--- Log opened Sat Aug 08 00:00:58 2009
20090808 00:01:28-!- Turuk_ [n=Turuk@wesnoth/forumsith/turuk] has quit [Read error: 104 (Connection reset by peer)]
20090808 00:01:43-!- Turuk [n=Turuk@74.83.233.84] has joined #wesnoth-dev
20090808 00:02:10< ilor> Soliton: can you take a look at http://www.wesnoth.org/wiki/MultiplayerRooms ?
20090808 00:14:50-!- silene1 [n=plouf@AMarseille-251-1-58-35.w83-197.abo.wanadoo.fr] has quit ["Leaving."]
20090808 00:16:46 * Soliton clicks.
20090808 00:21:19-!- Sirp [n=user@wesnoth/developer/dave] has joined #wesnoth-dev
20090808 00:21:25-!- grzywacz [n=grzywacz@wesnoth/developer/grzywacz] has quit [Remote closed the connection]
20090808 00:22:02< Sapient> yo Sizrp
20090808 00:23:04< shadowmaster> is that Sirp's lost twin brother from Poland?
20090808 00:23:16-!- elias [n=allefant@allegro/developer/allefant] has joined #wesnoth-dev
20090808 00:23:22-!- allefant [n=allefant@allegro/developer/allefant] has quit [Read error: 110 (Connection timed out)]
20090808 00:23:28< Sapient> ha
20090808 00:23:35< shadowmaster> no wait, it may need some extra consonants at the beginning.
20090808 00:23:49< Sapient> no, it's a really obscure rap reference
20090808 00:24:44-!- Turuk [n=Turuk@wesnoth/forumsith/turuk] has quit [Read error: 60 (Operation timed out)]
20090808 00:25:30< Soliton> ilor: shorting /query to /q as well sounds like a good idea.
20090808 00:25:47< crimson_penguin> Soliton: can't reproduce with a simple Isar's Cross game
20090808 00:26:20-!- Turuk [n=Turuk@wesnoth/forumsith/turuk] has joined #wesnoth-dev
20090808 00:26:25< Soliton> crimson_penguin: maybe there is some issue with some ghost units or something.
20090808 00:27:04< crimson_penguin> Soliton: I'm pretty sure there was only 1 ghost in this game, and though that side did hang, others did too
20090808 00:27:23< Soliton> crimson_penguin: units not players.
20090808 00:27:49< crimson_penguin> I know, but I meant the sides that never had a ghost
20090808 00:28:14< crimson_penguin> there were 2 undeads, I was one of them, and I never got any ghosts, the other one did, and it hung, but other ones than that hung too
20090808 00:28:58< Soliton> i mean "ghost" units.. not visible just in the unit_map for some reason. ;-)
20090808 00:29:23< Soliton> not the actual unit of type ghost.
20090808 00:29:23< crimson_penguin> OH
20090808 00:29:36< crimson_penguin> that could be it
20090808 00:29:59< crimson_penguin> well you could look at the replay of the game I just played (the Royal Rumble Classic)
20090808 00:30:21< Soliton> it wouldn't tell me much.
20090808 00:30:28< crimson_penguin> ok
20090808 00:31:01< Soliton> just maybe that there is other stuff between the [end_turn]s which means it was not issued immediately.
20090808 00:33:35< Soliton> ilor: well, looks good to me.
20090808 00:33:57< Soliton> ilor: i've not tried to imagine what functionality might be missing though.
20090808 00:34:19< ilor> Soliton: did /q, will commit when gna's up
20090808 00:34:19< fendrin> Is there a way to combine several scenario objectives with "and", "or" ?
20090808 00:34:32< Soliton> ilor: cool, thanks.
20090808 00:34:34< ilor> Soliton: I imagine stuff will crop up eventually
20090808 00:34:43< Soliton> yes, me too.
20090808 00:35:26< Soliton> ilor: did you check out my commit that removed admins_? i've changed a bit in the rooms implementation.
20090808 00:35:34< Soliton> trivial stuff though iirc.
20090808 00:35:55< Sapient> fendrin: for defeat conditions, just separate them with a newline to indicate "or"; for victory, newline indicates "and"
20090808 00:36:06< ilor> Soliton might have not noticed, lemme check if I fetched it before gna died
20090808 00:36:32< Soliton> probably not. i think it was one of the last commits. ;-)
20090808 00:36:56< Sapient> fendrin: see bay of pearls in HttT for an example of alternate victory conditions
20090808 00:37:36< fendrin> Sapient: thanks
20090808 00:38:17-!- noy [n=noy@wesnoth/developer/noy] has joined #wesnoth-dev
20090808 00:42:43-!- noy [n=noy@wesnoth/developer/noy] has quit [Client Quit]
20090808 00:46:10< fendrin> How can I filter for a the unit that stands nearest to a location?
20090808 00:48:10< Sapient> you could use a store_unit inside a while loop and let the filter contain a location filter where radius is a variable
20090808 00:48:44< Sapient> the put that inside a loop that increments the radius and have the loop terminate when $stored_unit.length is greater than zero
20090808 00:49:24< Sapient> when I say put it in the loop, I mean the while loop (it's not nested)
20090808 00:50:45< Crab_> fendrin: or you can try to use nearest_loc formula function, together with SUF 'formula' attribute
20090808 00:50:58-!- Turuk_ [n=Turuk@wesnoth/forumsith/turuk] has joined #wesnoth-dev
20090808 00:51:22< Crab_> haven't tried it, but, in theory, it should work (and if doesn't work, it can be fixed)
20090808 00:52:02< fendrin> Crab_: please tell me more
20090808 00:52:05-!- noy [n=noy@wesnoth/developer/noy] has joined #wesnoth-dev
20090808 00:52:16< Sapient> Crab_: formula= attribute in SUF must evaluate to bool type
20090808 00:52:50< Crab_> Sapient: yes, so, it can evaluate to true only for 'unit that is nearest to location'
20090808 00:53:05< Sapient> oh, ok
20090808 00:55:50< Sapient> http://www.wesnoth.org/wiki/FormulaAI_Functions#.27nearest_loc.27_function
20090808 00:55:53< Crab_> fendrin: well, SUF accepts a formula= attribute, which calls a formula to check if unit matches.
20090808 00:56:22< Crab_> fendrin: technically, this can allow to find that 'unit which is nearest to a location' using SUF + fai
20090808 00:56:27< Sapient> looks like the documentation is a little screwy on that
20090808 00:56:52< Crab_> fendrin: but, looking further, it seems that it will not work as intended. but, I can fix this
20090808 00:57:51< Crab_> fendrin: what is better to do -  a) fix formula= in SUF and allow filtering a unit using formula_ai    or  b) allow writing objective checks directly in formula_ai  ?
20090808 00:58:59< Sapient> I don't think objective checks in formula_ai are workable...
20090808 00:59:11< Crab_> Sapient: why not ?
20090808 01:00:08< Sapient> you might as well have generic even handler filtering
20090808 01:00:16< Sapient> *event
20090808 01:01:08< Sapient> [event] name=victory crab_idea="... formula ai ..."
20090808 01:01:09< fendrin> [event] [message] formula=nearest_loc(x,y) message="foo bar!" ?
20090808 01:02:13-!- Turuk [n=Turuk@wesnoth/forumsith/turuk] has quit [Connection timed out]
20090808 01:02:25< fendrin> Crab_: I don't know what "writing objective checks" means.
20090808 01:03:34< Crab_> fendrin: formula_ai check that calls an event with a given name if it evaluates to true
20090808 01:03:38< Sapient> Crab_: can't you write formula="this.loc = nearest_loc(loc($target_x,$larget_y), all_units)" ?
20090808 01:04:16< Sapient> note that I don't know the magic keywords so I put "this" and "all_units" instead
20090808 01:04:45< Crab_> Sapient: atm, you are not allowed to use formula_ai from WML. you can use formulas, but not formula_ai (this can be fixed). SO, nearest_loc is not available directly from WML
20090808 01:05:14< Soliton> why is nearest_loc ai specific?
20090808 01:05:41< Sapient> sounds kinda loco
20090808 01:06:00-!- ABCD [n=ABCD@gentoo/contributor/abcd] has joined #wesnoth-dev
20090808 01:06:06< Sapient> cya later ;)
20090808 01:06:09-!- Sapient [n=patrickp@wesnoth/developer/sapient] has left #wesnoth-dev []
20090808 01:07:03-!- ABCD_ [n=ABCD@gentoo/contributor/abcd] has quit [Read error: 104 (Connection reset by peer)]
20090808 01:07:48< Crab_> fendrin: so, what about 'formula_ai check that calls an event with a given name if it evaluates to true', or 'formula_ai check with result assigned to WML variable' ?
