[00:07]  * mgedmin blinks
[00:07] <mgedmin> your LCD monitor does only 60 fps, so why do you care?
[00:15] <MrChrisDruif> It's still a significant decrease in performance, don't you think mgedmin ? Performance which could be used elsewhere?
[00:39] <smspillaz> MrChrisDruif: in any case, vanvugt and I found a pretty big performance bottleneck yesterday (that I initially thought wasn't a problem) and should have a fix this week
[00:39] <MrChrisDruif> Ahh, good to hear =)
[00:40] <MrChrisDruif> Is someone still working on the workspace switcher in Unity?
[00:40] <smspillaz> I am, but I'm not working these next two weeks (exams)
[00:43] <MrChrisDruif> Ahh, great. Did you spot this simple addition for it? http://askubuntu.com/questions/35488/list-of-custom-launchers-quicklists-for-unity/46832#46832
[00:43] <smspillaz> that's not really part of the workspace switcher, more or less a .desktop file thing
[00:44] <smspillaz> hack / thing / whatever
[00:44] <MrChrisDruif> Yeah, because I didn't know how to hack it into the switcher =P
[00:44] <smspillaz> worth submitting to the design group to see what they think, though I'm pretty sure they wanted to keep it so that it was always 2x2
[00:46] <MrChrisDruif> Hmmm, I wouldn't like that. A lot of people either don't use them / often and those that use them might want to add more/reduce it. How can I submit it to them?
[00:46] <smspillaz> ayatana@lists.launchpad.net , or I'll ping JohnLea about it later today if I get time to
[00:47] <MrChrisDruif> That would be great, shall I also write a small post to the mailing-list?
[00:48] <smspillaz> sure
[01:15] <thumper> smspillaz: what was the problem?
[01:15] <thumper> smspillaz: I'm pretty curious
[01:17] <smspillaz> thumper: glBindBuffer flushes the pipeline, and we're doing that once per frame
[01:17] <smspillaz> pipeline flush == stalled CPU
[01:17] <thumper> ah
[01:17] <smspillaz> and then on machines which might only manage about 80FPS with no vsync, that means that you might dip below 60FPS
[01:18] <smspillaz> and then when that happens, you get throttled to 30FPS in order to keep up with the monitor
[01:19] <smspillaz> thumper: I initially had some code which only did glBindBuffer (eg, fbo updates) whenever blur updates were pending, but its tricky since if a plugin damages a region halfway through a paint, the blurs will lag behind by one frame
[01:19] <thumper> smspillaz: so do you have a sane fix?
[01:19] <smspillaz> I think we merged my branch into core now that allows us to get around that now
[01:20] <smspillaz> thumper: not right now no, I'm not working ;-)
[01:20] <thumper> smspillaz: just wondering, who knows what you do on your off hours
[01:20] <smspillaz> oh you know
[01:20] <smspillaz> nothing
[08:34] <filip> hi, I've just installed ubuntu 11.10 with Unity and I am a bit puzzled with the dash hiding behaviour - it sometimes won't show up even though I keep the mouse cursor on the left edge of the screen. And now I am having some maximized windows and the launcher is obstructing them and I have no ide when it is supposed to hide...
[08:44] <smspillaz> filip: its a bug thats being worked on
[08:46] <filip> smspillaz: oh, do you know the url of the bug in the tracker?
[08:48] <greyback> smspillaz: hey, at the start of this year, you wrote a patch for metacity to "Add UXD shadows & borders" - if I'm skimming it correctly, it add larger invisible grabbable borders to windows to allow resizing? Am I wrong?
[10:23] <smspillaz> greyback: yes
[10:23] <smspillaz> greyback: it adds support for specifying the shadow and border size in the metacity theme, but it doesn't add support in metacity itself.
[10:23] <smspillaz> its not relevant now though
[12:26] <pindonga> morning all. I have a question re: performance with unity. Since I upgraded to oneiric, performance has been woeful :( This is strange, specially as I get a big difference with other wms. I get ~200 fps with unity3d, ~900 fps with unity2d and ~2200 fps with gnome3
[12:26] <pindonga> any ideas why this could be?
