--- Log opened Fri Nov 09 00:00:13 2012
20121109 00:13:10-!- shadowm_desktop [shadowm@wesnoth/developer/shadowmaster] has quit [Quit: Reconnecting]
20121109 00:13:21-!- shadowm_desktop [shadowm@wesnoth/developer/shadowmaster] has joined #wesnoth-umc-dev
20121109 00:20:41-!- shadowm_desktop [shadowm@wesnoth/developer/shadowmaster] has quit [Quit: Reconnecting]
20121109 00:20:50-!- shadowm_desktop [ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-umc-dev
20121109 00:24:49-!- csarmi [~csarmi@178-164-130-120.pool.digikabel.hu] has quit [Ping timeout: 272 seconds]
20121109 00:40:47-!- irker431 [~irker@ai0867.net] has joined #wesnoth-umc-dev
20121109 00:40:47< irker431> AI-Demos: Alarantalara master * r55fc096 / lua/generic-recruit_engine.lua : Set up find_best_hex to look for villages too - http://git.io/dKNP0A
20121109 00:40:47< irker431> AI-Demos: Alarantalara master * r4afb68f / lua/generic-recruit_engine.lua : Factor out the actual recruit picking and place it in loop - http://git.io/Ja9WhA
20121109 00:47:52-!- csarmi [csarmi@178-164-141-13.pool.digikabel.hu] has joined #wesnoth-umc-dev
20121109 00:50:53< irker431> AI-Demos: Alarantalara master * read2aa1 / (lua/ai_helper.lua lua/generic-recruit_engine.lua): Factor out some generally useful code into ai_helper - http://git.io/fHII7Q
20121109 01:01:33< Alarantalara> skyfaller: A question about recruiting.
20121109 01:02:20< Alarantalara> If you are significantly outnumbered but still have income, would you prefer to recruit a cheaper but not optimal unit over the better unit a turn or 2 later?
20121109 01:02:42< skyfaller> that partly depends on how desperate the situation is...
20121109 01:03:07< skyfaller> like, if my army got wiped out in an assault on the enemy keep, so it will take him time to threaten my leader, I can afford to wait
20121109 01:03:21< skyfaller> if it is a bad time of day for the enemy to attack, then I can afford to wait
20121109 01:03:43< skyfaller> if night is about to fall and orcs are already in my face, then I need a unit NOW
20121109 01:04:02< skyfaller> it's basically a question of timing my next attack
20121109 01:04:12< skyfaller> that's how I always think about recruiting
20121109 01:04:32< skyfaller> "how many turns will it take this unit to reach the front, and will it arrive at the ideal time of day?"
20121109 01:05:12< skyfaller> for instance, playing as Drakes I'd want my drakes to hit at dawn and my saurians to hit at dusk
20121109 01:05:51< Alarantalara> So I could start with something like: if the enemy can attack my leader next turn and I'm outnumbered: recruit if at all possible
20121109 01:06:13< skyfaller> if I can buy a glider now and it would reach the front at dawn, that's probably better than a clasher that would arrive at dusk
20121109 01:06:13< skyfaller> unless I'm desperate for a tank
20121109 01:06:40< skyfaller> ... did that help? Or did I throw in too many conditions? :P
20121109 01:07:08-!- skyfaller [~skyfaller@wikipedia/Skyfaller] has quit [Read error: Connection reset by peer]
20121109 01:07:44-!- skyfaller [~skyfaller@ool-43551edd.dyn.optonline.net] has joined #wesnoth-umc-dev
20121109 01:07:44-!- skyfaller [~skyfaller@ool-43551edd.dyn.optonline.net] has quit [Changing host]
20121109 01:07:44-!- skyfaller [~skyfaller@wikipedia/Skyfaller] has joined #wesnoth-umc-dev
20121109 01:07:58< skyfaller> Alarantalara: did you get what I said? I disconnected
20121109 01:08:17< Alarantalara> that helps, though the glider/clasher comparison is perhaps more useful in situations where the AI isn't losing
20121109 01:09:03< skyfaller> well, I guess if I'm losing and I need tanks to hold a defensive line, then I would try to wait... hm
20121109 01:09:09< Alarantalara> I think I got everything you said if it ends with: "Or did I throw in too many conditions? :P"
20121109 01:09:16< skyfaller> yes :)
20121109 01:10:49< Alarantalara> I suppose the situation I'm thinking of is more like: I could recruit a skirmisher now and have 3 units to protect my leader from those 5 enemy units coming in that will attack me next turn, or wait one turn and if I survive have a clasher with the extra income
20121109 01:11:55< Alarantalara> it's even possible that the enemy is elvish fighters and it's day if that's not enough of a difference in quality
20121109 01:12:38< skyfaller> I mean, how dangerous are elf fighters to your leader?
20121109 01:13:00< skyfaller> if you've got a soothsayer leader then the situation is dire, you need defense now
20121109 01:13:36< skyfaller> if you've got a lvl2 clasher leader, you can probably take some hits from elf fighters
20121109 01:14:31< skyfaller> it's all about how desperate the situation is, and how long you can afford to wait
20121109 01:15:20< Alarantalara> This would be a lot easier if my current income is negative or 0
20121109 01:17:15< Alarantalara> What it sounds like is that it can make sense to go for recruiting anything now rather than something good later, but that it will be hard to find a clear set of conditions for what implies "do it now".
20121109 01:18:26< skyfaller> basically, if waiting means that your leader is seriously threatened, then buy suboptimal units now
20121109 01:18:34< skyfaller> in most other situations you can afford to wait
20121109 01:19:12< skyfaller> but I would not wait if a less powerful unit can reach the front at a good time of day while a more powerful unit would arrive at a bad time of day
20121109 01:19:44< skyfaller> you want as many units at the front as possible during your powerphase, even if some are "wussy"
20121109 01:20:19< skyfaller> when it's the enemy's powerphase you should be retreating and avoiding battle if possible anyway
20121109 01:20:41< skyfaller> the only time that doesn't apply is if they are threatening your keep
20121109 01:21:52< Alarantalara> Maximizing units for the powerphase would be a separate consideration though for recruitment, since that's more a relative quality question in which faster units occasionally get bonuses at some times of day
20121109 01:21:52< skyfaller> I see
20121109 01:22:09< Alarantalara> especially since the faster wussy unit could be more expensive
20121109 01:22:21< skyfaller> fair enough :)
20121109 01:24:57< Alarantalara> I suppose this means we're going to have to play a game on Elensefar Courtyard one day thanks to its wild time of day rules
20121109 01:29:19< skyfaller> oh gosh
20121109 01:30:07< skyfaller> have you played tombs of kesorak or whatever? It's got fires that are local daytime and crypts that are local nighttime
20121109 01:30:16< Alarantalara> Ah yes, that one too
20121109 01:30:35< skyfaller> that's a hilarious map, I had a really good game on there where my UD held out during the day by putting units in the crypts
20121109 01:30:42< skyfaller> or maybe I had orcs, can't remember
20121109 01:30:52-!- shadowm_desktop [ignacio@wesnoth/developer/shadowmaster] has quit [Read error: Connection reset by peer]
20121109 01:31:03< skyfaller> the fires aren't as numerous / strategically located IMHO
20121109 01:31:07< Alarantalara> As a bonus, in 1.11, Woses can now hide in the dead trees
20121109 01:31:27< Alarantalara> they can't in 1.10
20121109 01:31:27< skyfaller> huh
20121109 01:31:30< skyfaller> how does a living wose hide in dead trees?
