00:01:36 -!- Grivan has quit [Ping timeout: 256 seconds] 00:05:09 -!- GuraKKa has quit [Read error: Connection reset by peer] 00:05:43 -!- bh has joined ##crawl-dev 00:05:44 Unstable branch on crawl.s-z.org updated to: 0.13-a0-2156-ga9a2eb8 (34) 00:05:59 Unstable branch on rl.heh.fi updated to: 0.13-a0-2156-ga9a2eb8 (34) 00:07:17 -!- CKyle has joined ##crawl-dev 00:07:33 -!- kekekela has quit [Quit: Page closed] 00:13:17 Unstable branch on crawl.develz.org updated to: 0.13-a0-2156-ga9a2eb8 (34) 00:13:21 -!- Snarwin has quit [Quit: leaving] 00:18:59 -!- Sky__ has quit [Quit: Page closed] 00:19:18 ??twisted resurrection 00:19:19 twisted resurrection[1/4]: Converts all piles of corpses in LOS into crawling corpses or macabre masses, very weak attackless creatures that can merge to form abominations. Creates 1HD worth of creatures per 60 to 20 aum, depending on spell power (and with half efficiency past 15 HD). Small aboms require 6HD and two corpses, large 11HD and three corpses. 00:19:43 -!- Dixlet has quit [Ping timeout: 264 seconds] 00:21:31 -!- archaeo has quit [Ping timeout: 250 seconds] 00:22:58 -!- Fhqwhgads_ has quit [Remote host closed the connection] 00:23:01 03bh 07* 0.13-a0-2157-gfb30ef6: Make Polymoth remains mutagenic 10(33 seconds ago, 1 file, 1+ 1-) 13http://s-z.org/neil/git/?p=crawl.git;a=commitdiff;h=fb30ef67ae1a 00:24:07 -!- lasserith has quit [Ping timeout: 276 seconds] 00:24:07 -!- [1]lasserith is now known as lasserith 00:26:26 -!- jday_ has quit [Quit: Leaving] 00:27:33 how can i get build.h to build 00:27:54 have you tried running the line i pasted? 00:28:49 what was it again i restarted computer 00:28:58 perl.exe "util/gen_ver_msvc.pl" build.h 00:30:18 -!- dtsund has joined ##crawl-dev 00:32:09 yech. I don't miss windows at all :) 00:33:10 why is that an example of why windows is bad? 00:33:56 'cause you have to run perl! 00:37:05 heh 00:37:42 -!- rebthor has quit [Quit: ChatZilla 0.9.90 [Firefox 21.0/20130511120803]] 00:37:43 you mean instead of perl running you? 00:38:24 ...in Soviet Crawl Build, perl runs you? 00:39:58 -!- CKyle has quit [Quit: CKyle] 00:40:08 in soviet tiles, the tiles are all 34x34 00:40:33 weird 00:40:38 can anyone check the cszo lobby 00:40:46 and tell me what Game type they see for my GhEE 00:40:56 -!- Silurio has quit [Ping timeout: 240 seconds] 00:41:32 something to do with wrong imperial -> metric conversions when they were reading the schematics 00:43:15 well, the spoiler is that it shows my game type as 'tut-git' 00:43:32 and unless this game has actually been one incredibly complicated and deceptive tutorial 00:43:36 I'm pretty sure that's wrong 00:44:05 gammafunk: I see that 00:44:17 (tut-git, that is) 00:44:26 rchandra: ok, good to know it's not just my browser 00:44:34 Windows builds of master branch on crawl.develz.org updated to: 0.13-a0-2157-gfb30ef6 00:44:42 and it's only you among those playing 00:45:18 Maybe |amethyst is messing with me 00:46:33 oh, haha 00:46:37 I see how I did it 00:47:24 !tell |amethyst I'm not sure if you'd consider this a bug, but if I choose T)utorial from the trunk game menu, it starts my saved game instead of the tutorial 00:47:25 gammafunk: OK, I'll let |amethyst know. 00:47:54 -!- eggonlegs has quit [Changing host] 00:48:02 !tell |amethyst I only hit 'T' by accident, but my game plays normally, and it shows my game type as 'tut-git' in the crawl lobby 00:48:03 gammafunk: OK, I'll let |amethyst know. 00:48:50 -!- Nomi has quit [Quit: Lag] 00:51:25 -!- pelotron has quit [Ping timeout: 276 seconds] 00:52:43 -!- archaeo_ has quit [Ping timeout: 250 seconds] 00:53:07 C:\games\roguelikes and like\Crawl-Git\crawl\crawl-ref\source>perl.exe "util/gen _ver_msvc.pl" build.h 'git' is not recognized as an internal or external command, operable program or batch file. No Git, and util/release_ver doesn't exist. 00:53:48 seems to be only file i can't get to build 00:54:45 the generate all bat works though 00:54:55 so perl is workinh 00:54:58 so perl is working 00:55:22 -!- st_ has quit [] 00:55:32 Arendeth: It's not finding the git binary 00:57:16 now i reinstall to run from command prompt 00:57:38 reinstalling 00:58:25 -!- eggonlegs has quit [Quit: leaving] 00:58:36 think i got that file made 00:58:53 -!- eggonlegs has quit [Changing host] 01:00:23 good think that I'll get both windows 8 and linux when I build I gaming rig next year I'll have the advantage of both 01:00:57 -!- bh has quit [Quit: sleep] 01:01:37 how can I reach this by an irc client? 01:02:12 reach ##crawl-dev? 01:02:14 -!- eggonlegs has quit [Client Quit] 01:02:29 ya 01:02:32 -!- eggonlegs has quit [Changing host] 01:02:40 you just need to join chat.freenode.net 01:02:43 and join the room 01:03:04 probably also register your nick; see freenode for details on that, and it's not required 01:03:07 oh you're using freenode's lame web frontend? 01:03:19 or is it less lame now 01:03:25 It's still pretty lame 01:05:36 -!- edlothiol has joined ##crawl-dev 01:06:27 -!- Arendeth has quit [Quit: Page closed] 01:13:35 Monsters summoned with rod of swarms can follow player down stairs by ACG 01:14:53 -!- gammafunk has quit [Quit: gammafunk] 01:18:04 -!- Kalir has quit [Quit: CADENZA-CLASS CLOCKWORK KNIGHT--OPERATION TERMINATED] 01:22:57 -!- Pacra has quit [Quit: Pacra] 01:23:16 -!- Melum has quit [Ping timeout: 276 seconds] 01:30:37 -!- Thyme has quit [Ping timeout: 240 seconds] 01:33:31 -!- Silurio1 has quit [Ping timeout: 264 seconds] 01:39:55 -!- eggonlegs has quit [Ping timeout: 260 seconds] 01:45:32 -!- ontoclasm has quit [Quit: Leaving.] 01:46:23 -!- Arendeth has joined ##crawl-dev 01:47:00 still got error or 3 01:47:25 -!- Sizz has quit [Quit: Page closed] 01:47:31 Error 92 error LNK2001: unresolved external symbol _pcre_free C:\games\roguelikes and like\Crawl-Git\crawl\crawl-ref\source\MSVC\pattern.obj crawl 01:51:30 -!- gammafunk has joined ##crawl-dev 01:56:45 -!- gammafunk has quit [Quit: gammafunk] 01:59:30 -!- radinms has quit [Ping timeout: 252 seconds] 02:07:13 -!- Silurio has quit [Ping timeout: 248 seconds] 02:10:09 -!- MarvinPA has joined ##crawl-dev 02:17:13 -!- Flex has quit [Ping timeout: 246 seconds] 02:33:58 -!- Arendeth has quit [Quit: KVIrc 4.2.0 Equilibrium http://www.kvirc.net/] 02:38:06 -!- ophanim has quit [Read error: Connection reset by peer] 02:40:01 -!- Vizer has quit [Read error: Connection reset by peer] 02:41:37 -!- lasserith has quit [Read error: Operation timed out] 02:51:04 -!- Yermak has quit [Quit: Page closed] 02:53:11 -!- ussdefiant has quit [Ping timeout: 256 seconds] 03:01:30 -!- Pacra has joined ##crawl-dev 03:03:22 -!- radinms has quit [Ping timeout: 276 seconds] 03:07:20 -!- magicpoints has quit [Ping timeout: 240 seconds] 03:18:39 -!- Mu_ has joined ##crawl-dev 03:20:43 -!- tolly has quit [Ping timeout: 260 seconds] 03:23:57 -!- qoala has quit [Remote host closed the connection] 03:24:23 -!- qoala has joined ##crawl-dev 03:27:29 -!- Gastrox has quit [Quit: Page closed] 03:28:15 -!- rchandra has left ##crawl-dev 03:32:12 -!- Silurio1 has quit [Ping timeout: 268 seconds] 03:35:49 -!- Lightli has quit [Quit: zzz] 03:36:13 -!- simmarine has quit [Quit: Leaving] 03:38:04 -!- adityarajbhatt has quit [Quit: WeeChat 0.4.1] 03:45:29 -!- Daisuki has quit [Client Quit] 03:52:31 -!- ahahaha has quit [Quit: Page closed] 03:56:42 -!- bonghitz has quit [Remote host closed the connection] 03:57:41 -!- Aryth1 has quit [Client Quit] 03:57:58 -!- Aryth has quit [Ping timeout: 276 seconds] 04:08:46 -!- eith has joined ##crawl-dev 04:10:26 -!- Blazinghand has quit [Ping timeout: 268 seconds] 04:24:29 -!- mason- has quit [Quit: Leaving] 04:33:27 -!- alefury has joined ##crawl-dev 04:37:34 -!- mnoqy has joined ##crawl-dev 04:42:51 -!- ProzacElf has quit [Quit: Leaving] 04:44:07 -!- Helmschank has quit [Ping timeout: 276 seconds] 04:45:03 -!- scummos has quit [Ping timeout: 252 seconds] 04:51:39 -!- Duralumin has quit [Ping timeout: 252 seconds] 04:52:43 -!- scummos has quit [Ping timeout: 264 seconds] 04:58:57 -!- BizmarkRibeye has quit [Ping timeout: 248 seconds] 04:59:16 -!- santiago_ has quit [Remote host closed the connection] 05:01:23 -!- santiago_ has joined ##crawl-dev 05:04:00 -!- buppy has quit [Remote host closed the connection] 05:08:20 -!- buppy has joined ##crawl-dev 05:11:30 -!- Aponym has quit [Quit: KVIrc 4.2.0 Equilibrium http://www.kvirc.net/] 05:12:59 -!- eb has quit [] 05:23:51 -!- Grivan has quit [Ping timeout: 256 seconds] 05:26:48 -!- oberstein has quit [Ping timeout: 245 seconds] 05:28:40 -!- oberstein has joined ##crawl-dev 05:32:57 -!- dtsund has quit [Quit: dtsund] 05:42:30 Did I hear an idea forming about making summons stronger but with a penatly for them dieing? A mana ddoorish penalty, perhaps? 05:42:34 On the item destruction branch: strikes me as pointless and liable to make the game a little shallower. The effects themselves seem nice for some creatureness though. 05:58:15 I'm pretty sure this is a buff to a weapon-spell combo that doesn't need it, but what if one could reach past a spectral weapon without chance of accidentally hitting it? 05:59:29 I was reasoning that spectral weapon and battlesphere are extensions of the caster and thus shouldn't get in the way for other effects, and realized flavourwise it could be extended to polearms/reaching as well. 06:00:16 -!- sacje has quit [Quit: sacje] 06:00:40 damage sharing already provides some incentive not to stand behind the weapon, unlike normal summons/allies. 06:01:05 s/accidentally hitting it/"couldn't reach far enough"/ 06:01:33 (you can't accidentally hit allies) I should probably be asleep ;p 06:04:48 -!- halv has quit [Client Quit] 06:09:52 -!- blabber has joined ##crawl-dev 06:12:17 -!- dtsund has joined ##crawl-dev 06:12:18 -!- dtsund has quit [Client Quit] 06:16:42 Moth of Nihilism by pubby 06:21:17 -!- tcjsavannah has quit [Read error: Connection timed out] 06:26:03 -!- halv has quit [Client Quit] 06:29:18 patman (L13 DrIE) ERROR: range check error (-29 / 17) (D:17) 06:31:26 has there been any decision on dwants or those artefacts I submitted? 06:32:17 -!- santiago_ has quit [Ping timeout: 248 seconds] 06:43:12 Engwar (L14 DrWz) (Lair:8) 06:44:28 nobody online? I never know the best times for IRC 06:47:06 hi 06:49:04 hello friend 06:50:43 friend hi 06:50:53 I just realised you're the pubby I answered at least one Stack Overflow question by 06:50:59 i thought all of your artefacts seemed cool for what it's worth 06:51:32 elliott: yes, I post there. what's your username? 06:52:23 http://stackoverflow.com/users/1097181/ehird 06:53:13 oh! I know you as the really smart haskell guy 06:53:30 (and that's relative to other haskellers as pretty much all of them are smart) 06:53:45 :☺) 06:53:59 hehe, thanks :P 06:55:07 -!- st_ has joined ##crawl-dev 06:57:28 -!- qoala has quit [] 07:01:13 -!- halv has quit [Quit: Leaving] 07:11:00 -!- absolutego has quit [Quit: leaving] 07:16:58 -!- Vizer_ has quit [Quit: bye] 07:26:34 -!- archaeo has quit [Quit: Page closed] 07:31:44 -!- Bloax has joined ##crawl-dev 07:33:56 -!- radinms has quit [Ping timeout: 268 seconds] 07:36:23 -!- Roarke has quit [Quit: Page closed] 07:37:39 -!- nooodl has joined ##crawl-dev 07:38:00 -!- Daekdroom has quit [Quit: Saindo] 07:40:33 elliott: I take it SO is generally better than #haskell? :P 07:40:40 haha 07:41:04 I'm glad I asked 07:43:19 -!- Burer_ has quit [Quit: I'll give myself a bulldog out of winston churchill] 07:43:55 buppy: dwants seemed cool (although 1) I'm not an authority on crawl balance 2) I didn't really think about the implications 3) I have absolutely zero influence on whether anything gets included) 07:46:47 An ant-like race seems cool in itself. 07:47:03 But the concept of dwants was pretty damn terrible. 07:47:08 At least the one I recall. 07:47:56 Bloax: they're actually called formicids instead of dwants now.. if that matters. 07:48:15 Which is why I said that last line. :p 07:48:32 dwants? 07:48:44 dwant >>>> formicid 07:49:16 elliott: sadly, not everyone agreed :( 07:50:20 Eronarn: ask the learndb! (I'm asking you because I have no clue how to work the bots) 07:50:35 buppy: well not everyone agrees on formicid either :P 07:50:48 how about Dwant-Formicids. a compromise. 07:51:01 how about their scientific name is formicidae dwanticus 07:51:31 okay deal 07:51:40 put it in the x-v description 07:52:01 Dwanticids 07:53:48 their description better state that they are formidable 07:54:08 also myrmidons duh 07:54:47 -!- darkflagrance has quit [Ping timeout: 250 seconds] 08:09:42 -!