--- Log opened Thu Aug 27 00:00:51 2009
20090827 00:01:19-!- Blueblaze [n=nick@c-98-197-50-41.hsd1.tx.comcast.net] has quit [Remote closed the connection]
20090827 00:02:53-!- Sirp [n=user@wesnoth/developer/dave] has joined #wesnoth-dev
20090827 00:07:33< YogiHH> night everyone
20090827 00:07:57-!- YogiHH [n=chatzill@c129221.adsl.hansenet.de] has left #wesnoth-dev []
20090827 00:16:23-!- loonycyborg [n=sergey@wesnoth/developer/loonycyborg] has quit ["Zzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzz"]
20090827 00:16:52-!- loonybot [n=loonybot@wesnoth/bot/loonybot] has quit [Remote closed the connection]
20090827 00:18:47-!- happygrue_ [n=George@c-67-176-145-41.hsd1.in.comcast.net] has joined #wesnoth-dev
20090827 00:20:01-!- happygrue [n=George@wesnoth/developer/wintermute] has quit [Read error: 104 (Connection reset by peer)]
20090827 00:20:21-!- boucman [n=rosen@wesnoth/developer/boucman] has quit ["Leaving."]
20090827 00:22:32< Crab_> alink: well, it has some corner cases (i.e. "getting the list of enemies near path to target" and using that in calculations), but, in general, reducing the number of A* calls seems possible using your method.
20090827 00:22:34-!- shikadibot [n=shikadi@wesnoth/umc-dev/bot/shikadibot] has quit ["leaving"]
20090827 00:22:54-!- shikadibot [n=shikadi@wesnoth/umc-dev/bot/shikadibot] has joined #wesnoth-dev
20090827 00:23:09< alink> Crab_: good :)
20090827 00:23:42-!- happygrue_ [n=George@c-67-176-145-41.hsd1.in.comcast.net] has quit [Read error: 104 (Connection reset by peer)]
20090827 00:24:08< Crab_> I'll try to make a copy of ai_default targeting, by removing some stuff from it, adding your optimization, and making it 'simpler, faster but still good enough'
20090827 00:24:25< alink> Crab_: I initialy tried it with 1.6, but I relied on AI bugs, which was ugly
20090827 00:24:40-!- shadowmaster [n=ignacio@wesnoth/developer/shadowmaster] has quit ["killing all bash sessions for memory..."]
20090827 00:24:49< Crab_> alink: I'll write some parts from scratch
20090827 00:24:55-!- shadowmaster [n=ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20090827 00:25:02< Crab_> alink: to avoid relying on existing ai bugs too much..
20090827 00:25:10< alink> (like the fact that some AI 1.6 test was always false and things like that)
20090827 00:25:44< alink> Crab_: yeah that's why I abandoned the idea, knowing that my code will not work on 1.7
20090827 00:26:02< Crab_> ok. thanks for the idea.
20090827 00:26:03< Crab_> night
20090827 00:26:06< alink> or even stop working when someone will patch the 1.6 AI behavior
20090827 00:26:12< alink> Crab_: gn
20090827 00:26:23-!- Crab_ [n=Crab_@wesnoth/developer/crab] has quit ["Leaving."]
20090827 00:26:45-!- stikonas [n=and@wesnoth/translator/stikonas] has quit [Read error: 110 (Connection timed out)]
20090827 00:28:04< alink> btw I think it's what is know as a "greedy" algorythm 
20090827 00:29:11-!- ancestral [n=ancestra@97.116.111.138] has quit []
20090827 00:31:40< alink> mmh maybe not exactly
20090827 00:35:18-!- Baufo [n=thomas@wesnoth/developer/baufo] has quit ["Leaving"]
20090827 00:39:29< alink> in fact it's a bit like A* but work on targets instead of nodes: Use some quick-heuristic function to decice what is the best next target/step, check its real cost and if good, possibly ignore the others (only possible if the heuristic is 'admissible'=not underestimating cost )
20090827 00:42:00< alink> but one possible problem with the AI version is that since we use value, any cost function having some sort of "Oh a treasure is on the path, value is super high", will force use to use an heuristic always expecting this, and thus often overestimating
20090827 00:43:58< alink> mmh all the paths will be overestimated, so the best one is still findable. It's more a problem of weight between this potential treasure and other smaller values
20090827 00:49:43< alink> A last refinement, in a maze map (or map with various terrain), simply using the distance for evaluating the move cost of the path is a poor information
20090827 00:50:12< alink> a better choice is to use A* on the real terrains but ignoring units and ZoC
20090827 00:50:55< alink> in the past it was 2 times faster and possibly more when unit_map is heavily loaded
20090827 00:51:27< alink> but more important, it make your previous idea about caching previous A* easier
20090827 00:52:14< alink> an A* ignoring units must not be updated after each unit_map change (=unit move)
20090827 00:55:16< alink> OTOH we wil do more work (+ 1 cheap A*), so we need to check if it really allow to avoid enough costy A*
20090827 01:01:16-!- Espreon [n=espreon@wesnoth/developer/espreon] has quit ["WRYYYYYYYYYYYYYYYYYYYY!"]
20090827 01:13:51-!- ilor [n=user@wesnoth/developer/ilor] has quit []
20090827 01:17:00< alink> Another last idea: if we already have a good best target and we know that any other following targets must have a MP cost lower than some value to be able to beat our champion, then we can pass this max_cost to the next A*, thus aborting early some possibly long ones
20090827 01:18:34< Zarel> ...erm... pathing on a tile-based game with fairly small maps?
20090827 01:18:42< alink> that could potentially shorter all the A* if a good target is near our unit and we start with it
20090827 01:18:48< Zarel> Isn't this the kind of thing where you can just drop in A* and have it work? o_O
20090827 01:19:34< Zarel> Heck, you could probably just use breadth-first search and have it be plenty efficient.
20090827 01:19:39< alink> Zarel: not really, there is some things costy
20090827 01:20:07< alink> The AI currenlty do a *lot* of them when there is many units
20090827 01:20:41< alink> plus don't forget that it's not simple move, there is WML running to compute them
20090827 01:20:45-!- thespaceinvader [n=chatzill@wesnoth/artist/thespaceinvader] has quit ["night all"]
20090827 01:21:03< alink> a typical example is ZoC and "skirmisher" ability
20090827 01:21:14< Zarel> alink: Erm, shouldn't the AI only have to do it once for each unit? :/
20090827 01:21:43< Zarel> Like, you only need to recompute if one of the enemy units dies or something.
20090827 01:22:02< alink> no any change can trigger a WML change
20090827 01:22:12< Zarel> ?
20090827 01:22:22< alink> WML events, ability affecting adjacent units
20090827 01:23:03< Zarel> Well, other than the "distract" ability, you'd need to kill an enemy unit before ZoC changes...
20090827 01:23:04-!- BenUrban [n=benurban@unaffiliated/benurban] has joined #wesnoth-dev
20090827 01:23:40< alink> we also need to support any crazy UMC WML
20090827 01:24:21< alink> but there is also the problem that some unit have different ability at different location, which is important for a move
20090827 01:24:41< alink> example: you  can have skirmisher in forest at night only 
20090827 01:25:26< Zarel> Well, yes, but you still only have to recompute that at a maximum of once per turn.
20090827 01:25:42< Zarel> And only in situations where you're actually dealing with a unit with such an ability.
20090827 01:25:57< alink> and recompute it after each AI move
20090827 01:26:03< Zarel> Why?
20090827 01:26:40< alink> the Ai must check it the target is still alive before lauching its second unit to attack it
20090827 01:27:46< alink> depending of this, maybe rethink its strategy, choose other target, do new A* and compare them etc...
20090827 01:28:13< alink> i am not expert in wesnoth AI, but at least a good AI should do this
20090827 01:28:23< Zarel> Well, the target is nearly always still alive.
20090827 01:28:30< Zarel> And if it is, then it doesn't need to recompute.
20090827 01:29:01< alink> ok, then check if wounded/easy to kill/poisoned etc...
20090827 01:30:43< alink> in this case slowed is a better example
20090827 01:32:24< alink> anyway, caching is of course possible, but all cache has a cost(in CPU/memory and code simplicity), you must be sure that you will not end flush it most of the time
20090827 01:32:28< Zarel> Yes, but that doesn't require any pathing...
20090827 01:32:40< Zarel> (the checking doesn't)
20090827 01:33:38< alink> indeed, if all is similar enough, we can possibly reuse the same paths
20090827 01:34:25< alink> the trick is that you know to check every little change to see if it affect any of your decisions or path
20090827 01:34:54< alink> know/now need
20090827 01:47:44-!- happygrue [n=George@wesnoth/developer/wintermute] has joined #wesnoth-dev
20090827 01:49:42< Zarel> Well, yeah.
20090827 01:49:58< Zarel> Or you can just have a list of whitelisted changes (the majority of them), and just repath the others.
