--- Log opened Tue Oct 16 00:00:51 2012
20121016 00:45:52-!- mattsc [~mattsc@s142-179-107-97.bc.hsia.telus.net] has joined #wesnoth-umc-dev
20121016 01:15:45< mattsc> Hi, Alarantalara 
20121016 01:31:59-!- mattsc_ [~mattsc@s142-179-109-102.bc.hsia.telus.net] has joined #wesnoth-umc-dev
20121016 01:34:19-!- mattsc [~mattsc@s142-179-107-97.bc.hsia.telus.net] has quit [Ping timeout: 260 seconds]
20121016 01:34:20-!- mattsc_ is now known as mattsc
20121016 02:40:21-!- vultraz [~chatzilla@124.109.10.167] has joined #wesnoth-umc-dev
20121016 02:41:03-!- vultraz [~chatzilla@124.109.10.167] has quit [Changing host]
20121016 02:41:03-!- vultraz [~chatzilla@unaffiliated/vultraz] has joined #wesnoth-umc-dev
20121016 02:54:27-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20121016 03:06:23-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-umc-dev
20121016 03:16:15< Alarantalara> hello, mattsc
20121016 03:17:03< mattsc> Hey.
20121016 03:17:35< mattsc> Did you mean the goblin-recruit-area commit to be for the 1.10 branch?
20121016 03:17:40< Alarantalara> It appears you asked for me about 2 hours ago? I was going for a walk and left my computer on for experiments
20121016 03:17:56< mattsc> Walks are good!
20121016 03:18:24< mattsc> Yeah, I'm at the Vancouver airport with a lot of time ...
20121016 03:19:49< mattsc> As for the commit I asked about there, it doesn't really matter anyway if you look at the wesnoth.current.turn line.
20121016 03:20:25< mattsc> That was just my way of disabling goblin recruiting for the time being without actually deleting the code
20121016 03:22:06-!- mattsc_ [~mattsc@s142-179-107-97.bc.hsia.telus.net] has joined #wesnoth-umc-dev
20121016 03:23:41< Alarantalara> I only see the message "Hi", so I have no idea what commit you're referring to
20121016 03:24:12< mattsc_> The one I sent just after you said "hello, mattsc" and my "hey"
20121016 03:24:32< Alarantalara> ah, yes
20121016 03:24:47< Alarantalara> I realized that
20121016 03:24:58< mattsc_> Anyways, that's not qhat I pinged you about 2h ago...
20121016 03:25:06-!- mattsc [~mattsc@s142-179-109-102.bc.hsia.telus.net] has quit [Ping timeout: 245 seconds]
20121016 03:25:06-!- mattsc_ is now known as mattsc
20121016 03:25:06< Alarantalara> An considered changing the turn back as well
20121016 03:25:46< mattsc> Ok, that's up to you as you're doing the recruiting.  I just disabled it because the specific purpose for which the goblin was recruited does not exist in the code any more.
20121016 03:26:11< mattsc> As for my other question: We had been talking about doing high-variance attacks first, and then re-evaluating.
20121016 03:26:21< Alarantalara> yes...
20121016 03:26:36-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20121016 03:26:42< mattsc> Let's assume there's a 4-unit attack combo.  There are two ways to re-evaulate after the first unit attacks:
20121016 03:26:59< mattsc> 1. re-evaluate which unit if the other 3 goes next
20121016 03:27:30< mattsc> 2. re-evaluate all possible attack combos and decide if we maybe go for a different combo, given the outcome of that first attack
20121016 03:27:45< mattsc> Which one did you have in mind?
20121016 03:27:48-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-umc-dev
20121016 03:28:30< Alarantalara> The second, though if we can get the results of the attack and get better than average performance, then 1 would be enough
20121016 03:29:03< mattsc> Ok.  One of the commits I made since then uses the second method.