- radinms has quit [Ping timeout: 268 seconds] 08:09:58 -!- Poncheis has quit [Quit: ChatZilla 0.9.90 [Firefox 21.0/20130511120803]] 08:14:59 -!- evilmike has joined ##crawl-dev 08:22:09 -!- sbanwart has quit [Ping timeout: 248 seconds] 08:23:01 -!- lordfrikk has quit [Remote host closed the connection] 08:25:30 -!- mnoqy has quit [Ping timeout: 241 seconds] 08:25:58 -!- mnoqy has joined ##crawl-dev 08:32:45 -!- MaxFrost has quit [Ping timeout: 246 seconds] 08:36:07 03kilobyte 07* 0.13-a0-2158-gcbca859: Don't use exceptions for short reads. 10(21 minutes ago, 1 file, 15+ 6-) 13http://s-z.org/neil/git/?p=crawl.git;a=commitdiff;h=cbca8596bb62 08:41:24 -!- blackflare has quit [Ping timeout: 256 seconds] 08:47:06 -!- dagonfive has quit [Read error: Connection reset by peer] 08:52:21 -!- santiago_ has joined ##crawl-dev 08:54:35 -!- Daekdroom has quit [Quit: Saindo] 08:56:43 -!- Silurio has quit [Ping timeout: 260 seconds] 09:00:57 -!- ackack has quit [Ping timeout: 256 seconds] 09:05:18 -!- Kromgart has quit [Read error: Connection reset by peer] 09:07:47 -!- CKyle has joined ##crawl-dev 09:11:58 -!- Porost has quit [Quit: fgtuhy6YuivghuyYttFgbn] 09:12:39 -!- Fhqwhgads_ has quit [Ping timeout: 246 seconds] 09:12:48 -!- Fhqwhgads__ is now known as Fhqwhgads_ 09:18:39 -!- Datul has quit [Quit: Page closed] 09:25:34 -!- Hailley has quit [Ping timeout: 276 seconds] 09:32:07 -!- Porost has quit [Quit: Reconnecting] 09:44:42 -!- orionstein_away is now known as orionstein 09:44:43 -!- Nakatomy has quit [Ping timeout: 256 seconds] 09:45:54 -!- ZebTM has quit [Ping timeout: 252 seconds] 09:56:04 -!- rossi has quit [Ping timeout: 246 seconds] 09:56:18 -!- blabber has quit [Quit: leaving] 09:57:34 -!- orionstein has quit [Quit: ZNC - http://znc.in] 09:59:22 -!- santiago_ has quit [Ping timeout: 276 seconds] 10:00:42 -!- myrmidette has quit [Ping timeout: 268 seconds] 10:01:11 -!- orionstein_away is now known as orionstein 10:07:51 -!- Hailley has quit [Quit: Petra is closed!] 10:08:28 -!- ZebTM has quit [Read error: Connection reset by peer] 10:11:33 -!- Helmschank has quit [Ping timeout: 240 seconds] 10:12:55 !tell qoala there was some discussion yesterday (me and evilmike specifically) around whether SW actually needed damage sharing, maybe we should figure out exactly if / what it achieves first 10:12:56 mumra: OK, I'll let qoala know. 10:14:08 |amethyst: I'm guessing 6b712aaa may have to do with this? 10:14:14 (this being Sky's problem) 10:16:26 <|amethyst> you.transform_uncancellable is true 10:16:31 -!- eith has quit [Ping timeout: 260 seconds] 10:16:54 o_O 10:17:41 <|amethyst> _You feel yourself come back to life. You feel a surge of divine spite. You miscast Necromutation. You are contaminated with residual magics. You are stuck in your current form! 10:17:45 <|amethyst> _Your body is flooded with distortional energies! 10:18:43 SW? 10:18:50 oh, nm 10:22:02 !tell Nivim that was one of many suggestions about summons i think yes 10:22:03 mumra: OK, I'll let nivim know. 10:22:53 <|amethyst> hm, this case TRAN_PIG: thing in transform() looks suspicious 10:23:35 <|amethyst> it sets you.transform_uncancellable = true even if just_check is true 10:23:39 -!- st_ has quit [Read error: Connection reset by peer] 10:24:08 -!- st_ has joined ##crawl-dev 10:27:09 -!- Nexos_ is now known as Nexos 10:28:31 <|amethyst> !tell gammafunk re the tutorial thing, that's in Mantis... it's because crawl puts tutorial and regular saves in the same directory 10:28:32 |amethyst: OK, I'll let gammafunk know. 10:29:12 <|amethyst> !tell tenofswords re %git search and a commit in a branch... not sure what the best way to detect that is 10:29:12 |amethyst: OK, I'll let tenofswords know. 10:31:46 <|amethyst> hm, there are other problems with the transform stuff 10:32:22 <|amethyst> You might get prompted "Transforming will reduce your strength to zero. Continue?"; if you say no, you'll get a different form 10:34:41 -!- st_ has quit [Read error: Connection reset by peer] 10:34:57 -!- st_ has joined ##crawl-dev 10:41:12 03|amethyst 07* 0.13-a0-2159-g170a2ce: Don't set transform_uncancellable when checking pig form. 10(2 minutes ago, 1 file, 2+ 1-) 13http://s-z.org/neil/git/?p=crawl.git;a=commitdiff;h=170a2cef99fb 10:41:12 03|amethyst 07* 0.13-a0-2160-g3de8706: Fix saves with players stuck in their base form. 10(48 seconds ago, 1 file, 7+ 0-) 13http://s-z.org/neil/git/?p=crawl.git;a=commitdiff;h=3de87067b41d 10:41:59 -!- ground4 has quit [Quit: WeeChat 0.4.1] 10:44:15 -!- archaeo has quit [Quit: Page closed] 10:45:46 -!- bh has joined ##crawl-dev 10:47:45 |amethyst: involuntary transformations try hard to not be fatal, but with stat_zero nerfed that much there's little reason to care here 10:47:54 Unstable branch on crawl.akrasiac.org updated to: 0.13-a0-2160-g3de8706 (34) 10:47:55 it's not fatal, at least 10:48:02 morning 10:48:14 bh: evening! 10:51:38 kilobyte: what side of the item destruction holy war are you on? :) 10:55:04 I don't know. I strongly dislike removing challenges: death is annoying, most games have it non-fatal, people dislike it... But on the other hand, forcing stashing is a good argument. 10:55:36 we can murder the player in other ways! :) 10:55:37 so at this point, I'm not voting for either side 10:56:05 well, but this is one of rare permanent but recoverable losses 10:58:13 <|amethyst> what about the 52-item limite 10:58:18 <|amethyst> s/e$/?/ 10:58:25 <|amethyst> as far as forcing stashing goes 10:58:57 with item destruction gone, the 52 item limit will hurt a lot more 10:59:11 as scrolls are among the lightest items 10:59:49 -!- RZX has quit [Ping timeout: 252 seconds] 11:00:02 -!- randomizr is now known as RZX 11:00:04 players will still have to stash basically 11:00:41 with weak mages i'm usually thinking more about the weight limit than about actually trying to not carry many consumables anyway 11:01:32 tome has a lot of status effects that block certain abilities while active, it works a lot better than item destruction 11:01:55 item destruction is disproportionately irritating to players, compared to the gameplay benefits 11:02:42 <|amethyst> what to do about jiyva? 11:03:13 presumably jiyva would still block the destruction on use 11:03:17 Eronarn: a lot of players don't actually find irem destruction that irritating 11:03:26 <|amethyst> I mean destruction of floor items? 11:03:29 <|amethyst> s/?// 11:03:35 mumra: a lot do tho 11:04:12 Jiyva only blocks food destruction, I thought 11:04:19 no, all item destruction 11:04:47 I guess I misread the code. The message is "Jiyva protects some of your delectables from destruction." Given that jellies eat everything, I suppose that makes sense 11:04:57 it's there only to make up for being unable to stash 11:05:01 <|amethyst> I mean, if item destruction should be removed because it annoys people, shouldn't jellies stop eating items because that annoys people? 11:05:23 <|amethyst> and if that happens, what's the point of jiyva 11:05:25 huh? that makes no sense 11:05:28 |amethyst: stuff like jelly drop should be removed :) 11:05:35 |amethyst: no. They're different cases 11:05:37 Eronarn: why? 11:05:49 If a Jelly eats an item, it's because you didn't kill it 11:05:53 -!- kekekela has quit [Ping timeout: 250 seconds] 11:06:01 <|amethyst> or because it spawned 11:06:03 If an orc wizard incinerates your scroll of acquirement, it's because you were a dope 11:06:08 |amethyst: not on Lair:2 11:06:12 <|amethyst> err 11:06:15 kilobyte: non-local monsters doing harmful things to you sucks, in crawl's context 11:06:30 <|amethyst> bh: I thought we wanted to discourage stashing? 11:06:50 <|amethyst> and "stash on Lair:2" is quite spoilery 11:07:13 "stash on the floor" isn't very spoilery and is just as good 11:07:32 <|amethyst> MarvinPA: but people don't do that because "what if a jelly comes by" 11:07:35 jellies, wandering monsters 11:07:36 bh: jellies eat items on floors other than Lair:2 11:07:51 <|amethyst> someone was complaining just a week ago about a jelly getting into their Lair:1 stash 11:08:17 Why don't jellies spawn on lair:2 btw? 11:08:26 kryft: so that you can stash there! 11:08:26 wands, potions, weapons, armour, throwables, some scrolls, jewelry, rods, ... 11:08:29 <|amethyst> kryft: because they're not a lair monster 11:08:42 kryft: on Lair:1 they can come down the stairs from D 11:09:02 Oh, ok, so actually it's lair:x where x>1 11:09:19 <|amethyst> kryft: >1 and < wherever the Slime entrance is 11:09:27 Right, of course 11:09:31 -!- bh_ has joined ##crawl-dev 11:09:35 -!- dcss29681 has quit [Ping timeout: 240 seconds] 11:09:35 != the Slime entrance actually 11:09:42 <|amethyst> yeah 11:09:48 -!- bh_ has quit [Client Quit] 11:09:57 mumra: So can they come down the stairs while you're not on lair:1 11:10:17 <|amethyst> kryft: "coming down the stairs" really means "random spawn from the previous level" 11:10:19 also not the Shoals / Snake entry, as item-using monsters can loot it 11:10:33 -!- blabber has joined ##crawl-dev 11:10:47 -!- bh_ has joined ##crawl-dev 11:10:48 kryft: no, monsters don't get to act when you're not on the level 11:10:51 -!- bh has quit [] 11:10:55 <|amethyst> otherwise things can only take stairs when they follow you 11:10:55 All right 11:10:59 -!- bh_ is now known as bh 11:11:06 hrm 11:11:10 there we go. 11:11:11 I guess that explains why it's pretty unlikely to get your stash eaten by a jelly on lair:1 11:11:30 <|amethyst> yeah, you'd have to have a jelly spawn at the stairs and then wander over to your stash 11:11:39 it's pretty unlikely to get your stash eaten on any floor if you put it near some upstairs 11:11:43 <|amethyst> probably even less likely now that there's only one stair 11:11:48 most folks put the stash right on the stairs 11:11:55 since you'll never spend more turns on that floor than you're at your stash 11:12:28 <|amethyst> mumra: what if you're coming at it from below? 11:12:32 oh. I did find a small stairs bug. If you worship Fedhas, spawn an Oklob on some stairs. Then traverse the stairs into the level. The oklob gets shoved 11:12:36 -!- bonghitz has quit [Remote host closed the connection] 11:12:45 <|amethyst> is that a bug? 11:13:00 |amethyst: put it on the downstairs of course 11:13:18 bh: Can also happen with plants randomly spawning on stairs. 11:13:26 I *think* it's a bug. Fedhasites can occupy the same square 11:13:30 <|amethyst> ahh 11:13:30 At least it used to, guess some of that stuff was changed. 11:13:34 bh: Oh right. 11:14:33 <|amethyst> the random spawn bit might be more difficult to handle than the deliberately-placed oklob case 11:14:42 <|amethyst> since in the former case the plant might not be friendly yet 11:15:34 What's the point of converting plants and slimes when you see them? To stop them from fighting things off screen? 11:15:55 -!- Flex has quit [Ping timeout: 240 seconds] 11:17:01 sounds like a sensible reason 11:17:04 it's nice flavour too 11:17:12 -!- bmfx has quit [Ping timeout: 252 seconds] 11:17:12 it's problematic with holies 11:17:18 it would be nice if we could make it not interrupt travel 11:18:08 -!- ahahaha has quit [Quit: Page closed] 11:29:00 -!- ussdefiant has joined ##crawl-dev 11:36:49 -!- Flex has quit [Ping timeout: 256 seconds] 11:39:41 kilobyte, is there any decision on getting dwant species into trunk? everything you brought up on mantis has been fixed 11:40:32 Moth of Nihilism: Blocks all god effects. Ugh! 11:40:53 silence almost does this already, and its good enough 11:41:01 pretty soon players will be begging for a scroll of genocide moths 11:41:03 <|amethyst> kilobyte: since I don't think you were around for the discussion yesterday: what do you think about allowing summons to attack out of LOS again (but possibly having them expire more quickly) 11:41:52 |amethyst: That sounds like an improvement over the current state of affairs 11:41:56 instead of tweaking the expiration, they could also do less damage or become weaker as their distance out of los increases 11:42:11 the problem with increasing expiration time, is some spells are super short, others last a long time 11:42:35 overall though, the AI for temporary summons needs to be better with this rule too 11:42:48 expiring summons out of LOS has a few (possibly interesting) consequences: los-blocking clouds can suddenly decimate your allies; and moving around certain types of layout can put them out of LOS as well 11:43:40 what about a combination of LOS and distance? 