20090808 01:08:53< Crab_> fendrin: the reason for using fai is that writing those kind of checks in it is simple.
20090808 01:09:25< Crab_> for example,  " nearest_loc(loc(17,7),map(my_units,self.loc)) " - returns the location of a nearest of player's units to location (17,7)
20090808 01:10:55< Crab_> or, " unit_at(nearest_loc(loc(12,14),map(units,self.loc))).side " - returns the side number (0-based) of a unit which is nearest to location (12,14)
20090808 01:11:28-!- loonycyborg [n=sergey@wesnoth/developer/loonycyborg] has quit ["KVIrc 3.4.2 Shiny http://www.kvirc.net/"]
20090808 01:12:02-!- loonybot [n=loonybot@wesnoth/bot/loonybot] has quit [Remote closed the connection]
20090808 01:12:09-!- Turuk_ [n=Turuk@wesnoth/forumsith/turuk] has quit [Read error: 60 (Operation timed out)]
20090808 01:12:58-!- fabi [n=fabi@wesnoth/developer/fendrin] has joined #wesnoth-dev
20090808 01:14:10< fabi> Crab_: fendrin = fabi, can you make it usable in a SUF? I believe that is the best solution.
20090808 01:14:37< Crab_> fabi: yes, it's possible
20090808 01:15:19< Crab_> but note that formula in SUF is restricted to returning boolean
20090808 01:15:43< Crab_> so, you will need to use SUF to 'get' the unit(s) for which the formula returned true, and deal with them in WML
20090808 01:16:07< fabi> hmmm
20090808 01:16:44-!- Turuk [n=Turuk@74.83.233.84] has joined #wesnoth-dev
20090808 01:17:13< fabi> Crab_: Isn't that the normal behaviour of the SUF?
20090808 01:17:28< Crab_> yes, that is the normal behavior of SUF :)
20090808 01:17:52< fabi> :-) 
20090808 01:19:46< Crab_> so, if it's ok, I can do this (tomorrow)
20090808 01:20:23-!- ABCD_ [n=ABCD@gentoo/contributor/abcd] has joined #wesnoth-dev
20090808 01:22:06-!- ABCD [n=ABCD@gentoo/contributor/abcd] has quit [Read error: 104 (Connection reset by peer)]
20090808 01:23:15-!- BenUrban [n=benurban@c-68-50-54-86.hsd1.md.comcast.net] has joined #wesnoth-dev
20090808 01:25:56-!- noy [n=noy@wesnoth/developer/noy] has quit ["Get Colloquy for iPhone! http://mobile.colloquy.info"]
20090808 01:26:29-!- fendrin [n=fabi@wesnoth/developer/fendrin] has quit [Read error: 113 (No route to host)]
20090808 01:26:44-!- noy [n=noy@wesnoth/developer/noy] has joined #wesnoth-dev
20090808 01:29:23-!- Valkier [n=karasawa@c-71-225-240-156.hsd1.pa.comcast.net] has quit []
20090808 01:32:57< fabi> Crab_: Would be great. I see some usecases for that.
20090808 01:33:04< Crab_> ok
20090808 01:33:56< Crab_> note that 'direct action' formulas should not be used from that formula (e.g. recruiting, moving units, etc)
20090808 01:34:36-!- ilor [n=user@wesnoth/developer/ilor] has quit []
20090808 01:46:09-!- noy [n=noy@wesnoth/developer/noy] has quit ["Get Colloquy for iPhone! http://mobile.colloquy.info"]
20090808 01:49:53-!- BenUrban [n=benurban@unaffiliated/benurban] has quit ["Power failu"]
20090808 01:57:13-!- Blarumyrran [n=minaise@81-20-159-197.levira.ee] has joined #wesnoth-dev
20090808 02:03:14< Crab_> night
20090808 02:03:22-!- Crab_ [n=Crab_@wesnoth/developer/crab] has quit ["Leaving."]
20090808 02:07:44-!- Turuk_ [n=Turuk@wesnoth/forumsith/turuk] has joined #wesnoth-dev
20090808 02:11:35-!- ABCD_ [n=ABCD@gentoo/contributor/abcd] has quit [Read error: 104 (Connection reset by peer)]
20090808 02:12:08-!- ABCD [n=ABCD@gentoo/contributor/abcd] has joined #wesnoth-dev
20090808 02:17:11-!- lordbyron [n=analog@116.25.21.66] has joined #wesnoth-dev
20090808 02:17:47-!- Turuk [n=Turuk@wesnoth/forumsith/turuk] has quit [Read error: 110 (Connection timed out)]
20090808 02:23:38-!- deekay [n=dk@wesnoth/developer/dragonking] has quit []
20090808 02:24:49-!- ABCD [n=ABCD@gentoo/contributor/abcd] has quit [Read error: 60 (Operation timed out)]
20090808 02:43:08-!- noy [n=Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20090808 02:58:23-!- ancestral [n=ancestra@97-116-126-165.mpls.qwest.net] has joined #wesnoth-dev
20090808 03:14:06< ancestral> Any reason the Ogre is recruitable in Age of Heroes while not in Default?
20090808 03:14:16-!- ABCD [n=ABCD@gentoo/contributor/abcd] has joined #wesnoth-dev
20090808 03:14:33< Blarumyrran> Because noone cares about balance in Age of Heroes :>
20090808 03:14:39< Blueblaze> ^
20090808 03:14:42< Blueblaze> Maybe you should ask the question the other way around too ancestral
20090808 03:14:52< ancestral> Sure?
20090808 03:15:22< ancestral> Actually it was meant to be the other way around. Didn't know that Ogres were loyalists
20090808 03:15:56< Aethaeryn> Eastern Invasion
20090808 03:16:16< Aethaeryn> shows how ogres are loy
20090808 03:16:18< Aethaeryn> iirc
20090808 03:16:40< ancestral> So all the units in AoH are the same as in Default
20090808 03:16:43< ancestral> Except for the Ogre
20090808 03:18:36< Soliton> svn is back.
20090808 03:18:44< Aethaeryn> yay...
20090808 03:18:54< Aethaeryn> time to run update and find out that my revision is actually *ahead* of SVN ;)
20090808 03:19:02< Aethaeryn> and break things
20090808 03:19:17< Soliton> nah, nothing lost.
20090808 03:19:53< Aethaeryn> \o/
20090808 03:19:59< Aethaeryn> At revision 37534.
20090808 03:20:09< CIA-62> jetryl * r37535 /trunk/data/core/images/units/human-magi/ (112 files): Shadow updates for the human magi.
20090808 03:20:11< Aethaeryn> Because no one could upload all day I'm actually at the current revision
20090808 03:20:13< Aethaeryn> DAMN IT
20090808 03:20:20< Aethaeryn> I'm never current.
20090808 03:20:45< ancestral> Haha Jet for the win
20090808 03:20:57< Aethaeryn> omfg, that's a big update.
20090808 03:21:04< Aethaeryn> 112 files, right after I say I'm current. 10 seconds after
20090808 03:21:06< Aethaeryn> that *is* win
20090808 03:21:15-!- Turuk_ [n=Turuk@wesnoth/forumsith/turuk] has quit [Read error: 104 (Connection reset by peer)]
20090808 03:21:23< CIA-62> jetryl * r37536 /branches/1.6/data/core/images/units/human-magi/ (112 files): Shadow updates for the human magi.
20090808 03:21:30-!- Turuk [n=Turuk@wesnoth/forumsith/turuk] has joined #wesnoth-dev
20090808 03:22:03< Aethaeryn> omfg
20090808 03:22:14< Aethaeryn> why must someone always update while I'm running scons?
20090808 03:23:27< Blueblaze> Because someone is -always- updating Wesnoth
20090808 03:23:49< Blarumyrran> It's just jetrel using multiple accounts
20090808 03:25:56< CIA-62> jetryl * r37537 /branches/1.6/data/core/images/units/undead-necromancers/ (55 files): Shadow updates on the necromancer line.
20090808 03:26:33< CIA-62> jetryl * r37538 /trunk/data/core/images/units/undead-necromancers/ (55 files): Shadow updates on the necromancer line.
20090808 03:29:20< CIA-62> jetryl * r37539 /trunk/data/core/images/units/human-loyalists/ (112 files): Shadow updates on some of the loyalists.
20090808 03:30:00< CIA-62> jetryl * r37540 /branches/1.6/data/core/images/units/human-loyalists/ (112 files): Shadow updates on some of the loyalists.