20090827 01:51:20-!- crimson_p [n=irchon@64.201.60.211] has joined #wesnoth-dev
20090827 01:51:47-!- crimson_p [n=irchon@64.201.60.211] has quit [Remote closed the connection]
20090827 01:53:10< alink> I suspect that if there was so much whitelisted changes (not really changing anything or not random), the AI would be indeed simpler to code, but wesnoth would be more like chess
20090827 02:02:30-!- melinath [n=melinath@pool-72-72-200-231.altnpa.east.verizon.net] has joined #wesnoth-dev
20090827 02:03:35< Zarel> ...?
20090827 02:03:56< Zarel> Erm, which changes aren't whitelisted?
20090827 02:04:14< Zarel> All mainline attacks that don't kill are whitelisted.
20090827 02:04:26< Zarel> All movement of mainline units is whitelisted.
20090827 02:06:40< alink> I mean in chess I can think : I do A, then B, then C etc.. ok seems good, let's go.
20090827 02:07:01< alink> In wesnoth, I do A, if all my hits misses I do B, if at least one hit, he is wounded, then I do C, if all hits, he is almost dead, then I do D to level my other unit etc..
20090827 02:07:57< alink> or damage to my units so i need to cover/heal him, etc...
20090827 02:09:40< alink> or your whitelist change include things like "don't so more that some damage etc"? but then that start to be complex to decide dynamicaly
20090827 02:09:50< alink> so more/do more
20090827 02:10:13< Zarel> Chess has a lot of things that makes it more difficult than Wesnoth.
20090827 02:10:30< alink> I disagree
20090827 02:10:33< Zarel> For one thing, movement is a lot more complex.
20090827 02:10:40< alink> and even don't see your point
20090827 02:10:59< Zarel> Erm, I think my point was that your chess analogy is flawed.
20090827 02:11:09< alink> IMHO terrain defense and MP cost specific at each unit, make wesnoth move more complex
20090827 02:11:17< Zarel> Since chess is a discrete combinatorial game, and Wesnoth is far more continuous.
20090827 02:11:27< alink> + ZoC, fog etc...
20090827 02:11:41< alink> ambush, skirmish, ...
20090827 02:11:49< alink> and that is only for move
20090827 02:11:53-!- Chusslove [n=Chusslov@212.124.173.172] has quit [Read error: 131 (Connection reset by peer)]
20090827 02:12:35< Soliton> movement in chess is more complex?
20090827 02:13:00< alink> Zarel: My chess analogy was mean to show that in wesnoth the situation change a lot between each move, not in chess
20090827 02:13:08< alink> + the randomness of these changes
20090827 02:13:27< alink> move or turn is irrelevant here
20090827 02:13:52< Zarel> Soliton: In Wesnoth, there's a heuristic that if you can move somewhere far away, you can generally move somewhere closer, something that chess doesn't have.
20090827 02:14:30< alink> Zarel: teleport ability ;)
20090827 02:14:50< Soliton> i don't even see how such a heuristic is relevant.
20090827 02:15:14< Soliton> sure movement in chess and wesnoth is different so you can find things that.. differ.
20090827 02:16:49< Zarel> alink: Ah, yes, teleport messes it up. :(
20090827 02:17:04< alink> + most of the time each wesnoth units has more possible moves than a chess piece (which is what count for AI power)
20090827 02:17:19-!- Blueblaze [n=nick@c-98-197-50-41.hsd1.tx.comcast.net] has joined #wesnoth-dev
20090827 02:17:58< Zarel> Well, I dunno. I just disliked that Wesnoth wiki article that said "writing an AI is hard, since Wesnoth is more complicated than chess"
20090827 02:18:34< Zarel> The end of the article mentions "Wesnoth is more continuous, so it's not entirely a fair comparison", but that point kind of defeats the purpose of the entire article.
20090827 02:19:08< Zarel> Anyway, this digression seems a bit silly. Let's go back to pathfinding.
20090827 02:19:28< alink> well it still is more complicated (for human and AI, but in different manners)
20090827 02:20:19< alink> in fact, in some ways wesnoth is maybe simpler for humans than for AI
20090827 02:20:40< Zarel> Yeah, but it's, to use a cliche, like comparing apples and oranges. Chess AI writing can't really get much better than combinatorial-game-tree traversal and some very rough heuristics, while heuristics are easy to come by in a game like Wesnoth, and so the AI can be a lot more humanlike.
20090827 02:20:50< alink> in wensoth, instinct helps more for evaluate power and probabibility
20090827 02:21:56< Soliton> now you're saying being human like is the goal.
20090827 02:22:10< Zarel> No, but it makes for easier AI writing. :P
20090827 02:22:19< alink> "heuristics are easy to come by in a game like Wesnoth" if you ignore all the little detail needed by a good wesnoth player
20090827 02:22:21< Soliton> i don't see that.
20090827 02:22:23< Zarel> Well, at the very least, entirely different AI writing.
20090827 02:22:33< Zarel> alink: That's why they're called 'heuristics' and not 'exact measures'
20090827 02:23:02< Soliton> a heuristic that totally fails because of a little detail is useless i'd think.
20090827 02:23:34< Zarel> What sort of little detail are you talking about?
20090827 02:24:04< Soliton> all those you want to ignore.
20090827 02:24:18< alink> my old example of little detail, did i just (or can i) heavily wounded this unit in my last move
20090827 02:25:11< Zarel> Erm... what does that have to do with heuristics?
20090827 02:26:04< alink> well ok , only the "(or can i)" part
20090827 02:26:29< Soliton> the heuristic needs to take all details into account albeit obviously not exactly but in an approximative way.
20090827 02:26:30< alink> the rest was to make the link with the previous points to avoid repetition 
20090827 02:26:49< Zarel> Soliton: Yes, it does...
20090827 02:27:14< Soliton> right, and what alink mentioned is that they're not easy to come by then.
20090827 02:28:09< Zarel> Why not?
20090827 02:29:04< Soliton> because wesnoth is a lot more complex than chess.
20090827 02:29:55< Zarel> ...
20090827 02:30:25< alink> another example, multi-exploiting of leadership ability: need to decompose an unit move, order with the others etc.. relatively simple for human, very hard for AI
20090827 02:31:32< Zarel> ...it's more complex, but heuristics are easier. That's my entire point...
20090827 02:31:36< Zarel> You're speaking in circles.
20090827 02:31:58< Zarel> And this is hardly useful... I still want to get back to pathing, which might actually be relevant to improving Wesnoth.
20090827 02:32:15< Soliton> if you're saying that heuristics are easier than figuring situations out exactly then yes. duh.
20090827 02:33:04< xchat_sux> no, Zarel is saying that heuristics are easier in Wesnoth than in Chess
20090827 02:33:05< alink> maybe, depend your definition of heuristic. My point here was that power of move/attack are not independant, which make them harder to evaluate
20090827 02:33:18< Zarel> See, in Wesnoth, a human can see "oh, this unit is low on HP, and it's near a bunch of high level enemy units. It will probably die." A computer can see that fairly easily, as well.
20090827 02:33:32< Zarel> On the other hand, there's no corresponding heuristic in chess.
20090827 02:33:42< alink> harder to evaluate, means harder to have a useful heuristic
20090827 02:33:45-!- xchat_sux is now known as Elvish_Pillager
20090827 02:34:02< Soliton> "oh this unit is near a piece that can capture it it will probably be captured."
20090827 02:34:12< Zarel> A human can visualize a bunch of interrelations between chess pieces, but a computer has no choice but to traverse the decision tree to determine a chess piece's survivability.
20090827 02:34:22< Elvish_Pillager> Soliton: that's a great example of what Zarel is arguing for :)
20090827 02:34:37< Soliton> i don't see how.
20090827 02:34:44< Zarel> Soliton: Erm, pawn in front of a pawn. They're near each other, they can capture, and yet they're perfectly safe.
20090827 02:34:48< Soliton> since that was obviously an example from chess.
20090827 02:34:54< Elvish_Pillager> Soliton: it's an example of a Chess heuristic that doesn't make any sense
20090827 02:35:06< Elvish_Pillager> thus supporting Zarel's stance that Chess is harder to make heuristics for.
20090827 02:35:28< Soliton> it doesn't make much sense because you don't need a heuristic for it, maybe.
20090827 02:36:15< Soliton> Zarel: not sure if it is necessary to point out that those pawns can't capture each other.
20090827 02:36:18< Zarel> Erm, piece survivability is pretty important, in chess. Especially since you don't get recruits.
20090827 02:36:35< Zarel> Soliton: But they're near each other. Read what you wrote yourself. ;)
20090827 02:36:50< Zarel> Did you mean "Oh this unit is in the capture range of a unit that can capture it"?
20090827 02:36:57< Elvish_Pillager> My 1-HP Bishop is right next to their Bishop!
20090827 02:37:09< Soliton> "this unit is near a piece that can capture it"
20090827 02:37:13< Zarel> If that's your only chess heuristic for survivability, you'd be completely wrong.
20090827 02:37:22< Soliton> i guess i'm not writing clearly.
20090827 02:37:30< Soliton> oh boy.
20090827 02:37:34< Soliton> nevermind.
