Options:
- # Session Start: Sat Feb 28 00:00:00 2009
- # Session Ident: #html-wg
- # [00:16] * Quits: aroben (aroben@71.58.77.15) (Connection reset by peer)
- # [00:24] * Quits: heycam (cam@118.138.224.6) (Connection reset by peer)
- # [00:24] * Joins: heycam (cam@118.138.224.10)
- # [00:32] * Quits: tH (Rob@83.100.255.15) (Quit: ChatZilla 0.9.84-rdmsoft [XULRunner 1.9.0.1/2008072406])
- # [00:42] * Quits: Sander (svl@86.87.68.167) (Quit: And back he spurred like a madman, shrieking a curse to the sky.)
- # [00:50] * Joins: dbaron (dbaron@222.151.83.100)
- # [01:43] * Quits: hober (ted@206.212.254.2) (Connection reset by peer)
- # [02:16] * Quits: dbaron (dbaron@222.151.83.100) (Quit: 8403864 bytes have been tenured, next gc will be global.)
- # [03:21] * Quits: heycam (cam@118.138.224.10) (Quit: bye)
- # [03:22] * Quits: adele (adele@17.246.17.16) (Quit: adele)
- # [03:30] <pimpbot> bugmail: "[Bug 6610] add a preventable forced-fragment method" ( message in thread) <http://lists.w3.org/Archives/Public/public-html-bugzilla/2009Feb/0052.html> ** "[Bug 6630] New: Optionalness of start tags when element is empty" (1 message in thread) <http://lists.w3.org/Archives/Public/public-html-bugzilla/2009Feb/0051.html>
- # [03:36] * Joins: heycam (cam@124.168.12.212)
- # [03:51] * Quits: Dashiva (noone@80.202.223.46) (Quit: Dashiva)
- # [03:53] * Joins: Dashiva (noone@80.202.223.46)
- # [04:01] * Quits: ChrisWilson (cwilso@131.107.0.103) (Ping timeout)
- # [04:05] * Quits: gavin_ (gavin@99.226.207.11) (Ping timeout)
- # [04:10] * Joins: gavin_ (gavin@99.226.207.11)
- # [04:37] * Quits: gavin_ (gavin@99.226.207.11) (Connection reset by peer)
- # [06:01] <pimpbot> bugmail: "[Bug 6610] add a preventable forced-fragment method" (2 messages in thread) <http://lists.w3.org/Archives/Public/public-html-bugzilla/2009Feb/0053.html>
- # [06:05] * Quits: Dashiva (noone@80.202.223.46) (Quit: Dashiva)
- # [07:53] * Joins: annevk (opera@219.55.240.218)
- # [09:02] * Joins: Sander (svl@86.87.68.167)
- # [09:20] * Joins: dbaron (dbaron@222.151.83.100)
- # [10:08] <pimpbot> bugmail: "[Bug 6637] New: Put boolean and enumerated attributes next to each other" (1 message in thread) <http://lists.w3.org/Archives/Public/public-html-bugzilla/2009Feb/0054.html>
- # [10:35] * Quits: dbaron (dbaron@222.151.83.100) (Quit: 8403864 bytes have been tenured, next gc will be global.)
- # [10:52] * Joins: ROBOd (robod@89.122.216.38)
- # [12:14] * Joins: gavin_ (gavin@99.226.207.11)
- # [12:14] * Joins: myakura (myakura@123.225.72.182)
- # [13:13] * Quits: jmb (jmb@152.78.68.189) (Ping timeout)
- # [13:14] * Joins: dbaron (dbaron@222.151.83.100)
- # [13:16] * Joins: jmb (jmb@152.78.68.189)
- # [13:22] * Quits: dbaron (dbaron@222.151.83.100) (Quit: 8403864 bytes have been tenured, next gc will be global.)