20090808 03:30:05< Aethaeryn> lol
20090808 03:30:12< Aethaeryn> I think I'll wait before updating if he's doing *everything*
20090808 03:51:21-!- Blarumyrran [n=minaise@81-20-159-197.levira.ee] has quit []
20090808 04:00:28< corn> svn working again but commit emails broken... 1 step forward, 2 steps backward :)
20090808 04:04:33-!- Turuk [n=Turuk@wesnoth/forumsith/turuk] has quit [Read error: 60 (Operation timed out)]
20090808 04:04:52-!- Turuk [n=Turuk@wesnoth/forumsith/turuk] has joined #wesnoth-dev
20090808 04:15:05-!- noy [n=Noy@wesnoth/developer/noy] has quit []
20090808 04:23:10-!- Elvish_Pillager [n=eli@71-10-231-36.dhcp.oxfr.ma.charter.com] has quit [Read error: 110 (Connection timed out)]
20090808 04:51:48-!- Ivanovic_ [n=ivanovic@dtmd-4db26d70.pool.einsundeins.de] has joined #wesnoth-dev
20090808 05:08:38-!- Ivanovic [n=ivanovic@wesnoth/developer/ivanovic] has quit [Read error: 113 (No route to host)]
20090808 05:09:46-!- Ivanovic_ is now known as Ivanovic
20090808 05:33:18-!- lordbyron [n=analog@116.25.21.66] has left #wesnoth-dev []
20090808 05:57:04-!- BenUrban [n=benurban@c-68-50-54-86.hsd1.md.comcast.net] has joined #wesnoth-dev
20090808 05:59:09-!- crimson_penguin [n=ben@wesnoth/developer/crimsonpenguin] has quit []
20090808 06:16:31-!- stikonas [n=and@ctv-213-164-123-204.vinita.lt] has joined #wesnoth-dev
20090808 06:34:50-!- Netsplit over, joins: AI0867
20090808 06:34:50-!- Netsplit clarke.freenode.net <-> irc.freenode.net quits: corn
20090808 06:34:55-!- corn [n=cornmand@cornmander.com] has joined #wesnoth-dev
20090808 06:46:40-!- Aethaeryn [n=Michael@wesnoth/umc-dev/developer/aethaeryn] has quit ["quit"]
20090808 07:33:59-!- silene [n=plouf@AMarseille-251-1-58-35.w83-197.abo.wanadoo.fr] has joined #wesnoth-dev
20090808 08:05:03-!- Sirp [n=user@wesnoth/developer/dave] has quit [Read error: 113 (No route to host)]
20090808 08:18:01-!- conor [n=conor@c-67-180-158-41.hsd1.ca.comcast.net] has joined #wesnoth-dev
20090808 08:19:52-!- stikonas [n=and@wesnoth/translator/stikonas] has quit [Read error: 104 (Connection reset by peer)]
20090808 08:33:10-!- elias [n=allefant@allegro/developer/allefant] has quit [Read error: 110 (Connection timed out)]
20090808 08:33:39-!- elias [n=allefant@allegro/developer/allefant] has joined #wesnoth-dev
20090808 08:35:33-!- stikonas [n=and@ctv-213-164-123-204.vinita.lt] has joined #wesnoth-dev
20090808 08:43:47-!- busfahrer [n=busfahre@unixboard/user/busfahrer] has joined #wesnoth-dev
20090808 08:46:46-!- stikonas [n=and@wesnoth/translator/stikonas] has quit [Remote closed the connection]
20090808 09:12:17-!- mordante [n=mordante@wesnoth/developer/mordante] has joined #wesnoth-dev
20090808 09:12:57< mordante> servus
20090808 09:16:04< fabi> mordante: bonjour
20090808 09:16:31< mordante> hoi fabi 
20090808 09:18:56-!- stikonas [n=and@wesnoth/translator/stikonas] has joined #wesnoth-dev
20090808 09:21:03< CIA-62> silene * r37541 /trunk/src/menu_events.cpp: Fixed recruit and recall user actions not registering the proper position in the replay. (Fix for bug #14085.) Cleaned the code along the way.
20090808 09:34:47 * fabi can't compile
20090808 09:35:31< fabi> The libboost regex in jaunty isn't good enough for scons. (1.35)
20090808 09:41:03< mordante> 1.35 should work
20090808 09:42:27< fabi> No it didn't. But replacing it with 1.37 solved the problem.
20090808 09:43:45< Soliton> you were probably missing serialization, the package dependencies are not complete there.
20090808 09:44:11< Soliton> 1.35 is definately enough.
20090808 09:45:54< fabi> compilation takes forever lately.
20090808 09:46:14< fabi> Need more cores!
20090808 09:48:27< Blueblaze> how many do you have?
20090808 09:49:27< fabi> 2x1.8ghz
20090808 09:54:13< fabi> There is a new dwarven scout unit, is it a good thing to add it into a campaign recruit list?
20090808 09:54:42< Blueblaze> If it fits for a specific campaign
20090808 09:59:31< fabi> mordante: you solved the transparent campaign image in the campaign dialogue by downscaling?
20090808 10:01:40< mordante> I didn't do a thing, so if solved somebody else did
20090808 10:02:43 * fabi loves such things that fix themselves :-)
20090808 10:03:11 * mordante doesn't
20090808 10:07:12-!- zookeeper [n=l@wesnoth/developer/zookeeper] has joined #wesnoth-dev
20090808 10:26:22-!- giusef [n=giusef@unaffiliated/giusef] has joined #wesnoth-dev
20090808 10:31:16-!- mordante [n=mordante@wesnoth/developer/mordante] has quit ["Leaving"]
20090808 10:36:10-!- Noyga [n=lame-z@wesnoth/developer/noyga] has joined #wesnoth-dev
20090808 10:38:09-!- busfahrer [n=busfahre@unixboard/user/busfahrer] has quit ["leaving"]
20090808 10:41:01-!- lizard_r [n=Miranda@wesnoth/umc-dev/developer/lizard] has joined #wesnoth-dev
20090808 10:52:48-!- grzywacz [n=grzywacz@wesnoth/developer/grzywacz] has joined #wesnoth-dev
20090808 10:55:13-!- sorsis [n=Mikael_H@hoasnet-fe22dd00-90.dhcp.inet.fi] has joined #wesnoth-dev
20090808 10:58:36< sorsis> I have been thinking of setting up wesnoth-ranking server. what do you guys think of the idea?
20090808 11:00:46< sorsis> I could do the work related to the programming and databasemanagement
20090808 11:01:51-!- loonybot [n=loonybot@79.139.138.234] has joined #wesnoth-dev
20090808 11:02:38-!- loonycyborg [n=sergey@79.139.138.234] has joined #wesnoth-dev
20090808 11:05:41-!- deekay [n=dk@wesnoth/developer/dragonking] has joined #wesnoth-dev
20090808 11:08:42< sorsis> server could be used to balancing of sides and player skill-level-based automatching for "quickplay" or something.
20090808 11:11:21< Soliton> there is an unofficial ladder.
20090808 11:12:07< sorsis> where?
20090808 11:12:20< sorsis> so my work is not needed
20090808 11:16:53< Soliton> well, if someone wants to setup a server to better support ranking they might like that. it's unlikely to be an official one though.
20090808 11:17:11< Soliton> http://ladder.subversiva.org/
20090808 11:23:29< sorsis> what would need an official-ladder-server?
20090808 11:23:53< Ivanovic> moin
20090808 11:24:46< Soliton> convincing a lot of devs that it's a good idea.
20090808 11:24:52< Blueblaze> sorsis: The question isn't what is needed to make an official ladder server, the question is the do we want an official ladder server, and the general answer is no.
20090808 11:26:06< sorsis> ok...
20090808 11:26:17< sorsis> do you need hel in development? :)
20090808 11:26:20< sorsis> hel = help
20090808 11:26:40< Soliton> always. ;-)
20090808 11:26:49< sorsis> yay :)
20090808 11:28:12< sorsis> so if I click "Send a request to join this project" in source forge, my apply would be accepted probaply?
20090808 11:28:20< sorsis> what kind of help you neeD?
20090808 11:29:21< Soliton> well, we usually like to see two non-trivial patches before we grant commit access.
20090808 11:29:32< sorsis> ok...
20090808 11:29:46< Soliton> bugs.wesnoth.org is a good place to start, i suppose.
20090808 11:30:04< Soliton> or the easy coding page in the wiki.
20090808 11:32:21-!- ilor [n=user@wesnoth/developer/ilor] has joined #wesnoth-dev
20090808 11:35:19-!- ancestral [n=ancestra@97-116-126-165.mpls.qwest.net] has quit []
20090808 11:41:36-!- Turuk [n=Turuk@wesnoth/forumsith/turuk] has quit [Read error: 60 (Operation timed out)]
20090808 11:41:40-!- Turuk [n=Turuk@wesnoth/forumsith/turuk] has joined #wesnoth-dev
20090808 11:58:40< fabi> error audio: Could not load sound with empty filename -- Does this mean the file is not found?
