00:00:06 -!- sorear has quit [Read error: Connection reset by peer] 00:00:29 -!- sorear has joined ##crawl-dev 00:02:24 My steel bolts are mysteriously absent on the list of items to quiver. (https://crawl.develz.org/mantis/view.php?id=1154) by elliptic 00:14:51 -!- Twinge_ has quit [Read error: Connection reset by peer] 00:15:27 -!- Twinge has joined ##crawl-dev 00:20:12 -!- stabwound has quit [Ping timeout: 258 seconds] 00:24:04 -!- stabwound has joined ##crawl-dev 00:27:32 Traps and Doors worthwhile? (https://crawl.develz.org/mantis/view.php?id=1155) by dtsund 00:45:30 -!- ixtli has quit [*.net *.split] 00:45:30 -!- ivan has quit [*.net *.split] 00:52:40 -!- ixtli has joined ##crawl-dev 00:52:40 -!- ivan has joined ##crawl-dev 01:21:49 I'm playing 0.7 on the new server, and sometimes the screen isn't redrawing itself. 01:23:18 "sometimes"? 01:23:53 I can't find any pattern for when it doesn't 01:25:29 But it happens fairly frequently 01:30:48 -!- ortoslon has joined ##crawl-dev 01:32:32 I think it only happens when I autoexplore, but not every time I autoexplore :/ 01:42:48 -!- Zaba has quit [Ping timeout: 240 seconds] 01:46:50 -!- Zaba has joined ##crawl-dev 01:52:24 -!- Madtrixr has joined ##crawl-dev 01:58:02 overcautious inaccessibility (https://crawl.develz.org/mantis/view.php?id=1156) by nubinia 02:19:23 -!- murphy_slaw has joined ##crawl-dev 02:22:33 ogaz: That might be caused by the server setup, but I have no idea how 02:23:04 joosa: does it happen to you? 02:23:06 I followed your game for a while and I saw no redraw problems 02:23:16 it never happened to me 02:23:25 weird 02:25:03 It's almost like if it was dropping packets 02:25:07 well 02:25:16 on the login for rl.heh.fi it says "Crawl 0.7 broken" 02:25:21 what's that about then 02:25:39 "_You start resting. Something unseen opens the door." - currently, the only door I have in LOS looks closed to me 02:25:48 but the dropping packets thing would make sense 02:25:51 Ah, that was old news from yesterday when I had that sqlite problem 02:27:21 -!- stabwound has quit [Ping timeout: 252 seconds] 02:27:22 -!- by has joined ##crawl-dev 02:27:35 hi 02:27:54 hey :D 02:28:12 ixtli: when is Enne merging your changes? 02:28:18 About that. 02:28:39 sorear and neunon and I have been discussing the thing i talked to you about 02:28:49 ah, you're talking, good 02:28:57 ogaz: I think I read about doors not being opened by monsters in the log 02:28:57 using ABC to wrap SDL 02:29:10 for values of "discussing" equal to "waiting in a private channel while lots of people dodge initiative" 02:29:10 joosa: yeah, I think that's something unrelated now 02:29:14 They don't approve of wrapping a wrapper, afaict. I understand this and think that' it's overkill. 02:29:21 sorear: ;-) 02:29:26 sorear: I'm traveling! :P 02:29:32 I'll be back in a few days though. 02:29:38 So what's Japan like? :) 02:29:49 sorear: I will link you to my pictures of Kyoto. 02:30:02 I made the wrong decision in living in Tokyo for a year. 02:30:11 Kyoto is /gorgeous/ around this time of year. 02:30:17 And in the fall. 02:30:41 I hear tokyo has lots of fabulous concrete 02:30:46 Haha it does. 02:30:56 I also hear it has lots of ... people 02:31:06 It's an amazing city, but not for traditional japan. 02:31:21 Like, it's not very 'japanese' in the way that foreign people think. 02:31:23 ixtli: in my opinion, if your approach is good enough, do that instead of something perfect; also, they get no say if they don't actually help :) 02:31:37 * doy agrees with by 02:31:40 The issue is that I agree with them. 02:31:50 Especially for wrapping SDL_image 02:31:53 -!- stabwound has joined ##crawl-dev 02:32:01 Though we've agreed on the glwrapper and fontwrapper classes. 02:32:02 i'd rather have something that already works than spend weeks fiddling with details 02:32:25 -!- lorimer has quit [Read error: Operation timed out] 02:32:37 So, if you guys want my changes sooner or later I could cut out the controversial parts. 02:32:39 -!