On Wed, Nov 26, 2008 at 11:54 PM, the1geek <the1g
...@gmail.com> wrote:
> OK. Happy Emu Day, Pamela!
> Since v2.118, I applied elapsed time between the request of GMap API
> loader js and the firing of infowindowopen event on the frame-based
> test page below, and I've been monitoring the performance of v2.x and
> v2 against v2.118 (pre-modularization of GInfoWindow).
> http://genes.com/maps/gmap_test.htm
> I pay a lot of attention to the quickness of the bubble display
> because usually important info are contained in this, and speeding
> this up enhances user experience. And, it is very significant for some
> mobile devices especially those with javascript execution time
> limitation.
> The site will load test pages with different versions of the API
> loader into an iframe by clicking on the radio buttons. The actual
> version number and the elapsed time will be shown in the infowindow
> and recorded on the left panel.
> It also checks if the issues of content overflowing and the limitation
> of infowindow maxWidth have been fixed.
> I tested both fresh load and load from cache on IE7 and FF2 on a PC
> described below. For fresh load I made sure I cleared the browser
> cache before each load.
> The results below are in msec recorded from testing done within 1
> minute on
> WinXP
> 1.80GHz Pentium 4
> 1G RAM
> Fresh load (browser cache cleared)
> =============================
> IE7
> ===
> v2.x = v2.138f
> 6400 msec
> 5698 msec
> 5237 msec
> v2 = v2.137c
> 6489 msec
> 5328 msec
> 5899 msec
> v2.118 = v2.118
> 4076 msec
> 4066 msec
> 4987 msec
> FF2
> ===
> v2.x = v2.138f
> 2373 msec
> 2464 msec
> 2313 msec
> v2 = v2.137c
> 2584 msec
> 2383 msec
> 2604 msec
> v2.118 = v2.118
> 2343 msec
> 2174 msec
> 2473 msec
> Load from browser cache
> =======================
> IE7
> ===
> v2.x = v2.138f
> 3115 msec
> 2894 msec
> 2934 msec
> v2 = v2.137c
> 4387 msec
> 4186 msec
> 4186 msec
> v2.118 = v2.118
> 2214 msec
> 2194 msec
> 2574 msec
> FF2
> ===
> v2.x = v2.138f
> 2443 msec
> 2353 msec
> 2303 msec
> v2 = v2.137c
> 2584 msec
> 2364 msec
> 2454 msec
> v2.118 = v2.118
> 2894 msec
> 2624 msec
> 2594 msec
> In IE, 2.138f is about the same as 2.137c but slower than 2.118,
> specially with fresh load.
> Generally the same for all in FF and much quicker than in IE.
> ~newton
> On Nov 25, 2:21 pm, "pamela (Google Employee)" <pamela...@gmail.com>
> wrote:
>> Hey Newton-
>> We don't do turkeys in Australia. :)
>> There was a rollback for a day; it's all back to normal now. Sorry for any
>> inconvenience caused.
>> - pamela
>> On Wed, Nov 26, 2008 at 12:20 AM, the1geek <the1g...@gmail.com> wrote:
>> > Hey Pamela,
>> > I just checked. v2.x is still v2.137c and v2 = v2.136d.
>> > Happy Thanksgiving in NZ! I hope the turkey there is as big and
>> > succulent as in the US.
>> > We missed ya here.
>> > ~newton
>> > On Nov 23, 8:55 pm, "pamela (Google Employee)" <pamela....@gmail.com>
>> > wrote:
>> > > Hey folks-
>> > > We just pushed v2.138f as v2.x and pushed v2.137c to v2. We don't have
>> > > version aliasing implemented, so you'll need to specify the "c"/"f"
>> > > if you're hard-coding to those versions for whatever reason.
>> > > There were no bug fixes or features fulfilled in the newest version.
>> > > The full changelog is available here:
>> >http://code.google.com/p/gmaps-api-issues/wiki/JavascriptMapsAPIChang...
>> > > (If the groups interface cuts the link off, just find it in the wiki
>> > > tab of the issue tracker).
>> > > As always, please discuss anything in the forum that you think may be
>> > > a regression and report in the issue tracker if others confirm your
>> > > suspicions.
>> > > - pamela- Hide quoted text -
>> - Show quoted text -