20090827 02:38:02< Zarel> I mean, sure, it helps to know not to move into the threat range of opposing units.
20090827 02:38:03< Soliton> i didn't know that you meant silly discussion so literally.
20090827 02:38:10< Elvish_Pillager> Zarel: do you happen to be a programmer? Wesnoth could use some AI work :)
20090827 02:38:29< Zarel> Elvish_Pillager: I was planning on it, but AI work takes time, and I don't have enough of it. :(
20090827 02:38:36< Elvish_Pillager> Zarel: aww :(
20090827 02:38:40< Zarel> I do plan on working on the AI in the future when I do, though.
20090827 02:38:41< Soliton> but it's easy!
20090827 02:38:43< Elvish_Pillager> I've worked on wesnoth AI some
20090827 02:38:48< Elvish_Pillager> yeah, it's easy
20090827 02:38:53< Elvish_Pillager> ^_^
20090827 02:39:04< alink> maybe replace each HP of units by a piece and you have some sort of piece survavibility making it easier to compare and better showing that in wesnoth HP matters a lot
20090827 02:39:28< alink> making little details more important
20090827 02:39:33-!- Espreon [n=espreon@wesnoth/developer/espreon] has joined #wesnoth-dev
20090827 02:41:03< Elvish_Pillager> Zarel: if/when you want to work on the AI in the future, could you contact me? there's always going to be a chance that I'll be interested in that kind of project, so we might be able to work together.
20090827 02:41:39< alink> details like ability etc which make the succesion of moves more complex and more important, and not simply a locality problem like you seems to imply
20090827 02:42:14< alink> (ability linking pieces and actions together, not just special attack)
20090827 02:43:02< Zarel> Elvish_Pillager: To work together on AI in Wesnoth? Sure.
20090827 02:43:49< Zarel> alink: Any Wesnoth AI should probably be focused mostly around mainline abilities.
20090827 02:44:20< Zarel> Oh, nevermind; I misinterpreted.
20090827 02:44:24< alink> Zarel: probably but even just them are already hard for AI
20090827 02:44:36< Soliton> so we have to tweak it whenever we add a new ability.
20090827 02:44:36< Zarel> Yeah. Leadership, especially.
20090827 02:44:44< alink> healing too
20090827 02:44:54< alink> backstab
20090827 02:45:10< alink> illuminate
20090827 02:45:17< alink> you get the pictures
20090827 02:45:24< Zarel> Well, when I played 1.6.4, it seemed the AI just ignored those altogether
20090827 02:45:25< Elvish_Pillager> A really nicely written AI could probably handle new abilities with minimal additional coding.
20090827 02:45:54< Elvish_Pillager> yeah, an AI that acts like it actually knows its units have abilities would be quite nice.
20090827 02:45:58< alink> Zarel: it doesn't fully exploit them
20090827 02:46:06< Zarel> At least for healing, you'd think units could just treat areas next to +8 healers as villages with no evasion bonus
20090827 02:46:08-!- noy [n=Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20090827 02:46:27< wajimba1> Not to interrupt, but is there a way to get windows to download and install the latest dev version (short of compiling) on a schedule, or better yet, only when a new version is released?
20090827 02:46:54-!- noy [n=Noy@wesnoth/developer/noy] has quit [Client Quit]
20090827 02:46:57< Soliton> write a package manager for windows.
20090827 02:46:58< Zarel> Elvish_Pillager: I can think of arbitrary +healing abilities, invisibility abilities, weapon specials that modify damage or accuracy in certain cases...
20090827 02:47:10< Zarel> But more complex abilities would be pretty hard for any AI...
20090827 02:47:22< wajimba1> soliton:Xp
20090827 02:47:24< Elvish_Pillager> Zarel: what complex abilities do we have?
20090827 02:47:26< alink> Zarel: it doesn't place its unit to exploit i, but i think it handle them in its attack simulation, so sometimes a move is selected because it is good becasue of an ability from an adjacent unit
20090827 02:48:01< Elvish_Pillager> I guess Leadership is terribly complex
20090827 02:48:06< Soliton> wajimba1: doesn't matter what windows version.
20090827 02:48:11< Zarel> Elvish_Pillager: Well, you said "handle new abilities", so we're presumably referring to abilities that don't currently exist. ;)
20090827 02:48:41< Elvish_Pillager> Zarel: I was thinking of a lot of proposed abilities. Most of them are fairly simple.
20090827 02:48:46< Zarel> Leadership/illu can probably be treated similarly.
20090827 02:48:51< Elvish_Pillager> Or UMC abilities.
20090827 02:49:20< Zarel> All "Improve units' attacks near you" abilities can probably be handled similarly.
20090827 02:49:23< Elvish_Pillager> Well, Illuminates can also be used offensively
20090827 02:49:40< Soliton> wajimba1: alternatively you can maybe script something with some rss feed of sf.net. at least i guess they have such for download areas in projects.
20090827 02:50:07< wajimba1> soliton: hmmm
20090827 02:50:36< Zarel> Well, all "improve/hamper friendly/hostile units' attacks near you" abilities can be handled similarly.
20090827 02:51:00< Zarel> Well, no, split those into two groups.
20090827 02:51:09< Zarel> "improve/hamper friendlies" and "improve/hamper hostiles"
20090827 02:51:38< Zarel> "improve/hamper hostiles" would be a bit easier, since it can only be done at most twice, and at least some of the code could be recycled.
20090827 02:51:53< Zarel> Unless you have both that, and skirmisher.
20090827 02:51:56< Elvish_Pillager> Twice?
20090827 02:52:08< Zarel> If you start next to a hostile.
20090827 02:52:10< Elvish_Pillager> Beginning and end of your move, do you mean?
20090827 02:52:12< Zarel> Yes.
20090827 02:52:49< Zarel> Although I think Elysia or whatever-her-name-is elf-lady-of-light from IftU has both Skirmisher and Illuminates.
20090827 02:52:54< Elvish_Pillager> Elynia
20090827 02:52:58< Elvish_Pillager> and yes.
20090827 02:53:11< Zarel> Elynia also has Illuminates and an invisibility ability.
20090827 02:53:20< Zarel> I don't even know how Illuminates + invisibility would work.
20090827 02:53:29< Elvish_Pillager> shadowmaster denies loving Skirmisher, but I don't believe it :
20090827 02:53:30< Elvish_Pillager> :p
20090827 02:55:03< Zarel> Well, I don't think it'd be very easy to make an AI that uses a leader as best as possible.
20090827 02:55:13< Elvish_Pillager> Heh.
20090827 02:55:17< Zarel> But I think we should start with having the leader at least help a bit.
20090827 02:55:38-!- Blarumyrran [n=minaise@81-20-159-197.levira.ee] has quit []
20090827 02:55:57< CIA-62> fendrin * r38127 /trunk/data/campaigns/Legend_of_Wesmere/scenarios/18_Hour_of_Glory.cfg: 
20090827 02:55:57< CIA-62> LoW 18: Enabled the prestart save for scenario 19.
20090827 02:55:57< CIA-62> Forced a background image not to scale as requested by kitty.
20090827 02:56:10< Zarel> As opposed to current behavior, which to me seems more like "leader only helps if it gets lucky and a leader and low-level unit happen to move next to each other"
20090827 02:56:22-!- Blueblaze [n=nick@c-98-197-50-41.hsd1.tx.comcast.net] has quit [Remote closed the connection]
20090827 02:56:34< Elvish_Pillager> yeah...
20090827 02:57:00< Elvish_Pillager> Although I'd start out writing an AI that ignored abilities, except in intentionally being written in a modular way that would allow them to be added later.
20090827 02:57:01< CIA-62> fendrin * r38128 /trunk/data/campaigns/Legend_of_Wesmere/scenarios/19_Costly_Revenge.cfg: LoW 19: Code cleanups.
20090827 02:58:16< Zarel> Total AI restructuring? :/
20090827 02:58:30< Elvish_Pillager> uh, yeah.
20090827 02:58:44< Elvish_Pillager> I can't imagine trying to modify the current AI to be significantly better.
20090827 02:58:53< Elvish_Pillager> Do you have any idea how many horrible hacks are used in the current AI?!
20090827 02:59:36-!- Blueblaze [n=nick@c-98-197-50-41.hsd1.tx.comcast.net] has joined #wesnoth-dev
20090827 02:59:37< Zarel> No.
20090827 02:59:43< Zarel> I'm guessing that's probably a good thing? ;)
20090827 02:59:45< CIA-62> fendrin * r38129 /trunk/data/campaigns/Legend_of_Wesmere/scenarios/23_End_of_War.cfg: LoW 23: Fixed the setup of Landar's army.
20090827 03:00:04< Elvish_Pillager> (you're guessing that what is probably a good thing?)
20090827 03:00:35< Zarel> That I haven't experienced the horrors of the current AI.
20090827 03:00:43< Elvish_Pillager> ah, probably
20090827 03:00:50< Zarel> I mean, abilities and weapon specials that only modify the units own damage, accuracy, evasion, or resistance should be pretty simple to modularize in an AI.