20121016 03:29:12< Alarantalara> It's more revaluate the current attack, and if the CTK is now too low, try another
20121016 03:29:33-!- noy [~Noy@wesnoth/developer/noy] has quit [Client Quit]
20121016 03:29:45< mattsc> Right, that is essentially what's going to happen when you do it that way.
20121016 03:30:12< mattsc> Ok, I had two other things that are essentially comments on our previous conversation.
20121016 03:30:51< Alarantalara> It's mostly just not evaluating any other combinations if the current one is still on schedule
20121016 03:31:37< mattsc> Well, that could be done for computation speed reason, I guess.
20121016 03:31:50< mattsc> As for the other things, I found the part where I called recruiting from inside another CA.  It was in the now-deleted grab_villages CA.
20121016 03:32:07< mattsc> Hold on, looking up the commit ...
20121016 03:33:15< mattsc> https://github.com/mattsc/Wesnoth-AI-Demos/commit/a6034d68750876619b9e25ef8a40e2585e8ab781
20121016 03:33:34< mattsc> Search for 'recruit_loop' in that page
20121016 03:34:20< mattsc> So that's an easy way how that can be done (I'm not saying it's necessarily desirable, that depends, but it's easy)
20121016 03:34:40< mattsc> ... and it's even easier with the new generalized zone_control CA.
20121016 03:34:47< Alarantalara> So I see
20121016 03:35:41-!- Alarantalara [~Adium@173.33.158.188] has left #wesnoth-umc-dev []
20121016 03:35:59< mattsc> Secondly, if you want to see what I am imagining for doing my own CA controller, have a look at grunt_rush_FLS1:get_zone_action()
20121016 03:36:02-!- Alarantalara [~Adium@173.33.158.188] has joined #wesnoth-umc-dev
20121016 03:36:22< mattsc> (just sent you a message while you were out)
20121016 03:36:52< Alarantalara> I'll take a look at get_zone_action
20121016 03:36:59< mattsc> The second part of that function, after "This ends the common initialization for all zone actions"
20121016 03:37:49< mattsc> It's really just doing what the C++ CA controller does, but doing it myself so that I have more flexibility along the lines we discussed a couple days ago.
20121016 03:38:09< mattsc> Currently, this function does nothing that couldn't be done easily with normal CAs, but that will come soon.
20121016 03:43:45< mattsc> I have to be off to get on a plane in a few minutes.  Talk to you later.
20121016 03:47:27-!- mattsc [~mattsc@s142-179-107-97.bc.hsia.telus.net] has quit [Quit: bye]
20121016 04:03:32-!- loonybot [~loonybot@wesnoth/bot/loonybot] has quit [Remote host closed the connection]
20121016 08:59:10-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has joined #wesnoth-umc-dev
20121016 09:38:45-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-umc-dev
20121016 11:25:57-!- skyfaller [~skyfaller@wikipedia/Skyfaller] has quit [Remote host closed the connection]
20121016 12:35:05-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20121016 13:33:23-!- loonybot [~loonybot@ppp94-29-12-11.pppoe.spdop.ru] has joined #wesnoth-umc-dev
20121016 13:33:23-!- loonybot [~loonybot@ppp94-29-12-11.pppoe.spdop.ru] has quit [Changing host]
20121016 13:33:23-!- loonybot [~loonybot@wesnoth/bot/loonybot] has joined #wesnoth-umc-dev
20121016 13:33:23-!- mode/#wesnoth-umc-dev [+v loonybot] by ChanServ
20121016 17:01:48-!- mattsc [~mattsc@DHCP-211-241.caltech.edu] has joined #wesnoth-umc-dev
20121016 18:11:27-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-umc-dev
20121016 18:27:30-!- skyfaller [~skyfaller@wikipedia/Skyfaller] has joined #wesnoth-umc-dev
20121016 19:32:49< Alarantalara> mattsc, I'm wondering if it would be a good idea to release a new version. SeattleDad is expressing a preference for training against released versions rather than changing snapshots, and 0.11.3 Ron doesn't like to use rush units to capture villages which would make them appear worse — which seems unfair for training purposes.