20121109 01:31:35< skyfaller> I guess "ask a wose"
20121109 01:31:41< skyfaller> the wose knows
20121109 01:31:52< Alarantalara> The same way the wose hides in the deciduous winter trees I imagine
20121109 01:32:00< skyfaller> good point
20121109 01:32:04< skyfaller> maybe a lot of makeup
20121109 01:32:13< skyfaller> a LOT of makeup
20121109 01:32:25< Alarantalara> It pretends it's an evergreen?
20121109 01:32:34< skyfaller> "look, I'm a teapot!"
20121109 01:32:40-!- shadowm_desktop [ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-umc-dev
20121109 01:33:51< Alarantalara> I suppose I should say that's only in trunk right now. They still can't hide in 1.11.0
20121109 01:33:58< skyfaller> ah
20121109 01:34:19< Alarantalara> so 1.11.1 will be incompatible with 1.11.0
20121109 01:34:36< skyfaller> I guess the Rebels' costume department is still waiting for the right supplies to decorate their woses
20121109 01:34:45< Alarantalara> and ambushers
20121109 01:35:15< skyfaller> well, elves are skinny enough to hide behind trees of any sort
20121109 01:35:21< skyfaller> it's woses that have to hide in plain sight
20121109 01:35:43< skyfaller> woses hate hide and seek
20121109 01:35:43< Alarantalara> so why can't they hide behind dead great trees right now?
20121109 01:36:12< skyfaller> couldn't find enough makeup to make themselves look dead
20121109 01:36:26< Alarantalara> they = ambushers
20121109 01:36:29-!- ancestral [~ancestral@65-128-234-158.mpls.qwest.net] has joined #wesnoth-umc-dev
20121109 01:36:31< skyfaller> oh, ambushers
20121109 01:36:39< skyfaller> it just hasn't occurred to them yet
20121109 01:37:16< skyfaller> "what, hide behind that? But it's all dead and splinter-y. I could get infected and die."
20121109 01:37:40< Alarantalara> The desert palm makeup hasn't arrived yet either. (which is the reported bug that got me to notice the great trees in the first place)
20121109 01:37:46< skyfaller> ah, I see
20121109 01:37:55< skyfaller> maybe the elves climb the trees to hide
20121109 01:38:03< skyfaller> and they haven't learned how to shimmy up palm trees yet
20121109 02:00:45< irker431> AI-Demos: Alarantalara master * rd0711df / lua/ai_helper.lua : Fix variable naming error in get_closest_location and have it return distance as well - http://git.io/nvFhiQ
20121109 02:00:46< irker431> AI-Demos: Alarantalara master * refad1e0 / lua/generic-recruit_engine.lua : Use get_closest_hex instead of custom code - http://git.io/aGf5KQ
20121109 02:22:59-!- csarmi [csarmi@178-164-141-13.pool.digikabel.hu] has quit [Ping timeout: 256 seconds]
20121109 02:39:55-!- shadowm_desktop [ignacio@wesnoth/developer/shadowmaster] has quit [Read error: Connection reset by peer]
20121109 02:44:06-!- shadowm_desktop [ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-umc-dev
20121109 02:45:53-!- mattsc [~mattsc@d154-20-32-241.bchsia.telus.net] has joined #wesnoth-umc-dev
20121109 02:51:47< irker431> AI-Demos: Alarantalara master * rfd3de87 / lua/generic-recruit_engine.lua : Recruit close to target villages - http://git.io/0EayVQ
20121109 02:52:51< skyfaller> you guys commit so much code that I can't keep up in understanding what is happening!
20121109 02:53:25< Alarantalara> skyfaller: Play 1 turn of Arcanclave Citadel without fog and see where it recruits
20121109 02:53:46< Alarantalara> It doesn't pick the right units yet, but...
20121109 02:53:51< mattsc> skyfaller: you're assuming that we understand what we are doing
20121109 02:54:04< skyfaller> hahaha
20121109 02:54:18< Alarantalara> All of the commits of today have gone towards that in little steps
20121109 02:54:28-!- csarmi [csarmi@195-38-114-53.pool.digikabel.hu] has joined #wesnoth-umc-dev
20121109 02:54:42< Alarantalara> Now I just have to fix what is recruited and improve the position choice a bit
20121109 02:55:13< skyfaller> I originally intended to help with commenting the code as a prelude to improving it
20121109 02:55:38< skyfaller> but just comprehending the code as it flies by is a full-time job!
20121109 02:56:01-!- Blueblaze [~Blueblaze@67.214.66.174] has joined #wesnoth-umc-dev
20121109 02:56:12< skyfaller> I don't know how I'll ever reach the point where I can commit some code to the project myself
20121109 02:56:17< skyfaller> it is too much of a moving target
20121109 02:56:41< Alarantalara> Pick something that isn't changing right now. Understand it and go for it
20121109 02:56:58< Alarantalara> That's how I got started on this recruitment bit
20121109 02:57:36-!- Blueblaze [~Blueblaze@67.214.66.174] has quit [Client Quit]
20121109 02:58:15< Alarantalara> mattsc was looking at moving units and wasn't worried about recruitment and it was easy to understand as it was, so away I went
20121109 02:58:18< mattsc> skyfaller: yeah, if you want something that isn't going to be worked on for a while but could use some work, I'm sure Alarantalara and I can make plenty suggestions
20121109 02:58:31< skyfaller> ok, hit me
20121109 02:58:39< mattsc> right now I'll be afk for a little though.  bbl
20121109 02:58:57< skyfaller> I might not be ready to contribute for a while, but if I were up to speed today, where could I help?