20090827 03:01:10< Elvish_Pillager> If you're writing a modular AI, yeah.
20090827 03:01:32< Zarel> But I think it'd be better to hardcode an arbitrary "affect units near it" ability.
20090827 03:22:17-!- wajimba1 [n=Andrew_A@24-158-30-63.dhcp.dlth.mn.charter.com] has left #wesnoth-dev []
20090827 03:33:50-!- Espreon [n=espreon@wesnoth/developer/espreon] has quit ["WRYYYYYYYYYYYYYYYYYYYY!"]
20090827 03:39:06-!- Elvish_Pillager [n=eli@71-10-231-36.dhcp.oxfr.ma.charter.com] has quit [Read error: 110 (Connection timed out)]
20090827 03:42:21-!- Appleman1234 [n=Appleman@131.181.102.120] has joined #wesnoth-dev
20090827 03:54:58-!- Espreon [n=espreon@wesnoth/developer/espreon] has joined #wesnoth-dev
20090827 03:56:55-!- Appleman1234 [n=Appleman@131.181.102.120] has quit ["Leaving"]
20090827 03:57:56< alink> <Soliton> right, so how is it handled now when you move over a village?
20090827 03:58:02< alink> Seeing the code, it seems that we also avoid unwanted capture when spotting en enemy during a move passing by a not-owned village
20090827 03:58:22< alink> however i expect this case to be very rare
20090827 03:58:52< alink> (compared to the  friendly unit one, i mean)
20090827 04:00:25< Soliton> sure. it's still the same issue that you can't really solve.
20090827 04:00:56< Soliton> so there is not much need to try if it's at the expense of other issues.
20090827 04:02:12< alink> indeed, but note that we can just reverse the feature (since it's there now, instead of random) and have it to prefer go by friendly unit
20090827 04:02:57< alink> just need to check if it doesn't cause other problems
20090827 04:03:42< alink> + I need to check some numerical details (there is much more empty hexes, cost will add up more)
20090827 04:03:59< alink> but that's a detail
20090827 04:04:49< alink> And fix these damn blocked goto, we had a lot of complain about them too, i was happy to have reduced their frequency
20090827 04:05:04< Soliton> well, you already noted that it will ... exavtly.
20090827 04:06:01< Soliton> i wouldn't say there are a lot of complaints.. it's really not an integral feature most of the time but it'd be nice to have it work better certainly.
20090827 04:06:45< Soliton> you will have to check gotos every move to really solve it though, i'd think.
20090827 04:07:06< Soliton> also btw, gotos are not undoable atm.
20090827 04:07:33< alink> ideally we need to handle end of turn differently that normal move
20090827 04:07:52< alink> no idea baout the undo thing, i will check it quicly
20090827 04:08:20< alink> btw what is the status of the "pass-by" WML event? because it will need to deal with the shared hex case too
20090827 04:08:43< Soliton> nobody working on it, i think.
20090827 04:08:50-!- Ivanovic_ [n=ivanovic@77.178.197.81] has joined #wesnoth-dev
20090827 04:08:58< alink> lol, ok same status then :)
20090827 04:10:13< alink> Soliton: about your drawing path idea, i was thinking to maybe use the drag&drop for that
20090827 04:11:10< alink> and I think that other games use a sequence of shift clicks for this kind of thing
20090827 04:11:49< Soliton> in rts yeah, i think so.
20090827 04:12:21< alink> could also be a solution for indicating "long" goto
20090827 04:12:28< Soliton> using drag&drop sounds good to me.
20090827 04:13:10< Soliton> well, shift is a bit problematic for us but i guess for moving it's ok to use.
20090827 04:13:41< alink> well any key (without mouse click) works too
20090827 04:14:08< alink> a drop "pass here" kotkey
20090827 04:15:32< alink> wich is somehowo better than the drawing version, because you can allow the pathfinding to optimize your route between these points
20090827 04:16:24< alink> and maybe keep the key pressed for having the drawing path behavior
20090827 04:17:16< Soliton> yes, that seems friendlier.
20090827 04:18:39< alink> with some UI things (waypoints, special cursor), I think that can be intuitive enough
20090827 04:19:14< alink> of course, there is always the little details, like allowing or not loops etc
20090827 04:21:25< alink> seems a nice little project, not sure that i restart wesnoth-dev right now, but I like this pathfinding interface stuff. I may begin with that (later)
20090827 04:21:26< Soliton> it's the same as moving now. if the player wants to move back and forth for some reason he can do so. :-)
20090827 04:21:57< alink> does RC is planned soon ?
20090827 04:22:18< alink> no wait we are 1.7.small-number right
20090827 04:22:23< Soliton> aftr gsoc is over a new stable is planned soon, i think.
20090827 04:22:39< Soliton> and gsoc is pretty much over now afaik. :-P
20090827 04:22:55< alink> ok so no 1.x.16 like before
20090827 04:23:04< Soliton> there are still a lot of issues with the new gui for example though.
20090827 04:23:20< Soliton> so who knows how quickly things will really work out.
20090827 04:23:36< Soliton> what would be nice is to fix the issue with avoiding ambushes though.
20090827 04:23:53< Soliton> even if that is at the expense of gotos.
20090827 04:24:09< Soliton> you can always fix gotos by hand if you wish.
20090827 04:24:22-!- Ivanovic [n=ivanovic@wesnoth/developer/ivanovic] has quit [Read error: 110 (Connection timed out)]
20090827 04:24:23< alink> that part should be easy. Plus it's not very fragile/important
20090827 04:24:48-!- Ivanovic_ is now known as Ivanovic
20090827 04:25:15< Soliton> well, and if you do some experiments with drag&drop it's not so important if that breaks since you have other ways to move, no?
20090827 04:26:22< alink> mmh experimenting with drag&drop is much more bug-risky just before RC
20090827 04:26:56< alink> I remember that 1.6.RC had a lot of false-click bugs because of that kind of code
20090827 04:28:01< Soliton> if it messes up non-drag&drop moves then yes.
20090827 04:28:14< alink> always passing by friendly unit is done by changing one line, and check if not causing other UI problem of this type
20090827 04:30:51< alink> OTOH one source of the problem was that drag&drop share a lot of code with normal clicks becasue it's just a disguised 2 clicks. Startind to use different behaviors means using different code, reducing the risk of bug for standard moves
20090827 04:31:21< alink> but also killing its code simplicity :-/
20090827 04:32:28< alink> hehe i always remember the young alink proposing a 1-line patch introducing a working drag&drop
20090827 04:32:57< alink> good old time :)
20090827 04:35:33< alink> mmh indeed, no undo for goto
20090827 04:56:23-!- wesbot changed the topic of #wesnoth-dev to: 1.7.4 planned for Sunday | 82 bugs, 242 feature requests, 10 patches | logs: http://irclogs.wesnoth.org/ | Don't paste on IRC! Use a pastebin: http://wesnoth.pastebin.com | http://imagebin.org
20090827 05:07:55-!- happygrue_ [n=George@c-67-176-145-41.hsd1.in.comcast.net] has joined #wesnoth-dev
20090827 05:08:15< CIA-62> alink * r38130 /trunk/src/menu_events.cpp: Fix broken undo for goto, caused by a typo in r37801
20090827 05:08:34< alink> Soliton: ^
20090827 05:08:58-!- happygrue [n=George@wesnoth/developer/wintermute] has quit [Read error: 104 (Connection reset by peer)]
20090827 05:09:21< alink> another reason why reference are better than pointer, no unwanted casting between bool and pointer
20090827 05:11:05< alink> good, seems that my wesnoth-dev setup is almost ok now
20090827 05:11:48< Soliton> alink: ahh, excellent.
20090827 05:12:23-!- happygrue [n=George@wesnoth/developer/wintermute] has joined #wesnoth-dev
20090827 05:12:57-!- happygrue_ [n=George@c-67-176-145-41.hsd1.in.comcast.net] has quit [Read error: 104 (Connection reset by peer)]
20090827 05:13:09-!- happygrue [n=George@wesnoth/developer/wintermute] has quit [Read error: 104 (Connection reset by peer)]
20090827 05:13:12-!- happygrue [n=George@wesnoth/developer/wintermute] has joined #wesnoth-dev
20090827 05:13:54< Soliton> alink: btw, the random leader choice regression was because leaders where named after their faction and not the player's name. haven't looked into how to fix that mess though.
20090827 05:14:43< alink> Soliton: did you see wath i posted in the bugreport ?
20090827 05:15:00< Soliton> yeah, briefly.
20090827 05:15:08< alink> and indeed it is messy
20090827 05:15:53< alink> I was horrified when I saw that 1.6 handled incorrect and mixed WML keys, which change the UI etc, without anyone noticing it
20090827 05:16:56< Soliton> one issues is the the host and players have different UIs.
20090827 05:17:12< Soliton> something that should be abandoned at some point...
20090827 05:17:18< alink> yes there is also that
20090827 05:17:25< fabi> Is there a known bug that prevents the terrain masks to work?