- # [14:03] * Joins: maddiin (mc@87.185.202.161)
- # [14:12] * Quits: jmb (jmb@152.78.68.189) (Ping timeout)
- # [14:15] * Joins: jmb (jmb@152.78.68.189)
- # [14:25] * Joins: tH (Rob@83.100.255.15)
- # [14:37] * Quits: gavin_ (gavin@99.226.207.11) (Ping timeout)
- # [14:42] * Joins: gavin_ (gavin@99.226.207.11)
- # [14:57] * Quits: maddiin (mc@87.185.202.161) (Quit: maddiin)
- # [15:04] * Joins: maddiin (mc@87.185.202.161)
- # [15:10] * Quits: jmb (jmb@152.78.68.189) (Ping timeout)
- # [15:13] * Joins: jmb (jmb@152.78.68.189)
- # [17:27] * Joins: Dashiva (noone@80.202.223.46)
- # [19:41] * Parts: asbjornu (asbjorn@84.48.117.191)
- # [19:52] * Quits: maddiin (mc@87.185.202.161) (Quit: maddiin)
- # [20:49] * Joins: gavitron (gavitron@204.248.57.243)
- # [21:01] <gavitron> anyone know where there is information regarding <canvas> performance? In particular my code uses the canvas api to generate skinable information dialog popups. The skin is drawn using alpha-transparent png's. During the life of the widget that uses canvas, it is rotated and manipulated and I'm trying to figure out if there are any tricks to stop the manipulation from killing my cpu.
- # [21:02] <gavitron> the shape uses 4 bezier curves and a few straight lines, over there are 2 of those shapes being drawn for each widget, if that helps..
- # [21:03] <gavitron> *over all there are
- # [21:15] * Joins: gavitron_ (gavitron@98.226.63.24)
- # [21:17] * Quits: gavitron (gavitron@204.248.57.243) (Ping timeout)
- # [21:24] <Philip> gavitron_: I'm not aware of any significant documentation on canvas performance
- # [21:24] <Philip> gavitron_: It's going to vary hugely between different implementations, too
- # [21:25] <Philip> gavitron_: so probably the best information will come from just experimenting yourself
- # [21:25] <Philip> gavitron_: (e.g. disable the curve-drawing code and see if it's any faster)
- # [22:15] * Joins: asbjornu (asbjorn@84.48.117.191)
- # [22:46] <pimpbot> planet: Offered Without Comment <http://blog.mozilla.com/rob-sayre/2009/02/28/offered-without-comment-3/>
- # [22:49] * Quits: ROBOd (robod@89.122.216.38) (Quit: http://www.robodesign.ro )
- # [23:27] <gavitron_> Philip: okay thanks for the advice. Maybe i'll make some performance tests of my own.
- # [23:27] <gavitron_> I did happen to find an article comparing SVN performance to canvas performance tho: http://www.borismus.com/canvas-vs-svg-performance/
- # [23:27] <pimpbot> Title: Performance of Canvas versus SVG | Musings of B (at www.borismus.com)
- # [23:30] * Quits: myakura (myakura@123.225.72.182) (Quit: Leaving...)
- # [23:45] <Philip> gavitron_: That article probably isn't too useful unless you're trying to draw a lot of circles, I guess
- # [23:46] <Philip> I spent some time optimising http://canvex.lazyilluminati.com/ a while ago and it seemed the most expensive thing was just drawing images
- # [23:46] <pimpbot> Title: Canvex - canvas-based FPS game (at canvex.lazyilluminati.com)
- # [23:46] <Philip> (so I tried hard to minimise the number of drawImage calls, which helped a lot)
- # [23:47] <Philip> but then there were lots of weird browser-specific things, like some browsers were much happier with drawImage(canvas, ...) than drawImage(image, ...) and so it helped to copy all the images onto canvases at startup
- # [23:48] <gavitron_> i see
- # [23:48] <gavitron_> i wonder if i can use straight CSS for that then..
- # [23:50] <gavitron_> Philip: canvex is pretty sweet
- # [23:50] <Philip> CSS won't necessarily be faster, since the browser still has to draw the images somewhere
- # [23:50] <Philip> but it might be worth testing
- # [23:51] <gavitron_> yeah I don't know. It seems that VML might have trouble using rotated alpha-transparent PNG's anyway, so I may have to take the skinning portion out of the canvas element anyway
- # [23:51] <gavitron_> I'm using excanvas for IE
- # [23:51] * Joins: rubys1 (rubys@75.182.92.38)
- # [23:52] <Philip> Ah, I know very little about how VML works since I've never cared about IE compatibility
- # [23:52] <Philip> Performance of excanvas is likely to be hugely worse than performance of real canvas implementations :-(
- # [23:52] <gavitron_> yes, which makes me very sad :(
- # [23:54] <gavitron_> i'm trying to get this thing supported in a-grade browsers in time to make our app's next release, so i'll be doing a lot of tinkering i guess.
- # Session Close: Sun Mar 01 00:00:00 2009
The end :)