--- Log opened Sat Oct 09 00:00:08 2010
--- Day changed Sat Oct 09 2010
20101009 00:00:08< Ivanovic> using :droid i get perfectly fluent 27 to 35fps as normal value with the classic driver
20101009 00:00:19< Ivanovic> (gallium is basically an experimental, new driver basis)
20101009 00:01:51< alink> so it's a good change, I was a bit annoyed to lose some fps on empty map (which is not a good test), so good thing that it's globally better
20101009 00:02:11< Ivanovic> okay, i am only using two brothers as testbase
20101009 00:02:30< alink> don't forget to test fullscreen (switch and titlecreen)
20101009 00:02:36< Ivanovic> since the scenario has a rather small map but with some water right through it as well as lots of grass and forests plus some hills and mountains
20101009 00:03:28< alink> yeah that's ok, my difference with empty map was probably because of the crazy ~200fps rate
20101009 00:03:48< Ivanovic> WOW, holy shitheat!
20101009 00:04:02< Ivanovic> switching to fullscreen 1920x1200 does barely decrease performance!
20101009 00:04:38< alink> and it did before? interesting.
20101009 00:04:42< Ivanovic> yes, there are still some glitches left with the border stuff when scrolling
20101009 00:04:48< Ivanovic> before it went down to basically 0 fps
20101009 00:05:04< Ivanovic> okay, now isle of the damned
20101009 00:06:26< alink> the clipping will be fixed at some point. I just need to rewrite an ugly stack of tweaks from a clean scratch
20101009 00:06:31< Ivanovic> yeah, even isle of the damned is usable
20101009 00:06:45< Ivanovic> there are still glitches especialy with water and the top border
20101009 00:07:03< Ivanovic> but things seem to already be running nicely (at least on a very usable level) with the open source driver over here
20101009 00:08:10< Ivanovic> okay, whatever it is you do when scrolling, it is slow using the gallium driver but fast (enough) with the classic one
20101009 00:08:22< alink> maybe also check halo, I had a new glitch there (now fixed)
20101009 00:08:45< alink> scrolling didn't really changed, but I am about to try a new way
20101009 00:10:05< Ivanovic> even logo fade in is bearable with the gallium driver (was freaking slow before!)
20101009 00:10:26< alink> ah I never tested that :p
20101009 00:10:28< Ivanovic> scrolling is resulting in a huge frame drop for the gallium driver, down to some 10fps
20101009 00:10:47< Ivanovic> with the classic driver it is fast and stays well above 30fps
20101009 00:11:27< Espreon> Ugh, so much mess in UtBS... where to begin?
20101009 00:11:49-!- SpoOkyMagician [~chatzilla@74-138-202-211.dhcp.insightbb.com] has quit [Quit: be back later]
20101009 00:12:14< Ivanovic> but the general glitches are all the same (units deformed and stuff like this)
20101009 00:12:54< alink> Ivanovic: yeah just changed the double-buffering, the only possible new glitches would be alpha-blending stuff
20101009 00:12:59< Ivanovic> alink: you have really done some freaking cool stuff!
20101009 00:13:19< Ivanovic> and the darkens feature in the test scenario appears to work correctly
20101009 00:13:28< Ivanovic> afl now
20101009 00:13:31< alink> Ivanovic: I am happy that it's more portable, as I hoped 
20101009 00:13:32< Ivanovic> s/sfl/sfk
20101009 00:13:57< alink> too much afk already ;p
20101009 00:14:05< Ivanovic> uhm, yeah
20101009 00:17:53 * loonycyborg doesn't like the mention of FBOs since he had problems with them in the past
20101009 00:18:16< alink> FBO are nice
20101009 00:18:44< alink> but only needed by the SDL emulator, maybe later we can skip them
20101009 00:18:51< loonycyborg> Now if linux nvidia driver didn't have problems with them..
20101009 00:20:32< alink> well, it's openGL 1.1 stuff, and I use nothing fancy like renderbuffer, so I hope that it's ok
20101009 00:21:20< loonycyborg> Huh? I thought that those were added much later than 1.1.
20101009 00:22:12< alink> hmm yes it's an extension, so not sure about when it became official enough
20101009 00:22:31< loonycyborg> iirc at 2.0 or later
20101009 00:23:01-!- Blueblaze [~Blueblaze@adsl-99-148-247-46.dsl.hstntx.sbcglobal.net] has quit [Quit: Blueblaze]
20101009 00:23:22-!- Johannes13 [~Johannes@unaffiliated/johannes13] has quit [Ping timeout: 265 seconds]
20101009 00:26:40-!- mjs-de [~mjs-de@vpw.wh.uni-dortmund.de] has quit [Remote host closed the connection]
20101009 00:26:55< zookeeper> Espreon, the parts that don't work or could work better
20101009 00:27:02-!- iwontbecreative [~thibault@89-180-56-175.net.novis.pt] has quit [Ping timeout: 264 seconds]
20101009 00:27:31< alink> loonycyborg: yes you are right, much later, maybe even 3.0. I don't know where my mind got the 1.1
20101009 00:27:32-!- DesertPanther_ [~Khalid@41.234.232.239] has joined #wesnoth-dev
20101009 00:28:06< zookeeper> Espreon, no need to spend time rewriting things which work just fine atm when there's stuff that could actually be made to work better
20101009 00:28:10< Espreon> zookeeper: Would you please name at least three things that don't work?
20101009 00:28:36< alink> ah maybe some "OpenGL 1.1 is required" in the technical specifications
20101009 00:29:36< loonycyborg> alink: Anyway, gl branch still seems to work fine for me.
20101009 00:29:46< alink> hmm this OpenGL 3.0 thing, annoys me for older hardware
20101009 00:29:57< alink> loonycyborg: cool, thanks for testing
20101009 00:30:12< zookeeper> Espreon, well...not really. i don't remember any off the top of my head :P
20101009 00:30:19< Espreon> OpenGL 3.0? Right now?
20101009 00:30:21< Espreon> No....
20101009 00:30:25-!- DesertPanther [~Khalid@unaffiliated/desertpanther] has quit [Ping timeout: 252 seconds]
20101009 00:31:48-!- Vetinari [~lukjad@unaffiliated/lukjad] has quit [Ping timeout: 245 seconds]
20101009 00:31:55-!- Vetinari [~lukjad@unaffiliated/lukjad] has joined #wesnoth-dev
20101009 00:32:22-!- Appleman1234 [~Appleman1@CPE-60-226-176-19.qld.bigpond.net.au] has joined #wesnoth-dev
20101009 00:32:26< zookeeper> Espreon, well, there's probably a lot of moveto triggers with hard-coded location ranged which could be made to use things like [filter] side=1 [filter_location] x,y=1,1 radius="$($unit.moves + 1)" instead
20101009 00:32:33< zookeeper> s/ranged/ranges
20101009 00:32:51< zookeeper> s/unit/this_unit
20101009 00:34:22-!- Elvish_Pillager [~eli@dhip-164.foss.residences.colby.edu] has quit [Ping timeout: 240 seconds]
20101009 00:34:26< zookeeper> and one thing that could get rewritten more elegantly would be the increasing recruit costs
20101009 00:34:31< Espreon> zookeeper: ... I'll just clean things up and fix things as I go.
20101009 00:37:36-!- boucman [~rosen@wesnoth/developer/boucman] has quit [Quit: Leaving.]