20090808 12:00:11-!- Crab_ [n=Crab_@wesnoth/developer/crab] has joined #wesnoth-dev
20090808 12:00:15< Crab_> hi
20090808 12:02:55< silene> fabi: usually yes; it means that the sound part of the engine tried to load without first checking whether it existed or not
20090808 12:03:43< fabi> silene: could this be caused by using an ogg instead of an wav?
20090808 12:04:28< silene> fabi: no, the filesystem part doesn't care about extensions (and it's a message from the filesystem part, not from the sound part)
20090808 12:06:56< silene> fabi: i am about to commit something for making the error message less idiotic
20090808 12:08:00< silene> (once my computer recovers from overheating)
20090808 12:09:14-!- wesbot changed the topic of #wesnoth-dev to: 70 bugs, 239 feature requests, 14 patches | logs: http://irclogs.wesnoth.org/ | Don't paste on IRC! Use a pastebin: http://wesnoth.pastebin.com | http://imagebin.org
20090808 12:09:51< CIA-62> silene * r37542 /trunk/src/sound.cpp: Fixed idiotic error message, as the file name is usually not empty.
20090808 12:13:18< fabi> fabi: to print the name of the path that failed would help me.
20090808 12:13:28< fabi> silene: ^
20090808 12:13:46< fabi> I am talking to myself.
20090808 12:16:41< silene> fabi: that should be the case; if it still diplays no path, then the filename was really empty
20090808 12:18:52< fabi> hmm
20090808 12:19:52< fabi> silene: I have to go now. Will investigate later.
20090808 12:19:53< fabi> bye
20090808 12:28:45-!- Elvish_Pillager [n=eli@71-10-231-36.dhcp.oxfr.ma.charter.com] has joined #wesnoth-dev
20090808 12:30:04-!- giusef [n=giusef@unaffiliated/giusef] has quit ["exit (-1);"]
20090808 12:50:53< Soliton> wesnothd still has that mysterious memory allocation crash: http://nopaste.com/p/aW57nQ8hC
20090808 13:00:58< ilor> Soliton: that backtrace looks like it's missing a 7.5 frame, any ideas why it won't show the line where the new is called?
20090808 13:01:49< Soliton> inlined, i think.
20090808 13:05:03< Soliton> ah, it could also be in output() i'll make the index there size_t as well.
20090808 13:05:23< ilor> Soliton: does the crash contain any info at al other than new failed?
20090808 13:05:47< Soliton> well, what would you like to see?
20090808 13:07:47< ilor> Soliton: good question. I'd love to see which new actually fails ;)
20090808 13:08:03< Soliton> it pretty much has to be something like "new char[-1];"?
20090808 13:08:29< Soliton> and that can't happen in compress_buffer() anymore.
20090808 13:08:31< Crab_> ilor: this will not give you anything important, if it is a normal alloc which fails because something else leaks...
20090808 13:08:50< ilor> Crab_: we'd have noticed a massive leak I think
20090808 13:09:04< Soliton> the server is not actually out of memore.
20090808 13:09:04< Crab_> Soliton: do we have stats on server memory usage ?
20090808 13:09:09< Crab_> Soliton: ok
20090808 13:09:11< Soliton> collectd.wesnoth.org
20090808 13:09:34< Soliton> Crab_: that's actually not how memory allocation works on linux as i've learned.
20090808 13:09:58< Soliton> the OS always gives you memory no problem.
20090808 13:10:25< Crab_> Soliton: even if you hit a certain per-process limit ?
20090808 13:10:29< Soliton> if it gets into trouble oom-killer does its work.
20090808 13:10:53< Soliton> so it really has to be some insane request like with a negative value.
20090808 13:11:25< Soliton> Crab_: well, yeah if you really make a huge request it'll happen.
20090808 13:11:42< Soliton> but it's not a huge request here or shouldn't be.
20090808 13:12:22< ilor> Soliton: I'm out oif ideas other than add debug logging of allocation sizes around there
20090808 13:13:16< Soliton> ilor: the new in output() is a candidate, no?
20090808 13:13:33< ilor> Soliton: yes I definitely think so
20090808 13:14:00< Crab_> ilor: if we known potential offending places, can't we surround them with try { .. } catch { log, rethrow} blocks ?
20090808 13:14:00< CIA-62> ilor * r37543 /trunk/src/server/room_manager.cpp: fix default room creation policy on server
20090808 13:14:18< Soliton> though it looks like it's weird if output_size() returns a huge/negative value.
20090808 13:14:45< CIA-62> ilor * r37544 /trunk/doc/man/wesnothd.6: wesnthd manpage update
20090808 13:15:13< CIA-62> ilor * r37545 /trunk/ (86 files in 16 dirs): rename editor2 -> editor. Old editor is long dead, long live the new editor!
20090808 13:15:22< Soliton> \o/
20090808 13:15:32< ilor> mordante: r37545 :)
20090808 13:15:47< CIA-62> ilor * r37546 /trunk/ (changelog src/menu_events.cpp): add a /q alias for /query
20090808 13:16:01< ilor> I tried to be thorough in the change, even did the devcpp project files :D
20090808 13:17:12-!- Turuk [n=Turuk@wesnoth/forumsith/turuk] has quit [Read error: 104 (Connection reset by peer)]
20090808 13:17:27-!- Turuk [n=Turuk@wesnoth/forumsith/turuk] has joined #wesnoth-dev
20090808 13:24:50< Soliton> is a try catch block going to slow things down much?
20090808 13:27:29< ilor> Soliton: not that much so we can't afford it to test
20090808 13:27:42< ilor> Soliton: you can always use new (nothrow) and then check the pointer for null
20090808 13:29:34< Soliton> i guess if i try catch i can make the type signed again since that is faster?
20090808 13:30:34< ilor> Soliton: I don't think this makes any difference in the compiled code
20090808 13:43:38-!- ABCD [n=ABCD@gentoo/contributor/abcd] has quit [Remote closed the connection]
20090808 13:45:58< silene> Soliton: try catch doesn't cost anything as long as no exception is thrown; a thrown exception, however, is incredibly costly
20090808 13:49:42< Soliton> cool. when the exception is thrown i don't care much.
20090808 13:50:11< silene> Soliton: it depends, if the exception can be thrown on demand by a user, you get a DoS
20090808 13:51:07< Soliton> doesn't make much difference to the user being able to crash the server on demand.
20090808 13:53:24< silene> sure, i was mostly talking about ilor's suggestion of using the nothrow new; you have to keep in mind that the throwing new is about 20,000 slower in case of a failing allocation than the nothrow new
20090808 13:54:28< Crab_> silene: well, it is wanted to 'find out where it fails', not to 'gracefully handle failure'
20090808 13:54:29-!- Elvish_Pillager [n=eli@71-10-231-36.dhcp.oxfr.ma.charter.com] has quit [Read error: 104 (Connection reset by peer)]
20090808 13:54:57-!- Elvish_Pillager [n=eli@71-10-231-36.dhcp.oxfr.ma.charter.com] has joined #wesnoth-dev
20090808 14:02:15-!- sorsis [n=Mikael_H@hoasnet-fe22dd00-90.dhcp.inet.fi] has left #wesnoth-dev []
20090808 14:02:48< CIA-62> crab * r37547 /trunk/src/ai/ (configuration.hpp manager.cpp): Properly initialized command-line ai config. formula_ai console should be now fully operational.
20090808 14:03:53< CIA-62> crab * r37548 /trunk/ (21 files in 9 dirs): new formula_ai-related stages
20090808 14:11:08-!- lizard_r [n=Miranda@wesnoth/umc-dev/developer/lizard] has quit ["Saurian Augur - I'll heal you by 4 hp if you leave next to me"]
20090808 14:11:09-!- Elvish_Pillager [n=eli@71-10-231-36.dhcp.oxfr.ma.charter.com] has quit [Read error: 54 (Connection reset by peer)]
20090808 14:11:31-!- Elvish_Pillager [n=eli@71-10-231-36.dhcp.oxfr.ma.charter.com] has joined #wesnoth-dev
20090808 14:15:06-!- conor [n=conor@c-67-180-158-41.hsd1.ca.comcast.net] has quit ["Leaving."]