20121109 02:59:15< Alarantalara> two things come to mind
20121109 02:59:38< Alarantalara> better identification of when to attack/retreat (not making the attacks, but deciding if it's needed)
20121109 03:00:15< Alarantalara> right now Fred assumes he's chaotic, but we need some kind of better evaluation for time of day shifts that can handle lawful/neutral units
20121109 03:00:24< Alarantalara> and cases when the enemy is also chaotic
20121109 03:00:46< Alarantalara> 2) Logic to find and identify zones of concern
20121109 03:01:07< Alarantalara> Fred's are hard coded. Can you write something that will let Fred discover the same zones by himself
20121109 03:02:06< skyfaller> gosh
20121109 03:03:02< skyfaller> I kind of want to prioritize Fred in his current state, i.e. orcs side 1 freelands
20121109 03:03:20< skyfaller> both of those will be important for generalizing Fred
20121109 03:03:24< Alarantalara> Do both of those and Fred can play any map
20121109 03:03:33< Alarantalara> with any faction
20121109 03:03:52< skyfaller> but right now Fred is not playing excellently even in his very limited state
20121109 03:04:41< Alarantalara> the when to attack/retreat may be part of that. mattsc just guessed at the values to decide when
20121109 03:04:59< skyfaller> can you show me the code that decides attack/retreat in Fred currently?
20121109 03:06:42< Alarantalara> line 82ff of grunt-rush-Freelands contains the hard coded config
20121109 03:07:04< Alarantalara> including constants used to evaluate which choice to make
20121109 03:07:30< Alarantalara> You may want to change those constants and/or and new things to consider (like damage ratios)
20121109 03:08:26< Alarantalara> then grunt_rush_FLS1:get_zone_action choose the action based on the tables from line82ff
20121109 03:08:37< Alarantalara> it's at line 1586
20121109 03:09:17< Alarantalara> understand those two functions and everything they call and you're ready to go
20121109 03:10:38< skyfaller> ok, I'll do my best to comprehend those, thanks :)
20121109 03:11:18< Alarantalara> and if you don't like the subject, feel free to look for something else
20121109 03:11:23< mattsc> skyfaller: I'll have a look through the code and see if I can come up with something else.
20121109 03:11:39< mattsc> But for the next hour or so I am busy with other things and only come by here occasionally.
20121109 03:33:35-!- loonybot [~loonybot@wesnoth/bot/loonybot] has quit [Remote host closed the connection]
20121109 03:33:37-!- ancestral [~ancestral@65-128-234-158.mpls.qwest.net] has quit [Quit: And that’s the end of THAT chapter.]
20121109 03:54:22-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20121109 03:58:14-!- ancestral [~ancestral@65-128-234-158.mpls.qwest.net] has joined #wesnoth-umc-dev
20121109 04:05:39< Alarantalara> That's unfortunate. The AI is choosing units that are somewhat too expensive in Arcanclave citadel and so is unable to recruit enough units to capture all the villages. (It's two short, or one once I tell it to skip exactly one village the leader can reach)
20121109 04:07:07< irker431> AI-Demos: Alarantalara master * r39e6a2c / lua/generic-recruit_engine.lua : Lower variance of move scores to encourage AI no to recruit so many scouts in Arcanclave citadel - http://git.io/udpNrg
20121109 04:16:30< irker431> AI-Demos: Alarantalara master * r469df4c / lua/generic-recruit_engine.lua : Exclude one village the leader can reach from consideration - http://git.io/SzZN3Q
20121109 04:17:13< Alarantalara> Knalgans now capture every village in Arcanclave on turn 2
20121109 04:17:42< mattsc> Excellent!
20121109 04:18:55< Alarantalara> or they would if grab villages didn't put the unit that should go on one village to a different one
20121109 04:19:56< Alarantalara> I suppose I have to count units that can reach now, even though I know which unit should go there already
20121109 04:37:25< Alarantalara> skyfaller: do you know of any maps where there is a village that requires a unit that flys/swims to reach it?
20121109 04:38:08-!- csarmi [csarmi@195-38-114-53.pool.digikabel.hu] has quit [Ping timeout: 272 seconds]
20121109 04:56:20< skyfaller> Alarantalara: I can't think of any that literally require flying/swimming, but there are a number where using a land unit would take significantly longer
20121109 04:56:50< skyfaller> Alarantalara: how about that water village on Cynsaun Battlefield, for instance? Where your naga was hanging out?
20121109 04:57:23< Alarantalara> Ah yes, I'm going to have to hunt out villages like that for recruit purposes eventually
20121109 04:57:33< Alarantalara> Just to make sure that some water units are recruited for them
20121109 05:08:02< mattsc> skyfaller: I like Alarantalara's suggestion of auto-identifying zones ...
20121109 05:08:45< mattsc> It's something I definitely won't get to anytime soon.  I also think it's a conceptual problem more so than a coding problem, so you could "ease" into it that way.
20121109 05:09:25< mattsc> The attack/retreat decision is important too, of course, but that's close to what I am going to work on soon.
20121109 05:09:38< mattsc> Which doesn't mean that I can't use plenty help with that, just saying.
20121109 05:10:11< skyfaller> yeah, it's just that anything you are actually looking at you can probably code much quicker than I can :)
20121109 05:10:27< Alarantalara> goodnight all
20121109 05:10:47-!- Alarantalara [~Adium@CPEc0c1c09e8055-CM00252eac6d62.cpe.net.cable.rogers.com] has quit [Quit: Leaving.]
20121109 05:11:06< mattsc> but anything I don't have to do is time I have for something else.
20121109 05:15:48-!- ancestral [~ancestral@65-128-234-158.mpls.qwest.net] has quit [Quit: i go nstuf kthxbai]
20121109 05:16:04< skyfaller> well, I'll see what I can do
20121109 05:19:49< mattsc> If you come across something else that looks interesting, just let me know and I'll stay away from it.
20121109 05:34:10< mattsc> skyfaller: actually, there's a serious of much smaller tasks that also need to be done that might be an easier way to "get your feet wet"
20121109 05:34:39< skyfaller> mattsc: any chance you could write them up somewhere?
20121109 05:34:55< skyfaller> people other than me might be interested in them too :)
20121109 05:34:58< mattsc> hold on, don't be so impatient ;)
20121109 05:35:13< mattsc> Actually, part of the job is identifying them ...
20121109 05:35:21< skyfaller> hahahaha
20121109 05:35:38< mattsc> There's still a lot of little map-dependent bits in the code, like things where a northern position gets a slightly higher rating because (on Freelands) it's closer to the own keep, etc.
20121109 05:35:38< skyfaller> very meta ;-)
20121109 05:36:19< skyfaller> OK, I can probably identify the things that are hard-coded into Fred that will need to be generalized later, that doesn't sound ridiculously hard
20121109 05:36:21< mattsc> I am eliminating those as I come across them, but at some point we'll have to go through the code systematically and make sure we find them all.
20121109 05:36:40< mattsc> ... and change some of them.
20121109 05:37:14< skyfaller> actually doing anything to the code would probably take me a while, but finding what code needs to be changed sounds within my reach
20121109 05:37:29< mattsc> I think they range from trivial to ridiculously complicated, as far a generalizing them is concerned.
20121109 05:37:43< mattsc> You could start with the left category and move your way to the right.  :)
20121109 05:38:46< mattsc> Anyways, just something I was thinking about as I was cleaning up.  As I said, whatever sounds interesting to you, just let us know.