20101009 00:47:07< alink> ok FBO are part of OpenGL 3.0 but seems often supported by OpenGL 2.0+ driver/hardware via the extension
20101009 00:47:36< alink> still a bit annoying
20101009 00:53:09-!- zookeeper [~l@wesnoth/developer/zookeeper] has quit []
20101009 00:55:29-!- Upthorn [ogmar@adsl-75-26-200-206.dsl.scrm01.sbcglobal.net] has joined #wesnoth-dev
20101009 00:58:14-!- Upth [ogmar@adsl-75-26-200-206.dsl.scrm01.sbcglobal.net] has quit [Ping timeout: 264 seconds]
20101009 00:58:16-!- kitty_ [~kathrinpo@e180201082.adsl.alicedsl.de] has quit [Quit: kitty_]
20101009 01:05:53-!- shadowm_laptop [ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20101009 01:07:13< Ivanovic> alink: since it is opengl it is mainly a matter of the driver to support those nicely
20101009 01:07:25< Ivanovic> alink: especially since IIRC opengl es does come with this stuff, too
20101009 01:07:37< Ivanovic> which makes it a kind of "using it makes sense"-case
20101009 01:08:05< Ivanovic> even if there is/was no hardware support for FBOs it is only a matter of driver work to implement it in some software way
20101009 01:08:17< Ivanovic> i will test it on my laptop with rather ancient intel hardware tomorrow
20101009 01:08:43< Ivanovic> (using the not too new mesa 7.8.2 driver with the intel i945gm (or whatever it is exactly called) on board graphics
20101009 01:09:44< Ivanovic> in case you are interested: the glxinfo output from my card using mesa git with r600 classic: http://pastebin.com/XgET1s7B
20101009 01:09:56< Ivanovic> will paste the info for my laptop with ancient intel hardware tomorrow
20101009 01:10:20< alink> yeah, probably not a problem. Also, as I said, when the SDL emulator will be not needed anymore, this specific use of FBO will vanish
20101009 01:10:57< alink> thanks I am coding extension checking now
20101009 01:11:18< alink> In fact I can already commit the driver part
20101009 01:11:26< CIA-35> alink * r46972 /branches/ogl/src/video.cpp: Display driver info, OpenGL version, GPU name and vendor on standard output
20101009 01:12:17< Ivanovic> in case you are interested: here is the output of glxinfo using gallium: http://pastebin.com/AYJtM1wk
20101009 01:12:31< Ivanovic> as you see it barely differs in regards to extensions it announces as supported
20101009 01:12:53< Ivanovic> will be more interesting for my current laptop
20101009 01:13:06< Ivanovic> and yeah, i should get my ancient laptop out again, update the system and test things
20101009 01:13:27< Ivanovic> that one would have a mobility radeon 9000 (something r200 based, so *really* ancient)
20101009 01:14:18< Ivanovic> the hardware should in theory be opengl 1.4 capable, the driver level regarding the website is 1.3 only (cf http://wiki.x.org/wiki/RadeonFeature )
20101009 01:15:20< alink> http://www.opentk.com/node/831#comment-4456  this comment seems to indicate that it will be too old
20101009 01:15:52< Ivanovic> that entry is completely outdated
20101009 01:16:02< Ivanovic> the open source drivers progress rapidly for the r300 hardware
20101009 01:16:17< Ivanovic> those basically got opengl 2.1 support by now
20101009 01:16:31< Ivanovic> it is just that r200 is even older and won't get gallium support and stuff like this
20101009 01:16:33< alink> ok, I am a nvidia guy, so I don't really follow ATI stuff
20101009 01:17:23< Ivanovic> in general checking for the extensions that you want to use is the way to go
20101009 01:17:37< alink> it's true that outdated driver are only a problem for old window box
20101009 01:17:43< Ivanovic> yeah, some drivers do cheat in regards to what they advertise and what they support in a sane way, but better than anything
20101009 01:18:55< CIA-35> espreon * r46973 /trunk/data/campaigns/Under_the_Burning_Suns/ (2 files in 2 dirs): Moved the dehydration stuff to a separate file.
20101009 01:19:55< Ivanovic> the main benefit of the open source drivers is that you can easily talk to the devs if you work on something and it seems to be not working as expected
20101009 01:20:11< Ivanovic> they can usually tell you if you are doing some strange thinking or if there is a bug in the driver
20101009 01:21:58< alink> well, my main problem is that all always work fine here, I can quickly diagnose (and sometimes fix) glitches that I see myself.
20101009 01:22:31-!- Elvish_Pillager [~eli@dhip-164.foss.residences.colby.edu] has joined #wesnoth-dev
20101009 01:23:03< alink> But the slow method of pasting test code to someone else and check screenshot is really not efficient
20101009 01:23:19< Ivanovic> alink: are you using a desktop system or a laptop?
20101009 01:23:30< alink> desktop
20101009 01:23:36< Ivanovic> if you are using a desktop system, maybe get a cheap ati pci card for testing?
20101009 01:24:08< Ivanovic> ugh, okay, forget the idea
20101009 01:24:16< alink> mmh I get one but AGP :-/
20101009 01:24:26< Espreon> Awwww... the ogl branch doesn't start for me anymore...
20101009 01:24:31< Ivanovic> there is not much cheap stuff with reasonable recent hardware available
20101009 01:24:43< Ivanovic> wait, there are two cards with a 4350 as pci
20101009 01:25:07< alink> Espreon: any error ?
20101009 01:25:10< Ivanovic> alink: you can have two cards in your system
20101009 01:25:29< Ivanovic> and if you got a free pci slot it should not be much of an issue adding an aditional card
20101009 01:25:38< Espreon> alink: http://wesnoth.pastebin.com/cxrDvPWB
20101009 01:25:41< alink> Ivanovic: I mean I have an AGP ATI card, but my box don't have AGP port
20101009 01:25:44< Ivanovic> though it depends on your distribution if switching the card/driver is easy or not
20101009 01:26:35< alink> Espreon: thanks, glad that I added version info, now I know that you should support it
20101009 01:26:36< Ivanovic> alink: if you could work with switching pcie cards (or even got a free x16 slot) you could get a low budget ati card for maybe 40€
20101009 01:26:53< Ivanovic> considering what you are working on i'd guess that it should be no problem to use wesnoth funds to get you this card
20101009 01:26:59< Espreon> alink: OK... so... now what?
20101009 01:27:26< alink> Espreon: I will probably need to add more debug code
20101009 01:27:50< Espreon> OK.
20101009 01:28:31< Ivanovic> that is: there is even a pciex1 card for reasonable pricing
20101009 01:28:38< Ivanovic> http://geizhals.at/eu/a498687.html?v=k
20101009 01:28:39< alink> Ivanovic: well, that would be nice, but not sure yet if needed. And installing 2 GPU is really possible ?
20101009 01:28:47< Ivanovic> it is rather cheap and slow, but should work
20101009 01:28:52< Ivanovic> alink: of course this is possible
20101009 01:29:02< alink> I mean without the SLI thing
20101009 01:29:10< Ivanovic> you need to do some xorg.conf magic to switch to the correct output
20101009 01:29:21< Ivanovic> and you might need to do some manual symlink magic to switch the opengl driver
20101009 01:29:36< Ivanovic> but yeah, it should be possible getting two seperate cards to work
20101009 01:29:45< Ivanovic> (of course you have to change the display cable to the other card)
20101009 01:30:03< Ivanovic> you could ask in #radeon if they think such a setup is possible and if they can give you some hints
20101009 01:30:21-!- DesertPanther_ [~Khalid@41.234.232.239] has quit [Ping timeout: 250 seconds]
20101009 01:31:15< Ivanovic> i remember that in the old days it was possible to get more displays attached to your system by adding additional pci cards
20101009 01:31:21< Ivanovic> using some xorg.conf magic
20101009 01:31:26< Ivanovic> this should still be possible
20101009 01:31:37< alink> yeah I used AGP + PCI in the past
20101009 01:31:53< Ivanovic> it should as well be possible using two pcie cards
20101009 01:32:07< Ivanovic> (if you don't have two full slows, have a look at the x1 card i just posted a link to)
20101009 01:32:56< Ivanovic> s/slows/slots
20101009 01:34:03< alink> hmmm another problem is that my nvidia fan is already dead and replaced by a case-fan, not sure what is the space status
20101009 01:34:15< alink> I could use a smaller fan, though
20101009 01:34:20< Ivanovic> and yeah, those cards are freaking slow, basically on par with onboard graphics chips, but this should be enough and fine
20101009 01:35:57< Ivanovic> you should just check where your free slots are and if a card fits in there, those x1 cards fit into any pcie slots and just use one space (passive cooling and not producing much heat)
20101009 01:36:24-!- Upthorn [ogmar@adsl-75-26-200-206.dsl.scrm01.sbcglobal.net] has quit [Ping timeout: 272 seconds]
20101009 01:36:56< alink> I go checking
20101009 01:37:06< Ivanovic> then you could test a *really* wide range of stuff: nvidia proprietary driver, nvidia open source driver (yeah, not expected to work...), fglrx (ati proprietary, the card is new enough to still support it), mesa r600classic, mesa r600 gallium
20101009 01:37:11< alink> (box is in another room)
20101009 01:37:26< Ivanovic> i got to head off to bed now, will try to do some further testing tomorrow
20101009 01:38:40< Ivanovic> alink: if the money for the additional card would be an issue i am rather sure that it should be possible to handle it via some wesnoth funds (those 50€ should still be affordable for the project, right?)
20101009 01:41:37< alink> that would be nice, because I don't want to buy something that I don't really need myself
20101009 01:41:55< alink> but better wait that I am more advanced in the ogl branch
20101009 01:43:08< alink> because I am not sure that it's a good idea to lose even more time at installing complex setup, before having my base working
20101009 01:44:14< alink> that compatibility stuff already slow me too much, even if it's needed for early directions
20101009 01:46:39-!- Johannes13 [~Johannes@pD9502543.dip0.t-ipconnect.de] has joined #wesnoth-dev
20101009 01:46:39-!- Johannes13 [~Johannes@pD9502543.dip0.t-ipconnect.de] has quit [Changing host]
20101009 01:46:39-!- Johannes13 [~Johannes@unaffiliated/johannes13] has joined #wesnoth-dev
20101009 01:56:47< CIA-35> espreon * r46974 /trunk/data/campaigns/Under_the_Burning_Suns/_main.cfg: Changed UtBS's definition to 'CAMPAIGN_UNDER_THE_BURNING_SUNS'.
