--- Log opened Wed Feb 10 00:00:37 2010
20100210 00:01:05-!- deekay [~dk@wesnoth/developer/dragonking] has quit [Quit: deekay]
20100210 00:10:18-!- boucman [~rosen@wesnoth/developer/boucman] has joined #wesnoth-dev
20100210 00:19:13-!- elias [~elias@allegro/developer/allefant] has quit [Quit: Leaving]
20100210 00:20:24-!- YogiHH [~chatzilla@wesnoth/developer/yogihh] has left #wesnoth-dev []
20100210 00:22:16-!- boucman [~rosen@wesnoth/developer/boucman] has quit [Quit: Leaving.]
20100210 00:34:24-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20100210 00:36:00-!- Espreon [~espreon@wesnoth/developer/espreon] has joined #wesnoth-dev
20100210 00:39:19< Ivanovic> crimson_penguin: okay, no idea what went wrong on this system
20100210 00:39:24< Ivanovic> anyway, got to head off to bed now, n8
20100210 00:43:19< crimson_penguin> Ivanovic: yeah, odd
20100210 00:43:20< crimson_penguin> godnatt
20100210 00:58:05-!- Noyga [~noyga@wesnoth/developer/noyga] has quit [Remote host closed the connection]
20100210 01:05:44-!- noy [~Noy@wesnoth/developer/noy] has quit [Read error: Connection reset by peer]
20100210 01:07:59-!- k23z__ [~k23z__@86.121.72.119] has joined #wesnoth-dev
20100210 01:19:04-!- migge [~marc@ip-95-223-247-31.unitymediagroup.de] has quit [Quit: Ex-Chat]
20100210 01:24:38-!- Chusslove [~Chusslove@brsg-d9befba0.pool.mediaWays.net] has quit [Remote host closed the connection]
20100210 01:26:20-!- Chusslove [~Chusslove@brsg-d9befba0.pool.mediaWays.net] has joined #wesnoth-dev
20100210 02:01:04-!- Blarumyrran [~Blarumyrr@81-20-159-197.levira.ee] has quit [Quit: Lahkun]
20100210 02:17:19-!- meric_ [~Eric@124-168-253-152.dyn.iinet.net.au] has joined #wesnoth-dev
20100210 02:20:13-!- meric [~Eric@124-170-229-106.dyn.iinet.net.au] has quit [Ping timeout: 260 seconds]
20100210 02:20:13-!- meric_ is now known as meric
20100210 02:43:59-!- stikonas [~and@wesnoth/translator/stikonas] has quit [Read error: Connection reset by peer]
20100210 02:57:39-!- loonycyborg [~sergey@wesnoth/developer/loonycyborg] has quit [Quit: Zzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzz]
20100210 02:58:22-!- loonybot [~loonybot@wesnoth/bot/loonybot] has quit [Remote host closed the connection]
20100210 03:14:27-!- meric [~Eric@124-168-253-152.dyn.iinet.net.au] has quit [Quit: meric]
20100210 03:15:29-!- Mythological [Mythologic@77.28.122.19] has quit [Read error: Connection reset by peer]
20100210 03:50:16-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20100210 05:24:55-!- k23z__ [~k23z__@86.121.72.119] has quit [Quit: Leaving]
20100210 05:34:13-!- meric [~Eric@124-168-253-152.dyn.iinet.net.au] has joined #wesnoth-dev
20100210 06:00:24-!- Zarel [~Zarel@warzone2100/developer/Zarel] has joined #wesnoth-dev
20100210 06:06:53-!- meric [~Eric@124-168-253-152.dyn.iinet.net.au] has quit [Read error: Connection reset by peer]
20100210 06:07:15-!- meric [~Eric@124-168-144-138.dyn.iinet.net.au] has joined #wesnoth-dev
20100210 06:31:27-!- rosso_ [~rosso@dslb-088-070-233-077.pools.arcor-ip.net] has quit [Read error: Operation timed out]
20100210 06:34:05-!- rosso [~rosso@dslb-088-070-072-111.pools.arcor-ip.net] has joined #wesnoth-dev
20100210 06:43:39-!- noy [~Noy@wesnoth/developer/noy] has quit [Read error: Connection reset by peer]
20100210 06:44:36-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20100210 06:51:15-!- meric [~Eric@124-168-144-138.dyn.iinet.net.au] has quit [Quit: meric]
20100210 07:19:27-!- Zarel [~Zarel@warzone2100/developer/Zarel] has quit [Read error: Connection timed out]
20100210 07:20:26-!- Zarel [~Zarel@warzone2100/developer/Zarel] has joined #wesnoth-dev
20100210 07:22:45-!- rosso [~rosso@dslb-088-070-072-111.pools.arcor-ip.net] has quit [Read error: Operation timed out]
20100210 07:23:13-!- rosso [~rosso@dslb-088-070-084-020.pools.arcor-ip.net] has joined #wesnoth-dev
20100210 07:27:16-!- crimson_penguin [~ben@wesnoth/developer/crimsonpenguin] has quit [Quit: crimson_penguin]
20100210 08:01:28-!- Zarel [~Zarel@warzone2100/developer/Zarel] has quit [Quit: This computer has gone to sleep]
20100210 08:05:49-!- meric [~Eric@124-168-144-138.dyn.iinet.net.au] has joined #wesnoth-dev
20100210 08:18:41-!- zookeeper [~l@wesnoth/developer/zookeeper] has joined #wesnoth-dev
20100210 08:51:49-!- meric [~Eric@124-168-144-138.dyn.iinet.net.au] has quit [Quit: meric]
20100210 09:01:58-!- stikonas [~and@wesnoth/translator/stikonas] has joined #wesnoth-dev
20100210 09:26:25-!- rosso_ [~rosso@dslb-088-070-068-169.pools.arcor-ip.net] has joined #wesnoth-dev
20100210 09:28:18-!- rosso [~rosso@dslb-088-070-084-020.pools.arcor-ip.net] has quit [Ping timeout: 248 seconds]
20100210 09:31:17-!- stikonas [~and@wesnoth/translator/stikonas] has quit [Remote host closed the connection]
20100210 10:26:38-!- Espreon [~espreon@wesnoth/developer/espreon] has quit [Remote host closed the connection]
20100210 10:40:42-!- noy [~Noy@wesnoth/developer/noy] has quit [Ping timeout: 240 seconds]
20100210 10:42:28< Ivanovic> moin
20100210 10:51:04-!- mjs-de [~mjs-de@vpw.wh.uni-dortmund.de] has joined #wesnoth-dev
20100210 11:23:00-!- loonybot [~loonybot@ppp79-139-136-187.pppoe.spdop.ru] has joined #wesnoth-dev
20100210 11:23:00-!- loonybot [~loonybot@ppp79-139-136-187.pppoe.spdop.ru] has quit [Changing host]
20100210 11:23:00-!- loonybot [~loonybot@wesnoth/bot/loonybot] has joined #wesnoth-dev
20100210 11:23:59-!- loonycyborg [~sergey@wesnoth/developer/loonycyborg] has joined #wesnoth-dev
20100210 12:30:29< CIA-62> zookeeper * r41136 /trunk/data/campaigns/Under_the_Burning_Suns/scenarios/03_Stirring_in_the_Night.cfg: Prevented Garak's champion boost from changing his sword from arcane to fire in case he picked up the holy water in the previous scenario.
20100210 12:36:22< fabi> zookeeper, should the holy water not be limited to a single scenario in its affect?
20100210 12:36:42< zookeeper> fabi, no, it's the one you get from the second scenario's leader
20100210 12:38:07< fabi> zookeeper, I think Garak's death is some kind of tomato surprise. What do you think about preventing the player to give him any items/bonuses?
20100210 12:46:08< zookeeper> fabi, i guess it could be done. but it'd at least need some new strings if it was to be done in a nice manner
20100210 12:47:52< zookeeper> an alternative would be to allow the player to pick up whatever items he has been given, after he's death
20100210 12:48:23< Appleman1234> alternative sounds better
20100210 12:48:42< fabi> Right, better idea. But it will also need string changes.
20100210 12:49:41< zookeeper> yeah
20100210 12:50:14< zookeeper> but his arcane melee should still be slightly useful when he goes berserk, so it's not a total loss if you've happened to give the holy water to him
20100210 12:50:49< zookeeper> having characters permanently removed is always annoying to try to deal with
20100210 12:51:41< zookeeper> the best way IMO to handle it would be to have the character constantly nag about how he's sure he's gonna die soon and how you shouldn't waste your time on him, but that doesn't suit garak :P
20100210 12:52:02< zookeeper> anyway
20100210 12:53:01< zookeeper> wtf is this? the green undead gets a free shadow/nightgaunt for free at the beginning, meaning that it'll take the nearest player village unless the player blocks it
20100210 12:53:22< zookeeper> ...and since it's night, the player can't see that there's a shadow/nightgaunt there
20100210 12:53:26< zookeeper> -.-
20100210 12:54:57< zookeeper> on easy, it's a wrait
20100210 12:54:59< zookeeper> h
20100210 12:55:54< fabi> What is the problem with that?
20100210 12:56:57< zookeeper> with an invisible enemy you couldn't know was there at all taking your village(s) in a scenario where protecting the villages is your objective? :x
20100210 12:57:31< zookeeper> on my first turn 2 i thought it was a _bug_ when no enemy has taken my village, yet the dialogue suggested that and there was even a WC appearing out of nowhere
20100210 12:58:36< fabi> so you want to disallow all nightstalking units for the scenario?