- lorimer has joined ##crawl-dev 02:32:51 no, I was just worried you were being stalled 02:33:00 Not at all, everyone's been helpful. 02:33:05 great! 02:33:08 I've just been in Kansai. 02:33:09 :D 02:33:13 Nemelex should not accept gifts of gold (https://crawl.develz.org/mantis/view.php?id=1157) by MrMisterMonkey 02:37:03 Also, if you speak japanese, I've engineered a great way to stay in any major city for about 2000 yen / day as long as you're with two other people. 02:37:45 And they're nice hotels. 02:57:28 -!- ogaz has quit [Remote host closed the connection] 03:02:21 -!- murphy_slaw has quit [Quit: Colloquy for iPhone - http://colloquy.mobi] 03:03:20 -!- by has quit [Quit: Lost terminal] 03:09:57 Geez, that's cheap 03:11:29 -!- dpeg has joined ##crawl-dev 03:14:03 i know, right? 03:14:04 :) 03:14:47 define "stay in" 03:15:59 -!- ortoslon has quit [Quit: Leaving.] 03:24:44 -!- ortoslon has joined ##crawl-dev 03:28:24 日本語でスパッムできるよ！ 03:33:02 if you ask me, that's not much of a spam though 03:41:32 felirx: can you read that? 03:51:21 * Spads can :) 03:51:58 it's claiming that someone can spam in japanese 03:55:53 -!- ais523 has joined ##crawl-dev 03:57:58 -!- twb has joined ##crawl-dev 03:58:21 Hi, it's your friendly neighbourhood Q/A pedant here! 03:58:23 I: crawl: spelling-error-in-binary ./usr/games/crawl-curses Unkown Unknown 04:11:52 twb: you're spell-checking binaries? 04:12:15 It's just one of the checks that Debian provides. 04:12:29 * ais523 greps the source 04:12:29 It only checks for a blacklist of typos, though. 04:12:49 "crawl-curses"? Whose packaging is that? 04:13:12 kilobyte: mine. 04:13:23 "Unkown temple tag" is the context in the binary 04:13:27 Because I build both SDL and curses versions, as separate .debs 04:13:37 privately? 04:13:40 Yes. 04:13:45 so it looks like an internal error message 04:13:50 in mapdef.cc 04:13:59 line 2310 04:14:10 kilobyte: Real Soon NowTM, I'll get around to providing a deb-src repo 04:14:25 then why didn't you use the official packaging? 04:14:49 kilobyte: er, because Debian only ships releases 04:15:01 I usually build from git about once a week 04:15:15 Isn't there also some story involving you hating the official Debian maintainer and wanting to usurp the role? 04:15:28 sorear: oh yeah, for reals 04:16:10 Actually I just found the old packaging cringeworthy, so I pushed to get it cleaned up a bit (which has happened) 04:16:14 twb: from git? so you did rename it to crawl-curses then 04:16:21 Sarcasm detector returned "ambiguous". 04:16:53 kilobyte: I just change the name of the binary; crawl becomes a symlink to either crawl-curses or crawl-sdl 04:17:42 yeah, but the current packages in Debian can't be used for 0.6 since 1. they use an own makefile, and 0.6 does some heavy code generation so you can't really write your own one without reinventing a whole set of wheels, 2. it's not split 04:18:09 twb: instead of existing crawl and crawl-tiles? 04:18:14 kilobyte: yes. 04:18:31 kilobyte: if upstream builds a binary called "crawl-tiles", that post-dates what I did 04:18:48 uhm, you just said you build it from git 04:19:05 Yes. But when I *started* doing that, in 0.4, your makefiles were fucking awful 04:19:14 And I haven't bothered to drop my one yet. 04:19:37 the Debian packaging in git produces three .debs, crawl, crawl-common, crawl-tiles. "crawl" and "crawl-tiles" provide binaries with the same name. 04:19:54 Then that's something the Debian maintainer has done. 04:20:07 Oh, "the debian packaging in git" 04:20:12 hmm, are you using the debian alternatives mechanism? 04:20:25 uhm, Guus Sliepen hasn't done anything in ages 04:20:44 $ file /usr/games/nethack 04:20:46 /usr/games/nethack: symbolic link to `/etc/alternatives/nethack' 04:20:54 and presumably exactly the same trick works with Crawl 04:20:57 ais523: yes. 04:21:34 alternatives work only if you want to make only one version invokable 04:21:59 kilobyte: IME that is not correct. 04:22:17 alternatives make one version invokable under the default name, say "crawl" 04:22:25 ais523: right. 