**** BEGIN LOGGING AT Wed May 08 02:59:58 2013
May 08 07:05:42 <EricBlade-TP> i see a new machine definition has been added to the ports sysmgr config :-)
May 08 07:07:40 <ka6sox> :)
May 08 07:10:22 <morphis> EricBlade-TP: yes :)
May 08 07:11:26 <EricBlade-TP> v nice
May 08 07:13:15 <EricBlade-TP> have i missed any other developments? my email configs have been all sorts of out of whack since the buy
May 08 07:14:13 <morphis> EricBlade-TP: video acceleration works now
May 08 07:14:18 <EricBlade-TP> not to even consider the incredible amount of time ive been putting in lately
May 08 07:14:27 <EricBlade-TP> whoa?! nice
May 08 07:14:39 <morphis> it's limited and not optimized yet
May 08 07:14:50 <morphis> meant video acceleration for cards
May 08 07:14:55 <EricBlade-TP> its better than it was?
May 08 07:14:59 <morphis> yes
May 08 07:15:21 <morphis> a lot better than software rendering
May 08 07:15:39 <EricBlade-TP> i think that theres still going to be a pretty serious problem involving card view though
May 08 07:16:22 <morphis> which one?
May 08 07:17:18 <EricBlade-TP> rendering to cards in card view is a serious cpu sink
May 08 07:19:17 <EricBlade-TP> it looked like sending an entire screen of image data down the pipc channel was just killing the cpu
May 08 07:21:14 <EricBlade-TP> i have never researched it directly, but i believe that the method for rendering to non fulk screen cards may have changed pretty seriously between 305->ce->open
May 08 07:21:43 <morphis> EricBlade-TP: we're doing it a little bit different
May 08 07:21:50 <morphis> PIpc is not involved anymore
May 08 07:22:21 <morphis> we're rendering directly into a GPU buffer in webappmanager and then sending only the file descriptor over a second local socket to LSM
May 08 07:22:38 <morphis> so no sharing of the buffer content is involved
May 08 07:22:59 <EricBlade-TP> that sounds like a much better idea than jamming it down pipc
May 08 07:23:30 <EricBlade-TP> i was really surprised the other day when i noticed that fullscreen video played almost flawlessly in card view on 305 with ce
May 08 07:24:00 <EricBlade-TP> id forgotten that 305/ce had real time card updates
May 08 07:26:54 <morphis> anyway it still needs a lot of love
May 08 07:27:01 <morphis> but should work on any android device
May 08 07:28:37 <EricBlade-TP> very interesting. how stable?
May 08 07:29:20 <EricBlade-TP> imo stability is more important than performance on a first run
May 08 07:30:06 <morphis> not really yet
May 08 07:30:23 <morphis> as buffer management isn't finished yet
May 08 07:30:41 <EricBlade-TP> i wish i couldve provided more help when i had more time, but ive been involved with so many different components that my experience level is extremely broad and not at all deep
May 08 07:30:49 <morphis> currently we're synchronizing LSM/WAM with a QSystemSemaphore
May 08 07:31:31 <morphis> EricBlade-TP: no problem, my time is limited as well these days too
May 08 07:32:55 <EricBlade-TP> im in an interestingly unique position, with my experience as an app developer, ive got a breadth of knowledge about this, that, and the other systems, but very little specific to any one piece
May 08 17:22:03 <EricBlade> mornin peeps
May 08 17:22:14 <PatrickC> morning EricBlade
May 08 17:25:28 <morphis> EricBlade: morning
May 09 01:17:26 <jsea> I ported most of webos but have a serious performance issue at playing video through html5 video tag of webkit, is there anyone also having this problem?
May 09 01:22:49 <ka6sox> webkit1?
May 09 01:23:43 <jsea> ka6sox : yes
May 09 01:25:31 <jsea> as far as I searched webkit has/had a known issue about it
May 09 01:41:29 <ka6sox> I've seen some of this but haven't tried with OW yet
May 09 02:30:05 <bhuey> hey ka6sox
**** ENDING LOGGING AT Thu May 09 02:59:58 2013