20121109 05:38:52< mattsc> There's plenty work for everybody!
20121109 05:39:01< skyfaller> ok :)
20121109 05:41:06-!- csarmi [csarmi@217-197-190-10.pool.digikabel.hu] has joined #wesnoth-umc-dev
20121109 05:50:14-!- csarmi [csarmi@217-197-190-10.pool.digikabel.hu] has quit []
20121109 05:53:47-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-umc-dev
20121109 05:55:52-!- noy_ [~Noy@S0106602ad0727b65.vw.shawcable.net] has joined #wesnoth-umc-dev
20121109 05:55:52-!- noy_ [~Noy@S0106602ad0727b65.vw.shawcable.net] has quit [Changing host]
20121109 05:55:52-!- noy_ [~Noy@wesnoth/developer/noy] has joined #wesnoth-umc-dev
20121109 05:55:53-!- noy [~Noy@wesnoth/developer/noy] has quit [Read error: Connection reset by peer]
20121109 05:55:54-!- noy_ is now known as noy
20121109 05:56:19-!- noy_ [~Noy@d50-98-14-63.bchsia.telus.net] has joined #wesnoth-umc-dev
20121109 05:56:19-!- noy_ [~Noy@d50-98-14-63.bchsia.telus.net] has quit [Changing host]
20121109 05:56:19-!- noy_ [~Noy@wesnoth/developer/noy] has joined #wesnoth-umc-dev
20121109 06:00:48-!- noy [~Noy@wesnoth/developer/noy] has quit [Ping timeout: 246 seconds]
20121109 06:00:49-!- noy_ is now known as noy
20121109 06:17:59-!- ancestral [~ancestral@65-128-234-158.mpls.qwest.net] has joined #wesnoth-umc-dev
20121109 06:18:07-!- ancestral [~ancestral@65-128-234-158.mpls.qwest.net] has quit [Client Quit]
20121109 06:19:04-!- shadowm_desktop [ignacio@wesnoth/developer/shadowmaster] has quit []
20121109 06:44:18-!- SeattleDad [~SeattleDa@c-67-171-51-0.hsd1.wa.comcast.net] has joined #wesnoth-umc-dev
20121109 06:57:56-!- Blueblaze [~Blueblaze@67.214.66.130] has joined #wesnoth-umc-dev
20121109 07:02:28-!- Blueblaze [~Blueblaze@67.214.66.130] has quit [Ping timeout: 252 seconds]
20121109 07:16:12-!- shadowm_desktop [ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-umc-dev
20121109 07:17:57-!- ancestral [~ancestral@65-128-234-158.mpls.qwest.net] has joined #wesnoth-umc-dev
20121109 07:49:11-!- shadowm_desktop [ignacio@wesnoth/developer/shadowmaster] has quit []
20121109 08:01:02-!- ancestral [~ancestral@65-128-234-158.mpls.qwest.net] has quit [Quit: i go nstuf kthxbai]
20121109 08:43:16-!- irker431 [~irker@ai0867.net] has quit [Quit: transmission timeout]
20121109 08:47:43-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has joined #wesnoth-umc-dev
20121109 08:48:55-!- loonybot [~loonybot@46.138.65.68] has joined #wesnoth-umc-dev
20121109 08:48:56-!- loonybot [~loonybot@46.138.65.68] has quit [Changing host]
20121109 08:48:56-!- loonybot [~loonybot@wesnoth/bot/loonybot] has joined #wesnoth-umc-dev
20121109 08:48:59-!- mode/#wesnoth-umc-dev [+v loonybot] by ChanServ
20121109 09:30:52-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has quit [Ping timeout: 252 seconds]
20121109 11:01:41-!- Crendgrim [~crend@90-137.eduroam.rwth-aachen.de] has joined #wesnoth-umc-dev
20121109 11:01:49-!- Crendgrim [~crend@90-137.eduroam.rwth-aachen.de] has quit [Client Quit]
20121109 11:11:01-!- SeattleDad [~SeattleDa@c-67-171-51-0.hsd1.wa.comcast.net] has quit [Ping timeout: 255 seconds]
20121109 11:38:33-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20121109 13:36:18-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has joined #wesnoth-umc-dev
20121109 14:15:38-!- Crendgrim [~crend@port-92-204-114-76.dynamic.qsc.de] has joined #wesnoth-umc-dev
20121109 14:21:29-!- Alarantalara [~Adium@CPEc0c1c09e8055-CM00252eac6d62.cpe.net.cable.rogers.com] has joined #wesnoth-umc-dev
20121109 14:29:22-!- faabumc [~vcr@wesnoth/developer/faabumc] has joined #wesnoth-umc-dev
20121109 15:01:23-!- mattsc [~mattsc@d154-20-32-241.bchsia.telus.net] has quit [Quit: I'm asleep]
20121109 15:11:07-!- mattsc [~mattsc@207.230.251.234] has joined #wesnoth-umc-dev
20121109 15:30:44-!- faabumc [~vcr@wesnoth/developer/faabumc] has quit [Quit: Ex-Chat]
20121109 16:08:21-!- irker303 [~irker@ai0867.net] has joined #wesnoth-umc-dev
20121109 16:08:21< irker303> AI-Demos: mattsc master * r7f18160 / (4 files in 2 dirs): Move lua/CA test scenario into AI-Demos - http://git.io/xtGQGA
20121109 16:09:00< mattsc> skyfaller: This ^ is for you, if you want to start fiddling with coding
20121109 16:12:20< mattsc> Alarantalara: my .gitignore suddenly stopped working (as in, file names in there are no ignored any more).  Any idea what could be the cause of that?
20121109 16:14:07< mattsc> Hmm, not quite true.  It still ignores _server.pbl, but it does not ignore the test scenario file and .DS_Store.  Weird...
20121109 16:19:16< mattsc> Alarantalara: ignore that I asked.  I'm just being stupid ...
20121109 16:25:20< irker303> AI-Demos: mattsc master * rc93b70e / (lua/test_lua_template.lua lua/test_lua.lua): Rename test_lua.lua to test_lua_template.lua - http://git.io/gcfOSA
20121109 16:25:21< irker303> AI-Demos: mattsc master * r7e9fb29 / .gitignore : Ignore test_lua.lua - http://git.io/mcDrQg
20121109 16:26:26-!- mattsc [~mattsc@207.230.251.234] has quit [Quit: I'm asleep]
20121109 16:44:38-!- Crendgrim [~crend@port-92-204-114-76.dynamic.qsc.de] has quit [Ping timeout: 252 seconds]
20121109 16:51:59-!- mattsc [~mattsc@fw.hia.nrc.ca] has joined #wesnoth-umc-dev
20121109 16:53:32-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-umc-dev
20121109 17:18:36-!- ancestral [~ancestral@65-128-234-158.mpls.qwest.net] has joined #wesnoth-umc-dev
20121109 18:20:25-!- ancestral [~ancestral@65-128-234-158.mpls.qwest.net] has quit [Quit: i go nstuf kthxbai]
20121109 18:25:37< skyfaller> mattsc: can you summarize what that did for me?