04:22:41 e.g. with Debian nethack, I can type "nethack" to get the default version, but "nethack-qt" or "nethack-x11" or "nethack-console" to get a specific version 04:22:48 exactly 04:23:27 and the packaging has one package for each version, and one package for common files 04:23:35 so that they all share things like a save directory and level maps 04:23:51 and you can save in tiles then load in console 04:24:19 kilobyte: that's what I set up for my private packages, yes 04:24:37 can you save in tiles then load in console on Crawl, technically speaking? 04:24:50 Well, I only tested it briefly 04:24:55 It didn't have any obvious errors 04:25:05 I know that adding or deleting tiles screws up save data, which vaguely implies to me that it doesn't 04:25:12 ais523: only if the packaging supports that, or you copy the save manually 04:25:18 kilobyte: ah, ok 04:25:42 the only difference are doll files, which are not created by console builds 04:25:49 note: kilobyte maintains the debian branch in git 04:26:42 although some other people helped 04:28:26 twb: could I take a look at your packaging, so I can compare these two? 04:28:53 Sure. Lemme find somewhere to put it 04:30:08 twb: the 'debian' branch in git is at 0.6.0-a2 on purpose so it's not tied to either the 0.6 branch or trunk, but if you want to look at 0.6 packages: deb-src http://angband.pl/debian/ sid main 04:31:05 Righto. 04:32:56 dget http://angband.pl/debian/pool/main/c/crawl/crawl_0.6.0~rc3-1.dsc 04:34:13 -!- stabwound has quit [Ping timeout: 264 seconds] 04:35:52 kilobyte: dget http://code.haskell.org/~twb/tmp/crawl_0.7.0~alpha0+859-1.dsc 04:36:39 03kilobyte * r3594c072cc1f 10/crawl-ref/source/mapdef.cc: Typo fix: "Unkown". 04:38:19 kilobyte: that build has #TILES# stuck in front of some tile-y stuff, since I moved from a Core 2 Q9550 to an Atom n450 last Monday, and I was sick of waiting for g++ builds I wasn't going to use anyway. It should be obvious how to uncomment them. 04:39:55 -!- stabwound has joined ##crawl-dev 04:44:18 twb: xz is not a valid upload compressor, only gz, bzip2 and lzma are 04:44:36 I wonder what you build-depend on quilt for. 04:44:49 kilobyte: hysterical raisins 04:44:59 the patch "cheating-motherfucker" is nice :p 04:45:04 lzma isn't a valid upload compressor for the debian archive, but xz will be. 04:45:16 kilobyte: that's for when I get sick of losing all the time :-) 04:45:25 That is, losing before D:4 04:45:52 the official origs are .bz2, there's no reason to repack and some people shout at you if you do 04:46:05 Like I said, I'm building from git 04:46:10 !lg . s=place 04:46:12 5419 games for kilobyte: 1656x D:1, 1328x D:$, 711x D:2, 464x D:3, 308x D:4, 253x Abyss, 177x D:5, 118x D:6, 59x D:7, 50x D:8, 31x D:9, 19x Orc:3, 18x D:11, 15x Orc:2, 15x D:10, 11x Orc:1, 11x Lair:2, 10x D:13, 10x D:12, 9x Lair:5, 9x Orc:4, 7x Elf:7, 7x Elf:1, 7x Lair:8, 7x Lair:7, 7x Temple, 6x Snake:5, 6x Lair:4, 6x Elf:2, 5x Ossuary, 5x D:16, 4x D:15, 4x Slime:6, 4x Sewer, 3x Swamp:2, 3x Vault... 04:46:33 If I was rolling stable releases, I'd use the upstream tar.bz2 04:47:19 you can "make package-source", it creates a tarball the same way releases do 04:53:29 why do you build-depend on: libz-dev (used only transitively by libpng), libgl1-mesa-dev libglu1-mesa-dev libxext-dev (not used by Crawl directly in any way) 04:54:24 If it got in there, it was because it was needed at some time. 04:54:46 twb: are you using Crawl's versions of things like lua, or the Debian versions? 04:54:59 ais523: I'm not using convenience copies, because that is not the Debian Way 04:55:07 And to be honest I don't care if that upsets upstream. 04:55:11 if you're using the Crawl versions, I can see why something needed transitively by libpng would be needed in build-depend 04:55:25 but not really otherwise 04:55:31 CFLAGS += -march=atom -mtune=atom 04:55:40 kilobyte: that only got added a couple of hours ago 04:56:06 also, something looks wrong about specifying both march and mtune 04:56:13 ais523: yup 04:56:32 march means "build only for this platform", mtune means "build to be optimal on this platform whilst also working on similar platforms" 04:57:05 nowadays, mtuning for i686 is probably a sensible idea for x86 builds 04:57:55 ais523: I'm running x86-64 on my atom, not that it matters. 