20101009 02:20:34< CIA-35> ai0867 * r46975 /trunk/data/tools/wesnoth/wmlparser2.py: Make wmlparser2's jsonify option use a single dictionary entry with a list containing all tags with the same name
20101009 02:32:40< CIA-35> alink * r46976 /branches/ogl/src/video.cpp: Check for GL_EXT_framebuffer_object OpenGL extension
20101009 02:32:51< CIA-35> alink * r46977 /branches/ogl/src/video.cpp: More verbose error message for FBO setup error
20101009 02:33:05< alink> Espreon: ^could you retry with this (no fix, just debug info)
20101009 02:33:48< Espreon> alink: Sure.
20101009 02:34:55-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has quit [Remote host closed the connection]
20101009 02:34:58-!- Gambit [~Gambit@wesnoth/developer/grickit] has quit [Read error: Connection reset by peer]
20101009 02:36:45< Espreon> alink: http://wesnoth.pastebin.com/CcySSAy3
20101009 02:37:30< alink> hmm the most vague one :-/
20101009 02:37:35< alink> Espreon: thanks
20101009 02:37:43-!- Gambit [~Gambit@pa-67-234-122-243.dhcp.embarqhsd.net] has joined #wesnoth-dev
20101009 02:37:43-!- Gambit [~Gambit@pa-67-234-122-243.dhcp.embarqhsd.net] has quit [Changing host]
20101009 02:37:43-!- Gambit [~Gambit@wesnoth/developer/grickit] has joined #wesnoth-dev
20101009 02:37:57-!- vcap [~vcap@AReims-551-1-37-250.w86-215.abo.wanadoo.fr] has quit [Quit: leaving]
20101009 02:38:16< Espreon> alink: No prob.
20101009 02:39:00< CIA-35> alink * r46978 /branches/ogl/src/video.cpp: Possible fix for FBO error.
20101009 02:41:34< Espreon> alink: That fixed it. Thanks.
20101009 02:41:59< alink> Espreon: cool :)  thanks again for the multiple testing
20101009 02:42:07< Espreon> No prob.
20101009 02:43:09< Espreon> alink: The merfolk sprites get crushed...
20101009 02:43:52< alink> yeah, clipping temporary disabled, and it's used by submerge
20101009 02:44:01< Espreon> Ah, I see.
20101009 02:44:26 * alink should wait to fix all regression before commit code for testing
20101009 02:44:48< Espreon> There's also some tearing at hex borders.
20101009 02:44:55< alink> same problem 
20101009 02:45:12< Espreon> OK.
20101009 02:45:46< Espreon> And would the crushing of the sprites on the sidebar be caused by the same thing? Because they're crushed.
20101009 02:45:53< Espreon> Well...
20101009 02:45:56< Espreon> Downsized.
20101009 02:45:57< alink> the sad thing is that I took the time to fix this for the slow SDL emulator path, but later I added texture and break it again
20101009 02:46:22< alink> sidebar crushing is still mysterious but I didn't investigated it yet 
20101009 02:46:51-!- FAAB1 [~huajie@219.142.235.201] has joined #wesnoth-dev
20101009 02:46:54-!- k23z__ [~k23z__@188.26.49.33] has joined #wesnoth-dev
20101009 02:47:39< CIA-35> alink * r46979 /branches/ogl/src/video.cpp: Comment the previous FBO fix, since it was a bit tricky
20101009 02:47:40-!- k23z__ is now known as spetrea-home
20101009 02:51:22< alink> ok sidebar crushing is also clipping related
20101009 02:53:35-!- stikonas [~and@wesnoth/translator/stikonas] has quit [Remote host closed the connection]
20101009 02:53:43< CIA-35> espreon * r46980 /trunk/data/campaigns/Under_the_Burning_Suns/ (16 files in 2 dirs): Two new macros, UTBS_INCLUDE and UTBS_MAP.
20101009 02:55:39< Espreon> Note to self: simplify preprocessor stuff TSL inherited from IftU.
20101009 02:58:24-!- loonycyborg [~sergey@wesnoth/developer/loonycyborg] has quit [Quit: Zzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzz]
20101009 02:59:02-!- loonybot [~loonybot@wesnoth/bot/loonybot] has quit [Remote host closed the connection]
20101009 03:01:31-!- Johannes13_ [~Johannes@unaffiliated/johannes13] has joined #wesnoth-dev
20101009 03:04:48-!- Johannes13 [~Johannes@unaffiliated/johannes13] has quit [Ping timeout: 240 seconds]
20101009 03:21:05-!- vcap [~vcap@AReims-551-1-37-250.w86-215.abo.wanadoo.fr] has joined #wesnoth-dev
20101009 03:25:45< CIA-35> espreon * r46981 /trunk/data/campaigns/Under_the_Burning_Suns/ (9 files in 2 dirs): Replaced most MODIFY_UNIT calls with [modify_unit] equivalents.
20101009 03:28:34< CIA-35> espreon * r46982 /trunk/data/campaigns/Under_the_Burning_Suns/scenarios/09_Blood_is_Thicker_Than_Water.cfg: Replaced a MODIFY_UNIT call with a [petrify] equivalent.
20101009 03:32:24-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20101009 03:51:23-!- Johannes13_ [~Johannes@unaffiliated/johannes13] has quit [Ping timeout: 245 seconds]
20101009 03:52:16-!- spetrea-home [~k23z__@188.26.49.33] has quit [Ping timeout: 252 seconds]
20101009 03:53:19< CIA-35> alink * r46983 /branches/ogl/src/video.cpp: Fully specify texture parameters of the one attached to the FBO
20101009 03:53:23< CIA-35> alink * r46984 /branches/ogl/src/ (sdl_utils.cpp video.cpp): 
20101009 03:53:23< CIA-35> Use GL_RGBA8 instead of GL_RGBA for internal format of textures
20101009 03:53:23< CIA-35> (since it's what SDL uses anyway)
20101009 03:58:09-!- spetrea-home [~k23z__@188.26.49.33] has joined #wesnoth-dev
20101009 04:02:41-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20101009 04:11:18< CIA-35> espreon * r46985 /trunk/data/campaigns/Under_the_Burning_Suns/scenarios/01_The_Morning_After.cfg: Beginning porting AI configurations to their newer equivalents.
20101009 04:12:13-!- Blueblaze [~Blueblaze@adsl-99-4-146-84.dsl.hstntx.sbcglobal.net] has joined #wesnoth-dev
20101009 04:25:16-!- lgn [~debian@port-92-204-53-119.dynamic.qsc.de] has quit [Ping timeout: 265 seconds]
20101009 04:27:48-!- Upth [~ogmar@69.62.144.108] has joined #wesnoth-dev
20101009 04:27:48-!- Upth is now known as Upthorn
20101009 04:40:21-!- Ivanovic_ [~ivanovic@dtmd-4db2bf34.pool.mediaWays.net] has joined #wesnoth-dev
20101009 04:42:54-!- Ivanovic [~ivanovic@wesnoth/developer/ivanovic] has quit [Ping timeout: 265 seconds]
20101009 04:44:17-!- Ivanovic_ is now known as Ivanovic
20101009 04:45:50< CIA-35> alink * r46986 /branches/ogl/src/sdl_utils.cpp: 
20101009 04:45:50< CIA-35> Fix clipping (which was broken when adding texture support)
20101009 04:45:50< CIA-35> This fix: edges of screen, submerge, sidebar image and hp/xp bars.
20101009 04:45:50< CIA-35> Should be regression-free again (unless you have driver problem).
20101009 04:45:50< CIA-35> Please report any remaining glitch.
20101009 04:45:56< alink> \o/
20101009 04:46:05< Espreon> Yay...
20101009 04:46:45< alink> well the math there are still an unreadable mess, though
20101009 04:47:03< Espreon> Well, it's math... what do you expect?
20101009 04:48:47< alink> maths are always nice, or at least can always be written in a nice way
20101009 04:49:12 * Espreon rolls his eyes
20101009 04:49:24< Espreon> No, math is evil.