11:43:51 when I say LOS I mean LOS radius 11:44:03 you can fire conj spells through clouds, and this is fine 11:44:16 but not thru architecture 11:44:27 you can bounce a lightning bolt 11:44:33 currently if you go round a corner and this puts your summons out of LOS, they won't attack things 11:45:02 becoming useless because it stepped around a pillar sucks though 11:45:36 but you're right that corners etc could be abused 11:45:36 it makes total sense that "if you can't see it you can't control it" 11:46:00 well it should still ignore clouds and temporary stuff like that, I think 11:46:06 as far as i'm aware, the LOS changes had very little to do with balance anyway 11:46:20 it's more about making sure that interesting things are happening where you can see them 11:46:42 and the summons cap doesn't solve that situation for LOS 11:46:46 that was part of it, but it was a massive balance change 11:47:23 I'm going to assume balance was meant to be affected. If that's an unintended side effect, it was a huge one 11:47:30 <|amethyst> why weren't permanent allies affected? 11:47:49 it was fixing a particular bit of brokenness but i don't know if balance was changed all that much by it 11:47:53 <|amethyst> seems like only stopping summons from attacking out of LOS was an explicit summoning balance nerf 11:48:12 perma-allies are harder to get, tend to be tied to other resources 11:48:19 I think that's why 11:48:19 ok well the reason "it was to stop interesting things happening where you can't see them" was referenced many times 11:49:03 ...what if summons gave you detection within LOS? 11:49:39 <|amethyst> evilmike: I don't know about harder: one class starts with them 11:50:24 basically i think it's good that you can always see what your summons are doing, if they're doing anything interesting; so i would rather keep that but tweak their power and usefulness if it needs a buff 11:50:33 so, suppose your LOS is blocked by a cloud, but your summon can see something that's in _your_ LOS radius, allow you to detect it? That could reduce grousing about the summoning nerf 11:50:52 |amethyst: they are harder in that you can't exactly pull them up on demand. They're either gifts, or you convert monsters to your side, or they cost piety 11:50:54 Dragon: Hey Boss, I'm killing something over here, a little help! 11:51:01 Or they require corpses 11:51:58 <|amethyst> but what does that have to do with interesting things you can't see? 11:52:07 bh: if we're going with "summoning is some kind of mind link with the summon", maybe seeing through their eyes is justifiable 11:52:24 <|amethyst> I'm just wondering what are the real reasons for the LOS nerf and what are the reasons we tell players to make them stop complaining 11:52:25 mumra: I don't think it matters. While I think it's a huge balance-related issue, my complaint is just that it's an unfun game mechanic 11:52:28 mumra: extending LOS would be way bad 11:52:48 bh: only within your LOS radius 11:52:55 bh: but where blocked by clouds or geometry 11:53:11 bh: which is what i thought you were saying 11:53:14 -!- eith has joined ##crawl-dev 11:53:18 some weird thing where you can see what your summons see wouldn't help this either 11:53:20 <+bh> Moth of Nihilism: Blocks all god effects. Ugh! 11:53:22 please, yes 11:53:36 something with this effect would be great 11:53:40 yep, but detection rather than actual viewing, since it might be confusing "Your Dragon can see it, but you can't shoot it" 11:53:59 why would you want to shoot it if you have summon dragon? 11:54:04 <|amethyst> Eronarn: would also be a pain to implement 11:54:13 bh: well i was thinking just the immediately adjacent squares, maybe for only 1 specific summon type 11:54:15 and we already have silent spectres 11:54:24 It's not even about being able to see stuff, it's about summons becoming brain-dead at a certain distance, and the AI doesn't compensate for this, so it's annoying to manage 11:54:35 |amethyst: hasn't stopped suppression 11:54:37 I think this whole "it's more interesting the way it is now" justification is weak 11:54:40 Eronarn, |amethyst: is the joke that theres a patch on mantis for it 11:54:44 |amethyst: buppy implemented it, and it was a pain by the looks of it, it's an 85k patch 11:54:49 -!- RZX has quit [Ping timeout: 252 seconds] 11:54:51 <|amethyst> oh 11:54:54 buppy is my new hero 11:54:57 -!- randomizr is now known as RZX 11:55:14 -!- Kalir has quit [Changing host] 11:55:18 speaking of patches, how are lorcs going? 11:55:28 evilmike: things dying where you can't see them is uninteresting, do you disagree with this? 11:55:47 yes 11:55:52 Eronarn: well mine is going like this: 11:55:57 !lm . 11:55:57 1163. [2013-06-30 03:29:17] SamB the Axe Maniac (L17 LOBe) left the Swamp on turn 39268. (Swamp:1) 11:56:00 Here's what I do agree with: things dying when you can't see them is SOMETIMES uninteresting 11:56:10 SamB: i mean more like how is their balance, do people like them, are they fun 11:56:30 !lm . x=runes 11:56:31 No milestones for SamB. 11:56:36 !lm . rune 11:56:37 1. [2013-06-30 03:25:24] SamB the Axe Maniac (L17 LOBe) found a gossamer rune of Zot on turn 39071. (Spider:5) 11:57:17 <|amethyst> mumra: I think the bigger thing is "with no risk to the player" more than "where you can't see them" 11:57:20 the balance is not too great I suppose, but I'm so lousy that makes it more fun for me 11:57:21 if your summon is way across the other side of the level (bonus points if you abused disperal), it's uninteresting. If it's just a few spaces off screen and engaged with a monster that you were fighting, then I have no issue 11:57:36 Eronarn: Lava orcs? I see lots of people playing them and having fun, but many consider their aptitudes too strong and their heat mechanic way too abusable 11:57:37 Now, even if it's a few spaces out of view, it's abusable. But this is an AI problem 11:57:44 and making summons less fun to paper over some AI issue is a bad idea 11:57:47 mumra: it's not fun when it's *YOUR* things dying where you can't see them either 11:58:18 buppy: i have considered dropping their melee apts below hill orcs but people would scream murder... 11:58:29 the problem is, if it's allowed even just a little bit, it's always optimal to engineer things so this happens 11:58:32 IMO what summons need is to be more like transmutations. each summon spell should actually be usable in different situations 11:58:39 there is a patch for the heat thing, but nobody's even tried it out afaik :( 11:59:10 buppy: tmut isn't a great example of a school things should be more like 11:59:53 imo what summons need is to play more like tower defense and less like 555555555555 11:59:56 buppy: yes, i've been trying to think about summons that would have more varied usage, and there was some really good brainstorming here yesterday 12:00:24 at the very least, summons should NOT be sitting ducks against enemies out of LoS 12:00:38 they should, like, try to pull the enemies in 12:00:49 i think the right approach is to make them MP investments rather than delayed effect conj spells 12:00:52 they should retreat towards you or something yeah 12:00:55 they should try to run towards you as quickly as possible to get with LOS 12:00:57 yeah 12:01:02 and what about ranged enemies 12:01:11 I guess fleeing isn't enough, there'd need to be a new behaviour that pathfinds towards the player 12:01:30 mumra: any ideas that stuck out? 12:01:30 SamB: what about ranged enemies? 12:01:31 -!- Nexos has quit [Ping timeout: 260 seconds] 12:01:35 SamB: ta 12:01:42 evilmike: it'd be cool to have a magical tether than shows the fastest route between your summon and you, and vice versa 12:01:52 -!- Thyme has joined ##crawl-dev 12:01:52 Eronarn: sounds ugly 12:01:57 well it seems like they're a problem even if your summon is right next to you and has a ranged attack 12:02:08 if ranged enemies are shooting at your summons, and your summons are unable to retaliate, then they should retreat to safety, I guess 12:02:21 mumra: just halofloor? i think that'd be fine 12:02:25 i don't understand 12:02:31 -!- RZX has quit [Ping timeout: 252 seconds] 12:02:42 SamB: can't you just tell your summon to attack? or get closer to the ranged enemy yourself? 12:02:43 -!- randomizr is now known as RZX 12:02:51 -!- Silurio1 has quit [Ping timeout: 252 seconds] 12:02:55 Eronarn: what if there are other halos 12:03:09 mumra: this is the problem with all halos, yes 12:03:11 <|amethyst> I could see a key to show that 12:03:12 mumra: I suppose you could sometimes move in 12:03:20 <|amethyst> but not having it active all the time 12:03:32 but you unfortunately can't make your summon attack something you can't see 12:03:33 Eronarn: i don't know if covering the screen in halos every time you have summons out would be a good thing 12:03:44 that's because it's an insanely bad idea 12:03:48 SamB: no, so you have to get where you can see it 12:03:48 mumra: that would be a bad thing 12:03:52 how did you not know it would be 12:04:00 mumra: and that seems like a problem 12:04:06 SamB: i was being tactful 12:04:08 a key would be fine 12:04:27 SamB: so you mean when something ranged is attacking your summon but not you? 12:04:30 just run away... 12:04:34 of course, if no summons have ranged attacks anyway it's no big deal 12:04:43 and there is a 't' command to tell your summons to retreat 12:04:54 our one level 9 summon spell has ranged attacks 12:05:00 don't they go a bit far though when you do that? 12:05:06 <|amethyst> level 2 summon has ranged attacks 12:05:14 yeah that too :P 12:05:17 hmm 12:05:36 more ranged summons is a good idea 12:05:53 anyway sure: the AI could make them move away from the attacker, if they can't see it to attack it themselves 12:05:55 I think it would be okay to let summons attack one cell beyond LoS? 12:06:04 ranged only, possibly 12:06:18 I odn't know 12:06:24 <|amethyst> BTW, I thought of a fatal flaw with my "summon self illusion" idea... your self-illusion is almost certain to be a summoner 12:06:35 <|amethyst> so it just turns into cap-avoiding summon spam 12:06:44 |amethyst: this is why it should be a tmut 'clone self' spell instead :) 12:06:45 -!- gammafunk has joined ##crawl-dev 12:07:03 SamB: i thought we were trying to eliminate ways of killing things outside LOS (e.g. cloud nerf) 12:07:30 I guess the "run out of LoS of things the player can't see" idea works too 12:07:34 <|amethyst> one big reason for the cloud nerf is that monster are kind of stupid when it comes to LOS clouds 12:07:39 <|amethyst> err 12:07:43 <|amethyst> out-of-LOS clouds 12:08:15 the main reason stationary monsters are bad is because there are a (very small now) number of ways to abuse LOS to kill them 12:08:55 <|amethyst> mumra: well, even ignoring that, they also have to have a summon, max-range conjuration, or blink other close 12:08:59 but LCS was nerfed to prevent kills outside LOS too 12:09:07 and i'm sure other stuff 12:09:12 <|amethyst> lightning bolt was not 12:09:25 LCS? Or LRD? 12:09:36 sorry LRD yeah 12:09:39 <|amethyst> well 12:09:44 oh, does LRD respect wall surfaces now or something 12:09:47 lightning bolt _was_ nerfed 12:09:49 <|amethyst> "on the other side of a wall" and "out of LOS" are different things 12:09:51 it use to bounce more 12:10:15 <|amethyst> mumra: that affected in-LOS things more 12:10:18 |amethyst: but "on the other side of a wall" is the same with summons too 12:10:47 does scrying have any effect on summon behavior? 12:10:50 |amethyst: i'm pretty sure one of the reasons for lightning nerf was also due to out-of-LOS stuff too (but it was a long while ago now) 12:11:03 ps: please make monster summons work like this too 12:11:14 <|amethyst> %git 058394c 12:11:14 07kilobyte * 0.9-a1-867-g058394c: Kludge beams to never hit the same person more than twice. 10(2 years ago, 2 files, 17+ 0-) 13http://s-z.org/neil/git/?p=crawl.git;a=commitdiff;h=058394ca084f 12:11:31 |amethyst> so it just turns into cap-avoiding summon spam 12:11:32 nice 12:11:36 thats like a boggart bomb 12:12:19 what was that quote in relation to? 12:12:29 <|amethyst> mumra: summon self illusion 12:12:39 haha 12:12:40 btw, the cloud nerf isn't actually in trunk yet, right? 12:12:41 <|amethyst> I once got a blue death that cast shadow creatures and got a boggart 12:12:50 cloud nerf isn't coded yet no 12:12:52 because the idea wasn't to COMPLETELY prevetn clouds out of LOS, just to weaken them 12:12:59 I think the proposal is to make them disappear faster 12:13:10 <|amethyst> hm 12:13:18 or maybe they won't spread as much out of your vision, but they can still exist for a bit still 12:13:36 <|amethyst> this is intended primarily for fcloud/pcloud/fire storm, right? 