20090827 05:18:03< Soliton> fabi: well, dark forecast worked last time we played.
20090827 05:19:29< alink> another source of the problem was that the players UI didn't have headers columns, which is a UI flaw
20090827 05:21:23< alink> also the idea of using one icon for indicating both faction and leaders is bad and confusing
20090827 05:21:49< fabi> Soliton: Right, the normal mask feature seems to work. But the extra [rule] isn't doing what it should.
20090827 05:22:22< Soliton> i see. no haven't heard that before.
20090827 05:25:30< alink> anyways, i go to sleep now, this short hello to wesnoth-dev took me more time that I planned ;)
20090827 05:25:32< alink> bye
20090827 05:25:43-!- alink [n=alink@host-85-27-114-223.brutele.be] has quit [Remote closed the connection]
20090827 05:29:41-!- happygrue_ [n=George@c-67-176-145-41.hsd1.in.comcast.net] has joined #wesnoth-dev
20090827 05:29:55-!- happygrue [n=George@wesnoth/developer/wintermute] has quit [Read error: 104 (Connection reset by peer)]
20090827 05:56:23-!- crimson_penguin [n=ben@wesnoth/developer/crimsonpenguin] has quit ["Bed"]
20090827 06:22:52-!- Aethaeryn [n=Michael@wesnoth/umc-dev/developer/aethaeryn] has quit ["Yay."]
20090827 06:28:47-!- happygrue_ [n=George@c-67-176-145-41.hsd1.in.comcast.net] has quit [Read error: 104 (Connection reset by peer)]
20090827 06:29:54-!- ancestral [n=ancestra@97-116-111-138.mpls.qwest.net] has joined #wesnoth-dev
20090827 07:19:00-!- dfranke [i=dfranke@wesnoth/developer/dfranke] has joined #wesnoth-dev
20090827 07:37:58-!- Sirp [n=user@wesnoth/developer/dave] has quit ["leaving"]
20090827 07:46:15-!- noy [n=Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20090827 07:51:43-!- melinath [n=melinath@pool-72-72-200-231.altnpa.east.verizon.net] has quit [Read error: 113 (No route to host)]
20090827 07:59:54-!- zookeeper [n=l@wesnoth/developer/zookeeper] has joined #wesnoth-dev
20090827 08:00:31< Soliton> zookeeper: http://www.wesnoth.org/forum/viewtopic.php?f=21&t=26818  shouldn't that automatically invalidate undo?
20090827 08:03:18< zookeeper> Soliton, no
20090827 08:03:30< zookeeper> err, it had a kill=yes, so i suppose..
20090827 08:05:06< zookeeper> i mean, i'd have expected it to, but it doesn't really matter if it doesn't.
20090827 08:49:43-!- dfranke [i=dfranke@wesnoth/developer/dfranke] has quit [Read error: 60 (Operation timed out)]
20090827 08:50:38-!- euschn [n=chatzill@wesnoth/developer/euschn] has joined #wesnoth-dev
20090827 09:01:00-!- Crab_ [n=Crab_@wesnoth/developer/crab] has joined #wesnoth-dev
20090827 09:01:35-!- dfranke [i=dfranke@wesnoth/developer/dfranke] has joined #wesnoth-dev
20090827 09:03:47-!- Crab_ [n=Crab_@wesnoth/developer/crab] has quit ["Leaving."]
20090827 09:13:18-!- busfahrer [n=busfahre@unixboard/user/busfahrer] has joined #wesnoth-dev
20090827 09:19:12-!- YogiHH [i=d4ca9d15@wesnoth/developer/yogihh] has joined #wesnoth-dev
20090827 09:19:16< YogiHH> hello
20090827 09:26:29-!- EdB [n=edb@106.117.88-79.rev.gaoland.net] has joined #wesnoth-dev
20090827 09:32:51< euschn> hi YogiHH 
20090827 09:33:44-!- Espreon [n=espreon@wesnoth/developer/espreon] has quit ["WRYYYYYYYYYYYYYYYYYYYY!"]
20090827 10:35:38-!- loonybot [n=loonybot@79.139.138.162] has joined #wesnoth-dev
20090827 10:37:00-!- loonycyborg [n=sergey@79.139.138.162] has joined #wesnoth-dev
20090827 10:51:31-!- Chusslove [n=Chusslov@212.124.171.10] has joined #wesnoth-dev
20090827 10:54:20< Ivanovic> moin
20090827 10:56:23-!- wesbot changed the topic of #wesnoth-dev to: 1.7.4 planned for Sunday | 83 bugs, 242 feature requests, 10 patches | logs: http://irclogs.wesnoth.org/ | Don't paste on IRC! Use a pastebin: http://wesnoth.pastebin.com | http://imagebin.org
20090827 10:56:52< Soliton> YogiHH: (or anyone else) do you remember "4 years ago I wanted to join developers' team. When I implemented a really big patch - all I got was bashing and ridiculing." --sauron?
20090827 10:57:21< Soliton> did he ever post an actual patch to patches.wesnoth.org?
20090827 11:02:17-!- thespaceinvader [n=chatzill@cpc3-whit1-0-0-cust862.cdif.cable.ntl.com] has joined #wesnoth-dev
20090827 11:05:02-!- ancestral [n=ancestra@97-116-111-138.mpls.qwest.net] has quit []
20090827 11:07:23-!- busfahrer [n=busfahre@unixboard/user/busfahrer] has quit ["leaving"]
20090827 11:17:04-!- ilor [n=user@wesnoth/developer/ilor] has joined #wesnoth-dev
20090827 11:21:51-!- thespaceinvader [n=chatzill@wesnoth/artist/thespaceinvader] has quit [Read error: 110 (Connection timed out)]
20090827 11:23:51-!- elias [n=allefant@allegro/developer/allefant] has joined #wesnoth-dev
20090827 11:25:06< YogiHH> Soliton: To be honest, i can't recall the exact circumstances. All i got to know by that time was the forum communication. I haven't checked, but from that i don't remember it. Maybe there was some PM stuff going on in parallel.
20090827 11:25:52< YogiHH> Soliton: Or maybe sauron interpreted his effort to create the mod with the intention of getting it into the official source as "joining the dev team"
20090827 11:32:12-!- Espreon [n=espreon@wesnoth/developer/espreon] has joined #wesnoth-dev
20090827 11:40:11-!- Zarel [i=Zarel@warzone2100/developer/Zarel] has quit [Read error: 110 (Connection timed out)]
20090827 11:41:13-!- mjs-de [n=mjs-de@vpw.wh.uni-dortmund.de] has joined #wesnoth-dev
20090827 11:43:29-!- ilor_ [n=user@wesnoth/developer/ilor] has joined #wesnoth-dev
20090827 11:50:01-!- dfranke [i=dfranke@wesnoth/developer/dfranke] has quit [Read error: 60 (Operation timed out)]
20090827 11:52:57< CIA-62> thespaceinvader * r38131 /trunk/ (6 files in 5 dirs): Add and wire new Grand Knight portrait, update changelogs, portrait credits.
20090827 11:53:16-!- thespaceinvader [n=chatzill@wesnoth/artist/thespaceinvader] has joined #wesnoth-dev
20090827 11:54:46-!- euschn [n=chatzill@wesnoth/developer/euschn] has quit [Remote closed the connection]
20090827 11:55:42-!- ilor [n=user@wesnoth/developer/ilor] has quit [Read error: 110 (Connection timed out)]
20090827 12:05:35-!- ilor_ [n=user@wesnoth/developer/ilor] has quit []
20090827 12:06:22-!- Blueblaze [n=nick@c-98-197-50-41.hsd1.tx.comcast.net] has quit [Remote closed the connection]
20090827 12:08:39-!- noy [n=Noy@wesnoth/developer/noy] has quit []
20090827 12:26:25-!- dfranke [i=dfranke@wesnoth/developer/dfranke] has joined #wesnoth-dev
20090827 12:27:36-!- Elvish_Pillager [n=eli@71-10-231-36.dhcp.oxfr.ma.charter.com] has joined #wesnoth-dev
20090827 12:29:05< fabi> How can I reset the "used_items" in a savegame?
20090827 12:29:55< fabi> zookeeper: around?
20090827 12:30:32-!- dfranke [i=dfranke@wesnoth/developer/dfranke] has quit [Read error: 60 (Operation timed out)]
20090827 12:35:18< zookeeper> fabi, yes. what used_items?
20090827 12:35:39< zookeeper> some kind of a list of id's of objects picked up so far?
20090827 12:35:55< fabi> zookeeper: [object] looks for items to not give them a second time to a unit.
20090827 12:36:33< fabi> This is stored in the save as used_items. But this seems to be a variable that is behind the scope of wml.
20090827 12:36:45< zookeeper> yeah, you can't change that
20090827 12:37:05< zookeeper> if you need to do that, then you just have to do the object bookkeeping manually
20090827 12:37:13< fabi> I want a treasure chest that drops if the carrier dies. My code is :
20090827 12:38:41< fabi> http://wesnoth.pastebin.com/ddcb92d5
20090827 12:39:10< fabi> zookeeper: your answer suggests to avoid the [object] tag?
