~---------~--~----~------------~-------~--~----~ v8-users mailing list [email protected] http://groups.google.com/group/v8-users -~----------~----~----~----~------~----~------~--~---