04:58:13 twb: x64 builds are separate builds from x86 builds, aren't they? 04:58:20 at least in Debian packagin? 04:58:23 *packaging 04:58:23 ais523: yes. 04:58:42 ais523: I'm only building for one architecture; these are private packages. 04:58:56 ok 04:59:16 I just build Crawl and run without installing 04:59:27 which is actually kind-of nice, as it means I have separate saves and morgues for separate purposes 04:59:32 kilobyte: I suspect that libxext-dev and libz-dev are needed for steps that you do during "make package-source" 05:00:48 twb: 1. "make package-source" is done only on the upstream side, 2. it uses bzip2 not zlib 05:00:51 ais523: until this week, I was building for x86 on an x86-64 build host, so it was easier to make a .deb than to scp across 30MB of stuff 05:01:03 and libxext-dev is some random x11 library 05:01:04 ais523: march=i386 = runs on i386, optimal for (unspecified, possibly defaults to `arch`) 05:01:15 ais523: march=i386 mtune=i386 = optimal for i386 05:01:20 kilobyte: by that description I *am* the upstream side. 05:01:22 doesn't seem at all redundant to me 05:01:40 sorear: I disagree, let me check the gcc docs 05:02:03 twb: yeah, but Build-Depends state what is needed for the actual package build, on buildds 05:02:14 It might if I was building official packages. I'm not. 05:02:35 When I roll an upstream snapshot, I am not including stuff that isn't in the git archive. 05:02:37 yeah, but you' 05:02:53 yeah, but you're listed as an uploader in the official archive 05:02:58 ais523: "specifying march=X implies mtune=X" 05:02:59 I don't know why that was done. 05:03:06 AFAIK I've never uploaded crawl to Debian. 05:03:09 ais523: but nothing is said about march=X mtune=Y being invalid 05:03:09 so I'm assuming you're preparing for the official upload 05:03:21 kilobyte: that would be a bad assumption. 05:03:54 What I did was send an email to Guus saying something like "here is what I have, take whatever you want, I don't really care". 05:05:00 "Here is Sigmund. He will kill you. Thanks for playing the tutorial!" 05:05:12 felirx: :p 05:06:05 sorear: hmm, it seems that march selects instruction set and ABI, mtune selects everything else 05:06:15 at least on x86 and x64 05:06:22 the options also exist on ARM, where they do something different 05:07:58 And mips 05:09:23 official Debian packages on i386 need to run on i486 (as the name implies :p) 05:11:57 Which is one reason I'm running my netbook in native x86-64 mode 05:13:41 for an official package, or even a Crawl-devteam-official Linux package, -mtune=generic would be sensible, it asks to optimize for commonly-used processors (while leaving the program still working on weirder x86 processors) 05:14:25 I would usually advocate leaving it unspecified 05:15:13 I'm not entirely sure if the mtune default then is generic or native 05:19:30 generic, native produces different binaries 05:20:12 yes, and native normally makes more sense unless you're compiling for someone else 05:21:26 I tested this on several machines, and it appears that -mtune=generic produces same code as -mtune=i686, or -mtune=native on a real Pentium 2 05:21:57 the oldest machine I have access to right now is: model name: Pentium II (Deschutes) 05:23:25 the description of -mtune=generic implies that it's going to always be an alias for one of the others, but which will change from gcc version to gcc version as different processors become popular and unpopular 05:24:41 The documentation for -mtune=generic claims that it means "processors that were most common when that version of GCC was released", yet you can find anything kilobyte: I know a couple of people that deliberately stick to PIII because it has low power draw and heat emissions 05:29:52 Occasionally they go around farming PSUs from the rest of the user group 05:30:05 modern stuff like i7 doesn't use all that much power 05:30:19 What can I say? They're nuts. 05:30:36 I have a 430wat psu running hd5870 and i7-930 without trouble 05:30:45 I'm on a U2700 atm, it's a netbook 05:30:45 calculated it uses iirc 320~350 wats 05:31:17 powertop rates my power draw at 11 Watts 05:34:03 kilobyte: FYI, you can't "make package-source" from a HEAD checkout, because util/release_ver doesn't exist 05:42:58 twb: uhm, works for me 05:43:51 In a fresh checkout? 