20101009 04:49:50< alink> unless it's a dumb problem like mixing SDL clipping rectangle with Opengl texture coordinates ;)
20101009 04:52:11< alink> anyway it's messy because it started with something nice and simple, and then I added little tweaks here and there for each glitch, instead of rewriting it
20101009 04:53:50< alink> OtOH things like SDL and OGL not using the same coordinates system don't help 
20101009 05:03:58-!- alink [~alink@wesnoth/developer/alink] has quit [Remote host closed the connection]
20101009 05:16:16-!- Mica [~Mica@d27-96-76-1.nap.wideopenwest.com] has joined #wesnoth-dev
20101009 05:32:48-!- Gambit [~Gambit@wesnoth/developer/grickit] has quit [Remote host closed the connection]
20101009 05:53:53-!- Sirp [~me@wesnoth/developer/dave] has joined #wesnoth-dev
20101009 05:55:20< Sirp> Soliton: around?
20101009 06:15:58-!- Elvish_Pillager [~eli@dhip-164.foss.residences.colby.edu] has quit [Ping timeout: 252 seconds]
20101009 06:18:07-!- SpoOkyMagician [~chatzilla@74-138-202-211.dhcp.insightbb.com] has joined #wesnoth-dev
20101009 06:37:36-!- crimson_penguin [~ben@wesnoth/developer/crimsonpenguin] has quit [Quit: crimson_penguin]
20101009 07:08:41-!- spetrea-home [~k23z__@188.26.49.33] has quit [Ping timeout: 240 seconds]
20101009 07:09:21-!- Mica [~Mica@d27-96-76-1.nap.wideopenwest.com] has quit [Quit: Good night]
20101009 07:25:37< shadowmaster> Sirp: do you need anything in particular?
20101009 07:26:06-!- dtiger [~dtiger@dynamic-vpdn-93-125-64-24.telecom.by] has joined #wesnoth-dev
20101009 07:36:49-!- Mica [~Mica@d27-96-76-1.nap.wideopenwest.com] has joined #wesnoth-dev
20101009 07:41:20-!- freim [~hogne@feathers.os-tr.net] has quit [Ping timeout: 276 seconds]
20101009 07:41:37-!- freim [~hogne@feathers.os-tr.net] has joined #wesnoth-dev
20101009 07:58:48-!- Mica [~Mica@d27-96-76-1.nap.wideopenwest.com] has quit [Quit: Good night]
20101009 08:13:25< Sirp> shadowmaster: well, I was trying to test a program that uses UDP packets on wesnoth's server, but the packets don't seem to get through...
20101009 08:13:32< Sirp> I'm wondering if we have a firewall that blocks them
20101009 08:13:49< shadowmaster> ask Rhonda. She recently set-up a firewall.
20101009 08:14:49< shadowmaster> I mean, *of course* we have a firewall. >:)
20101009 08:16:10< Sirp> okay. :)
20101009 08:16:12< Sirp> I sent her a pm. :)
20101009 08:20:48-!- Bocom_ [~Bocom@c-8dcde255.013-31-6b736412.cust.bredbandsbolaget.se] has joined #wesnoth-dev
20101009 08:21:11-!- Bocom__ [~Bocom@c-8dcde255.013-31-6b736412.cust.bredbandsbolaget.se] has joined #wesnoth-dev
20101009 08:22:00-!- Bocom [~Bocom@c-8dcde255.013-31-6b736412.cust.bredbandsbolaget.se] has quit [Read error: Connection reset by peer]
20101009 08:25:17-!- Bocom_ [~Bocom@c-8dcde255.013-31-6b736412.cust.bredbandsbolaget.se] has quit [Read error: Connection reset by peer]
20101009 08:27:03-!- Bocom__ [~Bocom@c-8dcde255.013-31-6b736412.cust.bredbandsbolaget.se] has quit [Read error: Connection reset by peer]
20101009 08:27:38-!- Bocom__ [~Bocom@c-8dcde255.013-31-6b736412.cust.bredbandsbolaget.se] has joined #wesnoth-dev
20101009 08:28:48-!- Bocom_ [~Bocom@c-8dcde255.013-31-6b736412.cust.bredbandsbolaget.se] has joined #wesnoth-dev
20101009 08:30:38-!- Bocom__ [~Bocom@c-8dcde255.013-31-6b736412.cust.bredbandsbolaget.se] has quit [Read error: Connection reset by peer]
20101009 08:31:20-!- silene [~plouf@wesnoth/developer/silene] has joined #wesnoth-dev
20101009 08:34:49-!- SpoOkyMagician [~chatzilla@74-138-202-211.dhcp.insightbb.com] has quit [Quit: Later...]
20101009 08:37:03-!- dtiger [~dtiger@dynamic-vpdn-93-125-64-24.telecom.by] has quit [Remote host closed the connection]
20101009 08:57:42-!- shadowm_laptop [ignacio@wesnoth/developer/shadowmaster] has quit []
20101009 09:24:10-!- MeccaGod [~majs@h14n4fls31o279.telia.com] has joined #wesnoth-dev
20101009 09:30:06-!- boucman [~rosen@wesnoth/developer/boucman] has joined #wesnoth-dev
20101009 09:53:44-!- mjs-de [~mjs-de@vpw.wh.Uni-Dortmund.DE] has joined #wesnoth-dev
20101009 10:27:11-!- Blueblaze [~Blueblaze@adsl-99-4-146-84.dsl.hstntx.sbcglobal.net] has quit [Quit: Blueblaze]
20101009 10:34:02-!- Unnheulu [~ieuan@cpc1-pnth2-2-0-cust788.5-2.cable.virginmedia.com] has joined #wesnoth-dev
20101009 10:34:33-!- GNUtoo|laptop [~gnutoo@host227-139-dynamic.211-62-r.retail.telecomitalia.it] has joined #wesnoth-dev
20101009 10:59:11-!- Ivanovic [~ivanovic@dtmd-4db2bf34.pool.mediaWays.net] has quit [Changing host]
20101009 10:59:11-!- Ivanovic [~ivanovic@wesnoth/developer/ivanovic] has joined #wesnoth-dev
20101009 10:59:45< Ivanovic> moin
20101009 11:10:21-!- stikonas [~and@wesnoth/translator/stikonas] has joined #wesnoth-dev
20101009 11:13:07-!- Appleman1234 [~Appleman1@CPE-60-226-176-19.qld.bigpond.net.au] has quit [Ping timeout: 240 seconds]
20101009 11:13:47-!- Appleman1234 [~Appleman1@CPE-60-226-176-19.qld.bigpond.net.au] has joined #wesnoth-dev
20101009 11:18:23-!- zookeeper [~l@wesnoth/developer/zookeeper] has joined #wesnoth-dev
20101009 11:27:56-!- loonybot [~loonybot@wesnoth/bot/loonybot] has joined #wesnoth-dev
20101009 11:29:14-!- loonycyborg [~sergey@wesnoth/developer/loonycyborg] has joined #wesnoth-dev
20101009 11:32:06< Ivanovic> alink: hui, this is *fantastic*
20101009 11:32:11< Ivanovic> i see no more glitches atm!
20101009 11:32:42< Ivanovic> speed appears to be unchanged, so scrolling in gallium ain't too fast, using classic is perfectly possible
20101009 11:33:04< Ivanovic> eg isle of the damned is at basically fluent framerates over here in 1920x1200
20101009 11:33:26-!- Appleman1234 [~Appleman1@CPE-60-226-176-19.qld.bigpond.net.au] has quit [Ping timeout: 240 seconds]
20101009 11:33:39< Ivanovic> though the software mouse cursor does come with some impact on performance (that is the movement of it is more stuttering than hardware cursor)
20101009 11:33:56< boucman> Ivanovic: at one point before merging the branch we might want to do a 1.9.x-gl release to have all packagers check that ogl is not a problem
20101009 11:34:29< Ivanovic> boucman: i think that alink will do a whole lot more polising and testing that it really works before we can even consider merging it back
20101009 11:34:45< Ivanovic> but yeah, we should create a 1.9.x-gl release eventually to get this testing
20101009 11:34:58< boucman> yes, my idea was more about for cross platform testing
20101009 11:35:30< Ivanovic> for real cross platform some opengl es stuff has to be added, too
20101009 11:35:40< Ivanovic> no idea how this is done best, maybe Sirp can provide some insight
20101009 11:35:51< Ivanovic> (since frogatto has both, OGL and OGL ES)
20101009 11:36:06< zookeeper> so when the ogl branch is merged to trunk, opengl will become a system requirement?
20101009 11:36:13< Ivanovic> zookeeper: yes
20101009 11:36:34< zookeeper> okay. sounds perfectly reasonable to me, just checking.
20101009 11:36:36< Ivanovic> zookeeper: though it should still run rather nicely in software mode for those that don't have a decent graphics card
20101009 11:37:08< zookeeper> really? that's nice.