20090808 14:34:02-!- Blueblaze [n=nick@c-98-199-143-139.hsd1.tx.comcast.net] has quit [Remote closed the connection]
20090808 14:38:03< CIA-62> soliton * r37549 /branches/1.6/src/server/simple_wml.cpp: trying to catch a bad alloc request
20090808 14:38:38< CIA-62> soliton * r37550 /trunk/src/server/simple_wml.cpp: trying to catch a bad alloc request
20090808 14:42:50-!- lizard_r [n=Miranda@wesnoth/umc-dev/developer/lizard] has joined #wesnoth-dev
20090808 15:04:57-!- ancestral [n=ancestra@97-116-126-165.mpls.qwest.net] has joined #wesnoth-dev
20090808 15:12:31-!- Elvish_Pillage2 [n=eli@71-10-231-36.dhcp.oxfr.ma.charter.com] has joined #wesnoth-dev
20090808 15:16:41-!- Elvish_Pillager [n=eli@71-10-231-36.dhcp.oxfr.ma.charter.com] has quit [Read error: 60 (Operation timed out)]
20090808 15:31:42-!- fabi [n=fabi@wesnoth/developer/fendrin] has quit [Read error: 104 (Connection reset by peer)]
20090808 15:32:17-!- crimson_penguin [n=ben@64.201.60.211] has joined #wesnoth-dev
20090808 15:42:44-!- stikonas_ [n=and@ctv-213-164-123-204.vinita.lt] has joined #wesnoth-dev
20090808 15:54:11-!- Elvish_Pillage2 [n=eli@71-10-231-36.dhcp.oxfr.ma.charter.com] has quit [Read error: 113 (No route to host)]
20090808 15:55:54-!- stikonas_ [n=and@ctv-213-164-123-204.vinita.lt] has quit ["Konversation terminated!"]
20090808 15:56:37-!- stikonas [n=and@wesnoth/translator/stikonas] has quit [Read error: 110 (Connection timed out)]
20090808 15:56:50-!- stikonas [n=and@wesnoth/translator/stikonas] has joined #wesnoth-dev
20090808 16:05:42-!- Sirp [n=user@pool-71-164-166-178.dllstx.fios.verizon.net] has joined #wesnoth-dev
20090808 16:13:30-!- ancestral [n=ancestra@97-116-126-165.mpls.qwest.net] has quit ["And that’s the end of THAT chapter."]
20090808 16:15:09-!- BenUrban [n=benurban@unaffiliated/benurban] has quit [Remote closed the connection]
20090808 16:23:08-!- ilor [n=user@wesnoth/developer/ilor] has quit []
20090808 16:24:08-!- stikonas [n=and@wesnoth/translator/stikonas] has quit [Remote closed the connection]
20090808 16:27:10-!- crimson_penguin [n=ben@wesnoth/developer/crimsonpenguin] has quit [Read error: 104 (Connection reset by peer)]
20090808 16:28:28-!- allefant [n=allefant@allegro/developer/allefant] has joined #wesnoth-dev
20090808 16:28:48-!- elias [n=allefant@allegro/developer/allefant] has quit [Read error: 110 (Connection timed out)]
20090808 16:29:39-!- stikonas [n=and@ctv-213-164-123-204.vinita.lt] has joined #wesnoth-dev
20090808 16:57:13-!- crimson_penguin [n=ben@64.201.60.211] has joined #wesnoth-dev
20090808 16:57:48-!- silene [n=plouf@wesnoth/developer/silene] has quit ["Leaving."]
20090808 17:15:32-!- grzywacz [n=grzywacz@wesnoth/developer/grzywacz] has quit [Remote closed the connection]
20090808 17:39:31-!- mordante [n=mordante@wesnoth/developer/mordante] has joined #wesnoth-dev
20090808 17:39:45< mordante> servus
20090808 17:41:56< mordante> ilor thanks for finishing last years gsoc project ;-P
20090808 17:51:43-!- Elvish_Pillager [n=eli@71-10-231-36.dhcp.oxfr.ma.charter.com] has joined #wesnoth-dev
20090808 17:53:23-!- isaac_ is now known as isaac
20090808 17:54:51-!- ABCD [n=ABCD@gentoo/contributor/abcd] has joined #wesnoth-dev
20090808 18:02:13-!- ABCD_ [n=ABCD@gentoo/contributor/abcd] has joined #wesnoth-dev
20090808 18:07:15-!- ABCD_away [n=ABCD@gentoo/contributor/abcd] has joined #wesnoth-dev
20090808 18:07:43-!- ABCD_ [n=ABCD@gentoo/contributor/abcd] has quit [Read error: 104 (Connection reset by peer)]
20090808 18:08:43-!- ABCD [n=ABCD@gentoo/contributor/abcd] has quit [Read error: 104 (Connection reset by peer)]
20090808 18:08:56-!- ABCD_away is now known as ABCD
20090808 18:09:14-!- wesbot changed the topic of #wesnoth-dev to: 69 bugs, 239 feature requests, 14 patches | logs: http://irclogs.wesnoth.org/ | Don't paste on IRC! Use a pastebin: http://wesnoth.pastebin.com | http://imagebin.org
20090808 18:11:01 * Ivanovic tries to summon sapient
20090808 18:11:09< Ivanovic> Sapient: how is cjhopman doing?
20090808 18:11:23 * mordante joins the summoning
20090808 18:14:03-!- ilor [n=user@wesnoth/developer/ilor] has joined #wesnoth-dev
20090808 18:14:13< mordante> hi ilor 
20090808 18:14:24< ilor> hi mordante
20090808 18:14:28< mordante> congratulations with finishing your gsoc 2008 project ;-)
20090808 18:14:32< ilor> thanks :)
20090808 18:14:33< ilor> :P
20090808 18:15:22< ilor> there's a room system doc on the wiki at http://www.wesnoth.org/wiki/MultiplayerRooms and I did the server manpage (though you probably know that from the commit)
20090808 18:17:38< mordante> I read the commit message but haven't read the man page yet
20090808 18:17:43< mordante> nor the wikipage
20090808 18:23:35< mordante> ilor, both look nice only the wiki page /room <room> <message> should have some info that this could create a new room, but might be disabled by the server administrator
20090808 18:24:35< ilor> mordante this one won't create a room
20090808 18:25:03< mordante> it used to create a room didn't it?
20090808 18:26:15< mordante> btw I won't code to much this weekend way too hot, so will do some minor cleanups
20090808 18:26:33< mordante> hope it cools down again next week so I can work on the major bugs
20090808 18:34:06-!- Crab_ [n=Crab_@wesnoth/developer/crab] has quit ["Leaving."]
20090808 18:34:31-!- Crab_ [n=Crab_@wesnoth/developer/crab] has joined #wesnoth-dev
20090808 18:43:21< CIA-62> crab * r37551 /trunk/src/ai/ (5 files in 3 dirs): fixed a segfault, and added code to allow getting 'variant' from aspects, to be used by formula_ai
20090808 18:51:59< ilor> afk
20090808 20:07:06-!- Noyga [n=lame-z@wesnoth/developer/noyga] has quit [Read error: 110 (Connection timed out)]
20090808 20:24:45-!- Noyga [n=lame-z@wesnoth/developer/noyga] has joined #wesnoth-dev
20090808 20:25:29-!- Noyga [n=lame-z@wesnoth/developer/noyga] has left #wesnoth-dev ["Quitte"]
20090808 20:26:11< CIA-62> crab * r37552 /trunk/src/ (7 files in 3 dirs): removed prepare_move from formula_ai, it is not needed now
20090808 20:39:32-!- Crab_ [n=Crab_@wesnoth/developer/crab] has quit ["Leaving."]
20090808 20:43:20-!- BenUrban [n=benurban@pool-71-191-100-51.washdc.east.verizon.net] has joined #wesnoth-dev
20090808 20:48:22-!- Crab_ [n=Crab_@wesnoth/developer/crab] has joined #wesnoth-dev
20090808 20:49:20-!- lizard_r [n=Miranda@wesnoth/umc-dev/developer/lizard] has quit ["Saurian Augur - I'll heal you by 4 hp if you leave next to me"]
20090808 21:05:40-!- ABCD [n=ABCD@gentoo/contributor/abcd] has quit [Read error: 104 (Connection reset by peer)]
20090808 21:08:07-!- Tigge [n=tigge@bacchus.olf.sgsnet.se] has quit [Read error: 104 (Connection reset by peer)]
20090808 21:34:59-!- giusef [n=giusef@unaffiliated/giusef] has joined #wesnoth-dev
20090808 21:36:05< corn> mordante: my gsoc is mostly on track, I am adding kill event info to the upload logs so that I can visualize it on the wesnoth gmaps I made. I wanted to have this done a week ago but it turned out to be a bit more complex than I thought
20090808 21:39:35< mordante> corn, nice is there more to see at your prototype site at the moment?