20121109 18:26:01< mattsc> skyfaller: sure.  can you wait 5 min?
20121109 18:26:05< skyfaller> no prob
20121109 18:31:10< mattsc> skyfaller: ok, I'm back.  It's probably easiest if I just walk you through an example.
20121109 18:31:49< mattsc> So, first, go into directory lua/ and make a copy of test_lua_template.lua
20121109 18:31:59< mattsc> Call it test_lua.lua
20121109 18:32:42< mattsc> That one's your working file and it will not get pushed to the repository, so you can mess with it as much asyou want
20121109 18:32:42< skyfaller> ok
20121109 18:33:01< mattsc> Start Wesnoth and go into the switchboard scenario in AI-Demos
20121109 18:33:39< mattsc> Then move Grnk to the snowy mountain in the north, at 9,1
20121109 18:35:47< mattsc> This should get you into the test scenario
20121109 18:37:34< skyfaller> mattsc: I'm having trouble... it should show up as a campaign, right?
20121109 18:37:52< skyfaller> this is actually my first time using AI-demos in 1.11
20121109 18:38:13< skyfaller> I just made the alias to the Github copy in the add-ons folder...
20121109 18:38:25< mattsc> assuming you have it installed yes.
20121109 18:38:35< mattsc> or you have the link set up from the 1.11 directory
20121109 18:38:41< skyfaller> well, something is wrong then
20121109 18:38:45< skyfaller> I'm stumped
20121109 18:38:53< mattsc> but everything we're doing right now will work in 1.10 also, I think
20121109 18:39:10< skyfaller> ok, but I want to figure this out
20121109 18:39:43< skyfaller> if I've made the alias to the github directory from the add-ons directory but it is not showing up as a campaign, what could be wrong?
20121109 18:39:56< skyfaller> ... it is not showing up as an add-on I can remove, either
20121109 18:40:10< mattsc> basic question, from which directory did you make the link?
20121109 18:40:27< mattsc> did you restart wesnoth since you have done that?
20121109 18:40:54< skyfaller>  /Users/nelson/Library/Application Support/Wesnoth_1.11/data/add-ons
20121109 18:41:23< skyfaller> yes, I restarted Wesnoth
20121109 18:41:57< skyfaller> I see the add-on in 1.10.4, but not 1.11.0
20121109 18:42:15< mattsc> If you go into that directory and type
20121109 18:42:37< mattsc> ls -l AI-demos/_main.cfg
20121109 18:42:46< mattsc> do you see the file?
20121109 18:43:21< mattsc> or use 'less AI-demos/_main.cfg' 
20121109 18:43:54< skyfaller> oh, I think I figured out the problem, hold on
20121109 18:46:28< skyfaller> mattsc: I was an idiot... because when you use the Finder to look at a directory with a symbolic link in it, it displays an "alias", I thought that creating an alias through the Finder would also make a symbolic link
20121109 18:46:37< skyfaller> I was quite wrong
20121109 18:47:02< skyfaller> now everything works as expected
20121109 18:47:12< mattsc> ok, good.
20121109 18:47:16< mattsc> So move Grnk to snowy mountain
20121109 18:47:50< skyfaller> ok
20121109 18:47:51< skyfaller> done
20121109 18:48:04< skyfaller> I have two archers facing off
20121109 18:48:08< mattsc> ok, so you got into the test scenario?
20121109 18:48:13< mattsc> ok, good.
20121109 18:48:30< mattsc> Now, shut down Wesnoth, and open file test_lua.lua
20121109 18:48:38< skyfaller> ok
20121109 18:48:49< mattsc> At the beginning, there's a command-line syntax there
20121109 18:49:04< mattsc> type that in iTerm or whatever your preferred terminal is
20121109 18:49:41< skyfaller> wait, I don't have that .gz file
20121109 18:50:04< skyfaller> do I have to create that .gz file myself?
20121109 18:50:04< mattsc> getting into the test scenario should have created that
20121109 18:50:22< mattsc> It's the beginning of scenario savefile
20121109 18:51:21< skyfaller> mattsc: where exactly should the file be located?
20121109 18:51:46< mattsc> question back first, did you try the command and get an error message?
20121109 18:52:05< skyfaller> yes
20121109 18:52:12< mattsc> it is create in /Users/nelson/Library/Application Support/Wesnoth_1.11/saves
20121109 18:52:22< skyfaller> https://gist.github.com/4cdc6449101d6dd43772
20121109 18:52:31< mattsc> Do you have beginning of turn savefiles turned off in the preferences?
20121109 18:53:03< skyfaller> where is that in preferences?
20121109 18:53:20< mattsc> I'm looking...  I think there is an option like that
20121109 18:54:33< mattsc> oh, another option
20121109 18:54:52< mattsc> go into that saves directory please and type 'ls -lrt'
20121109 18:55:41< mattsc> Do you see anything like AI-demos-test* at the end of the list?
20121109 18:56:53< skyfaller> oh yes, I do
20121109 18:56:59< skyfaller> I found the file there
20121109 18:57:22< skyfaller>  /Users/nelson/Library/Application Support/Wesnoth_1.11/saves/AI-demos-test.gz
20121109 18:57:29< mattsc> In my case it looks like this: http://pastebin.com/hPhKARuf
20121109 18:57:45< skyfaller> yes, that is what mine looks like too
20121109 18:58:37< mattsc> ok, then the command should work...
20121109 18:58:41< mattsc> oh, oh ...
20121109 18:58:56< mattsc> oh, no.  
20121109 18:59:02< mattsc> :)
20121109 18:59:06< skyfaller> ?
20121109 18:59:16< mattsc> I thought I had figured it out, but I was wrong
20121109 18:59:43< mattsc> could you copy here what you're using to start at the command line?
20121109 19:00:11< mattsc> oh, actually, I have figured it out now...
20121109 19:00:24< mattsc> Duh!
20121109 19:00:45< mattsc> Your error message does not mean that wesnoth cannot find the file, but that iTerm cannot find the Wesnoth Application file.
20121109 19:00:58< skyfaller> oh!
20121109 19:00:59< skyfaller> ha
20121109 19:01:00< skyfaller> sorry
20121109 19:02:01< mattsc> Most likely your folder is Applications/Wesnoth-1.11.0.app or something like that
20121109 19:02:13< skyfaller> I actually moved the app to a stupid location
20121109 19:02:21< skyfaller> ok, everything works as expected now
20121109 19:02:22< mattsc> No, my fault actually, I rename my folders so that they are the same for each subversion
20121109 19:02:26< mattsc> I should have remembered that
20121109 19:02:48< mattsc> ok, cool
20121109 19:02:52< skyfaller> test: No objectives available
20121109 19:03:05< mattsc> So you're back in the test scenario...