20100210 12:59:03-!- Appleman1234 [~Appleman1@CPE-124-191-177-16.oxqn1.cha.bigpond.net.au] has quit [Read error: Connection reset by peer]
20100210 12:59:26< zookeeper> preferably, yes, because there's no way for the player to know if there are nightstalkers or not
20100210 12:59:49< zookeeper> but at least it's stupid to give the enemy a free nightstalker on the first turn
20100210 13:00:41< zookeeper> but what should i make it? wraith is fine for easy and normal, but on hard it should be a lvl3 (currently nightgaunt; the other undead pet is a draug)
20100210 13:00:50< zookeeper> the leader itself is a spectre, so that doesn't really fit..
20100210 13:01:01-!- Appleman1234 [~Appleman1@CPE-124-191-177-16.oxqn1.cha.bigpond.net.au] has joined #wesnoth-dev
20100210 13:01:11< zookeeper> ah well, easy fix, i'll see if i can just give it [status] hidden=no or whatever it's called now.
20100210 13:01:36< fabi> zookeeper, let it stay as it is. We can warn the player with a string after the frezze. 
20100210 13:02:41< fabi> zookeeper: yes, that is a good idea. But it will be invisible the turns that follow?
20100210 13:03:55< zookeeper> nah, didn't work, it seems.
20100210 13:13:00< CIA-62> fendrin * r41137 /trunk/data/campaigns/Legend_of_Wesmere/scenarios/19_Costly_Revenge.cfg: 
20100210 13:13:00< CIA-62> LoW 19:
20100210 13:13:00< CIA-62> Fixed the event order to do the outro speech before units get removed.
20100210 13:13:00< CIA-62> Set victory_when_enemies_defeated=no to make sure every saurian unit and
20100210 13:13:00< CIA-62> village gets destroyed.
20100210 13:46:56-!- stikonas [~and@bcm-131-111-247-5.girton.cam.ac.uk] has joined #wesnoth-dev
20100210 13:46:56-!- stikonas [~and@bcm-131-111-247-5.girton.cam.ac.uk] has quit [Changing host]
20100210 13:46:56-!- stikonas [~and@wesnoth/translator/stikonas] has joined #wesnoth-dev
20100210 14:12:01< Ivanovic> stikonas: you saw the update in the frogatto bugreport at freedesktop?
20100210 14:12:10< Ivanovic> the patch fixes the issues for me
20100210 14:12:27< stikonas> not yet
20100210 14:13:55< Ivanovic> speed is lovely on my 3850
20100210 14:15:39-!- meric [~Eric@124-171-36-3.dyn.iinet.net.au] has joined #wesnoth-dev
20100210 14:17:43< stikonas> indeed, it works nicely
20100210 14:22:06< Ivanovic> then i'd say "leave a comment and hope that it will be merged to 'upstream' shortly"
20100210 14:22:07< Ivanovic> ;)
20100210 14:28:27-!- ^y_emkidzu-it [~KVIrc@host37-65-dynamic.10-79-r.retail.telecomitalia.it] has joined #wesnoth-dev
20100210 14:35:03< stikonas> should I mark it as fixed?
20100210 14:35:12< stikonas> Ivanovic: ^^
20100210 14:46:47-!- meric [~Eric@124-171-36-3.dyn.iinet.net.au] has quit [Quit: meric]
20100210 14:49:31-!- valgor [~596fb03c@gateway/web/freenode/x-ailcukcqzcznzmfl] has joined #wesnoth-dev
20100210 15:05:01< Ivanovic> stikonas: as long as the patch is not commited upstream, it is not fixed!
20100210 15:05:36< Ivanovic> FIXED:      A fix for this bug is checked into the tree and tested. 
20100210 15:05:40< Ivanovic> cf https://bugs.freedesktop.org/page.cgi?id=fields.html#status
20100210 15:07:18< stikonas> OK, I agree. Though it doesn't says which tree ;). In git there are many trees.
20100210 15:08:39< Ivanovic> i vote for master
20100210 15:08:43< Ivanovic> and it is not in there yet!
20100210 15:08:45< Ivanovic> ;)
20100210 15:13:03< fabi> Ivanovic, do you release today?
20100210 15:13:26< Ivanovic> hmm, good question
20100210 15:13:29< Ivanovic> might be
20100210 15:14:14< Ivanovic> AI0867, Chusslove, esr, fabi, loonycyborg, shadowmaster, Soliton, zookeeper, everyone else who cares: what is left in your pipe that could/should be included in 1.7.13?
20100210 15:15:49-!- |^y_emkidzu-it| [~KVIrc@host249-66-dynamic.5-87-r.retail.telecomitalia.it] has joined #wesnoth-dev
20100210 15:15:56< esr> Ivanovic: Clear from here.
20100210 15:16:57< zookeeper> Ivanovic, i've got nothing important, i think
20100210 15:17:28< fabi> Ivanovic, I am fine as well.
20100210 15:18:10-!- ^y_emkidzu-it [~KVIrc@host37-65-dynamic.10-79-r.retail.telecomitalia.it] has quit [Ping timeout: 248 seconds]
20100210 15:27:29< esr> fabi: Thanks for wmllint-cleaning WML.  It makes my general maintainance work on mainline easier
20100210 15:29:41< fabi> esr: Sure, and my work as maintainer of LoW as well. It's a pitty that most cleaning in LoW is done by disabling wmllint for that code segment.
20100210 15:30:28< esr> Sigh...that's probably storing up trouble for later, you know.
20100210 15:31:20-!- valgor [~596fb03c@gateway/web/freenode/x-ailcukcqzcznzmfl] has quit [Quit: Page closed]
20100210 15:32:49< fabi> Yes I will need to clean most parts of LoW by hand. But I don't see way how to do it otherwise. For example the MP port of LoW needs a attribute in every side definition. I can do that because I have that character macros which wmllint doesn't like by the change of a single line.
20100210 15:33:23< Ivanovic> okay, looks like at least the mp lobby is good to go atm
20100210 15:33:26< fabi> Making the code wmllint checkable would make it much harder to maintain.
20100210 15:33:50< Ivanovic> (that is: a lot better than in the previous beta versions, less jumpy (some slight jumping does still exist) and basically working
20100210 15:37:43< CIA-62> fendrin * r41138 /trunk/data/campaigns/Legend_of_Wesmere/scenarios/02_Hostile_mountains.cfg: LoW 2: Changed the team of Olurf to player in some cases.
20100210 15:38:58< fabi> esr: Please do some string transformation magic for me. I need every team_name=kalenz converted to team_name=player in LoW.
20100210 15:39:25< esr> I can do that.  Hold on...
20100210 15:41:28< CIA-62> esr * r41139 /trunk/data/campaigns/Legend_of_Wesmere/scenarios/ (10 files): Text transformation as requested by fabi.
20100210 15:41:52< fabi> esr: Thank you very much.
20100210 15:43:03< fabi> esr: Have you noticed LoW being a multiplayer campaign now? It can be played by two humans one playing kalenz one playing landar. local or on multiplayer server.
20100210 15:43:46-!- |^y_emkidzu-it| [~KVIrc@host249-66-dynamic.5-87-r.retail.telecomitalia.it] has quit [Ping timeout: 248 seconds]
20100210 15:48:22-!- ilor [~user@wesnoth/developer/ilor] has joined #wesnoth-dev
20100210 15:51:50-!- rosso [~rosso@dslb-088-070-226-178.pools.arcor-ip.net] has joined #wesnoth-dev
20100210 15:52:32-!- rosso_ [~rosso@dslb-088-070-068-169.pools.arcor-ip.net] has quit [Ping timeout: 265 seconds]
20100210 15:55:42< Ivanovic> zookeeper, esr: starting liberty i get this message:
20100210 15:55:43< Ivanovic> 20100210 15:55:23 warning unit: Unknown attribute 'outlaw_type' discarded.
20100210 15:57:29< esr> Hm.
20100210 15:57:36< esr> Looking...
20100210 16:01:17< Ivanovic> zookeeper: many messages when playing through all of the tutorial (this was using lisar and doing exactly as asked to do): http://pastebin.com/m178b708f
20100210 16:02:25< Ivanovic> starting DM i get this one: 20100210 16:02:11 warning unit: Unknown attribute 'can_recruit' discarded.
20100210 16:02:42< Ivanovic> (right after starting when the story screen appears=
20100210 16:05:22-!- Appleman1234 [~Appleman1@CPE-124-191-177-16.oxqn1.cha.bigpond.net.au] has quit [Ping timeout: 272 seconds]
20100210 16:07:41< Ivanovic> besied those messages things look okay regarding getting 1.7.13 out
20100210 16:08:00< Ivanovic> esr, fabi, zookeeper: will you have a short look and tell me if things are ready or not?
20100210 16:12:00< Ivanovic> the message in liberty might need to the units not being "converted" correctly, not sure though
20100210 16:12:27< CIA-62> fendrin * r41140 /trunk/data/campaigns/Delfadors_Memoirs/scenarios/01_overture.cfg: Fixed wrong named attribute.