05:44:59 twb: and it creates it on the fly; util/release_ver is never in git, it's always created by "make package-source" itself 05:45:17 I'll try again 05:47:18 yes, I just made a fresh clone, and it worked 05:47:20 http://paste.debian.net/65844/ 05:47:21 -!- Ashenzari has quit [Ping timeout: 240 seconds] 05:48:00 -!- Ashenzari has joined ##crawl-dev 05:48:00 -!- The topic of ##crawl-dev is: DCSS development channel! | Logs: http://tozt.net/crawl | People with +v have commit access. | Please keep general crawl-related chit-chat to ##crawl. | Dev wiki: http://crawl.develz.org/wiki | Ashenzari source: http://github.com/doy/crawlbot 05:48:25 Yeah, it is. 05:49:10 The "else" part of the ifndef NO_YACC could have something like prebuildyacc: \n\t @echo "this target requires yacc"; false 05:49:21 good idea 05:49:33 Make that \n\t false # This target requires yacc 05:49:41 (Since I don't like @ hiding :-) 05:49:43 Complaining about util/release_ver because of no yacc is utterly wrong. 05:50:01 kilobyte: no, that's separate 05:51:03 It's complaining about that BEFORE it does any work, due to the := in SRC_VERSION 05:51:51 You could silence the || cat util/release_ver part, though that might end up making things MORE confusing 05:52:03 I don't know why git describe --tags --long failed, though 05:52:03 Oh, because of --depth 1 05:52:03 Mea culpa 05:52:17 Pity git doesn't have a mode for just-in-time checkout of objects 05:56:29 -!- by has joined ##crawl-dev 06:02:03 Just showing off here: Power usage (ACPI estimate): 0.5W (95.6 hours) (long term: 8.3W,/6.2h) 06:06:27 -!- hotsun has joined ##crawl-dev 06:11:59 353 calculated PSU "size" needed :( 06:12:55 -!- hotsun has quit [Quit: HydraIRC -> http://www.hydrairc.com <-] 06:14:53 -!- Mu_ has joined ##crawl-dev 06:19:19 its looks like all remaining Debian bugs are either fixed or wishlist FRs 06:20:34 including that one with xterm-256 and bright colours 06:21:32 hellooo 06:21:46 meep? 06:22:18 due! 06:26:00 due: att:neutral monsters are a faction on their own, hostile both to you and to att:hostile monsters, yet they don't follow you aggressively. I'd want a second fully hostile faction. Do you think that's a good idea? 06:26:14 and how should we name it? :p 06:26:25 um 06:26:40 is a D:1 imp out-of-depth? 06:26:40 hang on, brain->fingers lag 06:26:56 by: I would say so, Twinge would know for sure. 06:27:18 kilobyte: I'm not sure exactly what you want. this is my problem, sedatives blah blah. 06:27:39 @whereis due 06:27:39 No where information for due. 06:27:44 @whereis duelium 06:27:44 duelium the Charm-Maker (L3 HaEn) saved on D:2 on 2010-03-22 after 1574 turns. 06:28:04 We're all in the duelirium! 06:28:31 no, duelium = due + valium 06:28:34 it is an interesting combination 06:29:50 due: I'd want a holy Pan level, with angels fighting both demons and you (unless you're a goody goody two shoes follower of TSO/Ely/Zin) 06:29:58 For us Germans, it's dü anyway :) 06:30:18 kilobyte: ah, ask dpeg about that 06:30:24 I'm not too sure whether I like the idea 06:30:25 kilobyte: I like a holy Pan level, but I don't think that the angels there need to fight demons... 06:30:48 isn't TSO all about destroying all that's unholy? 06:31:51 yes, but TSO wouldn't be tha Pan lord :) 06:33:08 something is weird with monster attitudes right now, though I can't quite put my finger on it 06:33:23 maybe it's the mix between player-relative and absolute? 06:34:04 There are three levels of neutral though 06:34:46 I find this in itself inherently impossible to understand intelligently. 