[12:29] <mhr3> pindonga, that's fps of what?
[12:29] <pindonga> mhr3, running glxgears
[12:29] <mhr3> ah, i see
[12:29] <pindonga> I'm using nvidia drivers
[12:29] <pindonga> I know fps is not the best benchmark, but I don't know how else to measure this
[12:30] <pindonga> unity is extremely sluggish, and the dash almost never opens up when I hit super
[12:30] <pindonga> this was not so with natty
[12:30] <pindonga> so I'm trying to see how to make it work again
[12:31] <pindonga> at one moment I thought it could be some compiz setting in my account, so I created a new clean account, but I get the same results
[12:32] <mhr3> well for one, glxgears is not a benchmark, for example if i run it in unity i get 2200 and when i close everything and use metacity i get 1100
[12:33] <mhr3> so you really need to find different way to profile this
[12:33] <pindonga> any suggestions?
[12:33] <mhr3> ask compiz people :)
[12:40] <diwic> ronoc, heya - just finished off the PA patch that gnome-cc needs to get the new info :-)
[12:41] <ronoc> diwic, sweet
[12:42] <diwic> ronoc, how do I make it available to you?
[12:43] <ronoc> diwic, ppa for now would be wise
[12:43] <diwic> ronoc, oneiric or precise?
[12:44] <ronoc> diwic, oneiric please
[12:44] <diwic> ronoc, excellent. Out of curiousity, have you had time to start working on the GUI parts?
[12:45] <ronoc> diwic, no, starting this week I hope
[12:45] <diwic> ronoc, hmm, if it's that bad "this week I hope" I might beat you to some of the backend of gnome-cc
[12:46] <diwic> ronoc, how is your workflow when it comes to the work on gnome-cc ?
[12:46] <ronoc> diwic, have you thought about the questions raised at uds about some of the user journeys ?
[12:46] <ronoc> diwic, does this not need to be flushed out before code is written
[12:48] <diwic> ronoc, hmm, did you have a list of those user journeys?
[12:49] <diwic> ronoc, can't find them in etherpad nor blueprint
[12:49] <ronoc> diwic, this is what I'm starting on before I start coding
[12:49] <ronoc> diwic, can you email me your mock up again ?
[12:49] <diwic> ronoc, fair enough
[12:49] <diwic> ronoc, it's not even my mockup :-) it was harry's
[12:51] <ronoc> diwic, oh right of course.
[12:52] <diwic> ronoc, but it's here http://people.canonical.com/~diwic/sound-settings/gvc_ui_final.jpg , move the top list to the left and do not add the bottom two checkboxes
[13:06] <diwic> ronoc, ok, but if you're not ready to start coding yet I'll wait a little with the ppa upload, see if I can get an implementation of the new jack detection interface as well, that would be nice
[13:07] <diwic> ronoc, (i e the new jack detection interface between kernel and pulseaudio, so it does not affect you directly, but it might help the complete experience)
[13:10] <ronoc> diwic, sounds good. i'm a bit under the weather today so was thinking of starting in the morning properly. I have meetings at Millbank but will try to get that document underway in the next day or so. Will share with you once I have something worthwhile to show
[13:12] <diwic> ronoc, killall -9 meeting-daemon
[13:12] <diwic> :-)
[13:13] <ronoc> diwic, just for the next few weeks I will have meetings about stuff ...
[13:13] <diwic> ronoc, good to have synced up with you. Let's resync once there is some interesting progress.
[13:13] <ronoc> diwic, ok cool, will be in touch in the next day or so
[13:40] <andyrock> who's Cheakamus?
[13:40] <AlexzAK_> Please help me with my bug https://bugs.launchpad.net/ubuntu/+source/indicator-appmenu/+bug/877191
[13:41] <AlexzAK_> Kate (and Aptana 3) have main menu corrupted in both unity and unity 2d
[13:41] <mhr3> davidcalle, ping?
[13:41] <davidcalle> mhr3, pong
[13:42] <mhr3> davidcalle, you have experience with multiple scopes in a lens, right? is everything working fine when you have more scopes?