20090808 21:41:09< corn> http://cornmander.com/wesstats/killmaps
20090808 21:41:12< corn> not sure if you have seen that
20090808 21:41:50< corn> the site already generates tilesets for all the maps that get sent to it in upload logs, but I don't have a UI coded that lets you switch it on the gmap link
20090808 21:44:54< mordante> I haven't seen the popup for the hexes you click on, but a stupid question.. why is it called a killmap? Should I see something regarding kills already?
20090808 21:46:02< corn> no
20090808 21:46:06< corn> the name comes from the original idea
20090808 21:46:19< corn> the hexes are currently all just a shade of black, but they will be colored according to frequency of kills
20090808 21:51:53< mordante> ah ok, I already wondered about the name before
20090808 21:59:52-!- Aethaeryn [n=Michael@wesnoth/umc-dev/developer/aethaeryn] has joined #Wesnoth-dev
20090808 22:17:23-!- ilor [n=user@wesnoth/developer/ilor] has quit [Read error: 113 (No route to host)]
20090808 22:22:26-!- ilor [n=user@wesnoth/developer/ilor] has joined #wesnoth-dev
20090808 22:22:59< CIA-62> mordante * r37553 /trunk/src/gui/ (auxiliary/window_builder/button.cpp widgets/settings.cpp): Update button documentation.
20090808 22:23:09< CIA-62> mordante * r37554 /trunk/src/gui/ (auxiliary/window_builder/image.cpp widgets/settings.cpp): Update image documentation.
20090808 22:23:15< CIA-62> mordante * r37555 /trunk/src/gui/ (2 files in 2 dirs): Update horizontal scrollbar documentation.
20090808 22:23:19< CIA-62> mordante * r37556 /trunk/src/gui/widgets/listbox.cpp: 
20090808 22:23:19< CIA-62> Add an error message when using deprecated code.
20090808 22:23:19< CIA-62> Trunk isn't clean of this error yet, but will be fixed later.
20090808 22:23:34< ilor> mordante: btw iirc /room never created a room, it requires the client be in the room already or the message is discarded
20090808 22:24:02< mordante> how do you create a room then?
20090808 22:24:41< ilor> mordante: by joining one
20090808 22:25:02< mordante> btw did you find a way to reproduce the dynamic_cast problem reliable?
20090808 22:25:10< ilor> sadly not :/
20090808 22:25:30< mordante> :-(
20090808 22:25:46< ilor> lingering in 1.7 lobby did the trick a bunch of times, but never noticed what exactly triggers it
20090808 22:26:23< mordante> I think I have a fix, which I'm compiling at the moment, but can only test whether it doesn't cause other problems
20090808 22:26:58< mordante> if /join creates the room then my remark regarding /room <room> <message> applies to that ;-)
20090808 22:29:56-!- PK7 [n=pk@r74-192-44-206.vctrcmta01.vctatx.tl.dh.suddenlink.net] has joined #wesnoth-dev
20090808 22:33:38< ilor> mordante: oh and btw, I'm considering buying a proper cpu cooler and shipping it to your address :P
20090808 22:33:52< mordante> ilor, :-)
20090808 22:34:26< mordante> ilor, I bought a new computer... and returned it again since the shop sucked and made a mess of my order
20090808 22:34:42< mordante> so went looking for a new one again today
20090808 22:35:14< mordante> so I hope to have a new one again soon
20090808 22:35:55< ilor> nice :) I'm looking into spending some of this year's gsoc cash on a new computer, but I'll be just ordering parts
20090808 22:36:35< mordante> I considered that as well, but don't want to bother with the parts too much and buying in parts is often more expensive
20090808 22:36:53< ilor> parts is usually less expensive here
20090808 22:37:03< mordante> we have some shops that can (at least in theory) build the system to your wishes
20090808 22:37:18< corn> ilor, mordante: where is this, for both of you?
20090808 22:37:25< ilor> corn: Poland for me
20090808 22:37:26< mordante> the Netherlands
20090808 22:37:27< corn> in the US it is always cheaper to build yourself
20090808 22:38:00< mordante> often if you buy the parts at different shops to your wishes they all want to make profit
20090808 22:38:00< corn> I built a computer in may, I think I saved 15% or so over buying it from one of the brands (hp, dell, etc)
20090808 22:38:09< corn> can you order online?
20090808 22:38:37< mordante> I don't want to buy online (would be cheaper but more hassle if you have problems)
20090808 22:38:45< corn> ah, that's true
20090808 22:38:45-!- Tigge [n=tigge@bacchus.olf.sgsnet.se] has joined #wesnoth-dev
20090808 22:38:48< ilor> I'm generally ordering through a friend's shop
20090808 22:39:05< mordante> I don't need the cheapest system, I just want to have something good and reliable
20090808 22:39:19< mordante> but I guess some sales people don't understand that :-(
20090808 22:39:25< corn> heh
20090808 22:39:44< mordante> aint' funny :-(
20090808 22:40:01< ilor> mordante: what are you aiming at, hardware-wise? :)
20090808 22:40:28< mordante> AMD 905e 8GB RAM 2*500GB software RAID 1
20090808 22:41:05< ilor> mordante: 905e?
20090808 22:41:07< corn> oh, that is pretty nice
20090808 22:41:21< mordante> I really like the 905e since it only uses 45W so the fans can remain silent
20090808 22:41:40< corn> http://products.amd.com/en-us/DesktopCPUDetail.aspx?id=557
20090808 22:41:44< corn> says it uses 65W
20090808 22:41:55< mordante> ilor, http://products.amd.com/en-us/DesktopCPUDetail.aspx?id=557
20090808 22:41:58< ilor> mordante: ah, not available at the wholesaler I usually use
20090808 22:42:09< mordante> corn, ah yes typo
20090808 22:42:34< corn> heh, that is half the wattage of my i7 920
20090808 22:42:40< ilor> mordante: well you can always get a normal one and underclock and undervolt :P
20090808 22:43:01< ilor> corn: oh, you've got an i7, how are build times looking for you?
20090808 22:43:07< mordante> ilor, also possible but this is a rather nice CPI
20090808 22:43:10< mordante> CPU*
20090808 22:43:19< corn> doing scons -j5, I can build in >3 minutes
20090808 22:43:51< mordante> ha that AMD does it in less than 3 minutes with cmake -j5
20090808 22:43:55< ilor> corn: why not -j8 / -j9?
20090808 22:44:02-!- BenUrban [n=benurban@unaffiliated/benurban] has quit []
20090808 22:44:26< corn> I don't think I have hyperthreading enabled, and afaik it doesn't give a significant boost
20090808 22:44:28< mordante> building the kernel in about 12 minutes, but those **** put in the wrong hdd so the hdd instead of the cpu was the bottleneck
20090808 22:45:09< ilor> corn: I started respectnig HT after a raytracer I wrote for univerisity got a 30% performance boost with 2 threads on a P4 with HT
20090808 22:45:41< corn> ok. I will run benchmarks *again* on monday. right now I am developing on an old pentium m with 512 mb of ram
20090808 22:46:18< corn> my last set of benchmarks was regarding the difference between -j5 and -j4, and I found that -j5 does infact give a slight performance increase over -j4
20090808 22:46:21< mordante> IMO the old P4 hyperthreading sucked, slowed down the machine quite often. The new HT seems to be much better
20090808 22:46:27< ilor> I haven't really decided on the cpu I'm going to get, i7 looks nice but pricewise it's not stellar
20090808 22:46:28< Ivanovic> ilor: only the case for P4
20090808 22:46:28< corn> I think the compile times were 2:15 vs 2:45 or so
20090808 22:46:45< Ivanovic> they were inefficent as hell because the pipeline was too long resulting in too many stalls
20090808 22:46:53< corn> ilor: i7 920 has a lot of overclocking potential, people have been able to run it at i7 960 speeds
20090808 22:46:55< Ivanovic> that is basically when they could intorduce HT
20090808 22:47:26< ilor> corn: yeah I know that getting around 3.5GHz takes little effort
20090808 22:47:28< corn> do you guys run with dual/triple monitor setups?
20090808 22:47:29< mordante> well once I have another new machine I want to test with various -j values for Wesnoth and the kernel to find the optimum
20090808 22:48:08< corn> mordante: make sure to make clean && reboot after every compilation or your results will be skewed due to file caching ;)
20090808 22:48:22< mordante> one monitor, 1280x1024 my next one will be 1920x1200
20090808 22:48:40< ilor> corn: I might end up with two screens but usually use one, I got a 22" dell this time
20090808 22:48:53< corn> ah
20090808 22:48:54< mordante> corn, I know but for the kernel caching doesn't help that much (to my surprise)
20090808 22:48:59< ilor> I found that having a *separate* computer beats a second monitor for stuff like watching a movie in the background :P
20090808 22:49:14< corn> I run with 1680x1050 + 1280x1024, the productivity boost is amazing really
20090808 22:49:18< mordante> I did two compilations with 8 GB in the console and both took 12 minutes
20090808 22:49:41< ilor> corn: that'd be the combo I'd be aiming at if I can grab a 19" somewhere
20090808 22:49:44< Ivanovic> i just run with 1920x1200
20090808 22:49:51< Ivanovic> and that is already really nice
20090808 22:49:57< ilor> corn: right now I got a IPS dell and I love the quality
20090808 22:50:07< corn> IPS?