20100210 16:13:02< CIA-62> fendrin * r41141 /trunk/data/campaigns/Legend_of_Wesmere/scenarios/09_Bounty_hunters.cfg: code cleanup
20100210 16:13:05< fabi> Ivanovic, fixed the warning in DM. ^
20100210 16:13:22< Ivanovic> okay
20100210 16:13:43-!- Appleman1234 [~Appleman1@CPE-124-191-177-16.oxqn1.cha.bigpond.net.au] has joined #wesnoth-dev
20100210 16:15:57< zookeeper> Ivanovic, i'll take a look in a little bit, busy for a few minutes still..
20100210 16:16:03-!- Zarel [~Zarel@warzone2100/developer/Zarel] has joined #wesnoth-dev
20100210 16:16:24< Ivanovic> sure, no problem
20100210 16:33:12< zookeeper> okay
20100210 16:33:43 * zookeeper looks
20100210 16:33:58-!- |^y_emkidzu-it| [~KVIrc@host249-66-dynamic.5-87-r.retail.telecomitalia.it] has joined #wesnoth-dev
20100210 16:35:10< zookeeper> righty, the liberty one should be easy to fix, just something the recent changes broke
20100210 16:38:03< Ivanovic> okay, would probably be good to have that fixed in 1.7.13
20100210 16:38:16< Ivanovic> ilor: what are your plans atm?
20100210 16:38:34< Ivanovic> will you have something ready "really soon", or are your improvements/fixes taking some more time?
20100210 16:39:00< ilor> I'll try to have something really soon
20100210 16:39:07< ilor> unless you mean <1hr
20100210 16:39:18< Ivanovic> that is: what are you currently working on?
20100210 16:39:48< zookeeper> Ivanovic, err, yeah, somewhat soon. i'm trying to debug two games at once here, so it's going a bit slow :x
20100210 16:39:52< Ivanovic> hmm, i mean "when the wml of liberty is fixed, so that it is more sure to work in 1.7.13" plus "when the pot-update is done (should not take too long)
20100210 16:39:58< zookeeper> but should take no more than 30 mins max
20100210 16:40:19< Ivanovic> ilor: as you see, the estimate is something like one hour
20100210 16:40:25< CIA-62> fendrin * r41142 /trunk/data/campaigns/Legend_of_Wesmere/scenarios/11_Battle_of_the_book.cfg: LoW 11: Village assignements.
20100210 16:40:46< Ivanovic> zookeeper: take your time, no need to rush and get bugs in because not being carefull
20100210 16:41:03< ilor> Ivanovic: then probably no, I did do some more testing to make sure my cleanups didn't mess things up and it looks okay
20100210 16:43:58-!- Appleman1234 [~Appleman1@CPE-124-191-177-16.oxqn1.cha.bigpond.net.au] has quit [Ping timeout: 258 seconds]
20100210 16:44:33-!- dtiger [~dtiger@dynamic-vpdn-93-125-68-222.telecom.by] has joined #wesnoth-dev
20100210 16:47:31< zookeeper> err, i wouldn't have thought that any attribute given in a [unit_type] would get discarded from single units of that type
20100210 16:47:41< zookeeper> too bad silene's not around right now
20100210 16:47:42-!- ilor [~user@wesnoth/developer/ilor] has quit [Ping timeout: 252 seconds]
20100210 16:49:11< zookeeper> i'll take back what i said about it being an easy fix: i'm not sure how i should be working around it
20100210 16:49:35< zookeeper> i need the [unit_type]s to have some kind of a custom attribute i can access from the single units
20100210 16:49:53< zookeeper> so currently there's just a made-up outlaw_type= in each [unit_type]
20100210 16:50:25< zookeeper> but if that doesn't work anymore, i need to find some other attribute i can abuse...i guess i'll try if [variables] in [unit_type] works, it might.
20100210 16:51:34-!- ilor [~user@wesnoth/developer/ilor] has joined #wesnoth-dev
20100210 16:52:58< Ivanovic> outsch
20100210 16:53:40< zookeeper> (what's the extra set of scrollbar arrows in the campaigns menu?)
20100210 16:54:15 * Ivanovic has no extra set
20100210 16:55:36< zookeeper> maybe it's been fixed already then. there was an extra set in the middle on the right side of the campaign list scrollbar.
20100210 16:57:05< zookeeper> nope, wrapping outlaw_type= in [variables] doesn't work, it's never added to the single units...
20100210 16:58:15< zookeeper> i can write some kind of an obtuse hack though, but it'll take a little bit longer
20100210 16:59:03-!- Zarel [~Zarel@warzone2100/developer/Zarel] has quit [Quit: This computer has gone to sleep]
20100210 17:00:39-!- Appleman1234 [~Appleman1@CPE-124-191-177-16.oxqn1.cha.bigpond.net.au] has joined #wesnoth-dev
20100210 17:01:20< zookeeper> heh, well, actually that was fast. let's see if it works...
20100210 17:01:20< wesbot> zookeeper: Sometimes we are fast
20100210 17:05:28< zookeeper> cool, i think my hack workaround works! i'll commit in a few minutes..
20100210 17:06:57-!- Appleman1234 [~Appleman1@CPE-124-191-177-16.oxqn1.cha.bigpond.net.au] has quit [Ping timeout: 276 seconds]
20100210 17:07:31< Ivanovic> okay
20100210 17:08:27< zookeeper> done, maybe a good idea for someone else to test it quickly too
20100210 17:08:28< CIA-62> zookeeper * r41143 /trunk/data/campaigns/Liberty/ (scenarios/02_Civil_Disobedience.cfg units/Villagers.cfg): Made the peasants-turning-to-outlaws work again (in a rather more hackish way than it was before).
20100210 17:08:54< zookeeper> that is, just kill the leader of the second scenario and see if the units turn into outlaws without trouble
20100210 17:08:59-!- mordante [~mordante@87.215.201.26] has joined #wesnoth-dev
20100210 17:09:01< mordante> servus
20100210 17:09:29< mordante> zookeeper, those extra scrollbars are the hidden tree view (at some point I fixed the issue you observe)
20100210 17:11:57< zookeeper> ok, then i guess all the treeview flood in stderr is gone too now? :P
20100210 17:12:36< zookeeper> Ivanovic, i don't really have time to look at the tutorial thing...probably nothing serious if the scenario still works right
20100210 17:17:31< mordante> zookeeper, it should be ;-)
20100210 17:17:43< mordante> off again
20100210 17:18:17-!- mordante [~mordante@87.215.201.26] has quit [Quit: Leaving]
20100210 17:20:20-!- Appleman1234 [~Appleman1@CPE-124-191-177-16.oxqn1.cha.bigpond.net.au] has joined #wesnoth-dev
20100210 17:25:03< Ivanovic> uhm, Des just sent me a PM requesting some forum title
20100210 17:25:28< Ivanovic> he is listed in the credits as "English strings edition"
20100210 17:25:39< Ivanovic> which group to add him to?
20100210 17:28:56-!- Appleman1234 [~Appleman1@CPE-124-191-177-16.oxqn1.cha.bigpond.net.au] has quit [Ping timeout: 245 seconds]
20100210 17:29:28-!- Zarel [~Zarel@warzone2100/developer/Zarel] has joined #wesnoth-dev
20100210 17:33:20< Ivanovic> zookeeper: http://pastebin.com/m2ddfb1c1
20100210 17:33:23< Ivanovic> does not work!
20100210 17:34:28-!- k23z__ [~k23z__@188.26.194.154] has joined #wesnoth-dev
20100210 17:36:04< Ivanovic> sorry, this was due to my debug mode units...
20100210 17:36:15< k23z__> is Perl used inside wesnoth ?
20100210 17:36:43< zookeeper> Ivanovic, false alarm then? all right
20100210 17:36:51< Ivanovic> exactly
20100210 17:36:52< Ivanovic> k23z__: no
20100210 17:37:14< k23z__> good
20100210 17:37:21< k23z__> what are you using for tests purposes ?
20100210 17:38:08< Ivanovic> ilor: http://forums.wesnoth.org/viewtopic.php?f=4&t=28838&start=0
20100210 17:38:57< Ivanovic> okay, releasing 1.7.13 now since things seem to basically work
20100210 17:42:09-!- elias [~elias@allegro/developer/allefant] has joined #wesnoth-dev
20100210 17:43:43< CIA-62> ivanovic * r41144 /trunk/po/ (324 files in 23 dirs): 
20100210 17:43:43< CIA-62> pot-update (no new/changed strings, reference update only; preparing for 1.7.13)
20100210 17:43:43< CIA-62> regenerated doc files (no changes here...)
20100210 17:44:35< CIA-62> zookeeper * r41145 /trunk/data/campaigns/Liberty/scenarios/02_Civil_Disobedience.cfg: Prevented units other than the custom peasant types from being turned into outlaws, meaning that debug-created units or other funny cases can't break the scenario anymore.