[13:42] <mhr3> i just noticed that we may be emitting some extra signals when you're using multiple scopes
[13:46] <davidcalle> mhr3, everything is working apparently, with scopes in the same category or in several. I have some weirdness with scopes being reordered in one particular case where icons are local for one scope and remote for the other, as they take time to load, results of one scope jump before the other.
[13:47] <davidcalle> mhr3, what are the extra signals emitted?
[13:48] <mhr3> lens is emitting search-finished multiple times if multiple scopes are used, this shouldn't be a problem for lenses/scopes themselves, but it may screw up stuff inside the dash itself
[13:55] <angelo-c> didrocks, can you help me whit bug 773841?
[13:55] <angelo-c> it is bitesize job, I would to contribute
[14:25] <davidcalle> mhr3, I've just made some tests to try to mess with search finished. No problems here.
[14:26] <mhr3> davidcalle, you mean other than the search spinner disappearing before all the searches actually finish, right? :)
[14:50] <jbwiv_> guys, anyone know if the improved multi-monitor support is testable/installable on Oneiric? I had to switch over the KDE for now because unity wouldn't support my three monitor setup...I'd like to switch back sooner rather than later...
[14:51] <coz_> smspillaz,   hey guy,, are there any changes in compiz/Unity that may effect triple monitor/dual card setups?
[14:54] <jbwiv_> coz_ is asking because I was asking in #ubuntu+1 ;-) At the moment I can't get three monitors working well on 11.10 but kde4 works fine with them. I'm using two cards: nvidia GeForce GT440 and Geforce 9800GT
[15:10] <toumbo> Hi, I noticed than when I type to find an aplication on ubuntu's dash home it stops wokin for a while
[15:55] <davidcalle> mhr3, oh ok I see what you mean. Yes, the spinner stops before all searches finish. The scope should have a way to notify the lens and the lens should notify search finished itself.
[15:55] <davidcalle> scopes*
[17:47] <andyrock> jaytaoko1, hi
[17:47] <andyrock> does it make sense? https://code.launchpad.net/~andyrock/nux/it-s-a-scroll-not-a-button-released-event/+merge/82063
[17:47] <jaytaoko1> andyrock: hello
[17:48] <andyrock> of course you understand nux better than me :P
[17:48] <andyrock> i know that it should be a WIP: because there is not a test
[17:49] <jaytaoko1> andyrock: I will get back to you on this...
[17:52] <andyrock> jaytaoko1, ok thanks...
[21:34] <compa> didrocks: I'm intresting contributing to this bug 773841, can you mentoring me?
[21:42] <andyrock> jaytaoko1, i've just read your comment...
[21:44] <andyrock> So  "Let's talk about how we can test this" :)
[21:44] <andyrock> we should be sure that there are not other regressions...
[21:48] <andyrock> ping me when you are available
[21:51] <andyrock> maybe we can made a test to be sure that when a mouse state changes the right signal is emitted
[22:02] <thumper> hi andyrock
[22:02] <andyrock> hi thumper
[22:02] <thumper> andyrock: I was just talking with jaytaoko1 before about testing
[22:03] <thumper> suggesting that we have a simple app that we fake an X mouse event to and confirm that scroll events don't have mouse down set
[22:04] <andyrock> but we should test that this patch create regressions
[22:04] <andyrock> so IMO se should test all x mouse events
[22:04] <andyrock> *we
[22:05] <andyrock> *this patch doesn't create (sorry :) )
[22:06] <jaytaoko1> andyrock: brb
[22:06] <thumper> andyrock: agreed, yes it should
[22:11] <andyrock> thumper, and i'm wondering if using nux has a "tool" to create fake X mouse event
[22:11] <andyrock> *without using
[22:11] <thumper> does xtest do this?
[22:15] <andyrock> yeah, but something inside nux should help writing other tests IMO
[22:15] <thumper> andyrock: so... talk with jaytaoko1 :)
[22:16] <thumper> I do wonder who jaytaoko (without the 1) is