20101009 11:37:36< Ivanovic> rather nicely meaning that it is playable with some stutters, but still possible to use the game if you don't go to some large resolutions and whatnot
20101009 11:37:55< Ivanovic> (who with ancient hardware, no opengl drivers and whatnot does fire a large screen?)
20101009 11:38:08-!- timotei [~timotei@wesnoth/developer/timotei] has joined #wesnoth-dev
20101009 11:38:43< timotei> hello!:D
20101009 11:38:53< Ivanovic> hi timotei
20101009 11:39:03< timotei> how are things going? I've missed you
20101009 11:39:17< Ivanovic> timotei: check out branches/ogl !!!
20101009 11:39:35< Ivanovic> still early but already working nicely, alink has done a really good job so far
20101009 11:39:51< timotei> hehe, I've seen that alink started the ogl branch (I get emails for each commit :P )
20101009 11:39:55< timotei> but will checkout soon
20101009 11:40:13< timotei> though, I'm still on windows for most of the time, so I don't know how will compiling undergo
20101009 11:40:13-!- Appleman1234 [~Appleman1@CPE-60-226-176-19.qld.bigpond.net.au] has joined #wesnoth-dev
20101009 11:40:19< Ivanovic> over here it does work really decent with the open source ati drivers
20101009 11:40:41< timotei> I don't have the guts to use those after what happened to shadowmaster and Espreon 
20101009 11:41:04< timotei> especially when the fgrlx (proprietary) drivers works more than ok
20101009 11:41:09< timotei> wokr*
20101009 11:41:26< Ivanovic> ah, i prefer the open source drivers because of their Xv behaviour
20101009 11:41:31< Ivanovic> that is less than perfect in fglrx
20101009 11:41:32< timotei> Xv?
20101009 11:41:52< timotei> what's that?
20101009 11:42:15< Ivanovic> some format to do some tiny accelerations for video output (scaling and whatnot)
20101009 11:42:32< Ivanovic> basically required to eg make use of analog tv cards and the default mode in most media players
20101009 11:42:55< Ivanovic> and scaling in tvtime is basically borked with fglrx while it works perfectly with the open source driver
20101009 11:43:18< timotei> ohh
20101009 11:43:22< timotei> nice
20101009 11:43:30< timotei> well, I don't use tv cards, so until that :D
20101009 11:43:36< timotei> btw, I don't get it
20101009 11:44:00< Ivanovic> (borked meaning that overscan and cropping the edges so that only the center of the screen in 16:9 format is shown does not work)
20101009 11:44:09< timotei> why I don't get (at all) any bug reports for my plugin? Is that 'perfect'? Or not so many people use it? (are scared of eclipse and wait for the standalone version)
20101009 11:44:21< Ivanovic> most likely not many people are using it
20101009 11:44:50< Ivanovic> timotei: have you tried talking to them in the content dev sections of the forums?
20101009 11:45:02< timotei> Then I'll need to hurry finishing the standalone version, so nothing related to eclipse will appear
20101009 11:45:11< timotei> hm
20101009 11:45:15< timotei> http://forums.wesnoth.org/viewtopic.php?f=21&t=30880
20101009 11:45:16< timotei> yeah
20101009 11:45:50< timotei> 428 views...
20101009 11:46:26< ilor> Soliton: around?
20101009 11:47:49< timotei> oh yeah, and the fact I use 1.9
20101009 11:47:53< timotei> and that's dev release
20101009 11:48:27< Ivanovic> yeah, most content devs probably still rely on 1.8 and will continue doing so for a while
20101009 11:50:49< timotei> wow, so at last it seems I've found out. The autocrlf=input in git windows, will break eventually the git working copy
20101009 11:50:57< timotei> with a checksum mismatch
20101009 11:51:52-!- boucman [~rosen@wesnoth/developer/boucman] has quit [Ping timeout: 265 seconds]
20101009 11:51:59-!- FAAB1 [~huajie@219.142.235.201] has quit [Quit: Leaving.]
20101009 12:02:05-!- Unnheulu [~ieuan@cpc1-pnth2-2-0-cust788.5-2.cable.virginmedia.com] has quit [Quit: Ex-Chat]
20101009 12:03:33-!- boucman [~rosen@wesnoth/developer/boucman] has joined #wesnoth-dev
20101009 12:07:13-!- timotei [~timotei@wesnoth/developer/timotei] has quit [Read error: Connection reset by peer]
20101009 12:07:43-!- timotei [~timotei@wesnoth/developer/timotei] has joined #wesnoth-dev
20101009 12:13:26-!- stikonas [~and@wesnoth/translator/stikonas] has quit [Remote host closed the connection]
20101009 12:15:15-!- stikonas [~and@bcm-131-111-247-113.girton.cam.ac.uk] has joined #wesnoth-dev
20101009 12:15:15-!- stikonas [~and@bcm-131-111-247-113.girton.cam.ac.uk] has quit [Changing host]
20101009 12:15:15-!- stikonas [~and@wesnoth/translator/stikonas] has joined #wesnoth-dev
20101009 12:34:22-!- hhyloc [~hhyloc@113.166.153.34] has joined #wesnoth-dev
20101009 12:49:16-!- Johannes13_ [~Johannes@unaffiliated/johannes13] has joined #wesnoth-dev
20101009 13:06:48< CIA-35> zookeeper * r46987 /trunk/data/core/ (terrain-graphics/new-macros.cfg terrain-graphics.cfg): Fixed some castle glitches, added a probability-affecting variation of the new wall macro and used that to convert the sunken castle to use the new macros.
20101009 13:08:48-!- timotei [~timotei@wesnoth/developer/timotei] has quit [Quit: Leaving]
20101009 13:09:44-!- timotei [~timotei@193.34.191.4] has joined #wesnoth-dev
20101009 13:09:44-!- timotei [~timotei@193.34.191.4] has quit [Changing host]
20101009 13:09:44-!- timotei [~timotei@wesnoth/developer/timotei] has joined #wesnoth-dev
20101009 13:09:46-!- timotei21 [~timotei@193.34.191.4] has joined #wesnoth-dev
20101009 13:10:09-!- timotei [~timotei@wesnoth/developer/timotei] has quit [Read error: Connection reset by peer]
20101009 13:10:09-!- timotei21 [~timotei@193.34.191.4] has quit [Read error: Connection reset by peer]
20101009 13:14:47-!- timotei [~timotei@wesnoth/developer/timotei] has joined #wesnoth-dev
20101009 13:27:40-!- Johannes13_ is now known as Johannes13
20101009 13:28:44< boucman> zookeeper: I have a couple of terrain related bugs floating around, should I assign them to you ?
20101009 13:31:32< zookeeper> boucman, sure
20101009 13:31:40< boucman> k, will do thx
20101009 13:32:10-!- Gambit [~Gambit@pa-67-234-122-243.dhcp.embarqhsd.net] has joined #wesnoth-dev
20101009 13:32:10-!- Gambit [~Gambit@pa-67-234-122-243.dhcp.embarqhsd.net] has quit [Changing host]
20101009 13:32:10-!- Gambit [~Gambit@wesnoth/developer/grickit] has joined #wesnoth-dev
20101009 13:35:50-!- EdB [~edb@70.103.89-79.rev.gaoland.net] has joined #wesnoth-dev
20101009 13:36:42< Gambit> does opensuse now send out the same broken libsdl that ubuntu did? http://forums.wesnoth.org/viewtopic.php?f=4&t=31853
20101009 13:36:59< timotei> hey there Gambit 
20101009 13:37:07< timotei> moved from ubuntu to opensuse?:))
20101009 13:37:55< Gambit> timotei: No, but people are having that "can't click any buttons" bug
20101009 13:38:18< timotei> oh
20101009 13:39:37< Gambit> (and hi)
20101009 13:53:15-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has joined #wesnoth-dev
20101009 14:10:53-!- boucman [~rosen@wesnoth/developer/boucman] has quit [Read error: Connection reset by peer]
20101009 14:11:14-!- boucman [~rosen@wesnoth/developer/boucman] has joined #wesnoth-dev
20101009 14:30:54-!- EdB [~edb@70.103.89-79.rev.gaoland.net] has quit [Remote host closed the connection]
20101009 14:31:36-!- lgn [~debian@port-92-204-27-102.dynamic.qsc.de] has joined #wesnoth-dev
20101009 14:40:48< Ivanovic> boucman: have you already checked if latest ogl branch runs on your box?