12:13:47 yeah i think it was to dissipate them quickly 12:13:51 <|amethyst> because it also reduces the saftey buffer from conjure flames 12:13:59 they should disappear pretty fast otherwise it's kinda pointless 12:13:59 <|amethyst> s/ftey/fety/ 12:14:18 -!- ussdefiant has quit [Ping timeout: 246 seconds] 12:14:26 if you can still kite something through poison while keeping it out of los then the change doesn't achieve anything 12:14:34 <|amethyst> e.g. it would be difficult for a slow character to use CF to get away from an orc priest 12:14:50 they were supposed to disappear quickly, yeah 12:14:59 just not magically disappear instantly 12:15:08 so you couldn't just step back, step forward to somehow remove a cloud 12:15:46 <|amethyst> would this be all clouds, or player-created ones? 12:16:25 <|amethyst> and do player-started forest fires count as player-created? 12:16:34 <|amethyst> (after they spread) 12:16:46 sounds like it might be simpler to just track magical clouds from actors 12:16:56 <|amethyst> we do track actors I think 12:17:04 oh right, for determining killers 12:17:34 <|amethyst> We keep a kill_category, killer_type, and mid_t 12:17:48 ??mid_t 12:17:49 I don't have a page labeled mid_t in my learndb. 12:17:59 <|amethyst> bh: globally unique monster ID 12:18:36 <|amethyst> bh: (unlike mindex, which is only unique at a given point in time on a given floor) 12:18:38 gotcha. Seems like it should be straight forward to ask if a cloud is visible to its creator and make it decay faster if it isn't 12:18:39 -!- RZX has quit [Ping timeout: 252 seconds] 12:18:43 -!- randomizr is now known as RZX 12:18:59 <|amethyst> not sure if that's a good idea 12:19:10 <|amethyst> e.g. steam dragons 12:19:35 <|amethyst> I guess that's the only attack cloud that blocks LOS? 12:20:18 and in the worst case, it'll just blast you again! 12:20:31 <|amethyst> oh, catoblepas breath too 12:20:43 <|amethyst> and holy flames 12:24:23 -!- ackack has quit [Ping timeout: 260 seconds] 12:26:52 clouds pose the biggest problem for diffusion pathfinding, since you need to run a search for every cloud type because mobs have different resists 12:27:09 |amethyst: the idea was just for player spells btw 12:27:59 monsters really don't matter here, they should ignore these rules 12:29:37 https://crawl.develz.org/mantis/view.php?id=5548 12:30:17 as you can see the proposal isn't that fancy. if you want to go ahead with it, i'd keep it simple like it is here 12:31:17 i also dont think the "move clouds" bit is necessary there 12:31:30 <|amethyst> evilmike: player s*spells* specifically? 12:31:40 <|amethyst> evilmike: because forest fires are still an issue 12:31:41 the proposal was only for poison cloud and freezing cloud 12:31:59 <|amethyst> evilmike: granted, there are monster spells to somewhat mitigate that problem 12:32:12 forest fires are harder to control and you have a finite number of trees. i dont feel like they are THAT abusable inherently 12:32:43 -!- anidude has quit [Read error: Connection reset by peer] 12:32:46 <|amethyst> the enchanted forest lair end is kind of trivializable if you have a level 3 spell 12:32:54 but note the proposal doesn't cover conjure flame, or fog scrolls, or draconian steam breath, that was intentional 12:33:28 hm, well, I think it's just a problem with forest fires themselves. I wouldn't like it if the fire magically depends on anything to do with you 12:33:38 it's a forest fire, it spreads on its own 12:33:43 <|amethyst> I agree with that 12:34:04 <|amethyst> maybe that lair end just needs to get a monster with (what was the spell name?) 12:34:09 <|amethyst> control winds? 12:34:15 that new? never seen it 12:34:29 spriggan air mage 12:34:36 spriggan air mage (11i) | Spd: 16 | HD: 14 | HP: 32-52 | AC/EV: 1/25 | Dam: 16 | 10weapons, 10items, 10doors, spellcaster, see invisible, lev, DMsl | Res: 06magic(130), 10elec++ | Chunks: 07contam | XP: 1467 | Sp: airstrike (0-38), b.lightning (3d19), control winds, shock (d16) | Sz: little | Int: high. 12:34:36 <|amethyst> %??spriggan air mage 12:34:55 <|amethyst> well, that might be a little tough for Lair:$, but maybe give it to the young druids? 12:36:06 <|amethyst> oh hey 12:36:15 <|amethyst> 5548 proposes that spell :) 12:36:21 <|amethyst> also giving it to elves 12:36:59 -!- rossi has quit [Ping timeout: 256 seconds] 12:37:35 yes, one of the elves should get it too 12:37:48 it's not all that flavorful for underground elves 12:37:56 what about a different spell that also protects against clouds 12:38:22 <|amethyst> eh, deep elves don't exactly suck at air magic 12:38:25 what's wrong with gassing elf:3? It's fun 12:38:38 |amethyst: they're good at pretty much all magic though 12:38:44 !apt de 12:38:45 DE: Fighting: -2*, Short: 0, Long: -1, Axes: -2, Maces: -3*, Polearms: -3*, Staves: 0, Slings: -2, Bows: 1, Xbows: -1, Throw: 1, Armour: -2, Dodge: 2, Stealth: 3, Stab: 1, Shields: -2, Traps: 0, UC: -2*, Splcast: 3!, Conj: 1, Hexes: 3, Charms: 4!, Summ: 1, Nec: 2!, Tloc: 1, Tmut: 1, Fire: 1, Ice: 1, Air: 0, Earth: 1, Poison: 1, Inv: 1, Evo: 2, Exp: -1, HP: -2, MP: 3! 12:38:46 it's a gas, you might say 12:38:50 I always thought the whole point of elf was to toss a bunch of glass cannons at you, and let you figure out the best way to kill them 12:39:01 <|amethyst> !apt sp air 12:39:01 one of the common endings even gives you a moat 12:39:02 Sp (SK_AIR_MAGIC)=-1 12:39:05 <|amethyst> !apt de air 12:39:06 DE (SK_AIR_MAGIC)=0 12:39:12 Which should only really say that Elf should be a rainbow branch of whatsgoingon. 12:39:37 crawl has a distinct lack of rainbow spells 12:39:46 <|amethyst> Eronarn: there's one now! 12:39:59 evilmike: it wouldn't make clouds ineffective against them, and the challenge becomes to kill the dudes that cast control winds first 12:40:36 |amethyst: which one? 12:40:43 <|amethyst> Fulminant Prism 12:40:56 oh right, yeah 12:41:12 it doesn't look very rainbow-like in tiles tho ;P 12:41:19 <|amethyst> I mean, I guess it's not explicitly rainbow-flavoured 12:41:26 <|amethyst> but it has "prism" right in the name :) 12:41:33 <|amethyst> maybe the explosion could be rainbow 12:41:55 or it could send out little 1-tile dazzlnig rainbow rays while its charging up 12:42:15 <|amethyst> dazzling spray could be rainbow-coloured too :) 12:42:46 rename conjurer "rainbow warrior", reflavour all spells 12:43:09 <|amethyst> mumra: give them Enslavement too 12:43:13 <|amethyst> Care Bear Stare! 12:43:20 haha 12:44:26 Couldn't the trees in forest just be made into magical, 'unburnable' trees? Or at least trees where the fire didn't spread nearly as well 12:44:27 gammafunk: You have 1 message. Use !messages to read it. 12:44:30 jump -> gummiberry juice 12:44:31 !messages 12:44:32 (1/1) |amethyst said (2h 16m 1s ago): re the tutorial thing, that's in Mantis... it's because crawl puts tutorial and regular saves in the same directory 12:44:33 <|amethyst> it even fits their second scool 12:44:46 <|amethyst> gammafunk: if anything that makes more sense in Swamp 12:44:59 <|amethyst> since there the roots are waterlogged 12:45:24 |amethyst: I see that too, but if the aim is prevent forest fire abuse, we don't need much justification other than MAGIC! 12:45:35 spriggan firefighters 12:45:38 -!- Duralumin has quit [Read error: Connection reset by peer] 12:45:58 <|amethyst> we were just talking about those 12:45:59 bh: Like keystone kops? 12:46:09 spriggan air mage (11i) | Spd: 16 | HD: 14 | HP: 32-52 | AC/EV: 1/25 | Dam: 16 | 10weapons, 10items, 10doors, spellcaster, see invisible, lev, DMsl | Res: 06magic(130), 10elec++ | Chunks: 07contam | XP: 1467 | Sp: airstrike (0-38), b.lightning (3d19), control winds, shock (d16) | Sz: little | Int: high. 12:46:09 <|amethyst> %??spriggan air mage 12:46:14 They just appear if you start a forest fire, hit you with cream pies 12:46:15 keystone kaptains 12:46:21 apocalypse crabs already shoot ranbows at you 12:46:35 <|amethyst> BTW, is it possible with lua and map markers to make a single tree that can't burn 12:46:51 !lg * ckiller=apocalypse_crab 12:46:52 52. scummos the Cleaver (L8 LOAK), worshipper of Lugonu, slain by an apocalypse crab (summoned by Lugonu's corruption) on D:7 on 2013-06-30 16:48:33, with 1937 points after 6264 turns and 0:13:35. 12:46:55 <|amethyst> because a permatree with a 1-square flame generator would be a cool flavour vault for somewhere 12:46:56 !lg * ckiller=apocalypse_crab max=xl 12:46:57 52. coolrobin the Slayer (L24 MiEn), worshipper of Okawaru, slain by an apocalypse crab on Abyss:2 on 2013-04-16 11:40:51, with 474945 points after 55816 turns and 3:26:32. 12:47:23 |amethyst: 12:47:25 ??burning bush 12:47:26 burning bush[1/1]: Bush that throws flame, in Zot defence. Sees invisible. Also in sprint IV. Also in a Zot stairs vault. 12:47:42 <|amethyst> yeah, but those attack you :) 12:48:48 <|amethyst> btw, now that treants are in that means we can implement deep elm fire elementalists without making them immobile 12:49:02 did cleaving get nerfed recently? meatsprint feels harder 12:49:19 -!- ystael has quit [Ping timeout: 256 seconds] 12:49:32 <|amethyst> no 12:49:59 <|amethyst> maybe the monster turn-taking behaviour? though I wouldn't think that would be very noticeable 12:51:27 |amethyst: deep……elm? 12:51:38 meatsprint is so start scummy. Depending on where the monsters spawn you can memorize blink and/or passage of golubria 12:51:55 <|amethyst> gammafunk: yeah, a race of spellcasting trees led by the ancient wizard Elmminster 12:52:42 <|amethyst> gammafunk: (Elms were a fake player race on badwiki) 12:53:02 They're fake? 12:53:08 Did someone actually proposed a species called 'Elms' 12:53:15 <|amethyst> they're not there anymore either 12:53:23 I thought elms used to exist 12:53:37 buppy: You would be interested in an Elm species :) 12:53:38 <|amethyst> gammafunk: I don't know the history, but they were written up as though they had existed in an older version of DCSS 12:53:42 |amethyst: there might be... I'm not sure. 12:53:53 You might try something like veto_fire = "veto" or veto_burn = "veto" 12:54:00 you can do this with things like dig, shatter, etc already 12:54:06 I am still mad about elms being removed! 12:54:22 buppy: They were a perfect habitat for dwants! 12:54:58 On a different topic. 12:54:59 dck: I wonder what main stance of devs is on the mines 12:55:17 <|amethyst> re elms: https://crawl.develz.org/tavern/viewtopic.php?f=5&t=5418 12:55:20 I'm also wondering if there are any ideas for Orc. 12:55:45 Bloax: can you reflavour your comments into a question or statement that makes some kind of sense? 12:55:49 of course elms existed 12:56:01 people who say they don't exist are just trying to spread a dumb hoax 12:56:26 mumra: Are there any plans on spicing up Orc to make it less boring and rather bland? 12:56:27 <|amethyst> Not Orc the branch, but I still think we should make Beogh suppress heat aura so we can enable LO^Beogh 12:56:29 is that better 12:56:38 elms are a conspiracy to cover up the removal of grey elves 12:56:49 Bloax: yes, i understand that :) 12:57:12 Bloax: that's what i _thought_ you might have been asking but i felt like i was trying to solve a cryptic crossword puzzle :P 12:57:38 Well all the wordplay is getting out of hand lately. 12:57:47 Because I keep fudging around with prepositions too. 12:59:26 <|amethyst> Re LO^Beogh, would anyone object to me doing just that? I think it avoids the problems of "why doesn't Beogh always protect orcs from fire", because it's an effect e's having on you; and it's not inconsistent with Okawaru and demonic guardian 12:59:45 <|amethyst> Because if we don't do that we need to rename LO 12:59:54 <|amethyst> s/that/something/ 13:00:24 Bloax: anyway i think Orc is pretty good really, Swamp is a much better candidate for spicing up right now 13:00:37 Bloax: we did discuss some different layout ideas but that's about it that i know 13:00:46 Orc is alright. But it just seems pretty bland. 13:01:22 It's four floors of yellow-brown mostly-open layouts with lots of gold. 13:02:04 |amethyst: I want them to be able to use beogh, at least 13:02:35 also about heat aura being a bit abusable, I had a suggestion a while back: increase their speed penalty so they are always slow, except at max temp when they have a move delay of 1.0 13:02:53 this way they can't easily kite like they do now (you lose the move penalty really quickly) 13:03:12 they could use swiftness, but that's what the spell is good for 13:03:27 It's rather silly for Beogh to be the Hill Orc god when you're fighting Cave Orcs who worship him. 13:03:44 And then him not accepting Lava Orcs. 13:04:19 <|amethyst> Bloax: right, no one is happy with the current state of affairs 13:04:26 evilmike: I was thinking the same thing. 