20121109 19:03:10< skyfaller> yes
20121109 19:03:19< mattsc> Just to test if things are working, try the riht-click options
20121109 19:03:31< mattsc> And do a "relaod lua code"
20121109 19:03:48< mattsc> or something like that - assuming I could type correctly ...
20121109 19:03:48< skyfaller> ok, I reloaded it and it threw some gibberish on the screen
20121109 19:04:22< mattsc> It's not gibberish, it's valuable debug information  :D
20121109 19:04:27< skyfaller> :)
20121109 19:04:44< mattsc> It should also have thrown up some additional information in the terminal window
20121109 19:05:08< mattsc> In particular, some rating information at the end
20121109 19:05:18< skyfaller> oh yes
20121109 19:05:34< skyfaller> and you can copy and paste from the terminal, unlike wesnoth :)
20121109 19:05:46< mattsc> e.g
20121109 19:06:04< mattsc> Also, the wesnoth screen does not scroll, so if it's bigger than a screen, you cannot see the end of it
20121109 19:06:42< mattsc> So, yes, that's why I am doing this from the terminal (you could just as well start the savegame from the load dialog, but then you don't have the terminal information)
20121109 19:07:23< mattsc> Ok, so I'd suggest that you now have a look at test_lua.lua and figure out what's going on herre.
20121109 19:07:34< mattsc> Or I can walk you through, whichever you prefer
20121109 19:08:17< skyfaller> can you give me an executive summary?
20121109 19:08:28< mattsc> fyi, the right-click option is set up in the scenario file(scenarios/test.cfg) and it executes the content of test_lua.lua
20121109 19:08:45< mattsc> Sure, you have the file in front of you?
20121109 19:10:31< mattsc> test_lua.lua, I mean, nt the scenario file
20121109 19:15:14< skyfaller> mattsc: yes
20121109 19:15:35< mattsc> skyfaller: ok.  quick walkthrough?
20121109 19:15:41< skyfaller> sure
20121109 19:16:13< mattsc> ok, first we set up all the libraries, then Fred's functions are loaded into my_ai, then we clear the screen
20121109 19:16:34< mattsc> You don't need to touch any of that, except if you want to test a different AI
20121109 19:17:04< mattsc> After that, there are two different testing mechanisms set up.
20121109 19:17:35< mattsc> If test_CA is set to true, we test a CA's eval/exec.  zone_control in this case.
20121109 19:18:06< mattsc> Let's skip that for now.  What this is set up to do at the moment is what starts at line 45 (manual testing)
20121109 19:18:20< mattsc> ok, so far?
20121109 19:18:34< skyfaller> ok
20121109 19:19:02< mattsc> So, when you click on that option, it executes the code starting at line 46
20121109 19:19:26< mattsc> It loads the first of the AI's units into attacker, the first of the enemy units into defender
20121109 19:19:37< mattsc> Since there is only one unit on each side, these are the two archers
20121109 19:19:58< skyfaller> "click on that option"?
20121109 19:20:12< mattsc> When you click on 'reload lua code' in the scenario
20121109 19:20:17< skyfaller> ah, ok
20121109 19:20:19< mattsc> sorry, being lazy  :)
20121109 19:21:12< mattsc> Then it sets up a bunch of parameters.  Again, let's skip that for now.  Go to line 60 instead.
20121109 19:21:18< skyfaller> k
20121109 19:21:33< mattsc> we call BC.battle_outcome there.
20121109 19:21:47< mattsc> BC is defined at the beginning of the file, it's the battle_calcs library
20121109 19:22:13< mattsc> It's inside a loop for speed testing, that's what all the os.clock()'s are for
20121109 19:22:31< mattsc> currently there's only one repetition, as I don't do speed testing right now.
20121109 19:22:56< mattsc> But you could set 'i=1,10000' to see how long the execution of BC.battle_outcome takes
20121109 19:22:59< skyfaller> (I really like the phrase "battle calculations" for some reason... it's like, I put on my battle armor, then I whip out my battle calculator...)
20121109 19:23:22< mattsc> It's actually built into the forearm armor!
20121109 19:23:26< skyfaller> :)
20121109 19:23:51< mattsc> So, DBG.dbms produces the gibberish on the screen
20121109 19:24:09< mattsc> comment out one of the two DBG.dbms commands and right-click again
20121109 19:24:30< mattsc> See the difference?
20121109 19:24:44< skyfaller> I guess it would technically be "gobbledygook" according to Wikipedia, since it has a meaning but it is jargon / difficult to understand
20121109 19:24:54< skyfaller> gibberish is meaningless
20121109 19:25:13< mattsc> I see.  Learned something today.  :)
20121109 19:25:22-!- irker303 [~irker@ai0867.net] has quit [Quit: transmission timeout]
20121109 19:25:52< mattsc> So, for example, DBG.dbms(att_stats) shows the structure of the att_stats table
20121109 19:26:14< mattsc> which is one of the return values of battle_outcome()
20121109 19:26:20< skyfaller> one sec
20121109 19:26:28< skyfaller> I am commenting out the thing
20121109 19:26:36< mattsc> wesnoth.message() displays messages on the wesnoth screen
20121109 19:26:53< mattsc> print() puts messages into the terminal.
20121109 19:27:14< mattsc> I used a mix of the three here to demonstrate how to do each of them.
20121109 19:28:04< skyfaller> I see
20121109 19:28:30< skyfaller> "mix of the three"? I see wenoth.message and print, that's two
20121109 19:28:38< mattsc> For myself, I usually print things to the terminal, that way I can scroll, copy-and-paste etc.
20121109 19:28:42< mattsc> DBG.dbms is the third
20121109 19:28:49< skyfaller> oh
20121109 19:28:52< skyfaller> got it :)
20121109 19:29:02< mattsc> which displays both on the screen and the terminal, although it has options to change that
20121109 19:29:33< skyfaller> this is really cool that I can see my changes to that lua file in the game right away
20121109 19:30:03< mattsc> Ok, final step, the rating's you see in the terminal are the two different attack ratings at the end of the file.
20121109 19:30:03< mattsc> yes, it is
20121109 19:30:03< mattsc> It's one of the huge advantages of lua over wml
20121109 19:30:17< mattsc> No need to restart or clear the cache or whatever
20121109 19:30:48< skyfaller> OK, I see the ratings
20121109 19:31:25< mattsc> The first is by specifying the attack weapons (weapon #1)
20121109 19:31:27< mattsc> The second is by letting attack_rating choose its own weapons
20121109 19:31:39< mattsc> which, for orcish archers, obviously is not the first weapon
20121109 19:31:49< skyfaller> indeed
20121109 19:32:09< mattsc> However, the rating's actualy worst for best weapon in this case...