20101009 14:41:07< boucman> nope, i intend to do that after installing 10.10
20101009 14:41:11< boucman> so pretty soon :)
20101009 14:42:11< Ivanovic> ain't running nicely on my laptop with i945GM hardware
20101009 14:42:24< Ivanovic> the issue: when scrolling only the borders are updated, not the "middle"
20101009 14:44:03< Ivanovic> hmm, interesting, no such issues in fullscreen at 1400x1050 but those issues are very present in 1024x768 fullscreen as well as windowed
20101009 14:44:31< Ivanovic> in fullscreen 1400x1050 scrolling speed is bad unless there are some units in the area, really strange stuff
20101009 14:49:03-!- spetrea-home [~k23z__@188.27.126.116] has joined #wesnoth-dev
20101009 14:52:12-!- FAAB1 [~huajie@219.142.246.48] has joined #wesnoth-dev
20101009 14:57:47< Ivanovic> alink: no error messages at startup or anything like this, here is the glxinfo output for the laptop: http://pastebin.com/butGvm0X
20101009 15:04:14-!- timotei [~timotei@wesnoth/developer/timotei] has quit [Ping timeout: 264 seconds]
20101009 15:39:43-!- alink [~alink@wesnoth/developer/alink] has joined #wesnoth-dev
20101009 15:39:54< alink> hi
20101009 15:41:11< alink> yeah far away from replacing trunk, but branch is still young (<2 weeks) so I think that's ok
20101009 15:42:58< alink> a gl release would be nice at some point soon, but I have a lot of cleaning to do before that. For now, I just add error checking when a dev report a problem, but I will need more when some random window user will just say "it doesn't work" ;) 
20101009 15:44:32< alink> AFAIK OGL ES should be easy to get, I can already keep it mind if it already important (not sure what platform you are aiming)
20101009 15:46:09< alink> btw do we agree to stop supporting SDL in trunk? For the moment, because of the SDL emulator, it would be easy to have a configure switch and #ifdef SDL in source, but that will become more work with time
20101009 15:48:18-!- timotei [~timotei@193.34.191.4] has joined #wesnoth-dev
20101009 15:48:18-!- timotei [~timotei@193.34.191.4] has quit [Changing host]
20101009 15:48:18-!- timotei [~timotei@wesnoth/developer/timotei] has joined #wesnoth-dev
20101009 15:48:23-!- timotei21 [~timotei@193.34.191.4] has joined #wesnoth-dev
20101009 15:48:32< alink> Ivanovic: to explain your scrolling problem: SDL trunk and ogl SDL emulator do 2 things when scrolling: draw the new visible part and copy/paste the common part between the 2 map view.
20101009 15:49:53< alink> There are various way to do (or optimize) the copy/paste part with ogl, but my final goal is to get rid of it and redraw the whole map view when scrolling
20101009 15:50:18< timotei> I don't get it
20101009 15:50:22< timotei> xchat duplicated all stuff
20101009 15:50:54< alink> because, assuming that the goal is to have more animated terrain, then that copy/paste is often a waste, we copy/paste water and then redraw it because of the animation
20101009 15:51:08< alink> timotei: are you sure ? ;p
20101009 15:51:12< alink> timotei: are you sure ? ;p
20101009 15:51:19-!- timotei21 [~timotei@193.34.191.4] has quit [Client Quit]
20101009 15:51:22< timotei> ops
20101009 15:51:24< timotei> I know why
20101009 15:51:24< timotei> :))
20101009 15:51:31< timotei> I'm a bad pirate boy
20101009 15:51:42< timotei> brb
20101009 15:51:45-!- timotei [~timotei@wesnoth/developer/timotei] has quit [Client Quit]
20101009 15:51:55< alink> Ivanovic: but before doing that, I need decent :benchmark speed.
20101009 15:52:07-!- timotei [~timotei@193.34.191.4] has joined #wesnoth-dev
20101009 15:52:07-!- timotei [~timotei@193.34.191.4] has quit [Changing host]
20101009 15:52:07-!- timotei [~timotei@wesnoth/developer/timotei] has joined #wesnoth-dev
20101009 15:53:07< alink> well, I don't need to redraw everything for each frame, just when scrolling. So :benchmark will just be an rough indication of the fps during scrolling
20101009 15:53:41-!- timotei [~timotei@wesnoth/developer/timotei] has quit [Client Quit]
20101009 15:54:07-!- timotei [~timotei@193.34.191.4] has joined #wesnoth-dev
20101009 15:54:09-!- timotei [~timotei@193.34.191.4] has quit [Changing host]
20101009 15:54:09-!- timotei [~timotei@wesnoth/developer/timotei] has joined #wesnoth-dev
20101009 15:56:57< alink> btw parameters impacting scrolling: animated stuff (cause additional hex redraw in the middle of screen), screen resolution (more new visible hexes and bigger copy/paste) and scrolling speed (fast mean smaller copy/paste and more new visible hexes)
20101009 15:57:28< alink> I will add a switch to use full redraw when scrolling
20101009 15:58:50< alink> maybe some advanced preference or some ":gl <switch>", whatever is easier to code (only for debug/test anyway)
20101009 15:59:10< alink> but afk now, bbl
20101009 16:15:53-!- spetrea-home [~k23z__@188.27.126.116] has quit [Read error: Connection reset by peer]
20101009 16:21:34-!- Appleman1234 [~Appleman1@CPE-60-226-176-19.qld.bigpond.net.au] has quit [Ping timeout: 265 seconds]
20101009 16:23:38-!- Elvish_Pillager [~eli@dhip-164.foss.residences.colby.edu] has joined #wesnoth-dev
20101009 16:25:17< timotei> ah damn
20101009 16:25:22< timotei> can't find the todo's file :(
20101009 16:27:26< Ivanovic> alink: i asked for someone with r300 hardware in #radeon to check if ogl branch does work on their hardware, too
20101009 16:27:32< Ivanovic> [16:19:58] <Archon_> Ivanovic, I tried wesnoth with r300g and it seems to work fine
20101009 16:30:19< Ivanovic> [16:29:23] <nanonyme> Ivanovic: Is there some demo game you could use for benchmarking?
20101009 16:30:40< timotei> r300 isn't HD radeon 4570 ?
20101009 16:30:42< timotei> I have that
20101009 16:31:27< Ivanovic> hd4570 is r700 and done with the r600 driver
20101009 16:31:33< timotei> oh
20101009 16:32:31< silene> more precisely, it's rv630
20101009 16:33:41< Ivanovic> silene: no, the 45XX series should be RV710
20101009 16:33:46< silene> it's the 46xx cards that have the r700 generation
20101009 16:34:02< Ivanovic> cf http://de.wikipedia.org/wiki/ATI-Radeon-HD-4000-Serie#Modelldaten
20101009 16:35:02< silene> Ivanovic: i don't find the 4570 in your list
20101009 16:35:21< Ivanovic> silene: it should be a 4550 with more clockspeed
20101009 16:37:39< silene> Ivanovic: amd seems to agree with me: http://developer.amd.com/gpu_assets/ATI_Device_ID_List_July_2010.txt
20101009 16:38:13< timotei> :((
20101009 16:38:19< timotei> I've lost my file with the eclipse plugin ideas
20101009 16:38:19< timotei> damn
20101009 16:40:22< Ivanovic> {“ ATI Radeon HD 4300/4500 Series , R710, 954F “}
20101009 16:40:36< silene> {“ ATI Radeon HD 4570    , RV630 ,9598 “} 
20101009 16:40:46< Ivanovic> heh, intereresting stuff...
20101009 16:44:29< Ivanovic> [16:42:53] <Archon_> Ivanovic, the battle of wesnoth text that appears on first screen is painfully slow
20101009 16:45:45< silene> nothing new; alink fixed it, but mordante reverted the patch
20101009 16:52:42< Gambit> D:
20101009 16:59:35-!- crimson_penguin [~ben@wesnoth/developer/crimsonpenguin] has joined #wesnoth-dev
20101009 17:09:29-!- Blueblaze [~Blueblaze@adsl-99-171-160-107.dsl.hstntx.sbcglobal.net] has joined #wesnoth-dev
20101009 17:27:57-!- hhyloc [~hhyloc@113.166.153.34] has quit [Quit: Leaving]
20101009 17:45:34< Ivanovic> silene: with the ogl branch it is decent with r600c but really slow for r600g
20101009 17:46:02< alink> Ivanovic: if you start recruiting external testers, let me tag some micro-release thing, where I fix the most obvious errors and have decent error logging. Also, assuming a limited numbers of testing, I prefer more on a finished feature than between 2 WiP commits
20101009 17:46:41< Ivanovic> alink: ah, this is just to get an idea if it works at all with hardware that i don't have any access to
20101009 17:47:26< Ivanovic> and most people able to get stuff compiled and do so from some svn branch are aware that things are most likely not perfect and can include *many* bugs
20101009 17:47:42< timotei> loonycyborg: hey
20101009 17:47:54< timotei> loonycyborg: it's compiling of ogl branch supported on windows?