20100210 17:45:33< CIA-62> ivanovic * r41146 /trunk/ (changelog configure.ac players_changelog src/wesconfig.h): version bump to 1.7.13-1.8beta6
20100210 17:54:06-!- Sirp [~97c1dc1c@gateway/web/freenode/x-geiqamlsowvvethc] has joined #wesnoth-dev
20100210 17:55:20-!- crimson_penguin [~ben@wesnoth/developer/crimsonpenguin] has joined #wesnoth-dev
20100210 17:55:53-!- Ivanovic changed the topic of #wesnoth-dev to: releasing 1.7.13-1.8beta6, using r41146 | string/feature freeze active! | 63 bugs, 245 feature requests, 8 patches | logs: http://irclogs.wesnoth.org/ | Don't paste on IRC! Use a pastebin: http://wesnoth.pastebin.com | http://imagebin.org
20100210 18:00:00-!- Zarel [~Zarel@warzone2100/developer/Zarel] has quit [Quit: This computer has gone to sleep]
20100210 18:07:37-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20100210 18:08:38< CIA-62> ivanovic * r41147 /tags/1.7.13/: tagged 1.7.13 aka 1.8beta6 (using r41146)
20100210 18:09:11-!- loonycyborg [~sergey@wesnoth/developer/loonycyborg] has quit [Quit: KVIrc 3.4.2 Shiny http://www.kvirc.net/]
20100210 18:09:52-!- loonybot [~loonybot@wesnoth/bot/loonybot] has quit [Remote host closed the connection]
20100210 18:11:00< CIA-62> ivanovic * r41148 /trunk/ (changelog configure.ac players_changelog src/wesconfig.h): post release version bump to 1.7.13+svn
20100210 18:11:26-!- Ivanovic changed the topic of #wesnoth-dev to: tagged 1.7.13, announcing "soon" | string/feature freeze active! | 63 bugs, 245 feature requests, 8 patches | logs: http://irclogs.wesnoth.org/ | Don't paste on IRC! Use a pastebin: http://wesnoth.pastebin.com | http://imagebin.org
20100210 18:13:47< Rhonda> oh
20100210 18:14:12< Ivanovic> Rhonda: huh?
20100210 18:14:22< Rhonda> well. Release. :)
20100210 18:14:29< Ivanovic> jupp
20100210 18:14:32 * Rhonda heads home to prepare for it.
20100210 18:15:27< Ivanovic> :)
20100210 18:28:25-!- Appleman1234 [~Appleman1@CPE-124-191-177-16.oxqn1.cha.bigpond.net.au] has joined #wesnoth-dev
20100210 18:34:22-!- Appleman1234 [~Appleman1@CPE-124-191-177-16.oxqn1.cha.bigpond.net.au] has quit [Ping timeout: 256 seconds]
20100210 18:34:44-!- rosso_ [~rosso@dslb-088-070-236-072.pools.arcor-ip.net] has joined #wesnoth-dev
20100210 18:35:06-!- rosso [~rosso@dslb-088-070-226-178.pools.arcor-ip.net] has quit [Ping timeout: 240 seconds]
20100210 18:38:33< stikonas> Ivanovic: your last commit contains a typo
20100210 18:38:53< stikonas> there is a missing " in src/wesconfig.h that aborts compilation
20100210 18:39:16< stikonas> please fix it :)
20100210 18:41:09-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20100210 18:44:36< CIA-62> ivanovic * r41149 /trunk/src/wesconfig.h: typo fix (missing ")
20100210 18:45:16-!- deekay [~dk@wesnoth/developer/dragonking] has joined #wesnoth-dev
20100210 18:45:43< stikonas> thanks
20100210 18:48:24-!- loonybot [~loonybot@ppp79-139-136-187.pppoe.spdop.ru] has joined #wesnoth-dev
20100210 18:48:24-!- loonybot [~loonybot@ppp79-139-136-187.pppoe.spdop.ru] has quit [Changing host]
20100210 18:48:24-!- loonybot [~loonybot@wesnoth/bot/loonybot] has joined #wesnoth-dev
20100210 18:49:14-!- loonycyborg [~sergey@wesnoth/developer/loonycyborg] has joined #wesnoth-dev
20100210 18:54:40-!- Noyga [~noyga@wesnoth/developer/noyga] has joined #wesnoth-dev
20100210 19:06:06-!- loonybot [~loonybot@wesnoth/bot/loonybot] has quit [Read error: Connection reset by peer]
20100210 19:10:50-!- EdB [~EdB@tss37-1-89-84-18-220.dsl.club-internet.fr] has joined #wesnoth-dev
20100210 19:33:33< Sirp> Ivanovic: hi! :)
20100210 19:33:50< Ivanovic> Sirp: you probably read the short first summary about fosdem this year, right?
20100210 19:33:59< Ivanovic> mainly the part about the (possible) migration to opengl
20100210 19:34:34< Sirp> Ivanovic: yes, the discussions looked very exciting.
20100210 19:34:46< Sirp> I'm glad that that was dicussed at fosdem
20100210 19:34:51< Ivanovic> that is: what stopped us before (opengl not really available for all platforms supported) is not much of a problem anymore
20100210 19:35:01< Ivanovic> since atm we only really support windows, mac and linux
20100210 19:35:14-!- |^y_emkidzu-it| [~KVIrc@host249-66-dynamic.5-87-r.retail.telecomitalia.it] has quit [Ping timeout: 248 seconds]
20100210 19:35:36< Ivanovic> solaris as well as amigaos and whatnot have not been providing binaries for a while, so i assume that those ports are basically dead atm
20100210 19:36:17< Ivanovic> we assume that windows and mac users will basically be fine, leaving us with linux, where the nvidia proprietary drivers should be fine
20100210 19:36:35-!- boucman [~rosen@wesnoth/developer/boucman] has joined #wesnoth-dev
20100210 19:36:39< Ivanovic> so we evaluated the open source drivers for intel at fosdem, which did quite a good job
20100210 19:36:41< fabi> hi boucman 
20100210 19:36:44< Sirp> okay.
20100210 19:36:53< Sirp> Ivanovic: you got a bunch of different people to try frogatto, right?
20100210 19:36:57< Ivanovic> over the last days i had some short tests on my desktop rig with the open source driver for r6xx
20100210 19:37:30< Ivanovic> there is/was some texture problem, a bug in mesa (cf https://bugs.freedesktop.org/show_bug.cgi?id=26471 )
20100210 19:37:43< Ivanovic> with the patch posted in there speed is 100% fine for me, too
20100210 19:38:01< Ivanovic> using software mode though is not much of an alternative since framerates are really low then
20100210 19:38:07-!- EdB [~EdB@tss37-1-89-84-18-220.dsl.club-internet.fr] has quit [Remote host closed the connection]
20100210 19:38:15< Ivanovic> Sirp: basically that is what all of us did, play a little frogatto to see how well it works
20100210 19:38:36< Ivanovic> Sirp: might be a good idea to create a more recent windows and mac binary to be sure that it works well on those systems, too
20100210 19:38:57< Ivanovic> since the warzone dev we have in here every now and then claimed that the most problematic thing was the intel onboard stuff
20100210 19:39:00< Ivanovic> on windows that is!
20100210 19:39:03-!- Espreon [~espreon@wesnoth/developer/espreon] has joined #wesnoth-dev
20100210 19:40:17< boucman> hey all
20100210 19:40:24< Sirp> Ivanovic: we've run the Windows binary on Intel in board and it works well.
20100210 19:40:25< Espreon> Hello boucman.
20100210 19:40:32< Sirp> hi boucman 
20100210 19:40:46< Ivanovic> okay, good to know
20100210 19:40:53< Sirp> Ivanovic: so I think it would be nice if we had OpenGL as the most prominent suggestion for GSoC this year.
20100210 19:40:55< Ivanovic> so we should really be rather fine
20100210 19:41:14< Sirp> and if we don't make it into GSoC and/or to supplement it we could consider funding developers to work full time for a period on OpenGL ourselves.
20100210 19:41:21< Ivanovic> we came to the conclusion that the complete port to opengl is too much as SoC project
20100210 19:41:23 * Espreon hopes that his Radeon Mobility x1400 will be able to handle it.
20100210 19:41:46-!- Zarel [~Zarel@warzone2100/developer/Zarel] has joined #wesnoth-dev
20100210 19:42:03-!- Aethaeryn [~Michael@wesnoth/umc-dev/developer/aethaeryn] has joined #wesnoth-dev
20100210 19:42:06< Sirp> Espreon: try out frogatto (http://www.frogatto.com) -- it'll give you a really good idea if it can or not.
20100210 19:42:07< Ivanovic> Espreon: follow the instructions in this bugreport to get a checkout from frogatto and test it! https://bugs.freedesktop.org/show_bug.cgi?id=26471
20100210 19:42:42< Ivanovic> a list of dependencies is included (basically what you need for wesnoth plus some extra parts in boost and glew)
20100210 19:42:57< Espreon> OK.
20100210 19:42:57< Ivanovic> editing the makefile is not required anymore either
20100210 19:43:07< Ivanovic> i *think* you will be perfectly fine
20100210 19:43:31< Ivanovic> another thing that supports me in this asumption: http://www.phoronix.com/scan.php?page=article&item=ati_r600_mesa_3d&num=1
20100210 19:43:51< Ivanovic> basically a comparision how well various ati cards (recent cards) do with mesa on several games
20100210 19:44:07< Espreon> I believe that my card is part of the 500s series or something.
20100210 19:44:30< Ivanovic> and i don't think that wesnoth well use anywhere close to the graphics resources of some shooter like world of padman or tremulous or urban terror
20100210 19:44:32< Ivanovic> ;)
20100210 19:45:21< Ivanovic> Sirp: since you have some experience with getting an engine work with opengl, what do you estimate, how much work will the conversion to opengl be?