20090808 22:50:10< mordante> well I like the 1920 since 1280 is just to small for 2x80 chars in vim
20090808 22:50:11-!- giusef [n=giusef@unaffiliated/giusef] has quit ["exit (-1);"]
20090808 22:50:16< ilor> corn: panel type
20090808 22:50:27< corn> ah
20090808 22:50:30< mordante> much better as a TN?
20090808 22:50:43< ilor> well I only had a real look for around an hour
20090808 22:50:44< Ivanovic> yes
20090808 22:50:57< ilor> it's back in the box until I get my new comp (and a new desk)
20090808 22:51:10< Ivanovic> i got a mva (or is it pva? i always mix those up)
20090808 22:51:20< ilor> but the difference is MAJOR. my TN laptop screen next to it was embarassed
20090808 22:51:25< Ivanovic> and those are a whole lot better regarding colors compared to tn panels
20090808 22:51:39< ilor> Ivanovic: *VA are one thing, just different trademarks iirc
20090808 22:52:06< Ivanovic> yeah, the differences are close to zero
20090808 22:52:09< ilor> I had a PVA panel a while back, but it had some annoying ghosting issues, and it finally died half a year ago
20090808 22:52:12< corn> heh, I have absolutely no preference regarding improved color range/accuracy on LCD panels
20090808 22:52:42< Ivanovic> corn: the image just looks better, too
20090808 22:52:45< mordante> I had a look at some TN panels and no real problem with them
20090808 22:52:58< ilor> color accuracy is one thing, the vieving angles are the real difference
20090808 22:53:08< corn> Ivanovic: that is true... but after you sit at a monitor for a few hours you get used to the way it displays colors
20090808 22:53:17< ilor> I got used to the fact that tilting my head slightly changes colors on my laptop screen a bit
20090808 22:53:34< corn> ilor: that's true, one of my monitors has horrible color accuracy especially wrt vertical tilt
20090808 22:53:42< ilor> but on a IPS panel it stays constant in a really large angle
20090808 22:53:52< ilor> corn: vertiacl angles is where *all* TN panels suck
20090808 22:54:55< mordante> btw Ivanovic did you ever hear of the monitor brand AOC and if so you know something about its quality
20090808 22:55:01-!- Tigge [n=tigge@bacchus.olf.sgsnet.se] has quit [Read error: 104 (Connection reset by peer)]
20090808 22:55:01< corn> I thinkI just found a bug in trunk: checking the "do not show AI moves" checkbox has no effect - I still see AI units moving
20090808 22:55:12< Ivanovic> mordante: i heard about them in the good old crt days
20090808 22:55:18< Ivanovic> they are/were not too bad
20090808 22:56:07< mordante> the shop I went to today said it's a rather common brand in Germany, but didn't find too much C't articles about them...
20090808 22:57:05< Ivanovic> in general you can/should have a look at prad.de, they do really critical tests regarding displays
20090808 22:57:39< mordante> thanks
20090808 22:57:56< Ivanovic> (though they are *really* pedantic)
20090808 22:58:06< ilor> if going for a phenom, do you guys think ddr3 is worth it?
20090808 22:58:20< Ivanovic> ilor: that is a clear and dedicated: NO!
20090808 22:58:22< corn> ilor: no
20090808 22:58:26< mordante> I went for it
20090808 22:58:47< Ivanovic> (especially looking at the current prices for ddr2, 2gb sticks from a reasonable brand for 20€...)
20090808 22:58:54< corn> however... you should go for an SSD drive if possible
20090808 22:59:05< ilor> a high-end phenom with ddr3 looks just a bit cheaper than an i7
20090808 22:59:08< corn> I have made the switch and it is easily one of the biggest performance increases I have done ever
20090808 22:59:09< Ivanovic> corn: personally i would not trust in ssd yet
20090808 22:59:12< mordante> it not that more expensive here
20090808 22:59:25< Ivanovic> 1) it is more expensive (a lot more than is)
20090808 22:59:37< Ivanovic> and 2) (the most important part): how good are the life times?
20090808 22:59:56< corn> if you believe the mfrs they say the MTBF is something in the area of 1 million to 10 million hrs
20090808 23:00:00< corn> given "regular usage"
20090808 23:00:03< mordante> corn, I also wouldn't trust SSD, fast at the start but seems to slow down after a while due to erasing the blocks
20090808 23:00:08< ilor> Ivanovic: lifeime is not a real issue imo
20090808 23:00:24< ilor> especially since the failure mode is rather graceful compared to hdd
20090808 23:00:34< Ivanovic> ilor: that is right
20090808 23:00:55< Ivanovic> though the posibility of reduced size is not nice....
20090808 23:00:55< corn> mordante: write speed is not really so significant, I think most of the responsiveness concerns on a desktop are from reads
20090808 23:01:03< ilor> problem is that to get reliable performace you need to buy one of the *really* expensive ones
20090808 23:01:08< Ivanovic> (depending on the number of write/erase cycles)
20090808 23:01:14< mordante> corn, I compile a lot ;-)
20090808 23:01:28< ilor> cheap SSDs have horrid small file writes
20090808 23:01:54< ilor> lemme check ddr3 vs ddr2 prices here...
20090808 23:02:32< Ivanovic> mordante: regarding displays i can really recommend the non tn panels from dell
20090808 23:02:43< Ivanovic> that is i got this one and am really satisfied: http://www.prad.de/new/monitore/test-dell-2405fpw.html
20090808 23:03:10< Ivanovic> a friend bought this one (got a real bargain for 1.100€) http://www.prad.de/new/monitore/test/2008/test-dell-3008wfp.html
20090808 23:03:16-!- lizard_r [n=Miranda@wesnoth/umc-dev/developer/lizard] has joined #wesnoth-dev
20090808 23:03:18< Ivanovic> both are really great displays
20090808 23:03:49< ilor> two stick of reasonably-performing ddr3 from kingston - 85EUR, two stick of similar perf ddr2 from kingston - 50eur
20090808 23:04:01< Ivanovic> ilor: jupp
20090808 23:04:46< corn> what is the purpose of the '_' before certain wml string variable values
20090808 23:04:48< ilor> Ivanovic: where's the panel type in that 2405 review? :P
20090808 23:05:15< Ivanovic> ilor: on a later page
20090808 23:05:40< Aethaeryn> any estimations on 1.7.3?
20090808 23:05:52< Ivanovic> ilor: page 7
20090808 23:06:01< Ivanovic> it is a pva panel from samsung
20090808 23:06:03< mordante> Ivanovic, as long as those expensive panels also only want to certify 13406-2 I don't want one
20090808 23:06:37< mordante> I don't want to pay >= € 1000,= and hope it works as expected
20090808 23:06:44-!- Blarumyrran [n=minaise@81-20-159-197.levira.ee] has joined #wesnoth-dev
20090808 23:06:51< mordante> Aethaeryn, IIRWIIR ;-)
20090808 23:07:00-!- Turuk [n=Turuk@wesnoth/forumsith/turuk] has quit [Read error: 104 (Connection reset by peer)]
20090808 23:07:06< Ivanovic> mordante: the 24" with non tn start at 500
20090808 23:07:15-!- Turuk [n=Turuk@wesnoth/forumsith/turuk] has joined #wesnoth-dev
20090808 23:07:17< ilor> mordante: the 2205wa I got (22", 1650 by 1050) comes with a single bright pixel warranty
20090808 23:07:18< Ivanovic> and those already tend to be quite good
20090808 23:07:21< Aethaeryn> mordante: I figured I'd get that response, but I was hoping someone set a date already.