13:04:26 evilmike: right now, their temperature increases much faster/to higher amounts than was desired 13:04:44 <|amethyst> Bloax: but there are several ways to solve it (e.g. renaming them to "lava dwarves") 13:04:45 well the speed thing still bugs me for some reason anyway 13:04:48 i have a patch that makes them use exponential decay but it's far too severe, so it needs both code review and also balance fixes 13:04:57 <|amethyst> even if that would make Eronarn cry 13:04:58 and i dont think its a big deal if they stay slow until max temperature, helps balance the apts 13:05:15 I would go with the LO^Beogh route. 13:05:19 the ideal is that in most fights you'd hit normal speed towards the end if at all, and only hard fights get you fast 13:05:20 <|amethyst> Eronarn: where was that patch? 13:05:24 Because it's silly to have a god for only one specific race. 13:05:33 |amethyst: on my github - it's a megapatchrebasething so it's not really suitable for a mantis patch yet 13:05:38 <|amethyst> Eronarn: last I looked at it I had serious issues with the code complexity I think 13:05:58 <|amethyst> Eronarn: is it at all possible to do the decay change separately? 13:06:01 absolutely 13:06:05 <|amethyst> Eronarn: it would be a lot easier to evaluate that way 13:06:09 but also it fixes some other bugs 13:06:32 -!- bonghitz has quit [Remote host closed the connection] 13:06:40 <|amethyst> those separately too :) 13:07:06 since we happened to be talking about patches, has anyone had a chance to look at my jump patch yet? 13:07:10 <|amethyst> unless some of those are intimately tied with the changes in code structure 13:07:33 I updated the mantis a few days ago with a slight bugfix, and pointed out the 'jumping into dangerous clouds' issue 13:08:51 why is temperature automatic in the first place? using tension seems really unreliable 13:09:12 |amethyst: what're the structural issues you know of? most of the added complexity comes from better stubbing for monster temperature, which should happen eventually 13:09:46 well, and the split into temperature.cc, but that was something i did because it was requested... 13:10:49 i don't mind putting in more work on lorcs but i want to make sure it'll be something reasonably likely to get merged since i'm very time-limited nowadays 13:11:22 <|amethyst> Eronarn: I don't recall now, it was several weeks ago 13:12:19 <|amethyst> Eronarn: ISTR some of it was which things were implemented in which classes 13:13:35 -!- Arendeth has joined ##crawl-dev 13:14:55 ooh, swamp burning down 13:15:29 when compiling with cpu set to 64 bit release 13:15:38 i get Error 118 error C1083: Cannot open include file: 'png.h': No such file or directory C:\engines\Crawl-Git\crawl\crawl-ref\source\contrib\sdl-image\IMG_png.c 72 1 SDL_image 13:15:38 13:16:06 |amethyst: can you take a look at the current status of temperature.cc? it's less than 500 lines https://github.com/Eronarn/Crawling-Chaos/blob/7ae81a44037c76f9c3f6486f27dc6f0cc246b2f7/crawl-ref/source/temperature.cc 13:17:53 <|amethyst> Eronarn: for one, the player:: methods should probably not say "you." 13:18:35 <|amethyst> not that that matters in the usual case, but one could imagine a situation where there is a temporary copy of the player 13:20:34 <|amethyst> Eronarn: second, doesn't this calculate temperature for non-LO players? 13:20:51 <|amethyst> Eronarn: I guess that's not much extra calculation really 13:20:52 even placing that header file in sdl-image doesn't prevent that error 13:21:10 I saw 'you' being used a lot in player member functions while coding nihilist moth 13:21:15 at least I think I did 13:21:39 -!- Thyme has quit [Ping timeout: 240 seconds] 13:21:43 |amethyst: they have a temperature variable but it doesn't actually do calculation except on each player turn 13:21:48 and only if you have temperature 13:22:00 (unless I botched something) 13:22:11 <|amethyst> Eronarn: The other thing is, I'm not sure about putting all this stuff into actor when it's just stubs 13:22:25 <|amethyst> I guess the idea is that it can become not a stub 13:22:27 fleshlord (13@) | Spd: 10 | HD: 60 | HP: 3000 | AC/EV: 0/10 | Dam: 26 | 10weapons, 10items, 10doors, fighter, spellcaster, 07vault | Res: 06magic(240), 05hellfire | Chunks: 07contam | XP: 15000 | Sp: slow, haste | Sz: Medium | Int: high. 13:22:27 %??fleshlord 13:22:58 |amethyst: well, i actively want there to be monster lorcs, but i'm not sure on the timeline 13:23:42 most of the effects are easy to translate, but figuring out when they should be high temp is harder 13:23:59 anyone know how i can fix that error? 13:24:04 <|amethyst> Eronarn: I think it would be better to move those stubs from actor to monster, then, and to make the actor methods abstract: virtual float get_raw_current_temperature() const = 0; 13:24:09 also, where to store it 13:24:20 <|amethyst> Eronarn: could be stored in a prop 13:24:32 <|amethyst> Eronarn: we definitely don't want an extra field for every monster 13:24:45 number......... :) 13:24:49 (this is a bad idea) 13:25:00 <|amethyst> not terrible actually 13:25:20 <|amethyst> but a prop would probably be better 13:25:25 |amethyst: i'm just envisioning something like a hydra polying into a superheated lava orc 13:25:59 couldn't you just give them ring of flames as their espace spell? 13:26:13 buppy: the heat aura/passive heat isn't a problem at all, monsters can have auras 13:26:32 <|amethyst> heat aura is a problem for monster AI 13:26:56 i don't anticipate that one being as bad since mostly lorcs will show up near other lorcs 13:27:02 <|amethyst> true 13:27:06 <|amethyst> or other lava monsters 13:27:46 now that looks like a rune 13:28:01 !learn add lorc_todo remove 'you.' from player methods 13:28:02 lorc todo[6/6]: remove 'you.' from player methods 13:28:25 !learn add lorc_todo move stubs from actor to monster and make the actor methods abstract 13:28:25 lorc todo[7/7]: move stubs from actor to monster and make the actor methods abstract 13:28:27 Eronarn: eh? 13:28:34 oh 13:28:58 SamB: if you have any code review on temperature.cc that would be swell! 13:29:04 got it to compiled needed to add libpng's folder to the include field of SDL-image 13:29:06 you mean you want to call this-> instead of you. 13:29:13 or something? 13:29:16 https://github.com/Eronarn/Crawling-Chaos/blob/lorc_rebase_bh_wip/crawl-ref/source/temperature.cc 13:29:24 isn't this implicit 13:29:52 it *can* be 13:29:52 though oddly enough it compiles with no changes needed on the defualt settings 13:30:49 well i haven't heard anything really awful, code change wise 13:31:01 <|amethyst> SamB: right 13:31:09 if i squash it, it should be pretty simple to make a patch 13:32:19 -!- Silurio has quit [Ping timeout: 264 seconds] 13:32:31 -!- Nakatomy has quit [Ping timeout: 260 seconds] 13:33:16 <|amethyst> there are also save compat issues 13:33:36 <|amethyst> since old saves have temperature_last, not temperature_upcoming 13:33:56 <|amethyst> could lead to one-turn glitches when transferring 13:34:07 Error 18 error LNK1181: cannot open input file 'freetype.lib' C:\engines\Crawl-Git\crawl\crawl-ref\source\MSVC\LINK tilegen 13:34:38 <|amethyst> Arendeth: might have to wait until mumra shows up again 13:34:57 <|amethyst> Arendeth: I don't know if anyone else has built with msvc 13:35:18 <|amethyst> did the freetype contrib get built? 13:35:32 |amethyst: it won't be able to load it properly if it has a different name? i assumed it would work fine if it were the same type 13:39:48 hi 13:40:01 Arendeth: i think this comes back to the issue that your contribs were failing to build 13:40:33 ya i deleted the folder i got all the stuff again 13:40:47 but with more success this time 13:40:55 ok have you tried "rebuild all" (in Contribs.sln) 13:41:31 this time doing it under the build for 64 bit option 13:42:00 think i turn of link time generation for that file 13:42:14 well you shouldn't have to change any options 13:42:22 -!- simmarine has joined ##crawl-dev 13:42:50 can i get you to just get a new, clean source tree, and start from an unmodified version, just with all the contribs as we've supplied them, using "git submodule update --init"? 13:43:12 -!- Silurio1 is now known as Silurio 13:43:17 so, why is it safe for us to use the section symbol? is that in all of MS's console codepages? 13:43:38 -!- rchandra has joined ##crawl-dev 13:44:06 <|amethyst> SamB: it's in CP 437 anyway 13:44:11 <|amethyst> as well as Latin-1 13:44:17 yes I can see that 13:44:31 I mean latin1 is easy, just "man latin1" 13:44:49 latin1 is easy, just man up 13:45:19 Error 3 error C1189: #error : The C++ Standard Library forbids macroizing keywords. Enable warning C4005 to find the forbidden macro. C:\Program Files (x86)\Microsoft Visual Studio 11.0\VC\include\xkeycheck.h 242 1 crawl 13:46:06 Arendeth: just try the 32bit build for now 13:46:19 <|amethyst> where's xkeycheck.h ? 13:46:22 -!- ackack has quit [Ping timeout: 246 seconds] 13:47:50 |amethyst: it's part of MS's libraries 13:48:18 to be honest i've only been doing 32bit builds, i can't get contribs to build in 64bit either 13:48:31 and that xkeycheck error comes up trying to build crawl in 64bit 13:48:49 but if the contribs haven't been built in 64bit then i doubt it would work anyway 13:50:21 Arendeth: msvc build is mainly useful for developing and debugging; if you just want to build and play you're probably better off using msysgit, it's simpler 13:50:41 and also more reliable 13:51:03 i prefer msvs anyway plus its good to test out anyway 13:51:14 its seems no problems now 13:51:55 heh in fact, i can't seem to get a release build working even 13:52:02 msvs has additional potions that msysgit doesn't have 13:52:17 -!- ackack has quit [Ping timeout: 248 seconds] 13:52:29 options lol 13:52:44 potions of berserk rage, probably 13:52:51 Arendeth: options like what? 13:53:10 * mumra thinks msysgit has many more options than VS but then it's also a completely different thing 13:53:34 mmm, just like i thought... the exponential is far too severe :/ 13:53:40 Arendeth: msysgit isn't a development environment, it's a build environment 13:53:47 mumra: 64 bit Windows builds work just fine for me 13:53:58 kilobyte: in msysgit? 13:54:04 kilobyte: we're talking about msvc 13:54:07 would ® be a totally stupid default glyph for runes 13:54:15 link time code generation and avx machine code 13:54:36 <|amethyst> SamB: in unicode anyway I think it would be neat 13:54:39 SamB: The runes already 'sparkle' in console, isn't that enough? 13:54:39 Arendeth: what do you need these features for in the context of Crawl? 13:54:40 mumra: no idea about msys, I'm using mingw64 13:54:55 kilobyte: ok, well i'm saying that 64bit msvc builds don't work ;) 13:55:00 gammafunk: does that include the one in Spider? 13:55:23 only if you set cpu to 64 bit 13:55:32 Arendeth: all official Windows builds use link time optimizations for years :p 13:55:44 on options 13:56:05 of course why wouldn't you :P 13:56:07 |amethyst: I think it's in all of the DOS codepages listed in charmap, and it's also obviously in latin 1 13:56:24 Arendeth: builds that aren't done in msvc, so i think you're wrong about that 13:56:29 ¢ is another option 13:56:31 <|amethyst> huh? is it in CP 437? 13:56:47 oh darn 13:56:52 maybe all except that one 13:56:54 SamB: I'd look at the cross-section of CP437 and WGS4 13:56:57 <|amethyst> that's the important one 13:57:06 yeah I know :-( 13:57:12 <|amethyst> also, there's the ascii and dec charsets 13:57:22 SamB: Yes, gossamer does 13:57:30 -!- Silurio has quit [Ping timeout: 268 seconds] 13:57:38 Arendeth: anyway if you have it building, that's good (but the msvc builds are only recently supported so i guarantee they're not as stable as msysgit/mingw, i still hit occasional bugs noticably in wizmode) 13:57:43 "dec", ie, vt100 line-drawing, is not really used anymore 13:57:47 even if the 64 bit cpu option doesn't compile though you can still set the enhance language set to /AVX for a little faster code 13:57:55 well, I meant the default default 13:58:15 I think I might have dec set in my local rcfile for some reason 13:58:24 maybe I liked the glyphs? 