20090827 12:39:30< zookeeper> no, you can just remove the id=treasure and it should work
20090827 12:39:49< zookeeper> if there's no id=, then it can be picked up again
20090827 12:41:13< fabi> zookeeper: This will give me points where several chests can be picked up? Because the old points aren't disabled?
20090827 12:41:28< zookeeper> also better test whether duration=level still works if the unit advances...IIRC there's a bug which causes duration=level objects to disappear on advancement
20090827 12:42:23< zookeeper> yes, several chests can then be picked up
20090827 12:44:55< fabi> This is more difficult then I thought.
20090827 12:45:07< zookeeper> hmh?
20090827 12:45:24< fabi> To code the chest that can be dropped.
20090827 12:46:47-!- zookeeper2 [n=l@88.148.253.151] has joined #wesnoth-dev
20090827 12:48:19< zookeeper2> what's the problem? AFAICT you just need to have the death of a chest carrier spawn a new moveto event, like you have there...just remove the id= and first_time_only=no and add the object filters to the event filter so no one but a viable carrier can trigger it
20090827 12:50:06< zookeeper2> i'm not a fan of relying on cannot_use_message and all that, i like to do the checks myself with an [if] or the event filters
20090827 12:52:37< zookeeper2> alternatively you can just have each chest have a different id=. when the carrier dies, increment a counter and use id=treasure$counter or something.
20090827 12:53:40-!- dfranke [i=dfranke@wesnoth/developer/dfranke] has joined #wesnoth-dev
20090827 12:54:36< fabi> zookeeper2: Thank you.
20090827 12:55:00-!- zookeeper [n=l@wesnoth/developer/zookeeper] has quit [Read error: 145 (Connection timed out)]
20090827 12:55:46-!- zookeeper2 is now known as zookeeper
20090827 13:02:51-!- EdB [n=edb@106.117.88-79.rev.gaoland.net] has quit [Remote closed the connection]
20090827 13:07:25-!- Espreon [n=espreon@wesnoth/developer/espreon] has quit ["WRYYYYYYYYYYYYYYYYYYYY!"]
20090827 13:07:59-!- dfranke [i=dfranke@wesnoth/developer/dfranke] has quit [Read error: 104 (Connection reset by peer)]
20090827 13:12:29< fabi> zookeeper: Works now.
20090827 13:30:11-!- dfranke [i=dfranke@wesnoth/developer/dfranke] has joined #wesnoth-dev
20090827 13:36:37< fabi> zookeeper: Where is for example the "slow" weapon thing defined. I would like to have such a item with explanation for the carrier.
20090827 13:45:24< zookeeper> fabi, the what? weapon specials are in abilities.cfg and the special notes in special-notes.cfg
20090827 13:52:31-!- dfranke [i=dfranke@wesnoth/developer/dfranke] has quit [Read error: 54 (Connection reset by peer)]
20090827 14:03:01-!- thespaceinvader_ [n=chatzill@cpc3-whit1-0-0-cust854.cdif.cable.ntl.com] has joined #wesnoth-dev
20090827 14:06:35-!- Elvish_Pillage2 [n=eli@71-10-231-36.dhcp.oxfr.ma.charter.com] has joined #wesnoth-dev
20090827 14:06:35-!- thespaceinvader_ [n=chatzill@cpc3-whit1-0-0-cust854.cdif.cable.ntl.com] has quit [Read error: 104 (Connection reset by peer)]
20090827 14:06:55-!- thespaceinvader_ [n=chatzill@cpc3-whit1-0-0-cust854.cdif.cable.ntl.com] has joined #wesnoth-dev
20090827 14:20:27-!- thespaceinvader [n=chatzill@wesnoth/artist/thespaceinvader] has quit [Read error: 110 (Connection timed out)]
20090827 14:22:46-!- dfranke [i=dfranke@wesnoth/developer/dfranke] has joined #wesnoth-dev
20090827 14:22:58-!- Elvish_Pillager [n=eli@71-10-231-36.dhcp.oxfr.ma.charter.com] has quit [Read error: 110 (Connection timed out)]
20090827 14:28:18-!- thespaceinvader_ [n=chatzill@cpc3-whit1-0-0-cust854.cdif.cable.ntl.com] has quit [Read error: 110 (Connection timed out)]
20090827 14:29:55-!- dfranke [i=dfranke@wesnoth/developer/dfranke] has quit [Connection reset by peer]
20090827 14:35:19-!- Chusslove [n=Chusslov@212.124.171.10] has quit [Read error: 145 (Connection timed out)]
20090827 14:45:34-!- elias [n=allefant@allegro/developer/allefant] has quit [Read error: 60 (Operation timed out)]
20090827 14:50:33-!- Chusslove [n=Chusslov@212.124.171.10] has joined #wesnoth-dev
20090827 14:51:39-!- dfranke [i=dfranke@wesnoth/developer/dfranke] has joined #wesnoth-dev
20090827 15:10:23-!- dfranke [i=dfranke@wesnoth/developer/dfranke] has quit [Read error: 110 (Connection timed out)]
20090827 15:20:29-!- dfranke [i=dfranke@wesnoth/developer/dfranke] has joined #wesnoth-dev
20090827 15:21:27-!- BenUrban [n=benurban@unaffiliated/benurban] has quit ["Power failu"]
20090827 15:25:31-!- euschn [n=chatzill@wesnoth/developer/euschn] has joined #wesnoth-dev
20090827 15:26:09-!- Blarumyrran [n=minaise@81.20.159.197] has joined #wesnoth-dev
20090827 15:28:56-!- dfranke [i=dfranke@wesnoth/developer/dfranke] has quit [Read error: 104 (Connection reset by peer)]
20090827 15:47:22-!- fendrin [n=fabi@wesnoth/developer/fendrin] has joined #wesnoth-dev
20090827 15:48:25-!- dfranke [i=dfranke@wesnoth/developer/dfranke] has joined #wesnoth-dev
20090827 16:00:21-!- fabi [n=fabi@wesnoth/developer/fendrin] has quit [Read error: 110 (Connection timed out)]
20090827 16:01:31< YogiHH> Soliton: I tried to reproduce the corrupted replays of mp games that are loaded again. I tried with local games, i tried on a server, as host, as client, everything seems to work alright. Do you know of a method to reproduce?
20090827 16:09:21-!- dfranke [i=dfranke@wesnoth/developer/dfranke] has quit [Read error: 110 (Connection timed out)]
20090827 16:11:54-!- ancestral [n=ancestra@97-116-111-138.mpls.qwest.net] has joined #wesnoth-dev
20090827 16:16:33-!- Aethaeryn [n=Michael@wesnoth/umc-dev/developer/aethaeryn] has joined #Wesnoth-dev
20090827 16:17:05-!- dfranke [i=dfranke@wesnoth/developer/dfranke] has joined #wesnoth-dev
20090827 16:23:09-!- crimson_penguin [n=ben@wesnoth/developer/crimsonpenguin] has joined #wesnoth-dev
20090827 16:35:13-!- ancestral [n=ancestra@97-116-111-138.mpls.qwest.net] has quit []
20090827 16:37:18< CIA-62> fendrin * r38132 /trunk/data/campaigns/Legend_of_Wesmere/scenarios/05_Saurian_Treasury.cfg: LoW 05: Redesigned the scenario so that the player has to transport the treasure with a horserider.
20090827 16:40:18-!- lizard_r [n=Miranda@wesnoth/umc-dev/developer/lizard] has joined #wesnoth-dev
20090827 16:43:58-!- YogiHH [i=d4ca9d15@wesnoth/developer/yogihh] has quit [Ping timeout: 180 seconds]
20090827 16:44:40-!- thespaceinvader [n=chatzill@wesnoth/artist/thespaceinvader] has joined #wesnoth-dev
20090827 16:49:02-!- melinath [n=melinath@pool-72-72-200-231.altnpa.east.verizon.net] has joined #wesnoth-dev
20090827 16:56:23-!- wesbot changed the topic of #wesnoth-dev to: 1.7.4 planned for Sunday | 85 bugs, 242 feature requests, 10 patches | logs: http://irclogs.wesnoth.org/ | Don't paste on IRC! Use a pastebin: http://wesnoth.pastebin.com | http://imagebin.org
20090827 16:58:28-!- elias [n=allefant@allegro/developer/allefant] has joined #wesnoth-dev
20090827 16:58:36-!- dfranke [i=dfranke@wesnoth/developer/dfranke] has quit [Read error: 104 (Connection reset by peer)]
20090827 17:12:22< Ivanovic> nice bug: https://gna.org/bugs/index.php?14202
20090827 17:13:11< Ivanovic> zookeeper: for you? https://gna.org/bugs/index.php?14191
20090827 17:17:37-!- euschn [n=chatzill@wesnoth/developer/euschn] has quit ["ChatZilla 0.9.85 [Firefox 3.5.2/20090729225027]"]
20090827 17:18:05-!- dfranke [i=dfranke@wesnoth/developer/dfranke] has joined #wesnoth-dev
20090827 17:35:02-!- Kenpachi [n=chatzill@CPE-58-170-105-239.sa.bigpond.net.au] has joined #wesnoth-dev
20090827 17:53:57-!- fendrin [n=fabi@wesnoth/developer/fendrin] has quit [Remote closed the connection]
20090827 17:55:32-!- Chusslove [n=Chusslov@212.124.171.10] has quit [Read error: 145 (Connection timed out)]
20090827 18:09:15-!- Chusslove [n=Chusslov@212.124.171.61] has joined #wesnoth-dev
20090827 18:09:44-!- ABCD [n=ABCD@gentoo/contributor/abcd] has quit [Client Quit]
20090827 18:10:11-!- ABCD [n=ABCD@gentoo/contributor/abcd] has joined #wesnoth-dev
20090827 18:14:26-!- thespaceinvader_ [n=chatzill@cpc1-whit1-0-0-cust763.cdif.cable.ntl.com] has joined #wesnoth-dev
20090827 18:15:08-!- thespaceinvader [n=chatzill@wesnoth/artist/thespaceinvader] has quit [Nick collision from services.]