20090808 23:07:22< mordante> corn, _ "Foo" marks it as translatable
20090808 23:07:28< Aethaeryn> I follow svn and it looks like a lot has been done already
20090808 23:07:30< Crab_> corn: "checking the "do not show AI moves" checkbox has no effect " - thanks, noted, bug submitted (assigned to myself). it's rather easy to fix, I'll do that
20090808 23:07:32< ilor> for, uh, around 300-400eur
20090808 23:07:54< mordante> Ivanovic, I've seen 24" 1920x1200 for around € 250,= here
20090808 23:07:54< ilor> *2209wa
20090808 23:08:10< ilor> mordante: those are TN panels
20090808 23:08:12< Ivanovic> yeah, but often without digital input and with really crappy panels
20090808 23:08:43< mordante> well I rather buy a crappy panel now and wait until they start to act sane on pixels
20090808 23:09:16< corn> mordante: ok
20090808 23:09:18< corn> Crab_: ok
20090808 23:09:27< mordante> ilor, true 
20090808 23:10:19< mordante>  Dell 3008WFP 30" is really huge and sweet ;-)
20090808 23:10:28< corn> Crab_: I have finished the patch for the additions to the replay log, however I am not sure if I should keep the "turn" variable or try to learn it from reading the replay back when I generate the upload log
20090808 23:10:42< corn> Crab_: just storing it in the replay is much more convenient
20090808 23:10:52< Crab_> corn: just go for the simplest possible thing that will work :)
20090808 23:10:57< Ivanovic> mordante: http://www.prad.de/new/monitore/test/2008/test-hp-lp2475w.html
20090808 23:10:59< corn> ok
20090808 23:11:07< Ivanovic> that one is available for ~540 incl shipping inside germany
20090808 23:11:15< corn> I will generate a diff for review
20090808 23:11:21< Crab_> ok
20090808 23:12:16-!- Tigge [n=tigge@bacchus.olf.sgsnet.se] has joined #wesnoth-dev
20090808 23:13:40< ilor> Ivanovic: help me decide on phenom vs i7 ;P
20090808 23:13:58< Ivanovic> ilor: uhm, good question, no idea
20090808 23:14:15< ilor> Ivanovic: that's pretty much my conclusion so far
20090808 23:14:28 * loonycyborg uses phenom because it's freaking cheaper :P
20090808 23:14:50 * mordante prefers amd cpus better price prestation 
20090808 23:15:24< corn> Crab_, mordante: if I want to store the tod in the replay in a way that does not change dependent on translation, what value should I use in time_of_day?
20090808 23:15:45< ilor> I'm inclined to spend a bit more, given how often I replace cpu/mobo usually (i.e. not very often)
20090808 23:15:49< mordante> afaik the tod has an id
20090808 23:15:57< corn> yes, it does
20090808 23:16:04< Aethaeryn> The thing about hardware is, you don't need much to run Wesnoth ;)
20090808 23:16:15< Crab_> corn: tod_manager_.get_time_of_day().name
20090808 23:16:15< mordante> best use ids for storage
20090808 23:16:35< ilor> Aethaeryn: build times, on the other hand... ;)
20090808 23:16:45< corn> Crab_: I am using name right now and the string is stored with '_' in the front
20090808 23:16:46< Aethaeryn> eh, yeah
20090808 23:16:51< Aethaeryn> my compile time could be improved
20090808 23:17:06< mordante> ilor, I also don't change too often, but I like the price prestation and wattage of the 905e
20090808 23:17:25< mordante> I also believe some concurrence for intel doesn't hurt ;-)
20090808 23:17:30< Ivanovic> mordante: and a shop in austria seels the dell 2408wfp ( http://www.prad.de/new/monitore/test/2009/test-dell-2408wfp.html ) for 500€ +25€ incluing 5 year warranty and zero dead pixel gurantee
20090808 23:17:35< Ivanovic> http://cgi.ebay.de/Dell-Ultrasharp-2408WFP-24-Zoll-LCD-Flachbildschirm-A01_W0QQitemZ190303512466QQcmdZViewItemQQptZDE_Elektronik_Computer_Computer_Monitore_Flachbildschirme?hash=item2c4ef92792&_trksid=p3286.c0.m14#shId
20090808 23:17:50< corn> in general, are t_string translatable strings?
20090808 23:18:14< mordante> corn, yes
20090808 23:18:36< corn> ok
20090808 23:18:46< corn> then I should probably not rely on unit.type_name() for storage in replays?
20090808 23:19:11< mordante> in general don't use names as unique ids
20090808 23:20:15< corn> ok, I will use type_id() instead
20090808 23:22:47< Ivanovic> time to watch a movie and then head off to bed, n8
20090808 23:23:48< mordante> Ivanovic, where do you see the 100% guarantee? I only see 100% no bight pixel guarantee
20090808 23:25:26< corn> Crab_: recompiling one last time before I generate a patch... unfortunately recompiling takes 30 mins on this laptp
20090808 23:25:54< Crab_> corn: yes, slow compilation is a real time-killer..
20090808 23:25:55< corn> actually no, I didn't change any header files so it should be pretty quick :)
20090808 23:26:36< Crab_> gives incentive to reduce header dependencies, through :)
20090808 23:26:41< corn> yeah
20090808 23:28:57< mordante> Crab_, which needs a fast CPU to test ;-)
20090808 23:30:11< CIA-62> jetryl * r37557 /branches/1.6/data/core/images/units/human-loyalists/ (95 files): Shadow updates on the loyalist leaders.
20090808 23:31:16< CIA-62> jetryl * r37558 /trunk/data/core/images/units/human-loyalists/ (95 files): Shadow updates on the loyalist leaders.
20090808 23:31:37< Crab_> mordante: yes
20090808 23:32:01-!- Elvish_Pillager [n=eli@71-10-231-36.dhcp.oxfr.ma.charter.com] has quit [Read error: 110 (Connection timed out)]
20090808 23:32:07-!- Elvish_Pillage2 [n=eli@71-10-231-36.dhcp.oxfr.ma.charter.com] has joined #wesnoth-dev
20090808 23:34:49< CIA-62> mordante * r37559 /trunk/src/gui/widgets/ (event_handler.cpp event_handler.hpp widget.cpp): 
20090808 23:34:49< CIA-62> Attempt to fix the lobby dynamic_cast crashes.
20090808 23:34:49< CIA-62> Since I haven't been able to reproduce the problem I haven't been able
20090808 23:34:49< CIA-62> to test whether it really fixes the problem. Soliton/ilor please test.
20090808 23:41:12< corn> 15 minutes of linking so far with heavy swapping...
20090808 23:42:55-!- fendrin [n=fabi@g226194090.adsl.alicedsl.de] has joined #wesnoth-dev
20090808 23:43:08< mordante> autsch
20090808 23:43:17< mordante> corn, which linker do you use
20090808 23:43:19< mordante> ?
20090808 23:43:32< corn> ld
20090808 23:44:04< corn> I wasn't aware that there were other choices 
20090808 23:44:08< corn> for linux, that is
20090808 23:44:42< mordante> there's a new gold linker in binutils, which isn't as feature rich as ld, but much faster
20090808 23:45:21< mordante> I use gold here as well, except for the linux kernel
20090808 23:45:23< corn> 'gold'?
20090808 23:45:37< mordante> I didn't invent the name ;-)
20090808 23:46:39< mordante> which distro do you use?
20090808 23:46:44< corn> ubuntu 9.04
20090808 23:46:54< corn> I am looking up if there is a pkg of a more recent/experimental binutils
20090808 23:46:59< corn> that includes this
20090808 23:47:37< mordante> the package is named binutils-gold in debian
20090808 23:49:04< corn> ah, it is in ubuntu 9.10
20090808 23:49:10< corn> I will check for backports later
20090808 23:49:18< corn> for now I will avoid recompilation as much as possible :)
20090808 23:49:47< Sirp> corn: so hey how's your project going?
20090808 23:50:24< mordante> corn, I understand, but 'gold' shaved off several minutes of linking for me
20090808 23:50:33< corn> Sirp: currently finishing the stuff with killmaps, I spent the past week or so adding extra logging stuff to the wesnoth client
20090808 23:51:07< corn> Sirp: if I can get this done over the weekend, all of next week will be for adding new graphs and visualizatons + improving the look of the site
20090808 23:51:13< corn> then after that I finally put it on wesnoth.org
20090808 23:51:38-!- BenUrban [n=benurban@c-68-50-54-86.hsd1.md.comcast.net] has joined #wesnoth-dev
20090808 23:57:27< corn> Crab_: https://gna.org/patch/?1237
20090808 23:57:44< Crab_> I'll take a look...
20090808 23:57:52-!- Turuk_ [n=Turuk@wesnoth/forumsith/turuk] has joined #wesnoth-dev
20090808 23:59:07< Crab_> corn: '+	const unit *d_ = get_unit(get_info(),attacker_loc_);+	const unit *a_ = get_unit(get_info(),defender_loc_);'  shouldn't them be swapped ?
20090808 23:59:24< corn> oh, yes
20090808 23:59:46< corn> oops
--- Log closed Sun Aug 09 00:00:07 2009