06:36:05 dpeg: my idea is to have the player enter there in the middle of TSO's forces' raid on Pan. Burned features, traces of fighting, and so on. 06:36:33 so it would still be a demonic Pan lord, maybe, and possibly a few demons 06:36:36 but mostly angels? 06:36:58 exactly, with demons holed up in a small portion of the level that's mostly overrun by angels 06:37:03 ais523: see the 0.7 holy monsters overhaul 06:37:24 due: ah, I was using "angels" as a general term, rather than the monster "angel" in particular 06:37:26 angels are going, going, goooone. 06:38:11 actually, it would be good to have the run-of-the-mill angelic monster, but not necessarily named "angel" 06:38:29 s/angel/cherub/. 06:38:35 sling + bucklers. 06:38:48 ugh, the sling idea is awful 06:38:50 we don't have a "run of the mill demonic monster". 06:38:57 wat? why? 06:38:59 imp 06:39:13 by: ew. 06:39:21 "so what's this monster?" "basically the holy version of an imp". 06:39:28 dpeg would kill me :) 06:40:16 What crawl needs is thousands of unique monsters, so that when you get to level 50 you can just go between zot:5 and D:1 looking for them to kill! 06:40:23 Oh wait, that's zangband 06:40:36 the holy monsters proposal has both good and bad parts. The vast majority is good, but punk cherubs with slings are in the "very bad" section. 06:40:52 * due shrugs. 06:41:16 It was on the tracker and the email list etc for almost a month before it got moved to the wiki and nobody complained. 06:41:26 And it's been on-wiki for the past five months without complaint. 06:41:47 because everyone cared about 0.6? 06:42:05 and the wiki is crawling with bad ideas. Like that jelly race. 06:42:21 what I'm saying is that it got relative approval enoguh that I am going to implement it as-is. 06:42:21 Trees should be a playable race! 06:42:31 and most of holy monster ideas are simply kick ass 06:42:32 You sit there and monsters attack you, and you have to use your "drop acorn" ability 06:42:35 needs cherub that shoot heart branded arrows 06:42:50 felirx: what about arrows of enslavement? 06:43:02 (btw, what would enslaving the player actually do?) 06:43:19 mesmerisation? inability to attack the monster in question? 06:43:38 You spend every second turn composing a sonnet 06:44:24 -!- stabwound has quit [Ping timeout: 276 seconds] 06:44:39 Whenever you change level, a unique named Cyrano appears and offers you useless advice 06:44:47 :)) 06:47:37 what would be really neat would be swapping the ally/enemy status of every monster for a while, whilst also putting the player under AI control 06:47:46 but that would be rather hard to program 06:48:02 -!- stabwound has joined ##crawl-dev 06:49:10 another issue with holy monsters: if you follow Zin/Ely/TSO, a _part_ will convert to your side, beating on their brothers. 06:49:41 Not alll holy monsters are perfect. 06:49:45 Cf. orcs. 06:49:48 + beogh. 06:51:25 with beogh, isn't the flavour that some of the orcs believe you're the messiah, and some don't 06:51:38 and some see the error of their ways after being beaten to a pulp 06:51:43 some holy monsters beating up others is definitely a bug 06:52:02 Why? 06:52:14 Some daevas in the Abyss are worshippers of Xom. 06:52:22 isn't that just when a neutral one in LOS tries to walk into a not-as-yet neutral one outside of LOS ? 06:52:42 angels of Xom absolutely have to be added, now someone's mentioned them 06:53:32 ais523: They already exist. That's my point. 06:53:59 do they have particularly xommish stats, though? 06:54:24 due: but those holy monsters that kilobyte describes all worship TSO, and some convert to your "side", which is also TSO, to start beating up on the others 06:56:03 normal angels are att:hostile, if they recognize you as a friend, they go att:neutral-good 06:57:23 yet another interface enhancement: compare two items 06:57:55 by: hence why I'm not sure it would work. 06:58:15 What about a new attitude cathegory: hostile to the player (unless worshipping a good god, with 100% chance) and demons, not hostile to anything else? 06:58:29 s/demons/unholy/g 06:58:41 sorry, yeah. 06:58:58 -!