13:58:27 ncurses use UTF-8 in all cases for Crawl, because it doesn't work with some terminals like PuTTY 13:58:27 -!- myrmidette has quit [Ping timeout: 256 seconds] 13:58:29 <|amethyst> oh, you don't have to worry about what DOS supports for the default 13:58:38 kilobyte: indeed 13:58:46 yup 2 successful and 1 skipped so I got it to build :P 13:58:46 what's "DOS"? :p 13:58:54 <|amethyst> CP 437 13:58:59 why don't we? 13:59:15 Why is it a 'visual' compiler? Doesn't that just mean it has a fancy IDE? 13:59:19 <|amethyst> because we assume people have unicode fonts 13:59:28 |amethyst: I meant, DOS support has been dropped a few years ago because of toolchain problems, and good riddance 13:59:29 <|amethyst> if they're using charset=unicode 13:59:32 |amethyst: even on Windows? 13:59:42 I mean maybe they have them, but configured? 13:59:53 for the console? 14:00:09 also what exactly is a "unicode font" 14:00:14 I'd say, let's allow a decent set if they set charset=unicode, as it's not the default 14:00:27 what IS the default then 14:00:41 charset=default 14:00:49 <|amethyst> oh, default isn't unicode 14:00:54 <|amethyst> what is default? 14:01:01 "default" 14:01:07 -!- Arkaniad has quit [Ping timeout: 264 seconds] 14:01:24 <|amethyst> I mean, what does "unicode" do differently from "default" 14:01:26 where are the default glyphs assigned? 14:01:29 if (player_in_branch(BRANCH_ABYSS)) 14:01:29 { 14:01:29 if (tension < 2) 14:01:29 tension = 2; 14:01:29 else 14:01:31 um 14:01:34 what is this even 14:01:46 |amethyst: quite a lot, including walls, floor 14:02:03 IMO the inner conditional should be replaced by "tension += 2" 14:02:09 |amethyst: viewchar.cc has the tables 14:02:26 SamB: that's a partially-calculated tension number, not a multiplier 14:02:32 !source viewchar.cc 14:02:48 Lines pasted to http://pastie.org/8097592 14:02:52 <|amethyst> oh 14:02:55 0 * 3/2 tension for abyss works fine, so does 1 * 3/2 14:02:57 no reason at all for that path 14:03:19 ugh, that pastie thingy corrupts UTF-8 14:03:24 kilobyte: OIC, we can has different default for windows console anyway 14:03:46 can't seem to find where it placed the file though 14:04:01 http://sprunge.us/jAWV works (as does $EDITOR viewchar.cc) 14:04:28 SamB: it's already slightly different: ⌂ vs ∆ 14:04:41 -!- bonghitz has quit [Ping timeout: 256 seconds] 14:05:09 kilobyte: yeah, hence the OIC 14:08:12 so would â„‹ be a stupid default elsewhere? 14:08:21 it looks nice to me ;-) 14:08:34 <|amethyst> It's double-width for me 14:08:36 oh hey i found some problems, no wonder the rewrite wasn't working as well as i hoped... 14:08:38 ouch 14:08:45 <|amethyst> and we don't support double-width glyphs 14:09:13 <|amethyst> here goes nothing 14:09:15 |amethyst: I'll take that as indicating that we don't have good font coverage, or that wcwidth for this is actually ambiguous 14:09:39 03|amethyst 07* 0.13-a0-2161-g6885dd5: Give lava orcs bonuses with orcish launchers. 10(27 minutes ago, 1 file, 1+ 1-) 13http://s-z.org/neil/git/?p=crawl.git;a=commitdiff;h=6885dd52b8e1 14:09:39 03|amethyst 07* 0.13-a0-2162-gd41a293: Remove dead code. 10(27 minutes ago, 1 file, 0+ 3-) 13http://s-z.org/neil/git/?p=crawl.git;a=commitdiff;h=d41a293a3000 14:09:39 03|amethyst 07* 0.13-a0-2163-ge100185: Allow Lava Orcs to worship Beogh. 10(24 minutes ago, 5 files, 7+ 14-) 13http://s-z.org/neil/git/?p=crawl.git;a=commitdiff;h=e10018555c94 14:09:39 03|amethyst 07* 0.13-a0-2164-g75d496b: Turn off heat aura for Lava Orcs of Beogh. 10(3 minutes ago, 2 files, 8+ 1-) 13http://s-z.org/neil/git/?p=crawl.git;a=commitdiff;h=75d496b77f15 14:09:41 <|amethyst> oh 14:09:43 <|amethyst> actually now that I look at it 14:11:03 -!- dupo has quit [] 14:12:51 -!- blabber has quit [Quit: leaving] 14:13:57 |amethyst: now that you look at what? 14:14:09 <|amethyst> SamB: that script H 14:14:41 <|amethyst> SamB: it's two cells wide but overlaps with the next character 14:14:41 also what did you realize now that you've looked 14:14:45 ah 14:14:50 <|amethyst> SamB: oh, oops 14:14:56 <|amethyst> SamB: I left out that line 14:15:05 |amethyst: there's quite a few other places that would need to be edited to make them orcs 14:15:06 so, poor font coverage 14:15:07 <|amethyst> SamB: it appears to be double-width in my font but not my locale 14:15:13 or something 14:15:27 <|amethyst> kilobyte: for example? 14:15:41 <|amethyst> kilobyte: I looked over every use of SP_HILL_ORC 14:15:43 it looks nice in unifont ;-) 14:16:03 unifont also has a few chars that are doublewidth and should be single-width 14:16:23 <|amethyst> kilobyte: they do count as different species still for cannibalism purposes 14:16:29 a bunch of speech, for example 14:16:40 notably the block characters from cp437 ... 14:16:41 godspeak.txt, monspeak.txt 14:17:22 SamB: looks double-width for me too, even if it should be single 14:18:09 would £ be stupid or do we use that already? 14:18:21 <|amethyst> kilobyte: ah, I missed Roka 14:18:34 <|amethyst> kilobyte: I did change the idol hill orc to idol orc 14:18:39 <|amethyst> kilobyte: (and the related code) 14:19:49 Remove use of 'you' inside player member functions by pubby 14:20:42 ^I didn't really test it, but it compiled so I assume everything's fine 14:21:05 Also how about an octopode statueform tile: https://dl.dropboxusercontent.com/u/63152810/Octostoner.png 14:21:26 <|amethyst> kilobyte: hm, would it be a problem to make lava orc monsters have species MONS_ORC? 14:21:39 <|amethyst> kilobyte: that, plus one tweak to Roka's speech, should fix most of the remaining issues 14:21:50 Seems a bit silly for it to still be the old one. 14:22:14 <|amethyst> kilobyte: (also, Wiglaf and DDs aren't related) 14:22:15 well, they're not anything like hill orcs species-wise 14:22:22 -!- Thyme has joined ##crawl-dev 14:22:27 sure they are, they're both orcs :) 14:22:29 <|amethyst> kilobyte: but high elves and deep elves are? 14:22:33 -!- Silurio1 has quit [Ping timeout: 252 seconds] 14:22:48 Eronarn: elves and humans are far closer to hill orcs than lava orcs 14:22:55 why do you say that 14:23:19 also i've reconsidered, lava orcs should probably have bad tmut apt too 14:23:25 because beogh 14:24:11 Eronarn: because it's a bit hard to evolve between carbon and silicon base? 14:24:17 Hill Orcs, Humans, and High Elves are all basically the same thing but with different apts/stats and one of them can worship Beogh 14:24:52 like they all fit into the pattern of normal-ish species with a full set of normal humanoid equip slots 14:25:17 "genus" is grammar-based so it makes sense for both to share it 14:26:27 kilobyte: a wizard did it 14:26:58 (remember, in DS, the only justification needed is 'half of a demon did it') 14:28:01 -!- myrmidette has quit [Ping timeout: 248 seconds] 14:28:08 -!- anidude has quit [Remote host closed the connection] 14:28:48 <|amethyst> kilobyte: maybe the "related" tag should be using genus rather than species? 14:28:55 lava orcs not being orcs seems to make more sense 14:30:45 |amethyst: yeah, sounds good 14:31:03 |amethyst: although most uses of "related" probably should go 14:31:08 even with equal races 14:31:17 like that "family" references 14:34:41 <|amethyst> kilobyte: likewise cannibalism? 14:34:53 -!- Hisar has quit [Quit: Page closed] 14:34:58 <|amethyst> kilobyte: and pacifism 14:35:10 <|amethyst> kilobyte: because easiest would be to change is_player_race_same_species to *_genus 14:35:25 <|amethyst> it already checks genus when forms are relevant 14:36:17 yeah, it makes sense for hill dwarves to be punished for eating gully ones 14:37:01 <|amethyst> ohh 14:37:04 <|amethyst> it does 14:37:13 <|amethyst> it already uses genus, except for ghouls 14:37:18 <|amethyst> so just needs renamed :) 14:37:45 interesting 14:38:01 the tension code path has a 2* multiplier for invis enemies, but it never gets called because it backs out if you can't see them before it's called 14:39:11 feel strange to me: frantically looking for an invis foe seems pretty tense 14:39:29 kilobyte: i think the idea is that tension could reveal an invis enemy before you're aware of it 14:39:39 (assuming the enemy got to attack first, which may be problematic for lava orcs but not Xom) 14:39:40 in practice this doesn't work well; sigmund goes invis and suddenly your tension plummets 14:39:46 and tension is all about Xom 14:40:31 well, it's only unseen annoyances and lorocyprocas who get to go invis before you can see them 14:41:04 kilobyte: not necessarily... fighting something offscreen, sky beasts, etc. 14:41:28 can sky beasts even go invis when asleep? 14:41:35 -!- Silurio has quit [Ping timeout: 260 seconds] 14:41:49 also there's the info leak of whether a monster is still in LOS of you, for everything 14:43:36 i wonder about adding a flag that makes a monster tensionable only once you start seeing a disturbance or interacting with it, and reset the flag on some condition - not seeing the monster for > 1 turn? 14:44:33 tension should just be based on the previous turn's tension, so if a something goes invis or an invis monster hits you, get a tension spike: the tension fades over a few turns if nothing else happens 14:44:43 this is also true, yes 14:45:05 if tension worked like that, you could tie temperature directly to the tension value instead of all this mucking around 14:45:12 https://dl.dropboxusercontent.com/u/63152810/UglyThing.gif 14:45:29 mumra: well the good news is that's how temperature works, so if we can get it working well, we can just reuse the code for tension 14:45:35 Bloax: that's pretty cool 14:45:44 03|amethyst 07* 0.13-a0-2165-gcd744e6: Rename a misleading function. 10(67 seconds ago, 7 files, 10+ 9-) 13http://s-z.org/neil/git/?p=crawl.git;a=commitdiff;h=cd744e6bb17f 14:45:44 03|amethyst 07* 0.13-a0-2166-g9b1ab93: Correct a St Roka line. 10(36 seconds ago, 1 file, 2+ 2-) 13http://s-z.org/neil/git/?p=crawl.git;a=commitdiff;h=9b1ab93b4666 14:45:55 Well certainly better than whatever the previous attempt was. 14:47:16 different colour ugly things could look more significantly different than just palette swaps too 14:47:28 -!- riot has quit [Ping timeout: 256 seconds] 14:47:33 Yeah I think I might hand ontoclasm this. 14:47:41 Because he seems quite good at that. 14:48:06 gammafunk: btw, i will definitely get around to taking a look at jump patch at some point if nobody else does, there are just quite a few new things going on at the moment ;) 14:49:27 mumra: Thanks. I know you're busy with tons of good stuff as-is, so no worries 14:50:02 Though I have to say that https://dl.dropboxusercontent.com/u/63152810/UglyThingCyan.png looks quite decent. 14:50:32 hmmm, log(tension) works reasonably well as a temp value to increase towards 14:51:31 problem though: it's hard to ensure that at every XL, it's reasonably possible to hit all temp values in a fight, up to max temp... 14:51:45 we really need some data collection about this 14:52:08 Bloax: yeah palette swaps will look decent, i'm just saying it'd be cool if they all had a slightly different pose too 14:52:23 Different features too. 14:52:51 (Note the little stack of horns on the (our) left side of its head.) 14:52:57 poor ontoclasm. Then he'd have to do it for *very* ugly things as well 14:53:22 heh 14:53:23 Bloax: even with palette swaps, small variations would be good (e.g. electric ones have electricty crackles around them, fiery ones have flames, etc.) 14:53:32 Then I'm reminded of this: https://dl.dropboxusercontent.com/u/63152810/VeryUglyThingy.png 14:53:52 gammafunk: yeah i'm not sure how he's going to make the very ugly things uglier; maybe these ones need toning down! 14:54:01 that thing certainly is ugly 14:54:01 and then we get a mother in law 14:54:41 -!- Silurio1 has quit [Ping timeout: 248 seconds] 14:54:42 fr: ExtremelyUglyThing 14:55:05 fr: merge abominations and ugly things 14:55:10 the ugliest thing 14:55:14 a unique 14:55:18 -!- bonghitz has quit [Remote host closed the connection] 14:55:23 ? 14:55:44 I love the 'ugliest thing unique' idea 14:55:46 This.. "thing" is so ugly, your eyes can't even comprehend what it is. 14:55:53 Bloax: especially with those attempts to make player-created and naturally spawning aboms distinct 14:56:04 I really hate the healing/non-healing distinction 14:56:25 either both heal or both don't 14:57:07 -!- knaveightt has quit [Quit: Give a man a fish and he will eat for a day. Teach him how to fish, and he will sit in a boat and drink beer all day] 14:59:06 kilobyte: you mean the branch i pushed? 