20121109 19:32:37< mattsc> That's because we're attacking an archer on hills from grassland.
20121109 19:33:02< skyfaller> ??
20121109 19:33:54< mattsc> Actually, let's skip that for a moment (it's more complicated than that)
20121109 19:34:07< skyfaller> k
20121109 19:34:12< mattsc> Why don't you change the HP of the Side 2 archer to, say, 8
20121109 19:34:21< mattsc> Remember how to do that?
20121109 19:35:20< skyfaller> mattsc: apparently yes
20121109 19:35:43< mattsc> cool.  right-click again and see the rating change
20121109 19:36:10< skyfaller> Best weapon looks a lot better now?
20121109 19:36:16< mattsc> Yes.
20121109 19:36:27< mattsc> That's because we have a much better CTK with that weapon now.
20121109 19:36:46< mattsc> Another thing to test, change the enemy archer to a grunt
20121109 19:36:59< mattsc> full-HP grunt
20121109 19:37:17< skyfaller> ok
20121109 19:37:54< mattsc> Hmm, and there things crash ...  :(
20121109 19:38:02< skyfaller> ? :(
20121109 19:38:21< mattsc> oh, i know why ...
20121109 19:39:00< mattsc> actually, since we don't need this right now, just comment out line 61
20121109 19:39:15< mattsc> and 63 and 64
20121109 19:39:33< skyfaller> ok
20121109 19:39:55< mattsc> Now if you right-click, you also get a much better score with the best weapon than with weapon 1, right?
20121109 19:39:57< skyfaller> yes
20121109 19:40:18< mattsc> that's because an archer should not attack a grunt with his dagger
20121109 19:40:41< skyfaller> yes 
20121109 19:40:43< skyfaller> :)
20121109 19:41:00< mattsc> Anyways, all of this is just playing around to show you how you can use the test scenario
20121109 19:41:18< mattsc> Do you want to make a useful contribution to the code?
20121109 19:41:27< skyfaller> maybe :)
20121109 19:41:36< mattsc> It's something I've been too lazy for so far, but it'll have to be done.
20121109 19:41:56< mattsc> And yes, I could probably do it in a few minutes myself, but you need to get started somewhere (if you want to, that is)
20121109 19:42:19-!- irker751 [~irker@ai0867.net] has joined #wesnoth-umc-dev
20121109 19:42:20< irker751> AI-Demos: Alarantalara master * rb24a66f / lua/generic-recruit_engine.lua : Set a max gold cost for recruited units and find shortest route via bat instead of straight line distances - http://git.io/EBCDVw
20121109 19:42:32< mattsc> so open file lua/battle_calcs.lua
20121109 19:43:06< skyfaller> ok
20121109 19:43:16< mattsc> line 867
20121109 19:43:37< mattsc> wow, really?  I only started this file a week ago and it already has over 1000 lines!
20121109 19:43:50< skyfaller> yes, that was what I was complaining about the other day ;-)
20121109 19:43:56< mattsc> :)
20121109 19:44:17< mattsc> So this line is inside the attack rating function
20121109 19:44:46< mattsc> why dont you comment out line 879, and right click on your scenario again
20121109 19:45:37< skyfaller> um, I don't have anything on 879
20121109 19:45:42< mattsc> it should now display the defender_level_penalty in the terminal (the first number in that line)
20121109 19:45:47< skyfaller> wait
20121109 19:45:53< skyfaller> something is wrong hold on
20121109 19:45:53< mattsc>     --print('  defender level penalty:', defender_level_penalty, value_fraction)
20121109 19:46:04< skyfaller> ok, yes
20121109 19:46:58< mattsc> and, actually, go up to line 810 and uncomment printing of the level bonus there also
20121109 19:47:12< skyfaller> 879 was already commented out
20121109 19:47:52< skyfaller> did you mean un-comment it?
20121109 19:48:00< mattsc> oops, sorry, I meant uncomment it ...
20121109 19:48:10< mattsc> yes
20121109 19:48:38< skyfaller> ok
20121109 19:48:45< mattsc> So you have two lines of extra output per rating in your terminal now, right?
20121109 19:49:00< skyfaller> yes
20121109 19:49:11< mattsc> and the first number is 0 in each case?
20121109 19:49:13< irker751> AI-Demos: Alarantalara master * r69f7445 / lua/generic-recruit_engine.lua : Only need one test bat - http://git.io/qz_Pag
20121109 19:49:14< irker751> AI-Demos: Alarantalara master * r130d679 / lua/generic-recruit_engine.lua : Revert movement score adjustment - http://git.io/eL-Jrw
20121109 19:49:23< skyfaller> yes
20121109 19:49:31< mattsc> Alarantalara's busy!
20121109 19:49:37< skyfaller> yes :)
20121109 19:49:39-!- shadowm_desktop [ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-umc-dev
20121109 19:49:52< mattsc> Ok, give the enemy grunt 1 HP, the run the code again
20121109 19:50:51< skyfaller> ok
20121109 19:51:08-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20121109 19:51:18< mattsc> doesn't do anything different, of course.  grrr.
20121109 19:51:29< mattsc> Also give the archer 1 XP less that max
20121109 19:51:42< skyfaller> I got different results...
20121109 19:52:02< skyfaller> https://gist.github.com/553f791a090a18443646
20121109 19:52:35< mattsc> yes, sorry, I meant the level bonus is still at 0
20121109 19:53:39< mattsc> so, if you give the archer 23/24 XP (or whatever one less than total is for yours), level bonus should go up to 1, right?
20121109 19:55:55< skyfaller> nope
20121109 19:56:14< skyfaller> oh
20121109 19:56:16< skyfaller> sorry
20121109 19:57:10< skyfaller> yes, it does
20121109 19:57:28< mattsc> ok, one more "exercise" (or two)
20121109 19:57:45< mattsc> make it archer vs. archer again
20121109 19:58:05< skyfaller> k
20121109 19:58:15< mattsc> give both archers 1 hp
20121109 19:58:48< mattsc> what do you get for level bonus now?
20121109 19:58:53< skyfaller> and still one with almost full XP?
20121109 19:59:00< mattsc> yes
20121109 19:59:31< skyfaller> https://gist.github.com/d46c2b91131343e0c653
20121109 20:00:26< mattsc> ok, so look at line 803
20121109 20:01:03< mattsc> The "level bonus" is defined as the chance of leveling up (which is one) times the chance of not dying
20121109 20:01:39< skyfaller> that makes sense... I think
20121109 20:02:05< skyfaller> what is the number that comes after the level bonus?
20121109 20:02:36< skyfaller> is that the value_fraction ?