20121016 19:34:44< mattsc> Alarantalara: I just saw his PM.  Just let me know when you think the recruiting algorithms are ready for release and I'll do that.
20121016 19:35:40< mattsc> What I am doing with Fred doesn't change the final behavior much (neither for better nor for worse), so I can release at (almost) any time.
20121016 19:37:38< Alarantalara> I haven't actually changed recruiting much. It's really only that one change to allow grunts to take villages that I'm concerned about here. I've been spending my time looking at improving leader behaviour rather than recruiting recently.
20121016 19:38:39< Alarantalara> That said, it should be better than the 0.11.3 version, even if loyalists, northerners and undead are pretty terrible at the moment
20121016 19:40:49< mattsc> Well, it looks like your doing better and more things than I ...
20121016 19:41:24< mattsc> If you could just check that the changelog is up-to-date cc. your changes, I'll push out 0.11.4.
20121016 19:41:45< Alarantalara> Not really, I just have the benefit of doing things with visible effects at the moment. You have the much less exciting refactoring which will lead to awesomeness.
20121016 19:42:29< mattsc> Well, we'll have to see about that, but thanks anyway.  :)
20121016 19:42:41< mattsc> I just did my part of the changelog, so you might want to sync first.
20121016 19:44:15< Alarantalara> I've already committed all changes I wanted to me
20121016 19:44:18< Alarantalara> *make
20121016 19:44:29< Alarantalara> including those to the changelog
20121016 19:45:01< mattsc> Ok, cool.  I'll upload right away then.
20121016 19:46:31< Alarantalara> I don't see any update from you concerning the changelog though.
20121016 19:47:29< mattsc> No, we did it at the same time (to within a minute) and I am trying to figure out how to deal with this in github for Mac...
20121016 19:47:45< Alarantalara> Open the file, edit as desired and save
20121016 19:48:18< Alarantalara> Once all the HEAD===Mine=== lines are gone, it will let you commit
20121016 19:52:15< mattsc> Yeah, I made kind of a mess of that ...  But I got it all fixed now, I think.
20121016 19:55:44< mattsc> Ok, v0.11.4 is up
20121016 19:56:07< Alarantalara> Excellent, thank you
20121016 20:00:02-!- mattsc_ [~mattsc@BeaverNet-166.caltech.edu] has joined #wesnoth-umc-dev
20121016 20:04:08-!- mattsc [~mattsc@DHCP-211-241.caltech.edu] has quit [Ping timeout: 260 seconds]
20121016 20:04:08-!- mattsc_ is now known as mattsc
20121016 20:06:02-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20121016 20:30:25-!- mattsc [~mattsc@BeaverNet-166.caltech.edu] has quit [Quit: bye]
20121016 22:06:30-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-umc-dev
20121016 22:27:16-!- irker679 [~irker@ai0867.net] has joined #wesnoth-umc-dev
20121016 22:27:16< irker679> wesnoth-umc-dev: shikadilord * r16079 /trunk/After_the_Storm/mainline-strings.cfg:
20121016 22:27:17< irker679> AtS: add proxy macros for mainline unit type names
20121016 22:27:32< irker679> wesnoth-umc-dev: shikadilord * r16080 /trunk/After_the_Storm/TODO:
20121016 22:27:32< irker679> AtS: update TODO
20121016 22:27:47< irker679> wesnoth-umc-dev: shikadilord * r16081 /trunk/After_the_Storm/BUGS:
20121016 22:27:47< irker679> AtS: update contact information
20121016 23:32:56-!- vultraz [~chatzilla@unaffiliated/vultraz] has quit [Read error: Connection reset by peer]
20121016 23:41:24-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has quit [Ping timeout: 260 seconds]
20121016 23:46:06-!- shadowm_laptop [ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-umc-dev
20121016 23:56:16-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
--- Log closed Wed Oct 17 00:00:07 2012