14:59:27 it was an attempt to balance the spell, not an actual attempt to distinguish player vs natural aboms 14:59:43 (although yes this would be simpler if they weren't called the same thing) 14:59:47 -!- buppy has quit [Quit: Leaving] 15:00:23 stonewall (L18 MuAs) ERROR: range check error (-44 / 17) (D:13) 15:00:52 then either we could look at merging aboms and uglies, or at least make all aboms non-healing 15:01:30 -!- absolutego has quit [Quit: leaving] 15:05:46 i suppose for something that only spawns in abyss/pan, non-healing isn't really a problem 15:06:40 although i also wanted to try making the player creations actually lose HP slowly over time 15:07:01 It makes sense for a creature made of rotting corpses to degenerate. 15:07:23 i think some difference between natural ones and player-created ones is easy to justify flavourwise anyway 15:09:09 anyone willing to add tension collection to the servers for the purposes of lorc improvement? 15:09:25 -!- raskol has quit [Ping timeout: 256 seconds] 15:09:35 -!- Silurio has quit [Ping timeout: 260 seconds] 15:12:07 -!- Silurio1 has quit [Ping timeout: 246 seconds] 15:13:31 -!- blabber has joined ##crawl-dev 15:14:55 -!- Silurio has quit [Ping timeout: 264 seconds] 15:15:08 Eronarn: it would require a new system for data of this kind, but then, I already had several cases I wanted something of this kind 15:16:13 probably only such case it actually got added: # of trivial/easy/hard/nasty monsters at every XL 15:20:00 Oh, nice. You can use the wizmode lua interpreter to test tutorial messages. 15:20:42 03|amethyst 07* 0.13-a0-2167-g3126737: Correct a few more related orc speech lines. 10(36 seconds ago, 1 file, 10+ 2-) 13http://s-z.org/neil/git/?p=crawl.git;a=commitdiff;h=312673711a2d 15:23:37 <|amethyst> BTW, lots of that speech will need to be changed if LO monsters become a thing 15:23:43 -!- Silurio1 has quit [Ping timeout: 276 seconds] 15:23:46 <|amethyst> since it assumes monster orcs are cave orcs 15:24:14 <|amethyst> though in some cases it only applies to priests 15:26:51 -!- raskol_ has quit [Ping timeout: 260 seconds] 15:27:53 Hm, you know what. 15:27:58 Let's just have recolors for now. 15:35:26 kilobyte: i think just having games per-turn randomly sample some metrics at a very low rate would be the right way to approach it 15:37:56 -!- ShadeTornado has quit [Quit: Leaving] 15:41:33 I guess rebranding lamps of fire to lava lamps is a bad idea 15:43:50 http://imgur.com/a/0WUrh#0 15:44:29 -!- dtsund has joined ##crawl-dev 15:44:35 03|amethyst 07* 0.13-a0-2168-g6c4f435: Account for LO^Beogh in the changelog. 10(19 minutes ago, 1 file, 2+ 2-) 13http://s-z.org/neil/git/?p=crawl.git;a=commitdiff;h=6c4f43570ade 15:44:35 03|amethyst 07* 0.13-a0-2169-gd690064: Manual sync from the wiki (LO^Beogh). 10(14 minutes ago, 1 file, 6+ 2-) 13http://s-z.org/neil/git/?p=crawl.git;a=commitdiff;h=d69006466fb7 15:44:35 03|amethyst 07* 0.13-a0-2170-gc3cec7c: Update a translation whose key changed. 10(12 minutes ago, 1 file, 2+ 2-) 13http://s-z.org/neil/git/?p=crawl.git;a=commitdiff;h=c3cec7c6d975 15:44:35 03|amethyst 07* 0.13-a0-2171-g6558d3f: Correct a few comments. 10(11 minutes ago, 3 files, 3+ 3-) 13http://s-z.org/neil/git/?p=crawl.git;a=commitdiff;h=6558d3f7e469 15:44:35 03|amethyst 07* 0.13-a0-2172-g3dbd8d1: Make trap_god_beogh welcome Lava Orcs. 10(34 seconds ago, 1 file, 7+ 3-) 13http://s-z.org/neil/git/?p=crawl.git;a=commitdiff;h=3dbd8d124c34 15:44:44 -!- Yermak has quit [Quit: Page closed] 15:44:46 -!- six40sword has quit [Quit: ThrashIRC v2.9 sic populo comunicated] 15:45:23 -!- myrmidette has quit [Ping timeout: 240 seconds] 15:46:19 -!- Aryth has quit [Read error: Connection reset by peer] 15:47:23 -!- Silurio has quit [Ping timeout: 240 seconds] 15:48:04 -!- Arendeth has quit [Quit: KVIrc 4.2.0 Equilibrium http://www.kvirc.net/] 15:48:31 -!- Silurio1 has quit [Ping timeout: 256 seconds] 15:50:37 <|amethyst> (btw, should LO still have +3 invo) 15:50:40 <|amethyst> ? 15:51:25 <|amethyst> it would be nice to differentiate their aptitudes from HO a little more 15:51:29 <|amethyst> maybe 15:53:26 -!- Silurio has quit [Ping timeout: 268 seconds] 15:54:57 -!- myrmidette has quit [Ping timeout: 252 seconds] 15:55:33 there ought to be some race that's good at invo besides HO. (Ds doesn't count because of god restrictions). Perhaps it should be Gr though, not LO 15:57:16 rchandra, uh, 15:57:18 !apt inv 15:57:19 Inv: Ds: 3!, HO: 3!, LO: 3!, DD: 3!, Ha: 1, Op: 1, Dj: 1, Dr: 1, Hu: 1, Ko: 1, Gh: 1, HE: 1, Mf: 1, Gr: 1, DE: 1, Ce: 1, Na: 1, Og: 1, Mi: 0, Sp: 0, Fe: 0, Mu: -1*, Vp: -1*, Te: -1*, Tr: -1*, Dg: N/A 15:57:42 ...you're forgetting someone :b 15:57:54 DD aren't a real race. 15:58:18 -!- mumra has quit [Ping timeout: 246 seconds] 15:58:32 DD's being DD overwhelms their apts such that they can't be compared to other races that wayu 16:03:12 -!- Silurio1 has quit [Ping timeout: 246 seconds] 16:03:21 Invocations is sort of a weird apt in general 16:03:36 most gods don't care about it enough that the apt makes a real difference, it's only a very few 16:04:21 I think of maxing invoc with Sif, TSO, and Fedhas 16:04:27 I'm not sure if there are any others 16:04:42 you missed the one it's most important on 16:05:23 that's Elyvilon incidentally 16:06:00 Figures, I don't like Elyv and haven't played a He 16:06:01 -!- dcss36606 has quit [Ping timeout: 246 seconds] 16:06:17 That's where you have the strongest incentive to do it anyway, it's certainly defensible on those other 3 - there are perks for doing so but depending on how you're playing you don't need to 16:06:43 it's really good with zin 16:07:05 I'm not sure I'd say it's "really" good 16:07:13 you do want 15 or so at least though 16:07:45 -!- Kromgart has quit [Quit: Leaving] 16:07:50 good enough that I can use recite on LO of zin throughout the game, where I usually can't on others. 16:08:26 training invo super high for recite is more of a comedy option than something you should really do 16:08:43 then again it might not be worst than, say, high-power cflame 16:09:57 There's also Cheibriados for bend time, but otherwise yeah. 16:10:53 I mean each god sort of has a different profile for how much you get out of them at what invo level 16:11:00 -!- reaver has joined ##crawl-dev 16:11:20 excepting the ones that don't use it at all of course 16:11:24 Aptitudes aren't only for training things super high though. 16:11:41 Early game is the hard part, that's where getting it cheaper to 10 matters. 16:11:42 yeah, they're for getting things where you want them faster. 16:12:19 I mean on one hand that's true 16:12:20 I really think +inv could help Gr especially since some people think they don't have enough of a niche 16:12:39 and LO certainly has its own wackiness 16:12:49 "Trolls but actually good" 16:12:53 Wait what. 16:12:57 no sorry those are Trolls 16:14:31 anyway, there are certainly some gods where you want 10 or more early but the problem here is that even then for most of them this isn't your first, second, maybe even third priority 16:15:02 for Healers it's definitely an overriding priority over basically everything... 16:16:01 and then there are a few like Lugonu where you need to weigh the 10-12 invo with some other stuff, but most gods with early abilities aren't very demanding on the invo rank to use them 16:16:38 Like, even with a good apt, chances are that piety requirements are slowing you down enough that the apt isn't making a huge difference 16:18:42 Basically what I'm saying is that to me the other things weighing on how quickly you can get a certain god ability to be strong mean that an invo apt difference isn't as much of a niche as it would seem like to me 16:19:05 -!- ystael has quit [Quit: Lost terminal] 16:19:15 saves you a little XP by the end but that kind of disappears into all your other apts by the time you're likely investing that XP on most gods 16:19:23 it's not huge, but it's non-zero and I think a bunch of little things like that add up 16:20:44 !apt Gr 16:20:44 Gr: Fighting: 1, Short: -1, Long: -1, Axes: -1, Maces: 1, Polearms: -1, Staves: 1, Slings: -1, Bows: 0, Xbows: 0, Throw: 0, Armour: -2, Dodge: -1, Stealth: 2, Stab: 0, Shields: -2, Traps: -1, UC: 1!, Splcast: -1, Conj: 1, Hexes: -1, Charms: -2, Summ: -1, Nec: -3*, Tloc: -1, Tmut: -1, Fire: -1, Ice: 1, Air: -3, Earth: 3!, Poison: 0, Inv: 1, Evo: 0, Exp: 0, HP: 1, MP: 0 16:20:54 -!- mumra has joined ##crawl-dev 16:21:16 -!- raskol_ has quit [Quit: Konversation terminated!] 16:21:22 hmm, I wonder if the +Inv would even fit in well there 16:21:44 (not evoke invis, I mean the apt, heh) 16:23:36 If you end up giving Gr a good Invocations apt, what exactly does this enable that's significantly different from doing the same kind of thing as HO? 16:23:51 their apts are really quite similar in a lot of ways 16:24:38 also different in a lot of ways, but most of them don't suggest anything particularly unique to them as I see it 16:25:08 |amethyst: i was thinking to nerf lorc melee relative to HO 16:25:21 they should still have the same invo, though 16:25:46 !apt HO 16:25:46 HO: Fighting: 2, Short: 0, Long: 1, Axes: 3!, Maces: 1, Polearms: 1, Staves: -1, Slings: -1, Bows: -1, Xbows: -1, Throw: 0, Armour: 1, Dodge: -2, Stealth: -1, Stab: 2, Shields: 1, Traps: 0, UC: 1!, Splcast: -3, Conj: 0, Hexes: 0, Charms: -1, Summ: 0, Nec: 0, Tloc: -2, Tmut: -3, Fire: 1, Ice: -1, Air: -2, Earth: 0, Poison: -1, Inv: 3!, Evo: 1, Exp: 0, HP: 1, MP: 0 16:25:48 !apt LO 16:25:48 LO: Fighting: 2, Short: 0, Long: 1, Axes: 2, Maces: 1, Polearms: 1, Staves: -1, Slings: -2, Bows: -2, Xbows: -2, Throw: -1, Armour: 1, Dodge: -2, Stealth: -1, Stab: 2, Shields: 1, Traps: 0, UC: 1!, Splcast: -3, Conj: 0, Hexes: 0, Charms: -1, Summ: 0, Nec: 0, Tloc: -2, Tmut: 3!, Fire: 2, Ice: -4*, Air: -3, Earth: 1, Poison: -1, Inv: 3!, Evo: 1, Exp: -1, HP: 1, MP: 0 16:26:29 hmm, wonder what to change there... 16:26:48 fighting:1 is probably good 16:27:10 i dunno, that just feels like a flavorless nerf 16:27:25 -!- raskol has quit [Ping timeout: 276 seconds] 16:27:33 maybe axes to 1, make them no better with them than other weapons 16:27:44 plausible 16:28:06 I'd suggest hitting armour but there really aren't enough races with -good- armour apt 16:28:20 tmut could be lowered... the idea is they're made of something that is often fluid, so it could be easier to change form 16:28:22 compared to almost everything being good at dodging for some reason 16:28:38 lowering dodging a bit more would be fine i think 16:28:57 -2 is already bad enough to be quite notable 16:29:14 setting it to -3 or something doesn't even change too many characters I would suspect 16:29:29 they also have bad dex, so dodging isn't something you do unless you must I think 16:30:04 problem i'm having is that HO apts are sort of weird, there isn't much room to make them meaningfully worse 16:30:12 which is why i focused on differentiation by abilities 16:30:33 making LO flatter than HO would make sense 16:30:55 well not with the elemental schools 16:31:09 lava humans 16:31:24 yeah, but with weapons/melee. 1 fighting 0 weapons 1 armour -1 dodging etc 16:31:26 -!- pantaril has quit [Remote host closed the connection] 16:31:27 hmmmn, come to think 16:31:34 -!- blabber has quit [Quit: leaving] 16:31:39 how well would juggling their spellcasting, etc around do to make them different 16:32:11 they could probably be less terrible at that 16:32:14 the gimmick behind HO is supposed to be strong melee + bad spellcasting + ok schools 16:32:24 yes not for HO 16:32:26 LO don't need to have the same gimmick, but it should probably be similar 16:32:50 I kind of think they could plausibly get -1 spellcasting 16:32:51 or an inverse, or something 16:33:11 -!- mnoqy has quit [Quit: hello] 16:33:18 -1 spc/conj/summ/necro 16:33:21 who made meatsprint? 16:33:24 if you're hitting some of their other good points anyway 16:33:34 nerfing summ really doesn't work, how many LOSu do you expect to even see 16:33:39 -!