20100210 19:45:35< Ivanovic> yes, opengl would become mandatory then, not keeping two render paths
20100210 19:45:43< Sirp> Ivanovic: well I don't think it will be too bad.
20100210 19:45:53< Sirp> Ivanovic: yes, opengl -> mandatory. We will NOT have two render paths.
20100210 19:45:54< Ivanovic> and it would have to start "soon", shortly after the release of 1.8
20100210 19:46:18< Sirp> Ivanovic: I think the main coding should be doable in a month by a few good developers.
20100210 19:46:39< Sirp> I think that ilor and deekay in particular should consider submitting a GSoC application 
20100210 19:46:56< ilor> Sirp: I'm not sure if I'll be applying this year
20100210 19:46:56< Ivanovic> the obvious target will be to have feature parity with the 1.8 render engine for 1.10
20100210 19:47:10< Ivanovic> ilor: you want to mentor instead?
20100210 19:47:11< Sirp> ilor: okay. I would like you to, but if you can't or don't want to I understand. :)
20100210 19:47:16< Ivanovic> great, we can need some more mentors!
20100210 19:47:18< Ivanovic> ;)
20100210 19:47:20< Sirp> ilor: or if you'd like to simply help port to OpenGL. :)
20100210 19:47:26< Rhonda> oh, mentors.
20100210 19:47:49< ilor> it seems I'm getting employed at my uni to work on a project, which will be related to my thesis eventually, and this might take time
20100210 19:48:20< Sirp> ilor: and this is more important than porting Wesnoth to opengl how exactly? :p
20100210 19:48:20< ilor> I'm not sure if I can reliably pull off both this and gsoc and have a life
20100210 19:48:30< deekay> Sirp: I got a feeling lat year that newer applicatns have priority over "old" contributors.. I haven't though about GSoC2010 for Wesnoth because of that, bit it's not like it is set in stone either
20100210 19:48:38-!- silene [~plouf@wesnoth/developer/silene] has joined #wesnoth-dev
20100210 19:48:44< Ivanovic> hi silene
20100210 19:48:48< silene> hi
20100210 19:48:55< Sirp> ilor: excellent. it comes down to deciding whether you don't want to have a life or you don't want to do the uni project.
20100210 19:49:00< Sirp> :)
20100210 19:49:01< Sirp> hi silene 
20100210 19:49:06< ilor> Sirp ;)
20100210 19:49:08< Ivanovic> silene: you read in the logs that zookeeper had to use some "hack" for making something in liberty work?
20100210 19:49:21< Sirp> deekay: I don't think I would consider a new contributor for opengl porting. It's too hard.
20100210 19:49:44< Sirp> deekay: and just remember, an excellent chance to learn opengl. :)
20100210 19:50:11< deekay> Sirp: You know, it's along my specialization at my UNI
20100210 19:50:18< Ivanovic> :)
20100210 19:50:30< Ivanovic> hey, then you are the perfect candidate to get it done!
20100210 19:50:31< Rhonda> Oh, sirp around. :)
20100210 19:50:31< Ivanovic> ;)
20100210 19:50:40< Sirp> deekay: really? even better reason. :)
20100210 19:50:48-!- grzywacz [~grzywacz@wesnoth/developer/grzywacz] has joined #wesnoth-dev
20100210 19:51:20< loonycyborg> Sirp: IMO using a 2d library with gl backend would make more sense.
20100210 19:51:34< deekay> Sirp, Ivanovic - at march I'm starting specialization - it is "Modern computern graphics" and includes lectures about game programming, openGL etc
20100210 19:51:47< deekay> Tho not sure if OpenGl is this on next term
20100210 19:51:52< boucman> hehe
20100210 19:51:54< deekay> s/on/or on/
20100210 19:52:18< fabi> hmmm frogatto is segfaulting every time I want to start a new game.
20100210 19:52:31< Sirp> loonycyborg: if someone has a good one, perhaps, though I think opengl directly is nice enough, especially since you can put a thin layer between most of the time.
20100210 19:52:36< Ivanovic> fabi: it is normal that it segfaults "often"
20100210 19:52:39< Ivanovic> just retry...
20100210 19:52:40< Ivanovic> ;)
20100210 19:52:46< silene> Ivanovic: hum... not sure why the hack is needed; i didn't change the way unit types are handled; in particular, unit types are copied verbatim inside the unit cfg; it's unrelated to [variables]
20100210 19:52:47< Sirp> Ivanovic: uhh really?
20100210 19:52:52< Sirp> it doesn't segfault for me.
20100210 19:52:56< Sirp> oh except
20100210 19:52:58< Sirp> try --nosound
20100210 19:53:02< Sirp> there are sound issues on Linux
20100210 19:53:08< Ivanovic> Sirp: at our tests at fosdem there were several segfaults
20100210 19:53:30< boucman> and we felt out of the map a couple of times
20100210 19:53:33< Sirp> fabi: if --nosound doesn't work, go to #frogatto and complain.
20100210 19:53:40< loonycyborg> Sirp: A library could provide non-gl rendering path.
20100210 19:54:06< Sirp> loonycyborg: is there a library you had in mind?
20100210 19:54:06< Ivanovic> loonycyborg: do you think this software path is much different from opengl in software?
20100210 19:54:20< loonycyborg> Sirp: cairo.
20100210 19:54:31< loonycyborg> We're already using it :P
20100210 19:55:01< Sirp> loonycyborg: it lists opengl as an "experimental backend"
20100210 19:56:12< loonycyborg> It has glitz backend which went unmaintained, but there was a FOSDEM presentation about a new backend that uses gl directly.
20100210 19:57:13< Sirp> well, we could consider it if it's any good.
20100210 19:57:16< silene> loonycyborg: what's wrong with keeping using sdl as the alternate backend? for themanaworld, we were using both sdl and opengl (and one couldd switch from one to the other at runtime without any difficulty)
20100210 19:57:26< Ivanovic> noone of us has seen this presentation (was not announced much in front...) and the slides are not *this* helpfull regarding questions like "is it already available" and "how good is it for a game project?"
20100210 19:57:36< Sirp> silene: I think that just adds unnecessary code complexity.
20100210 19:58:04< Sirp> I don't think anyone has demonstrated need for a non-opengl path
20100210 19:58:09< Ivanovic> silene: one reason for the switch is to *drop* stuff like "okay, which part do i really *have* to invalidate?"
20100210 19:58:27< Sirp> yes, we want to get rid of all this dirty rectangle stuff.
20100210 19:59:02< Ivanovic> that is basically one of the huge reasons, and of course making use of available computation power at the side of the gpu
20100210 19:59:03< silene> Sirp, Ivanovic: i suggest you take a look at the code of themanaworld, it is actually simpler than the one from wesnoth, yet the framerate is better, even for the sdl backend
20100210 19:59:25< Sirp> silene: okay, I will look at it.
20100210 19:59:35< loonycyborg> Ivanovic: 1.7.13 windows installer is uploaded.
20100210 19:59:42< Ivanovic> loonycyborg: great!
20100210 20:00:33< boucman> i'm not sure TMW has the same level of coplexity wrt invalidation as we have...
20100210 20:00:51-!- Zarel_ [~Zarel@warzone2100/developer/Zarel] has joined #wesnoth-dev
20100210 20:00:54< Ivanovic> crimson_penguin: new release available!
20100210 20:00:56< Ivanovic> ;)
20100210 20:01:05< zookeeper> silene, the problem in liberty was that the custom [unit_type]s had a custom outlaw_type= key inside them, which a scenario tried to read from the single units. the recent changes caused it to be dropped from the single units as an unrecognized value, hence the code which tried to read and use it broke.
20100210 20:01:08-!- Zarel [~Zarel@warzone2100/developer/Zarel] has quit [Ping timeout: 240 seconds]
20100210 20:01:15< crimson_penguin> Ivanovic: haha, oh yeah, I saw that and then forgot - but I did start svn up'ing
20100210 20:01:37-!- Zarel_ is now known as Zarel
20100210 20:01:39< silene> zookeeper: i don't see anything in the code that would drop it
20100210 20:01:39< Ivanovic> crimson_penguin: yeah, my great mind reading abilities now got an irc plugin, so i could know
20100210 20:01:41< Ivanovic> ;)
20100210 20:01:48< crimson_penguin> haha
20100210 20:02:24< silene> zookeeper: it's dropped only if it is in [unit], not if it is in [unit_type]
20100210 20:02:49< zookeeper> silene, but they were in [unit], because they were in [unit_type]
20100210 20:02:56< silene> zookeeper: what?!
20100210 20:03:38< zookeeper> (previously) if i out foo=bar in a [unit_type], then all single units of that type would have foo=bar as well.
20100210 20:03:42< zookeeper> s/out/put
20100210 20:03:54< silene> zookeeper: this is still true!
20100210 20:04:00< zookeeper> it is?
20100210 20:05:08-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20100210 20:05:31< zookeeper> they caused "warning unit: Unknown attribute 'outlaw_type' discarded." and $aunitofthattype.outlaw_unit was empty
20100210 20:06:42< silene> zookeeper: the warning is not necessarily an issue: if you create a unit, store it, unstore it, then the engine will complain and overwrite the "outlaw_unit" attribute from the unit with the one from the unit type, hence the warning
20100210 20:07:14< zookeeper> silene, hmmh...