20101009 17:48:00-!- shadowm_laptop [ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20101009 17:48:08< alink> Ivanovic: yeah, that is a good thing, just let's pick the best commit for test
20101009 17:48:18< Ivanovic> :)
20101009 17:48:21< loonycyborg> timotei: Not sure. Haven't tried yet.
20101009 17:48:24< alink> btw could be nice that such micro release would be build for Mac and Windows and posted on wesnoth/hidden, so other #wesnoth-dev users could test it
20101009 17:48:28< Ivanovic> i'd say that current head appears to be rather good
20101009 17:48:38< Ivanovic> (ogl head that is)
20101009 17:48:48< Ivanovic> maybe with trunk merged in so that all data is really recent
20101009 17:48:52< shadowmaster> timotei: I'm still running with radeon KMS and it hasn't burned my hardware yet
20101009 17:48:58< shadowmaster> ;)
20101009 17:49:12< timotei> shadowmaster: well.. I've had really enough bad stuff with notebooks this holiday
20101009 17:49:18< timotei> and I don't want to put them to risk anymore
20101009 17:49:19< timotei> :D
20101009 17:49:26< alink> Ivanovic: well, the FBO stuff is really fresh, and Espreon already spotted some stupid error (now fixed)
20101009 17:49:44< timotei> for a simple dell bios upgrade, I've thrown my notebook 6 times in service
20101009 17:49:56< Ivanovic> alink: i ain't able to start when using LIBGL_ALWAYS_INDIRECT=1 wesnoth-ogl
20101009 17:50:21< alink> Ivanovic: error message ?
20101009 17:50:30< Ivanovic> http://pastebin.com/QEhT17FD
20101009 17:50:40< Ivanovic> exactly the same when setting sw rendering to classic or gallium
20101009 17:51:17< alink> hehe as I said FBO is still very new ;)
20101009 17:51:33< Ivanovic> this is glxinfo when setting this somehow software rendering mode: http://pastebin.com/DamSna8U
20101009 17:53:03< Ivanovic> got to hunt for some food now, should be back in about an hour
20101009 17:53:14< alink> I will move that FBO code in proper functions and check all the OpenGL checking required by the specification (instead of just make it work on mine)
20101009 17:53:15< Ivanovic> (though i am not sure how much i will be around this evening)
20101009 17:54:22< alink> Ivanovic: btw for you scrolling bug with fullscreen, perhaps check if there is difference when starting up, instead of switching at runtime. Maybe I don't clean well between resolution change
20101009 17:54:41< Ivanovic> scrolling bug in fullscreen?
20101009 17:54:50< Ivanovic> in fullscreen at 1400x1050 it does *not* happen
20101009 17:54:51< alink> on your laptop, IIRC
20101009 17:54:59< Ivanovic> but when starting windowed in 1024x768 it happens
20101009 17:55:17< Ivanovic> when switching to fullscreen 1024x768 (since there was no other config yet) it happens, too
20101009 17:55:31< Ivanovic> after switching to 1400x1050 while in fullscreen things are rendered correctly
20101009 17:56:30< alink> ok, so it happens at start too, that's what I needed to know
20101009 17:57:26< Ivanovic> it happens when starting windowed, yeah
20101009 18:03:59< loonycyborg> alink: I'm getting compile errors when cross-compiling gl branch.
20101009 18:04:06< loonycyborg> src/sdl_utils.cpp: In constructor 'texture::texture(SDL_Surface*)':
20101009 18:04:12< loonycyborg> src/sdl_utils.cpp:69: error: 'GL_BGRA' was not declared in this scope
20101009 18:04:49< alink> GL_RGBA need some mininum OpenGL version IIRC
20101009 18:04:51< loonycyborg> Mingw *does* offer gl headers, they're just very old.
20101009 18:04:58< alink> I mean GL_BGRA
20101009 18:05:50< loonycyborg> iirc they're 1.1
20101009 18:06:04< alink> "GL_BGRA are available only if the GL version is 1.2 or greater"
20101009 18:07:17< alink> I could maybe change that, but it's currently fixed by the SDL code, and a lot of our SDL functions rely on BGRA
20101009 18:08:23< alink> also, IIRC some hardware prefer RGBA (and translation has a speed cost), so maybe it's worth it having a switch for that
20101009 18:09:29< alink> loonycyborg: Replacing it by GL_RGBA will works, but with weird colors. Still good for tests, though
20101009 18:12:36< loonycyborg> Even with that src/video.cpp:330: error: 'glDeleteFramebuffersEXT' was not declared in this scope
20101009 18:12:41< alink> I suppose I could use #ifdef and version checking, and add conversion code, it will just slow texture loading
20101009 18:13:23< loonycyborg> Either use OpenGL's extension mechanism to get FBO stuff or use GLEW.
20101009 18:13:30< alink> yeah very old OpenGL header might choke on FBO function
20101009 18:13:50< alink> I like GLEW but it will be even more work for builder :-/
20101009 18:14:42< loonycyborg> iirc it's not only mingw, microsoft's headers are 1.1 too.
20101009 18:14:56< silene> alink: i don't think it is just a matter of old headers; i may be wrong, but i seem to remember that you are supposed to always call ext functions through the ext mechanism, never directly
20101009 18:15:19< alink> loonycyborg: probably for its DirectX propaganda ;)
20101009 18:15:42< alink> silene: ok I will check
20101009 18:16:18< alink> it's true that with OpenGL 3.0 it's part of core, I think, so things are simpler only for that
20101009 18:17:00< alink> I only used the "#define GL_GLEXT_PROTOTYPES" for now
20101009 18:17:38< alink> which is needed before the #include <GL/gl.h> if it's relevant for building
20101009 18:18:50< alink> hmm in fact I should add it in sdl_utils.cpp too
20101009 18:20:29< CIA-35> alink * r46988 /branches/ogl/src/sdl_utils.cpp: Define GL_GLEXT_PROTOTYPES everywhere
20101009 18:33:07< alink> hmmm maybe just need to include <GL/glext.h> and get a recent version of glext.h (http://www.opengl.org/registry/api/glext.h) in mingw\include\GL\GLext.h
20101009 18:33:18< alink> but that's only for header, though
20101009 18:33:34< alink> anyway, sorry need to go, I investigate more later
20101009 18:42:38-!- Elvish_Pillager [~eli@dhip-164.foss.residences.colby.edu] has quit [Ping timeout: 264 seconds]
20101009 18:54:41-!- alink [~alink@wesnoth/developer/alink] has quit [Remote host closed the connection]
20101009 19:03:47-!- Aethaeryn [~Michael@wesnoth/umc-dev/developer/aethaeryn] has joined #wesnoth-dev
20101009 19:03:49-!- CIA-35 [~CIA@208.69.182.149] has quit [Ping timeout: 252 seconds]
20101009 19:25:01-!- CIA-88 [~CIA@208.69.182.149] has joined #wesnoth-dev
20101009 19:30:03< fendrin> zookeeper: Do you see any problem with delaying the moves left till after the attack event is fired?
20101009 19:30:17< fendrin> delaying the moves left reduction to zero
20101009 19:35:36< zookeeper> yes, the fact that moves are zeroed before the attack events are fired?
20101009 19:35:47< zookeeper> err
20101009 19:35:50< fendrin> Yeah, that is the question.
20101009 19:35:51< zookeeper> what are you asking?
20101009 19:36:19< zookeeper> oh, right. no, i guess that'd be fine. but it's still silly if you want to do that only because you're trying to make the ability which doesn't make any sense.
20101009 19:36:21< fendrin> I want to alter the c++ code so that in the attack event the unit still has it's moves left.
20101009 19:36:44< zookeeper> i've wished for that feature a couple of times myself
20101009 19:37:39< fendrin> I can see many more use cases for this. Maybe an ability that gives extra damage for having moves left.
20101009 19:37:56< fendrin> Not that I suggest anything like this for mainline.
20101009 19:38:23< fendrin> Except usage in UtBS which has always been a special campaign with altered gameplay.
20101009 19:39:28< Rhonda> shadowmaster: Do you know which port sirp wanted to use?
20101009 19:39:44< shadowmaster> no. I thought he messaged you.
20101009 19:40:40< Rhonda> Yes, but didn't mention a port, and I don't find anything running currently listening on a udp port.
20101009 19:41:54< silene> fendrin: you want a unit to not consume its movement points when it attacks? then why don't you set movement_used=0 for the attack instead of modifying the engine
20101009 19:42:36< fendrin> silene: Let me check. I hadn't noticed that attributed.