20090827 18:15:10-!- thespaceinvader_ is now known as thespaceinvader
20090827 18:19:23-!- alink [n=alink@wesnoth/developer/alink] has joined #wesnoth-dev
20090827 18:32:30-!- noy [n=Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20090827 19:23:30-!- melinath [n=melinath@pool-72-72-200-231.altnpa.east.verizon.net] has quit [Read error: 113 (No route to host)]
20090827 19:23:50-!- zookeeper [n=l@wesnoth/developer/zookeeper] has quit []
20090827 19:28:48-!- xchat_sux [n=eli@71-10-231-36.dhcp.oxfr.ma.charter.com] has joined #wesnoth-dev
20090827 19:31:47-!- Elvish_Pillage2 [n=eli@71-10-231-36.dhcp.oxfr.ma.charter.com] has quit [Read error: 60 (Operation timed out)]
20090827 19:33:31-!- noy [n=Noy@wesnoth/developer/noy] has quit []
20090827 19:42:54-!- ilor [n=user@wesnoth/developer/ilor] has joined #wesnoth-dev
20090827 19:50:49-!- Chusslove [n=Chusslov@212.124.171.61] has quit [Read error: 145 (Connection timed out)]
20090827 19:50:55-!- zookeeper [n=l@wesnoth/developer/zookeeper] has joined #wesnoth-dev
20090827 19:54:45-!- stikonas [n=and@ctv-213-164-125-176.vinita.lt] has joined #wesnoth-dev
20090827 20:03:38-!- Kenpachi [n=chatzill@CPE-58-170-105-239.sa.bigpond.net.au] has quit [Remote closed the connection]
20090827 20:05:27-!- grzywacz [n=grzywacz@wesnoth/developer/grzywacz] has joined #wesnoth-dev
20090827 20:05:41-!- Chusslove [n=Chusslov@adsl-ppp-3216.yubc.net] has joined #wesnoth-dev
20090827 20:08:06< Ivanovic> AI0867, Alesis-Novik, corn, deekay, dfranke, elias, esr, grzywacz, ilor, loonycyborg, shadowmaster, Soliton, thespaceinvader, zookeeper, everyoneelsewhocares: please keep in mind that i plan to release 1.7.4 on sunday
20090827 20:08:25< Ivanovic> if you got any blockers or work on things that i should wait for please make sure to ping me "in time"
20090827 20:08:42< thespaceinvader> thanks Ivanovic - with any luck, i should have one more portrait and one more drake fully animated for that time
20090827 20:08:51< Ivanovic> thespaceinvader: cool
20090827 20:13:09-!- noy [n=Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20090827 20:26:42< CIA-62> thespaceinvader * r38134 /trunk/ (6 files in 5 dirs): Add and wire new Merman Hunter portrait, update changelogs, portrait credits.
20090827 20:26:43< CIA-62> thespaceinvader * r38133 /trunk/data/core/units/drakes/Burner.cfg: Minor fix to Burner WML
20090827 20:31:03-!- mordante [n=mordante@wesnoth/developer/mordante] has joined #wesnoth-dev
20090827 20:31:06-!- ancestral [n=ancestra@97-116-111-138.mpls.qwest.net] has joined #wesnoth-dev
20090827 20:31:22< mordante> servus
20090827 20:32:09-!- giusef [n=giusef@unaffiliated/giusef] has joined #wesnoth-dev
20090827 20:32:18 * mordante is semi afk, playing with new rig :-)
20090827 20:33:16< Ivanovic> hui, so you finally got it!
20090827 20:33:27< Ivanovic> debian is already (pre)installed?
20090827 20:33:46< mordante> yes got it :-)
20090827 20:34:10< Ivanovic> don't forget that the current plan is to get 1.7.4 out on sunday
20090827 20:34:22< mordante> of course no OS preinstalled, nobody but me can do it the way I want it ;-)
20090827 20:34:41< mordante> I've no 1.7.4 blockers ;-)
20090827 20:34:55< Ivanovic> uhm, you could give them an image and have them do it...
20090827 20:34:57< Ivanovic> ;)
20090827 20:35:25< mordante> I've no proper image for an 64 bit version of Debian
20090827 20:35:38< Ivanovic> ah, okay
20090827 20:38:22< Aethaeryn> How often does units.wesnoth.org trunk update?
20090827 20:39:10-!- Sirp [n=user@wesnoth/developer/dave] has joined #wesnoth-dev
20090827 20:41:28-!- melinath [n=melinath@pool-72-72-200-231.altnpa.east.verizon.net] has joined #wesnoth-dev
20090827 20:45:37-!- fendrin [n=fabi@wesnoth/developer/fendrin] has joined #wesnoth-dev
20090827 20:53:10< CIA-62> ai0867 * r38135 /trunk/src/ (multiplayer_create.cpp team.cpp team.hpp): Remove broken enemy attribute of [side].
20090827 20:54:42< CIA-62> ai0867 * r38136 /trunk/data/ (3 files in 3 dirs): Remove enemy key from [side] tags.
20090827 20:54:47< CIA-62> ai0867 * r38137 /trunk/ (changelog data/scenario-test.cfg src/team.cpp): [side] team_name is now a comma-separated list of teams the side is on
20090827 20:56:17< fendrin> AI0867: Do you know the corresponding FR?
20090827 20:57:25-!- Crab_ [i=crab@wesnoth/developer/crab] has joined #wesnoth-dev
20090827 20:57:29< CIA-62> alink * r38138 /trunk/src/astarsearch.cpp: 
20090827 20:57:29< CIA-62> Fix pathfinding heuristic for teleport (broken by the A* rewrite, r35140)
20090827 20:57:29< CIA-62> Bug was: pathfinding never tried small detour to find useful teleport.
20090827 20:57:38< fendrin> AI0867: https://gna.org/bugs/?12776
20090827 21:01:42< AI0867> there was a corresponding feature request?
20090827 21:02:00< Soliton> Aethaeryn: once a day.
20090827 21:03:22< AI0867> fendrin: marked FR fixed
20090827 21:04:44< fendrin> AI0867: nice, I will recode LoW 2 at once.
20090827 21:05:16< AI0867> I added that feature because I removed the (long broken) enemy= thing
20090827 21:05:52< fendrin> AI0867: Does it work in modify_side as well?
20090827 21:06:13-!- jzmer [i=815dcecd@gateway/web/freenode/x-oipmuebaiuujdkho] has joined #wesnoth-dev
20090827 21:06:23< AI0867> yes, I only changed the interpreting of the string
20090827 21:06:44< AI0867> not the setting of it
20090827 21:16:09-!- stikonas [n=and@wesnoth/translator/stikonas] has quit [Read error: 104 (Connection reset by peer)]
20090827 21:20:23-!- stikonas [n=and@ctv-213-164-125-176.vinita.lt] has joined #wesnoth-dev
20090827 21:21:43-!- stikonas [n=and@wesnoth/translator/stikonas] has quit [Remote closed the connection]
20090827 21:22:47< CIA-62> mcshark * r38139 /trunk/projectfiles/ (CodeBlocks/wesnoth.cbp CodeBlocks-SCons/wesnoth.cbp): Update according to recent file changes: added src/gui/auxiliary/window_builder/horizontal_listbox.cpp+hpp
20090827 21:27:16-!- stikonas [n=and@wesnoth/translator/stikonas] has joined #wesnoth-dev
20090827 21:31:41< CIA-62> thespaceinvader * r38140 /trunk/data/core/images/portraits/humans/ (peasant.png transparent/peasant.png): Update Peasant portraits.
20090827 21:38:53-!- alink_ [n=alink@host-85-27-114-223.brutele.be] has joined #wesnoth-dev
20090827 21:43:21< shadowmaster> mordante: don't bother looking at the gui2 unit create dialog yet.