20121109 20:02:44< mattsc> Line 806 does the same, just that in that case the leveling up chance is proportional to the CTK on the enemy
20121109 20:02:57< skyfaller> yes, it must be the value fraction
20121109 20:03:08< mattsc> That's the added up total rating, whereas the first number is just the rating for potentially leveling up
20121109 20:03:28-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-umc-dev
20121109 20:04:04< skyfaller> ok
20121109 20:04:06< mattsc> I can justify why I have chosen all the rating functions like that, but I don't have time for that right now.  I need to leave in ~15 min.
20121109 20:04:13< skyfaller> ok, np :)
20121109 20:04:23< mattsc> So, do you want to know what you could do?
20121109 20:05:23< mattsc> I think it's essentially a one-line thing, but then it needs to be tested for all kinds of combinations of units ...
20121109 20:05:29< skyfaller> sure, although I won't make any promises
20121109 20:06:11< mattsc> Well, it's because you've repeatedly voiced interest in getting into the coding...
20121109 20:06:49< mattsc> So, the conditions in line 802 and 805 won't work for L0 enemies
20121109 20:07:27< mattsc> And the same goes for 871/874 the other way around
20121109 20:08:13< irker751> AI-Demos: Alarantalara master * rddb4330 / (lua/ai_helper.lua lua/generic-recruit_engine.lua): Factor function finding the cost of the cheapest unit into ai_helper - http://git.io/MMFrMg
20121109 20:08:14< irker751> AI-Demos: Alarantalara master * r8ba7490 / lua/generic-rush_engine.lua : Don't worry about extra time to find village if it's only one turn and we can't recruit again this turn - http://git.io/IJYqLA
20121109 20:08:15< mattsc> The exercise: make it work with L0's as well, test using the test scenarios for L0s (and that it still works with L1s and L2s), and commit  :D
20121109 20:08:17< skyfaller> I won't make any promises b/c I have still yet to write a single line of lua code outside of a tutorial :P
20121109 20:08:42< mattsc> But the nice thing with this test scenario is that you can write something, and then test it directly
20121109 20:09:00< mattsc> (rather than having to figure out during gamplay if it does what you want)
20121109 20:09:04< skyfaller> oh, this does sound relatively straightforward
20121109 20:09:06< mattsc> gameplay
20121109 20:09:12< mattsc> indeed
20121109 20:09:48< mattsc> And yes, explaining this to you took me longer than doing this bit myself.  It's the long-term payoff we're looking at here.  :)
20121109 20:09:56< skyfaller> :)
20121109 20:10:15< mattsc> Anyway, interested?
20121109 20:10:37< skyfaller> yeah, sure, this seems like something I should be able to do, although it might take me a while
20121109 20:11:09< mattsc> no worries.  I have other things to do and won't touch this part until you're done.
20121109 20:11:33< mattsc> Right now I have a meeting with a nice person who wants to drill holes into my head.
20121109 20:12:00< skyfaller> ah, I just had a dentist appointment myself
20121109 20:12:09< mattsc> :)
20121109 20:12:16< skyfaller> unless you have a trepanning appointment
20121109 20:12:31< mattsc> nope.  just dentist
20121109 20:13:19< mattsc> So, one more thing, I don't know if you saw that:
20121109 20:13:35< mattsc> test_lua.cfg is not synced with the repository
20121109 20:13:52< mattsc> only test_lua_template.lua is.
20121109 20:14:08< mattsc> So you can mess with the file as much as you want, without changing the original
20121109 20:14:21< mattsc> And if you mess it up too badly, you just copy it over again.  :)
20121109 20:14:21< skyfaller> do you mean test.lua.lua is not synced?
20121109 20:14:39< mattsc> Argh!  yes.
20121109 20:17:30< mattsc> Ok, so I'm off now.  Hope you'll find this useful ...
20121109 20:17:38< skyfaller> ok, have fun!
20121109 20:17:50< mattsc> I'll try
20121109 20:20:06-!- mattsc [~mattsc@fw.hia.nrc.ca] has quit [Quit: bye]
20121109 20:31:31< irker751> AI-Demos: Alarantalara master * r666622b / lua/generic-recruit_engine.lua : Require units fast enough to reach target villages - http://git.io/ziAdaQ
20121109 20:40:22< irker751> AI-Demos: Alarantalara master * r268078d / lua/generic-recruit_engine.lua : Recruit near close enemies, not just the leaders - http://git.io/CFF2qw
20121109 20:52:44< irker751> AI-Demos: Alarantalara master * r24de51a / CHANGELOG : Update changelog - http://git.io/ZxcQrQ
20121109 21:22:47-!- mnewton1 [~mnewton1@c-76-99-170-226.hsd1.de.comcast.net] has joined #wesnoth-umc-dev
20121109 21:37:57< Alarantalara> Hah! I now have a 100% victory rate over the RCA AI in Arcanclave Citadel
20121109 22:13:47-!- loonybot [~loonybot@wesnoth/bot/loonybot] has quit [Remote host closed the connection]
20121109 22:17:36-!- shadowm_desktop [ignacio@wesnoth/developer/shadowmaster] has quit [Ping timeout: 265 seconds]
20121109 22:19:17-!- shadowm_desktop [ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-umc-dev
20121109 22:28:27-!- shadowm_desktop [ignacio@wesnoth/developer/shadowmaster] has quit [Read error: Connection reset by peer]
20121109 22:31:10< skyfaller> and by you, you mean Ron? :)
20121109 22:31:50-!- shadowm_desktop [ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-umc-dev
20121109 22:32:40< Alarantalara> I mean Ron's recruiter without any other part of Ron
20121109 22:32:58< skyfaller> hey, that's pretty good
20121109 22:33:03< skyfaller> recruitment matters!
20121109 22:34:35< Alarantalara> Yes, the recruiter manages to place units that will capture every village on its side on turn 2 in almost every case
20121109 22:35:00< Alarantalara> Exceptions are occaisionally recruiting slow goblins to capture villages 5 hexes away
20121109 22:35:13< Alarantalara> and drakes not having enough gold to get 12 units
20121109 22:36:15< Alarantalara> Drakes require a minimum of 180 gold to do it and only 175 is available
20121109 22:53:27-!- mattsc [~mattsc@d154-20-32-241.bchsia.telus.net] has joined #wesnoth-umc-dev
20121109 23:01:24-!- loonybot [~loonybot@46.138.65.68] has joined #wesnoth-umc-dev
20121109 23:01:25-!- loonybot [~loonybot@46.138.65.68] has quit [Changing host]
20121109 23:01:25-!- loonybot [~loonybot@wesnoth/bot/loonybot] has joined #wesnoth-umc-dev
20121109 23:01:27-!- mode/#wesnoth-umc-dev [+v loonybot] by ChanServ
20121109 23:10:12-!- shadowm_desktop [ignacio@wesnoth/developer/shadowmaster] has quit [Quit: system reboot in progress]
20121109 23:11:52-!- shadowm_desktop [ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-umc-dev
--- Log closed Sat Nov 10 00:00:37 2012