20100210 20:07:33< zookeeper> i'll revert my commits locally and try it out one more time myself, but if it still breaks, i'll let you have a look ;)
20100210 20:08:08< silene> yes, if it is really empty, there may be an engine bug somewhere; either in my changes or Crab's ones
20100210 20:08:36< silene> (or perhaps even older and it went unnoticed till now)
20100210 20:08:46< zookeeper> no, that same WML was definitely working before
20100210 20:09:08< zookeeper> so whatever "broke" it (quotes since it relied on undefined behaviour...) must have been relatively recent
20100210 20:09:58< Ivanovic> it was most likely not broken in 1.7.12, since in my release tests i start all campaigns and look for errors/warnings
20100210 20:10:01-!- Zarel [~Zarel@warzone2100/developer/Zarel] has quit [Ping timeout: 264 seconds]
20100210 20:11:59< Rhonda> Ivanovic: The upload to files is final?
20100210 20:12:01< zookeeper> uh. it worked now...
20100210 20:12:06< silene> Ivanovic: how far do you play? because the warning doesn't appear if i just start the campaign here
20100210 20:12:14< Ivanovic> Rhonda: jupp
20100210 20:12:19< zookeeper> silene, i already applied a fix to it
20100210 20:12:21< Ivanovic> Rhonda: i uploaded that file to sf.net
20100210 20:12:32< silene> zookeeper: i know, but i haven't applied it yet
20100210 20:12:34< Rhonda> Already fetched from files. :)
20100210 20:12:41< Ivanovic> silene: it appeared here when starting, but zookeeper commited a hack/fix
20100210 20:12:59< zookeeper> revert to 41142, 41143 was the first part of my fix
20100210 20:13:30< silene> Ivanovic: it doesn't appear for me, and i'm using revision 41133
20100210 20:13:37< zookeeper> odd
20100210 20:14:16< zookeeper> anyways, yes, it looks like it doesn't actually affect the outlaw transformation. the problem i remember might have been when i tried to put [variables] outlaw_type= in the [unit_type]
20100210 20:14:27< Ivanovic> my first tests were with 41139, and there i got this warning when starting liberty
20100210 20:14:32-!- Zarel [~Zarel@warzone2100/developer/Zarel] has joined #wesnoth-dev
20100210 20:15:16< zookeeper> anyway, i'm not too unhappy with having made that fix, since now it's at least not relying on some undefined behaviour like before even though it's a bit more verbose.
20100210 20:15:50< silene> zookeeper: right, [variables] in unit types don't work (and i don't think they ever worked)
20100210 20:16:30< zookeeper> silene, i figured that if i put a [variables] in [unit_type] then that'd be put into single units of that type just like made-up keys are.
20100210 20:21:02< silene> zookeeper: as a rule of thumb, [variables], canrecruit, and a few others, are properties that are never overridden by unit types, they are intimately linked to a given unit; on the contrary, [resistance], [movement], and so on, are always overridden by unit types when a unit advances (though modifiers can change them afterwards)
20100210 20:23:00< silene> and (as strange as it may seem) a created unit does advance (!) from nothing to the given unit type
20100210 20:23:21< zookeeper> sure. i guess the contents of [variables] could be merged to the single unit, but doesn't sound worth it if it'd require extra code.
20100210 20:25:57< zookeeper> silene, anyway, the build i was testing with was 41001, but even that still spits out some "warning unit: Unknown attribute 'x' discarded." (same for y) warnings. been fixed since?
20100210 20:28:48< zookeeper> ah well, i'm testing that atm...
20100210 20:28:52< silene> zookeeper: yes, this one should have been fixed
20100210 20:29:15< zookeeper> great
20100210 20:45:44< Ivanovic> Espreon: so have you already tested frogatto?
20100210 20:46:04< shadowmaster> Ivanovic: It Got Worse for me.
20100210 20:46:12< Ivanovic> shadowmaster: read the bugreport!
20100210 20:46:13< shadowmaster> now no sprites are drawn either here :/
20100210 20:46:25< Ivanovic> shadowmaster: yes, things got worse, but a mesa patch is attached that fixes the issues
20100210 20:46:27< shadowmaster> yes, I was reading the bug report, and stikonas intervention confused me
20100210 20:46:38< shadowmaster> okay, I read it last night, not today
20100210 20:46:43< Ivanovic> yes, stikonas mixed things up, he meant worse, not better
20100210 20:47:00< Ivanovic> and in the attached screenshot you will see that it is worse
20100210 20:47:33< shadowmaster> I am also guessing you mean "textures are now displayed correctly" in #15
20100210 20:47:36< shadowmaster> meant
20100210 20:49:12< shadowmaster> too bad the patch will fail on 7.7
20100210 20:55:10< Espreon> Ivanovic: It seems to perform well here.
20100210 20:55:52< Ivanovic> Espreon: feel free to send a message with this content to the -dev ml (that is: mesa version used, kernel used, distri used, graphics card, cpu)
20100210 20:56:01< Ivanovic> since so far <r600 was not mentioned
20100210 20:56:31-!- Blarumyrran [~Blarumyrr@81-20-159-197.levira.ee] has joined #wesnoth-dev
20100210 20:56:38< Espreon> Oh, I'd have to turn off my machine and wait awhile for accurate CPU usage reports.
20100210 20:56:49< Ivanovic> no need for cpu usage
20100210 20:56:55< Espreon> After compiling something large, the usage statistics fluctuates violently.
20100210 20:57:00< Ivanovic> just which cpu you have and which graphics card
20100210 20:57:04< Espreon> OK.
20100210 20:57:17< Espreon> Wesnoth's -dev ml?
20100210 20:57:22< Ivanovic> since if hardware accelleration is not available mesa tends to do stuff in software
20100210 20:57:25< Ivanovic> Espreon: exactly
20100210 20:57:37< Ivanovic> there is already a thread you can reply to
20100210 20:58:00< Espreon> OK.
20100210 20:58:11< Ivanovic> just reply to "[Wesnoth-dev] Extra OGL measurement WAS(Re: FOSDEM 2010 - First results)"
20100210 20:59:47< Espreon> How does one reply to an ml thread?
20100210 20:59:57< Ivanovic> in your mail prog just hit "reply"
20100210 21:00:06< Ivanovic> or "answer" or however it is called
20100210 21:00:08-!- Appleman1234 [~Appleman1@CPE-124-191-177-16.oxqn1.cha.bigpond.net.au] has joined #wesnoth-dev
20100210 21:01:02< Ivanovic> it will send the mail to the dev-ml then (as you can see in the "To" entry in your compose view
20100210 21:01:51< Espreon> OK, then how does one get an e-mail client to interface with an ml?
20100210 21:09:31< Ivanovic> send a mail to the email address of the ml!
20100210 21:10:00< Ivanovic> easiest way: open the mail from mordante about his system with opengl (cf subject above)
20100210 21:10:11< Ivanovic> there click "reply", type your text, be done
20100210 21:11:12< Sirp> Ivanovic: yeah I think anyone for whom Frogattois segfaulting for should use --nosound
20100210 21:12:27< fabi> Sirp, It's even segfaulting with the --nosound option.
20100210 21:50:21-!- Zarel [~Zarel@warzone2100/developer/Zarel] has quit [Quit: This computer has gone to sleep]
20100210 21:51:24-!- Zarel [~Zarel@warzone2100/developer/Zarel] has joined #wesnoth-dev
20100210 21:56:02-!- noy [~Noy@wesnoth/developer/noy] has quit [Read error: Connection reset by peer]
20100210 21:56:34-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20100210 22:00:44-!- noy [~Noy@wesnoth/developer/noy] has quit [Ping timeout: 246 seconds]
20100210 22:11:14-!- dtiger [~dtiger@dynamic-vpdn-93-125-68-222.telecom.by] has quit [Remote host closed the connection]
20100210 22:12:22-!- Espreon [~espreon@wesnoth/developer/espreon] has quit [Remote host closed the connection]
20100210 22:14:47< shadowmaster> I want mesa master... :s
20100210 22:15:05< shadowmaster> why can't I have nightly snapshots or something!
20100210 22:16:07< crimson_penguin> Ivanovic: uploading
20100210 22:16:16< Ivanovic> crimson_penguin: cool!
20100210 22:16:32< Ivanovic> crimson_penguin: then i can announce tomorrow, lovely
20100210 22:16:33< shadowmaster> or anyone who can run the make target they use for making them
20100210 22:16:49< crimson_penguin> yup :)
20100210 22:17:10< Ivanovic> shadowmaster: are the "debian project/build files" not available?
20100210 22:17:19< shadowmaster> er?
20100210 22:17:39< shadowmaster> hey, I could compile it if I could donwload the git repository to my laptop
20100210 22:17:43< Ivanovic> shadowmaster: normally the files to create a deb of some package is basically one plain folder
20100210 22:17:50< shadowmaster> but > 100 MB... ew
20100210 22:17:51< Ivanovic> and what stops you from doing so?
20100210 22:18:16-!- Crab_ [~Crab_@wesnoth/developer/crab] has joined #wesnoth-dev
20100210 22:18:25< shadowmaster> er, connection? would take days? I'm not going to be sitting on this place for days
20100210 22:20:25< shadowmaster> sure I could do make tarballs on the VPS but I don't have aclocal and who knows what esle is required
20100210 22:23:32 * boucman is seeing beaches
20100210 22:25:40-!- Aethaeryn [~Michael@wesnoth/umc-dev/developer/aethaeryn] has quit [Quit: quit()]
20100210 22:29:50< Ivanovic> boucman: what? where?