20090827 21:43:42< shadowmaster> I suspect that I'm victim of memory corruption in my custom code in charge of filling the unit type and race lists
20090827 21:43:56< mordante> shadowmaster, ok, well won't have time tonight ;-)
20090827 21:44:13< shadowmaster> I'll disable it for the release. I'm sick atm and don't think I can concentrate on fixing the issue
20090827 21:44:16< ilor> mordante: I'm semi afk for the same reason :P
20090827 21:44:28< mordante> cool ilor :-)
20090827 21:49:22-!- Netsplit lindbohm.freenode.net <-> irc.freenode.net quits: Crab_, Chusslove
20090827 21:50:47-!- Netsplit over, joins: Crab_, Chusslove
20090827 21:56:23< ilor> only one issue so far is that the fan on the cpu cooler is vibrating somewhat
20090827 21:57:50< ilor> the vibrations make the entire cooler vibrate slightly, which I did't like very much so I'm running it with the fan off for now. There's some airflow in the case so the temps are bearable
20090827 21:58:17-!- Chusslove [n=Chusslov@adsl-ppp-3216.yubc.net] has quit [Operation timed out]
20090827 22:09:17-!- Chusslove [n=Chusslov@adsl-ppp-1074.yubc.net] has joined #wesnoth-dev
20090827 22:10:56-!- Amu [i=smar@smar.fi] has joined #wesnoth-dev
20090827 22:14:07-!- noy_ [n=noy@24.114.236.37] has joined #wesnoth-dev
20090827 22:15:18< mordante> shadowmaster, why is this? "You may embed only 3 quotes within each other."
20090827 22:17:10 * mordante knows why he hates forums again :-/
20090827 22:17:15-!- 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"]
20090827 22:21:18-!- noy_ [n=noy@24.114.236.37] has quit ["Get Colloquy for iPhone! http://mobile.colloquy.info"]
20090827 22:22:19-!- alink [n=alink@wesnoth/developer/alink] has quit [Read error: 104 (Connection reset by peer)]
20090827 22:22:51-!- noy_ [n=noy@24.114.236.37] has joined #wesnoth-dev
20090827 22:25:23-!- Smar [i=smar@freenet/translator/finnish/Smar] has quit [Read error: 104 (Connection reset by peer)]
20090827 22:26:22< Soliton> mordante: yeah, that hit me lately, too. 3 quotes sounds pretty limited. would be nice if that can be increased.
20090827 22:29:28-!- noy [n=Noy@wesnoth/developer/noy] has quit [Connection timed out]
20090827 22:29:31-!- noy_ is now known as noy
20090827 22:36:41-!- thespaceinvader_ [n=chatzill@cpc1-whit1-0-0-cust763.cdif.cable.ntl.com] has joined #wesnoth-dev
20090827 22:40:51-!- thespaceinvader_ [n=chatzill@cpc1-whit1-0-0-cust763.cdif.cable.ntl.com] has quit [Read error: 104 (Connection reset by peer)]
20090827 22:42:32-!- ancestral [n=ancestra@97-116-111-138.mpls.qwest.net] has quit ["And that’s the end of THAT chapter."]
20090827 22:43:59< Soliton> http://gna.org/bugs/?14196  is that a bug?
20090827 22:44:18< Soliton> i mean just don't reuse variables that way or clear them inbetween.
20090827 22:44:36-!- alink_ [n=alink@host-85-27-114-223.brutele.be] has quit [Read error: 110 (Connection timed out)]
20090827 22:45:41< mordante> not sure, I guess Sapient can answer that question
20090827 22:50:23-!- noy [n=noy@24.114.236.37] has quit ["Get Colloquy for iPhone! http://mobile.colloquy.info"]
20090827 22:50:35-!- thespaceinvader [n=chatzill@wesnoth/artist/thespaceinvader] has quit [Read error: 110 (Connection timed out)]
20090827 22:51:01-!- noy [n=noy@24.114.236.37] has joined #wesnoth-dev
20090827 22:51:12< Soliton> anyone know a tool for windows that creates patches?
20090827 22:51:30< Soliton> i guess tortoise does it but maybe better something indpendent.
20090827 22:52:03< Soliton> maybe there is some windiff program that can do it?
20090827 22:52:51< mordante> diff ;-) http://gnuwin32.sourceforge.net/packages/diffutils.htm
20090827 22:55:49< Soliton> thanks.
20090827 22:56:23-!- wesbot changed the topic of #wesnoth-dev to: 1.7.4 planned for Sunday | 85 bugs, 241 feature requests, 10 patches | logs: http://irclogs.wesnoth.org/ | Don't paste on IRC! Use a pastebin: http://wesnoth.pastebin.com | http://imagebin.org
20090827 23:01:06-!- happygrue [n=George@wesnoth/developer/wintermute] has joined #wesnoth-dev
20090827 23:01:50-!- thespaceinvader_ [n=chatzill@cpc1-whit1-0-0-cust763.cdif.cable.ntl.com] has joined #wesnoth-dev
20090827 23:01:55-!- thespaceinvader_ is now known as thespaceinvader
20090827 23:04:20< mordante> corn, how are things going with converting the canvas to pango?
20090827 23:05:29-!- Blueblaze [n=nick@c-98-197-50-41.hsd1.tx.comcast.net] has joined #wesnoth-dev
20090827 23:10:01-!- noy [n=noy@24.114.236.37] has quit ["Get Colloquy for iPhone! http://mobile.colloquy.info"]
20090827 23:12:31-!- Blueblaze [n=nick@c-98-197-50-41.hsd1.tx.comcast.net] has quit [Remote closed the connection]
20090827 23:18:24-!- deekay [n=dk@wesnoth/developer/dragonking] has quit [Remote closed the connection]
20090827 23:18:24-!- elias [n=allefant@allegro/developer/allefant] has quit [Remote closed the connection]
20090827 23:21:04-!- deekay [n=dk@wesnoth/developer/dragonking] has joined #wesnoth-dev
20090827 23:22:22-!- Appleman1234 [n=Appleman@131.181.100.37] has joined #wesnoth-dev
20090827 23:23:13-!- noy [n=noy@24.114.237.32] has joined #wesnoth-dev
20090827 23:24:46-!- noy [n=noy@24.114.237.32] has quit [Client Quit]
20090827 23:25:09-!- noy [n=noy@24.114.237.32] has joined #wesnoth-dev
20090827 23:26:13-!- zookeeper [n=l@wesnoth/developer/zookeeper] has quit []
20090827 23:27:05< Soliton> grzywacz: btw, you can never get the completed icon for DiD since you have to lose the last scenario. not sure if anything should be done about that though...
20090827 23:28:17< Blarumyrran> giving it for second-to-last wouldnt be too strange
20090827 23:29:38< grzywacz> Soliton, thanks, didn't know that.
20090827 23:30:29< Soliton> i think the ending might need to be modified since you also don't get the credits like usual since you lost.
20090827 23:33:28 * fendrin thinks that Chuck Norris can beat DiD.
20090827 23:33:57< Soliton> you can easily play it forever, sure.
20090827 23:34:55< loonycyborg> Soliton: Maybe have some wml magic to record it as won?
20090827 23:35:09< mordante> fendrin, he can't loose, nobody dares to oppose him :-P
20090827 23:35:50< fendrin> mordante: The ai won't even dare to recruit.
20090827 23:36:09< grzywacz> loonycyborg, probably the only way. That or WML magic to make you win when you lose. More generic. ;)
20090827 23:36:29< mordante> in fact the AI suddenly learns how to retre^Wflee
20090827 23:37:10< loonycyborg> iirc there are some scenarios where you're expected to be defeated and continue.
20090827 23:38:32< fendrin> loonycyborg: They are made to punish the save loaders.
20090827 23:39:13-!- noy [n=noy@24.114.237.32] has quit ["Get Colloquy for iPhone! http://mobile.colloquy.info"]
20090827 23:41:48< grzywacz> lol?
20090827 23:43:50< loonycyborg> There's one such scenario in middle of IftU. Though it doesn't seem to be very good at punishing saveloaders.
20090827 23:45:03 * loonycyborg even managed to defeat the enemy commander :P
20090827 23:50:49< loonycyborg> Walking corpse rushes can be very devastating against someone who immediately destroys them with a counter-attack..
20090827 23:51:55< stikonas> but if you have a hight defence you can kill lots of corpses in one turn with good counter-attack
20090827 23:52:05< stikonas> s/hight/high/
20090827 23:52:40< loonycyborg> Yes. So those hexes will be able to hold more walking corpses.
20090827 23:53:16 * stikonas never played DiD
20090827 23:54:02< Soliton> a 4mp unit rush? sounds fearsome.
20090827 23:54:03< loonycyborg> If none of your units are destroyed by the couterattacks then you're limited to only 6 attacks per turn maximum.
20090827 23:58:44< mordante> I'm off night
20090827 23:59:14-!- mordante [n=mordante@wesnoth/developer/mordante] has quit ["Leaving"]
20090827 23:59:47-!- Sapient [n=patrickp@wesnoth/developer/sapient] has joined #wesnoth-dev
--- Log closed Fri Aug 28 00:00:00 2009