- syllogism has joined ##crawl-dev 06:59:00 I would prefer to keep hostile/friendly separate from good/neutral/evil etc 06:59:46 is the centaur in hunters_booth_kilobyte supposed to be able to shoot past the bush? 07:00:09 by: so you'd want to rethink the current attitude system? Indeed it is a mess. 07:00:22 everyone can shoot past bushes, including centaurs. 07:00:57 this one can't 07:01:22 The centaur shoots an arrow of frost. The arrow of frost hits the bush. 07:01:25 should I file? 07:01:54 uh oh 07:03:36 "oops" 07:03:53 I think better to go "god conduct" route 07:03:56 give monsters gods 07:04:01 and a "god_attitude" 07:05:29 that might work 07:06:36 demons worship makhleb, angels worship one of the good gods, etc. 07:06:42 but anyway, sleep time 07:06:46 will take a look in the morning 07:06:52 03kilobyte * r06b78b9472d5 10/crawl-ref/source/makefile: Give a better fail message if yacc is not installed in targets that need it. 07:06:55 03kilobyte * r594a1edd234a 10/crawl-ref/docs/options_guide.txt: Describe various cset_ encodings. 07:08:00 ciao 07:08:35 good night 07:26:18 back 07:26:53 wb 07:27:36 dpeg, http://www.algorithmicartisan.com/gostones/pictures/cracked%20crystal.big.html 07:27:47 dpeg, gonna buy a new set of go stones? :) 07:29:49 cbus: no, I have my trusty Korean stones. 07:30:00 hehe 07:30:06 dpeg, does stones look like the moon :) 07:30:35 dpeg, http://www.algorithmicartisan.com/gostones/goStones.html 07:30:38 mwhaha, now I can pick my race with a mouse 07:30:40 cheap ;) 07:30:42 felirx: yes! 07:32:59 dpeg, nice stones? ;) 07:33:31 also clicking the "space for species only" entry works... 07:33:38 just random selection to go 07:33:44 felirx: commit? (y/y) 07:34:00 dpeg, shouldn't that be (Y/y) ;) 07:34:04 N! 07:34:14 it's only the species screen 07:34:20 needs background too! 07:35:17 felirx: Is your system flexible, for when we axe yet another species? 07:35:44 you just add it to the menu 07:35:54 you have total freedom where in (x,y) the entry is listed at 07:36:07 felirx: awesome 07:36:22 you might have to stab the formatting code but ;) 07:38:21 -!- bhaak has quit [Ping timeout: 276 seconds] 07:48:23 -!- bhaak has joined ##crawl-dev 07:54:26 by: you added to beam.cc line 3976 a condition: && (flavour == BEAM_MISSILE || flavour == BEAM_MMISSILE) 07:54:31 by: do you remember why? 07:54:35 no 07:54:51 arrows of frost will have flavour of BEAM_COLD, and so on 07:54:57 let me check the commit 07:57:30 d47e840a "Fix bushes not being affected by explosions. This also fixes Cerebov not fire-storming bushes in the arena." 07:57:31 d47e840a 07:57:41 this was reported as a bug 07:57:50 adding && !is_explosion looks ok 07:58:35 so maybe just remove the BEAM_xx line? 07:59:05 not sure about Cerebov and firestorm, 07:59:14 btw, is BEAM_COLD even accurate now that arrows of cold behave like missiles? 07:59:42 not sure 08:02:20 he hit it with his sword a few times then walked away 08:02:47 probably, wand of fire should hit the bush (and go through), right? 08:03:09 but he does the same with other plants 08:03:21 I think so. 08:03:48 that bug was a tracer bug 08:03:55 if I remeber correctly 08:04:08 of course, it's not important that cerebov firestorm bushes 08:04:23 it might be, if you lured him into a level with bushes on as a fedhasite 08:04:31 admittedly that's a little implausible 08:04:34 to clarify the options in the menu 08:04:49 # - Random species should pick a random (available) specie 08:05:02 # - Viable Charactor shoudl pick a recommended combo of anything 08:05:13 ! - random charactor should pick any allowed combo 08:05:17 am I correct? 08:07:08 -!- ortoslon has quit [Quit: Leaving.] 08:07:59 felirx: yes, but you use # twice 08:08:20 that vault is quite broken now, though, because the centaur tries to kill the bush to get to me when not shooting 08:08:43 whoops, * should be the first one 08:08:54 is there no entry to pick a random viable specie for current job? 08:09:07 wands of fire do not ignore bushes, though 08:10:01 03kilobyte * r0017779adcb1 10/crawl-ref/source/beam.cc: Fix bushes stopping branded arrows. 08:10:09