20100210 22:29:58< boucman> on my screen :)
20100210 22:30:11< Ivanovic> with animated water at the shore?
20100210 22:30:12< boucman> it's really ugly at this point, but i'm working on it ;)
20100210 22:30:17< boucman> yup
20100210 22:30:21< Ivanovic> cool!
20100210 22:30:40< Ivanovic> boucman: have you already found out the reason for the animated water progs on mac?
20100210 22:31:24< boucman> nope
20100210 22:31:35< Ivanovic> :(
20100210 22:31:42< boucman> i need some time with crimson_penguin for that, and unfortunately it won't be tonight
20100210 22:31:52< Ivanovic> do we maybe use some mac specific code somewhere?
20100210 22:31:53< shadowmaster> ~/git/mesa $ du -sh .git
20100210 22:31:53< shadowmaster> 105.2M  .git
20100210 22:32:26< Ivanovic> shadowmaster: the whole checkout (including the .git folder) has some 205MB over here
20100210 22:32:48< shadowmaster> I know very well
20100210 22:33:05< shadowmaster> and tar fails ("Broken pipe" indeed)
20100210 22:34:18< shadowmaster> um, make tarballs would probably die like that too anyway
20100210 22:37:55< Sirp> boucman: btw are you interested in helping on a move to opengl? :)
20100210 22:38:27< boucman> Sirp: yes and no
20100210 22:38:49< boucman> i'd love to helpp, esp with my knowledge of the unit animation engine
20100210 22:39:24< boucman> however i'm not sure I would have the time to spearhead the effort especially since i'm trying to learn my way in the terrain WML
20100210 22:39:35< shadowmaster> I am guessing that if we move to opengl, the next time I buy a computer I won't be able to use Wesnoth with Linux
20100210 22:39:52< shadowmaster> uunless it's an old second-hand computer
20100210 22:41:15< Sirp> shadowmaster: ummm huh?
20100210 22:41:29< Sirp> the newer a computer is the more likely you'll be able to use Wesnoth on any platform
20100210 22:41:36< shadowmaster> I was given this computer on Xmas 2008.
20100210 22:41:49< shadowmaster> I didn't get 3D accel until Mesa 7.7 was released one year later
20100210 22:42:06< Sirp> shadowmaster: you will not need 3d acceleration for Wesnoth
20100210 22:42:09< shadowmaster> not to talk about Suspend to RAM, etc.
20100210 22:42:26< boucman> zookeeper: at some point (not tonight) i'd like some advice on some macro-justu
20100210 22:42:28< boucman> jutsu
20100210 22:42:44< Sirp> boucman: being involved doesn't mean you have to spearhead it. :)
20100210 22:42:51< shadowmaster> Sirp: er, yeah... I'll need a überCPU then? It's not like I have a dual core AMD Athlon64 that barely runs Frogatto with Mesa's software render at 1-5 FPS
20100210 22:43:00< zookeeper> boucman, all right
20100210 22:43:07< boucman> in that case, yes, i'm definitely interested
20100210 22:43:21< shadowmaster> Wesnoth might not be a sidescroller, but it does have a lot of animations
20100210 22:43:54< shadowmaster> and then I need to work with animations as a pixel artist for my campaigns :/
20100210 22:44:01< boucman> zookeeper: basically with animated terrain, the terrain macroes need to be expanded and what I did so far is enough to test the graphics but will quickly become messy, so I wonder if there are macro tricks I could use to spear my sanity
20100210 22:44:28< zookeeper> great, i actually forgot to commit the fix to that one UtBS bug reported on #wesnoth today -.-
20100210 22:44:41< crimson_penguin> loonycyborg: http://forums.wesnoth.org/viewtopic.php?p=409096#p409096
20100210 22:44:45< zookeeper> i just made it locally and told the guy how to apply it himself, but forgot to commit it
20100210 22:45:25< shadowmaster> *AMD Athlon(tm) X2 Dual-Core QL-62
20100210 22:45:39< shadowmaster> 2000 MHz each core, according to software
20100210 22:45:50< CIA-62> zookeeper * r41150 /trunk/data/campaigns/Under_the_Burning_Suns/scenarios/03_Stirring_in_the_Night.cfg: Fixed a case of broken variable substitution.
20100210 22:46:14< zookeeper> boucman, yeah, well...depends on what you want to do
20100210 22:46:22< shadowmaster> and putting this much power on this laptop is already problematic
20100210 22:46:25< shadowmaster> overheating, etc.
20100210 22:46:36< zookeeper> silene, without that commit i just made, you should get some kind of a lua error at that point
20100210 22:46:49< boucman> honestly I don't really know the possibilities, so i'll show you what I have done and you tell me if there a preproc feature that could help me
20100210 22:47:00< loonycyborg> Sirp: Honestly, I don't see any benefit of moving wesnoth to opengl other than making it faster on platforms that it's already fast enough on.
20100210 22:47:07< shadowmaster> and there are pretty good reasons for not having a desktop machine with a good 3d -capable graphics controller; I can't support an Internet connection of my own either,
20100210 22:47:18< zookeeper> that is, without that commit that [gold] side= will be empty, and results in "error scripting/lua: ...le for Wesnoth 1.7.12-1.8beta5/data/lua/wml-tags.lua:126: bad argument #1 to 'get_side' (number expected, got string)"
20100210 22:47:52< zookeeper> i checked the lua code briefly and it looked ok to me, but i don't really know lua, so..
20100210 22:48:00< loonycyborg> If it ain't broken, don't fix it :P
20100210 22:49:24< Noyga> loonycyborg, from en engines point of view it makes a lot of things simplier
20100210 22:49:59< silene> zookeeper: the lua code is fine, though i could probably improve it so that it gives a better error message when the tag is given crap
20100210 22:50:38< Noyga> currently there is a lot of calculation to know which part to redraw. with opengl we not longer have to care
20100210 22:50:59< zookeeper> silene, hmmh, but in that case, AFAIK side=$nonexistantvariable should be equal to just not having a side= there at all
20100210 22:51:59< Noyga> btw drivers are not alway ok everywhere (on my desktop frogatto runs à ~10 fps with some artifacts because of a crappy driver)
20100210 22:52:00< zookeeper> i didn't really check if $unit.side really returns empty at that point though, but unless i'm missing something, it should.
20100210 22:53:52< loonycyborg> Noyga: How exactly opengl solves this issue?
20100210 22:55:13< Noyga> opengl always redraws everything
20100210 22:55:28< Noyga> stg like this (i'm not an opengl specialist)
20100210 22:55:35< loonycyborg> Indeed. That explains why software rendering is so slow :P
20100210 22:55:58< silene> zookeeper: from the c++ engine point of view (and the wml engine too sometimes, hence the usual trick of $empty), an empty attribute is not the same as a missing attribute; and so i kept this behavior for the lua engine: if the user explicitly puts an empty attribute, a lua script can notice it
20100210 22:56:39< silene> zookeeper: by the way, it helped found the error, so it's not that bad a behavior ;-)
20100210 22:56:49< crimson_penguin> boucman: what are the Mac problems?
20100210 22:56:55< Sirp> loonycyborg: (1) less memory usage; (2) more possible graphical effectes etc
20100210 22:57:24< boucman> crimson_penguin: i don't know, but I don't have a mac, and what I concluded at fosdem is that I won't know without some printf sprinkled around
20100210 22:57:34< boucman> so i'll need your help :)
20100210 22:57:36< crimson_penguin> k
20100210 22:59:44< zookeeper> silene, right...i might have confused it with the handy behavior where foo={BAR} is the same as omitting foo= if {BAR} is empty/undefined
20100210 22:59:54-!- yann [~dwitch@nan92-1-81-57-214-146.fbx.proxad.net] has quit [Ping timeout: 276 seconds]
20100210 23:00:17< crimson_penguin> loonycyborg: it's also annoying to have Wesnoth use way more CPU than it needs to, even if it works fine
20100210 23:00:52< crimson_penguin> I mean Wesnoth is fast on my computer, but it uses something like 25% CPU I think
20100210 23:01:28< loonycyborg> You probably have multiple cores too..