20101009 19:42:55 * shadowmaster isn't sure how someone would request a port to be opened without specifying the port number :\
20101009 19:43:10< timotei> maybe it's talking about THE port
20101009 19:43:54< fendrin> silene: Looks fine but does it avoid the attacking unit being zoced after the attack?
20101009 19:44:26< zookeeper> ...
20101009 19:44:41< zookeeper> units don't get ZoCed after they've attacked.
20101009 19:45:00< fendrin> But they get zoced when they move in to attack, don't they?
20101009 19:45:25< silene> fendrin: no, that's orthogonal; and modifying the attack event won't change anything, since it isn't related to attacks
20101009 19:46:22< fendrin> Refreshing their movement points like discussed yesterday does give the unit a full new move, not being zoced. So the modified event can be used for what I want to achieve.
20101009 19:47:19 * zookeeper orders what fendrin has had
20101009 19:47:27< silene> how you will refresh the movement points? by putting the value you read during the attack event? it will be zero!
20101009 19:48:02< fendrin> Right, that is what I want to change.
20101009 19:48:24< silene> fendrin: you don't understand; even after your change, it will still be zero!
20101009 19:48:45< silene> the unit didn't lose its move points because it attacked; it lost them because it moved in a zoc
20101009 19:49:06< silene> come on, you know that if you move in a zoc, even if the zoc disappears later, your unit will still have zero points
20101009 19:49:14< silene> that doesn't have anything to do with attacks!
20101009 19:49:46< fendrin> Skirmishing is handled extra? The unit ignores zoc but still can't move after the attack.
20101009 19:50:28< silene> fendrin: your solution would only work for skirmishers, but movement_used=0 also works for skirmisher, no need to modify the engine
20101009 19:54:46< fendrin> zookeeper: I had basically an attack_end event that gave the unit some extra moves. The only problem I had was determining the moves left but with a constant it worked quite fine. The unit is not a skirmisher but could move the newly gained moves. What silene told about how movement works does explain that perfectly.
20101009 19:55:38< fendrin> silene: Still let me ask why skirmishing units can't move after the attack, is that hardcoded?
20101009 19:56:20< fendrin> s/can't/can
20101009 19:56:41< fendrin> sorry, ignore the last line, I am getting confused.
20101009 19:57:32< silene> fendrin: movement_used=10000, how could they move after attacking?!
20101009 19:59:40< fendrin> So there are two systems active, one that leeches all movement points when entering zoc, what is a prediction for performing an attack. If the unit is a skirmisher the movement_used does leech the moves left to zero. So movement_used=0 will have no effect on non skirmishers. I think I am slowly going to get how it works.
20101009 20:00:07-!- Mica [~Mica@d27-96-76-1.nap.wideopenwest.com] has joined #wesnoth-dev
20101009 20:00:35< fendrin> For the except that some units don't have a zone of control.
20101009 20:01:12< fendrin> So a non skirmisher with an attack that uses movement_used=0 will still be able to move after attacking a walking corpse or goblin for example.
20101009 20:01:52< zookeeper> presumably movement_used=0 will have effect on non-skirmishers who don't move before they attack.
20101009 20:02:03< silene> exactly
20101009 20:02:08< fendrin> Ah yes.
20101009 20:02:43< fendrin> So I guess movement_used=0 is not doing well for me.
20101009 20:03:37< fendrin> I will need to go with modifying the moves manually. And it seems I can only go with a constant because there is no chance of getting a value != 0 for moves within any event.
20101009 20:03:59-!- Blueblaze [~Blueblaze@adsl-99-171-160-107.dsl.hstntx.sbcglobal.net] has quit [Ping timeout: 240 seconds]
20101009 20:05:24< fendrin> After moving in combat position of course and if the attacked unit does have zoc.
20101009 20:11:11-!- Blueblaze [~Blueblaze@adsl-99-186-65-218.dsl.hstntx.sbcglobal.net] has joined #wesnoth-dev
20101009 20:12:25-!- Mica [~Mica@d27-96-76-1.nap.wideopenwest.com] has quit [Quit: Good night]
20101009 20:13:53< fendrin> silene: Thank you for your explanation, saved my a lot of time reading code.
20101009 20:24:05-!- alink [~alink@wesnoth/developer/alink] has joined #wesnoth-dev
20101009 20:26:05-!- SpoOkyMagician [~chatzilla@74-138-202-211.dhcp.insightbb.com] has joined #wesnoth-dev
20101009 20:27:44-!- Mickcy [~mickcy@ip13-170-173-82.adsl2.static.versatel.nl] has joined #wesnoth-dev
20101009 20:38:33-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20101009 20:45:25< Gambit> alink: http://forums.wesnoth.org/viewtopic.php?f=4&t=31858
20101009 20:45:36< Gambit> Wouldn't your fix have been in 1.9.1?
20101009 20:45:49< Gambit> IIRC it was between 1.9.0 and 1.9.1
20101009 20:46:40< Gambit> Err nevermind. It's broken here in trunk too.
20101009 21:05:43-!- Elvish_Pillager [~eli@dhip-164.foss.residences.colby.edu] has joined #wesnoth-dev
20101009 21:25:24-!- rigved [~rigved@116.72.163.243] has joined #wesnoth-dev
20101009 21:46:52< alink> Gambit: no, I discussed about a possible fix (which was working IIRC) here with someone else (you?). But I thought it was assigned to boucman, so I forgot about it. I will assign it to me and retry my idea since in the mean time I worked on this code (for performance and randomization reasons)
20101009 21:47:32< alink> it was https://gna.org/bugs/?16529
20101009 21:47:34-!- timotei21 [~timotei@193.34.191.4] has joined #wesnoth-dev
20101009 21:47:36< Gambit> alink: Ah. Yeah that was me. I remember you giving me a patch that fixed it. I guess I just assumed it made it to mainline.
20101009 21:48:07-!- Octalot [~noct@host86-166-163-240.range86-166.btcentralplus.com] has joined #wesnoth-dev
20101009 21:50:03< alink> ah I thought it was you, I was just confused by "grickit", but I see now that it's you too
20101009 21:51:52-!- timotei [~timotei@wesnoth/developer/timotei] has quit [Ping timeout: 264 seconds]
20101009 21:53:52-!- silene [~plouf@wesnoth/developer/silene] has quit [Quit: Leaving.]
20101009 22:00:22-!- alink [~alink@wesnoth/developer/alink] has quit [Remote host closed the connection]
20101009 22:11:18-!- FAAB1 [~huajie@219.142.246.48] has quit [Quit: Leaving.]
20101009 22:16:54-!- SpoOkyMagician [~chatzilla@74-138-202-211.dhcp.insightbb.com] has quit [Quit: be back later]
20101009 22:26:54-!- Dalton [~~@warzone2100/moderator/Dalton] has quit []
20101009 22:33:50-!- Dalton [~~@warzone2100/moderator/Dalton] has joined #wesnoth-dev
20101009 22:34:08-!- rigved [~rigved@116.72.163.243] has quit [Quit: Bye]
20101009 22:42:49-!- SpoOkyMagician [~chatzilla@74-138-202-211.dhcp.insightbb.com] has joined #wesnoth-dev
20101009 22:51:36< CIA-88> timotei * r46989 /trunk/utils/java/org.wesnoth/ (5 files in 2 dirs): eclipse plugin: first try to remove the run and search menus``
20101009 22:53:29-!- timotei21 [~timotei@193.34.191.4] has quit [Quit: Leaving]
20101009 22:54:06< GNUtoo|laptop> hi, I've a bit off-topic question: I've a font issue with my system, that affected wesnoth too, here: http://scap.linuxtogo.org/files/ed08dba6b6953655225b85ba29df3631.png on bottom-left there is a box in wesnoth
20101009 22:54:12< GNUtoo|laptop> what font system is used ?
20101009 22:54:17< GNUtoo|laptop> truetype?
20101009 23:11:00< GNUtoo|laptop> sorry but I'll go to sleep, I'll be back tomorrow
20101009 23:11:09-!- GNUtoo|laptop [~gnutoo@host227-139-dynamic.211-62-r.retail.telecomitalia.it] has quit [Quit: Program received signal SIGSEGV, Segmentation fault.]
20101009 23:28:08< CIA-88> espreon * r46990 /trunk/utils/java/ (208 files in 48 dirs): Ran dos2unix and umcpropfix.
20101009 23:54:55-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: GO, GET TO THE CHOPPAH!!!]
20101009 23:55:08-!- Upthorn [~ogmar@69.62.144.108] has quit [Ping timeout: 245 seconds]
20101009 23:55:26-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20101009 23:57:19-!- Mickcy [~mickcy@ip13-170-173-82.adsl2.static.versatel.nl] has left #wesnoth-dev []
--- Log closed Sun Oct 10 00:00:02 2010