20100210 23:02:05< boucman> opengl should reduce CPU use since invalidating is responsible for a large part of the calculations
20100210 23:02:21< silene> zookeeper: actually, it depends on "foo"; for some attributes, empty is the same as missing, for others it is not; in fact, for some, even crap is seen as missing (for instance, the old [gold] tag was handling side="crap" the same as side=1)
20100210 23:02:46< crimson_penguin> yeah, actually testing now, it does go up to 30% sometimes, but it is normally 10-20% which is fine for me (that's of one core)
20100210 23:03:23< zookeeper> silene, i'm pretty sure the preprocessor gets rid of foo={BAR} if BAR is empty/undefined
20100210 23:03:40< crimson_penguin> ohhh, but a constant 18% when there's 3 bats on the screen
20100210 23:03:43< silene> zookeeper: i'm pretty sure you re wrong ;-)
20100210 23:03:51< loonycyborg> crimson_penguin: For me it 2-7 %
20100210 23:04:01< zookeeper> like if i call {DEBUG_MSG ()} then the preprocessor is just going to _not_ place any message= within that [message]
20100210 23:04:15< crimson_penguin> wow
20100210 23:04:15< crimson_penguin> 2.4Ghz Core2Duo here
20100210 23:04:24< silene> zookeeper: it does
20100210 23:04:29< crimson_penguin> I think it has a lot to do with OpenGL and stuff
20100210 23:04:34< crimson_penguin> errr, SDL
20100210 23:04:53< crimson_penguin> switching to SDL 1.2.14 made it go up to 100% and be really slow
20100210 23:04:57< zookeeper> silene, uh oh. why are you breaking my most basic assumptions of how WML works :|
20100210 23:05:27< loonycyborg> 7-14 % in the test scenario
20100210 23:05:28< zookeeper> but i'm still not convinced. i'll test that before going to bed
20100210 23:06:10< crimson_penguin> loonycyborg: I was playing on Rooting Out A Mage
20100210 23:06:24 * crimson_penguin goes to rejoin the unicycle party
20100210 23:06:42-!- Espreon [~espreon@wesnoth/developer/espreon] has joined #wesnoth-dev
20100210 23:07:28< zookeeper> silene, no, it looks like i was right
20100210 23:07:40< zookeeper> {DEBUG_MSG ()} results in a [message] without any kind of message=
20100210 23:07:49< silene> zookeeper: i wrote the preprocessor, please have a bit of confidence in what i say
20100210 23:08:31< zookeeper> (that behaviour is quite commonly used when calling unit-creating macros, such as if i want a named unit but without an id, i'll just leave the ID argument empty)
20100210 23:08:44< silene> the message= attribute is there, empty but there nonetheless
20100210 23:09:15< zookeeper> silene, where? i'm looking at a savefile which shows me there really isn't any message= there
20100210 23:10:04< zookeeper> what i can see in a savefile should be all there is (not counting default values provided by the engine of course)
20100210 23:16:03< silene> zookeeper: i must admit that i'm confused; i just checked and the attribute is present at execution time; perhaps the empty attributes are discarded at save time
20100210 23:19:52< silene> zookeeper: i confirm, empty attributes are not put into the savefile but they exist
20100210 23:20:21< zookeeper> O.o
20100210 23:20:35-!- stikonas [~and@wesnoth/translator/stikonas] has quit [Read error: Connection reset by peer]
20100210 23:20:42< zookeeper> silene, sounds bizarre.
20100210 23:21:20< silene> yes, and that's a bug bound to happen, since saving and reloading a game should never change the behavior of a scenario
20100210 23:21:40-!- stikonas [~and@wesnoth/translator/stikonas] has joined #wesnoth-dev
20100210 23:21:50< zookeeper> so...if i have a macro like this:
20100210 23:21:52< zookeeper> #define MYUNIT ID
20100210 23:21:52< zookeeper>     [unit]
20100210 23:21:52< zookeeper>         id={ID}
20100210 23:21:54< zookeeper> (etc)
20100210 23:22:39< zookeeper> then you're saying that there's an empty id= attribute visible at execution time if i call the macro like this: {MYUNIT ()}?
20100210 23:23:02< zookeeper> that is, in a different manner than if the macro didn't have that id={ID} line at all
20100210 23:24:14< silene> yes, there is an empty attribute; but as far as the "id" attribute itself behaves, that doesn't make a single bit of difference
20100210 23:24:55< Sirp> the thing is, Wesnoth is really not that fast on even a good computer. It's often embarassingly slow. We should be able to make it silky smooth, and provide any reasonable effects we want. Instead our artists feel crippled by all the limitations imposed on them.
20100210 23:25:09< zookeeper> right. so i've just been lucky enough to never do that on any attribute for which there's a difference between non-existing and empty?
20100210 23:25:24< zookeeper> (no sarcasm, that sounds entirely possible)
20100210 23:26:48< silene> zookeeper: that's right; on the top of my head, i can think of unit filters for instance; they make a difference between a missing attribute (not tested) and an empty attribute (tested for emptyness)
20100210 23:27:21< zookeeper> wow.
20100210 23:29:44< zookeeper> silene, i guess which way to change that might need some thinking?
20100210 23:30:32< loonycyborg> Sirp: Probably at cost of dropping support for platforms without hardware acceleration.
20100210 23:30:55< Ivanovic> shadowmaster: in fact i expect close to first day 3d support to be truth for both, intel and amd chips in something like one year
20100210 23:31:06< Ivanovic> shadowmaster: the matter is that they need the infrastructure in place first
20100210 23:31:14< zookeeper> i'd be in favour of changing it to work the way i thought it does, because otherwise there's this extra concept of an "existing but empty attribute" which i don't recall ever really needing
20100210 23:31:22< Sirp> loonycyborg: not at all. you can rotate and scale and flip and do basic effects like that easily. We do it in frogatto all the time.
20100210 23:31:25< zookeeper> anyways, i gotta go for tonight..
20100210 23:31:34-!- zookeeper [~l@wesnoth/developer/zookeeper] has quit []
20100210 23:31:35< Ivanovic> shadowmaster: and amd will release a cpu with integrated gpu in something like a year, this one will most likely have open source drivers with full support directly available
20100210 23:33:10< loonycyborg> Sirp: Wesnoth's current mode is tailored for rendering without acceleration, while opengl is more brute-force. Performance with software gl renderer may be brutal.
20100210 23:33:51< Sirp> loonycyborg: if anyone can produce a platform which gets acceptable performance in Wesnoth now but doesn't in Frogatto now we will look into it. But we've tested a bunch of platforms and they all seem fine with opengl.
20100210 23:34:37< loonycyborg> [00:42:19] <shadowmaster> Sirp: er, yeah... I'll need a überCPU then? It's not like I have a dual core AMD Athlon64 that barely runs Frogatto with Mesa's software render at 1-5 FPS
20100210 23:35:49< Sirp> loonycyborg: I'd have to look at the specifics of shadowmaster's platform. Ivanovic tested on a variety and said it works well.
20100210 23:37:07-!- elias [~elias@allegro/developer/allefant] has quit [Quit: Leaving]
20100210 23:37:22-!- silene [~plouf@wesnoth/developer/silene] has quit [Quit: Leaving.]
20100210 23:38:32< Sirp> shadowmaster: do you have the latest mesa?
20100210 23:43:54-!- YogiHH [~chatzilla@c214216.adsl.hansenet.de] has joined #wesnoth-dev
20100210 23:44:00< loonycyborg> Sirp: Why did you stop your previous opengl port of wesnoth?
20100210 23:44:06-!- YogiHH [~chatzilla@c214216.adsl.hansenet.de] has quit [Changing host]
20100210 23:44:06-!- YogiHH [~chatzilla@wesnoth/developer/yogihh] has joined #wesnoth-dev
20100210 23:44:14< YogiHH> hello
20100210 23:48:03-!- Noyga [~noyga@wesnoth/developer/noyga] has quit [Remote host closed the connection]
20100210 23:48:20-!- grzywacz [~grzywacz@wesnoth/developer/grzywacz] has quit [Remote host closed the connection]
20100210 23:48:24-!- k23z__ [~k23z__@188.26.194.154] has quit [Read error: Connection reset by peer]
20100210 23:50:32< fabi> YogiHH, hello
20100210 23:51:13< YogiHH> fabi: I got a fix so that sides keep their colour across scenarios of mp campaigns.
20100210 23:51:39< YogiHH> fabi: Is there anything else i should have a look at atm?
20100210 23:51:52< fabi> YogiHH, Very nice. Is it the color chosen in the game creation dialog?
20100210 23:52:41< YogiHH> yes, should be, though i haven't tested changing it yet
20100210 23:54:03< fabi> YogiHH, In singleplayer the new turn dialog still says: It's Landar's turn or it's cleodils turn. That is quite confusing when it changes during the campaign when getting more leaders.
20100210 23:55:16< YogiHH> fabi: It's supposed to say "It's Kalenz' turn" then?
20100210 23:55:20< fabi> In multiplayer it should say the players login name or even say "it's your turn". 
20100210 23:55:47< fabi> in singleplayer I would vote just for Player's turn.
20100210 23:56:19< fabi> But that shouldn't fixed to my tastes alone.
20100210 23:56:30< Ivanovic> Sirp: the problem shadowmaster currently has is this one: https://bugs.freedesktop.org/show_bug.cgi?id=26471
20100210 23:56:57< Ivanovic> Sirp: untill this report is solved for real, he won't have hardware accel (unless he is fine with seeing just a black/gray/blue screen
20100210 23:57:52< Ivanovic> Sirp: he is using the mesa version as supplied in debian testing (7.7), which in theory would most likely be fine since it should give him some very basic acceleration, not much but enough
20100210 23:58:29< Ivanovic> his card is some integrated chip (hd3200) which is roughly comparable to intel 4500 chips, most likely stronger from the hardware side, but hardware accelerated 3d is not far enough yet
20100210 23:58:33< Espreon> I thought that shadowmaster used mesa from git.
20100210 23:58:36-!- Aethaeryn [~Michael@wesnoth/umc-dev/developer/aethaeryn] has joined #wesnoth-dev
20100210 23:58:54< Ivanovic> no
20100210 23:59:12< Espreon> Meh... oh well...
20100210 23:59:16< Ivanovic> and even mesa from git requires the patch attached to the report i pointed out to show textures on r6xx/r7xx based cards
--- Log closed Thu Feb 11 00:00:20 2010