- mumra has quit [Ping timeout: 246 seconds] 16:33:40 that's an st_ creation pretty sure 16:34:22 !apt fire 16:34:23 Fire: Dj: 3!, Dr[red]: 2, LO: 2, Dr[mottled]: 1, Dr[pale]: 1, HO: 1, Te: 1, DE: 1, Ha: 0, Op: 0, Dr: 0, Hu: 0, Ko: 0, HE: 0, Na: 0, Ds: -1, Dg: -1, Gr: -1, DD: -1, Ce: -1, Fe: -1, Mu: -2, Sp: -2, Dr[white]: -2, Gh: -2, Vp: -2, Mi: -3*, Mf: -3*, Tr: -3*, Og: -3* 16:34:25 well they already have reasons to use fire (the enhancers) so why give 2 fire? 16:34:29 sigh, dj :( 16:34:44 rchandra: because they're made of lava 16:34:49 dj is pretty weird by now 16:34:53 ideally they won't have the enhancer most of the time 16:35:01 temp is pretty badly implemented as is 16:35:02 it's true that maybe too many races are good at fire 16:35:20 excluding the dracs ofc 16:35:32 i don't feel that pale needs 1 16:35:59 dj shouldn't really have 3, if anyone should it'd make the most sense for lorcs 16:36:02 !apt earth 16:36:03 Earth: Gr: 3!, DD: 3!, Dr[grey]: 2, Gh: 1, DE: 1, LO: 1, Ha: 0, Op: 0, Dr: 0, Hu: 0, Ko: 0, HO: 0, Vp: 0, Na: 0, Ds: -1, Sp: -1, Dg: -1, Ce: -1, Tr: -1, Mu: -2, Mi: -2, Dr[black]: -2, HE: -2, Mf: -2, Fe: -2, Dj: -3*, Te: -3*, Og: -3* 16:36:05 compare with earth: 16:37:05 -!- sbanwart has quit [Ping timeout: 248 seconds] 16:37:34 !apt spellcasting 16:37:35 Splcast: DE: 3!, Sp: 2, Dj: 1, Dr[purple]: 1, HE: 1, Og: 1, Mu: -1, Ds: -1, Op: -1, Dr: -1, Hu: -1, Ko: -1, Vp: -1, Mf: -1, Gr: -1, Te: -1, Na: -1, Fe: -1, Gh: -2, Dg: -2, DD: -2, Ha: -3, HO: -3, LO: -3, Ce: -3, Mi: -4, Tr: -5* 16:38:16 0 spc, 0 fighting, less HP apt, more MP apt, lose axes bonus? make them more magically inclined orcs? 16:39:12 I was kind of thinking that a bit, then you should compare them to HE though 16:39:22 i like the idea of them being strong at low levels defensively just because of perma stoneskin, but once they run into harder enemies, they're more like glass cannons 16:39:52 their defensive apts aren't terrible enough to be a glass cannon unless you neglect them 16:40:02 also their Int probably isn't high enough for that qualification 16:40:29 SwissStopwatch: being at HP 0 fighting 0 would make them worse than humans defensively because of XP apt 16:40:40 yes, but not all that much so 16:40:58 though i guess they do have good armor/shield apts... maybe those could be reduced to 0 along with axes getting -1 16:41:03 well 16:41:18 I wouldn't hit those because being a strong spellcaster with decent apts there is sort of uncommon 16:41:31 * SwissStopwatch vanish 16:49:35 -!- myrmidette has quit [Ping timeout: 256 seconds] 16:51:17 -!- Nightbeer has quit [Quit: Verlassend] 16:51:37 https://dl.dropboxusercontent.com/u/63152810/BogBody.png 16:51:40 Hey guys. 16:52:01 Now you can spawn more of these without being ashamed of the tile. 16:52:59 tile? what's a tile 16:54:00 something fancy 16:56:01 -!- jjpalen has quit [Ping timeout: 250 seconds] 16:58:01 so is £ a good default character for runes or not? 16:58:13 -!- LogicNinja has quit [Quit: Page closed] 16:59:54 -!- reaver has quit [] 17:02:42 -!- mikee_ has joined ##crawl-dev 17:03:35 it's in CP437 and WGL4 anyway 17:04:12 or maybe φ is better 17:04:17 conflicts with some folks (let's think what country) glyph for gold :p 17:04:35 that seems to be in both too 17:04:57 -!- Kellhus has quit [Quit: Page closed] 17:05:03 why not 😸 ? 17:05:12 <3 17:05:13 I can't even see that 17:05:27 ☃ this could be the ice beast glyph :D 17:05:37 bah, 💩 17:05:42 SamB: http://www.fileformat.info/info/unicode/char/1f638/index.htm 17:05:45 I think that's double-wide in my font 17:06:04 -!- Arkaniad has quit [Read error: Connection reset by peer] 17:06:36 mine too, sadly. And like all the rest in that block, it would be hard to cram into the space and still be readable. 17:07:09 kind of makes sense, as it's an union of a bunch of Japanese standards, who have all native characters double width 17:07:19 but sadly Unicode didn't set that flag :( 17:07:24 yeah 17:07:36 they really need to get better about that :-( 17:08:14 also they should totally feel free to set it when they hadn't set it before if all the good fonts do it that way ... 17:08:16 aren't that cat's eyes frowning? :p 17:08:40 no one in the committee gives a damn, terminal support was pretty much bolted on by Markus Kühn 17:08:58 well they should let him control that flag then 17:09:01 Medar: kind of hard to tell even with a double-wide glyp 17:09:03 h 17:09:41 needs more futhark 17:09:41 actually maybe snowman isn't double-wide for me 17:11:34 U+2603, so it's old Unicode rather than emoji 17:11:37 -!- sudoninja has quit [Ping timeout: 250 seconds] 17:11:55 -!- raskol has quit [Ping timeout: 264 seconds] 17:12:13 maybe there's another snowman I was thinking of 17:12:17 Eronarn: support for it is quite rare :( 17:12:26 -!- nooodl has quit [Ping timeout: 240 seconds] 17:13:05 Eronarn: I think only Windows has any futhark glyphs in the default install, but on an obscure font, and no popular Windows terminal can substitute glyphs 17:13:13 kilobyte: huh? i've never had a problem doing it in linux 17:13:53 Eronarn: anything but xterm and rxvt will substitute glyphs properly, but you need to have a font installed 17:14:06 -!- Vandal has quit [] 17:14:09 kilobyte: and FWIW glyph substitution is usually pretty shitty in terminals unless the substitute fonts are manually chosen? 17:14:26 kilobyte: i don't go around installing a ton of fonts, are you sure it's not in the default anywhere? 17:15:15 ??potion of speed 17:15:16 haste[1/1]: Reduces action delay by 33%. 17:15:26 -!- tJener has quit [Read error: Connection reset by peer] 17:15:50 http://en.wikipedia.org/wiki/List_of_Unicode_characters#Ogham 17:15:53 oh, bitmaps fonts used by xterm have furthark now 17:16:01 -!- minqmay has quit [Ping timeout: 246 seconds] 17:16:05 yeah ogham and runic both work for me on stock mint 17:16:25 -!- nooodl has joined ##crawl-dev 17:16:33 did you know xterm supports freetype now? 17:17:41 SamB: oh, I see: you can select "truetype fonts", off by default 17:17:44 however, freetype's auto-bolding works really badly on pixelfonts 17:18:21 Eronarn: what terminal do you use? 17:18:32 kilobyte: gnome-terminal, though i was looking at it in a web browser 17:23:49 -!- mikee_ has quit [Quit: HydraIRC -> http://www.hydrairc.com <- In tests, 0x09 out of 0x0A l33t h4x0rz prefer it :)] 17:27:44 -!- Helmschank has quit [Quit: Leaving] 17:27:45 -!- y2s82 has quit [Quit: WeeChat 0.4.1] 17:28:46 -!- Thyme has quit [Ping timeout: 240 seconds] 17:30:23 -!- Thyme has joined ##crawl-dev 17:32:05 -!- Villadelfia has quit [Ping timeout: 256 seconds] 17:35:47 -!- Villadelfia has joined ##crawl-dev 17:37:43 -!- edlothiol has quit [Ping timeout: 264 seconds] 17:37:43 -!- Kalir has quit [Ping timeout: 246 seconds] 17:38:16 -!- magicpoints has quit [Ping timeout: 268 seconds] 17:40:09 -!- sbanwart has quit [Ping timeout: 256 seconds] 17:43:44 Eronarn: that's fonts-freemono-ttf 17:44:05 -!- Nexos has quit [Read error: Operation timed out] 17:44:29 i'm actually going to try addressing some of this stuff within my game 17:44:36 er, freefont 17:44:38 let people modify glyphs that show up broken for them, in-place 17:46:07 -!- scummos has quit [Ping timeout: 264 seconds] 17:47:16 yay fontforge integration 17:47:42 SamB: <3 17:53:46 -!- Blazinghand has quit [Ping timeout: 256 seconds] 17:54:52 Ramc: They will replace the Spider Branch with the Moth Branch one day 17:54:52 Ramc: mark my words 17:56:06 but it will still be covered in moth webs 17:56:43 Of course. 17:58:25 -!- Nexos is now known as fdel 17:58:29 * SamB wonders if he should cut down on his wands 17:58:31 covered in spiders really needs to be a status effect 17:59:05 And then psychology becomes part of crawl. 17:59:10 So that one can have psychological trauma. 17:59:27 -!- myrmidette has quit [Ping timeout: 260 seconds] 18:00:07 I once wrote an Unicode map with a heavy stress on tickling every known bug in popular terminals: invalid wcwidth, corrupted display, glyphs out of bounds 18:00:18 perhaps I should share it... need a name, though :p 18:03:01 kilobyte: rtl chars? 18:03:10 and combining chars! 18:05:16 Eronarn: rtl chars are shown, combining ones not 18:05:24 <|amethyst> hm 18:05:35 out of scope for what I had then 18:05:36 <|amethyst> looks like this commit broke saves 18:05:44 <|amethyst> %git 6cfaa00 18:05:45 07qoala * 0.13-a0-2055-g6cfaa00: Moved the setting of Spectral Weapon stats to ghost code, like dancing. 10(5 days ago, 7 files, 56+ 7-) 13http://s-z.org/neil/git/?p=crawl.git;a=commitdiff;h=6cfaa0000909 18:08:53 -!- qoala has joined ##crawl-dev 18:10:13 speak of the devil? :) 18:11:21 -!- bmfx has quit [Quit: leaving] 18:11:29 03pubby 07* 0.13-a0-2173-gfaf8416: Remove use of 'you' inside player member functions 10(4 hours ago, 12 files, 153+ 150-) 13http://s-z.org/neil/git/?p=crawl.git;a=commitdiff;h=faf8416790ef 18:11:34 -!- mumra has joined ##crawl-dev 18:12:08 <|amethyst> and there's the committer :) 18:12:18 <|amethyst> I was just saying, saves with SWs were broken by 18:12:24 <|amethyst> %git 6cfaa00 18:12:24 07qoala * 0.13-a0-2055-g6cfaa00: Moved the setting of Spectral Weapon stats to ghost code, like dancing. 10(5 days ago, 7 files, 56+ 7-) 13http://s-z.org/neil/git/?p=crawl.git;a=commitdiff;h=6cfaa0000909 18:13:21 <|amethyst> not sure how to handle those this late 18:13:34 How broken? Now that I think about it I can see problems, but do they not evaporate when the weapon dies? Or does it crash on something 18:13:52 <|amethyst> qoala: completey corrupt when loading 18:14:01 <|amethyst> qoala: because it tries to unmarshall the ghost_demon structure 18:14:04 <|amethyst> which wasn't there 18:14:06 ah 18:14:14 <|amethyst> s/tey/tely/ 18:15:26 |amethyst: would that not have crashed already though? 18:15:37 <|amethyst> SamB: #7276 18:15:40 Yeah, I didn't understand all the implications of the ghost demon when I was working on it. 18:16:10 the problem is with saves between TAG_MINOR_WAR_DOG_REMOVAL and TAG_MINOR_CANARIES 18:16:19 <|amethyst> kilobyte: only some of them 18:16:40 <|amethyst> %git c2ce035 18:16:40 07DracoOmega * 0.13-a0-1816-gc2ce035: Bump up wolf stats to match war dogs, replace war dogs with wolves 10(2 weeks ago, 24 files, 68+ 79-) 13http://s-z.org/neil/git/?p=crawl.git;a=commitdiff;h=c2ce03562a48 18:16:42 |amethyst: the problem is, we don't know which side of the bad commit a given save is at 18:16:46 <|amethyst> %git b487634 18:16:47 07kilobyte * 0.13-a0-2141-gb487634: Pepper "you" and level chunks of save files with canaries. 10(2 days ago, 2 files, 81+ 1-) 13http://s-z.org/neil/git/?p=crawl.git;a=commitdiff;h=b487634b5f02 18:16:58 <|amethyst> kilobyte: yeah 18:18:01 I'm not sure, but I didn't set any part of the ghost that isn't copied by ghost_demon_init 18:18:27 <|amethyst> kilobyte: the next byte could be either a held_type or part of a string length 18:18:35 -!- eith has quit [Ping timeout: 260 seconds] 18:18:59 <|amethyst> qoala: the problem is what happens when a SW was saved without a ghost_demon then loaded into a version where it has one 18:19:03 yeah 18:19:14 <|amethyst> kilobyte: which isn't helpful, since it's likely to be zero either way 18:20:00 You all probably have a better understanding how to detect the problem. I was musing on what to do when we determine the ghost demon is missing. 18:20:20 <|amethyst> ah 18:20:25 <|amethyst> yeah, that would have to be solved too 18:20:52 this is why we should use protobufs for save files :) 18:22:43 bh: they would fail to initialize the spectral weapon 18:23:06 -!- nooodl_ has joined ##crawl-dev 18:23:14 <|amethyst> well, presumaby after you unpack you check validity 18:23:27 in this particular case they would make this error slightly easier to recover, but that's not the case in general 18:23:29 <|amethyst> since if we did use protobufs most things would have to be optional 18:23:37 <|amethyst> because you can't add non-optionals 18:23:52 <|amethyst> AIUI 18:23:54 -!- Mu_ has quit [Read error: Connection reset by peer] 18:24:52 -!- Mu_ has joined ##crawl-dev 18:25:40 -!- GuraKKa has quit [Ping timeout: 246 seconds] 18:25:42 <|amethyst> Mu_: btw, I've always wondered: Greek letter, Zen koan, or none of the above? 18:26:01 -!- nooodl has quit [Ping timeout: 246 seconds] 18:26:22 -!- valrus has quit [Ping timeout: 246 seconds] 18:26:23 -!- valrus_ is now known as valrus 18:27:03 <|amethyst> mumra, qoala: Is there a reason #7285 is still open? 18:28:05 <|amethyst> qoala: also, since you weren't around when I added it: is "Ed Gonzales" good for credits.txt? 18:28:13 |amethyst: no, i just forgot to close it 18:28:48 my last name ends in a 'z', but otherwise that's fine. 18:28:59 <|amethyst> qoala: gah 18:29:09 <|amethyst> qoala: I even saw that in your commits but typed it wrong twice :) 18:29:18 * qoala is trying to figure out what actually gets marshalled for monsters. Did old spectral weapon even get the right stats upon being loaded? 18:31:37 03|amethyst 07* 0.13-a0-2174-g3a0f4b0: Spell qoala's name correctly (mea maxima culpa). 10(35 seconds ago, 1 file, 1+ 1-) 13http://s-z.org/neil/git/?p=crawl.git;a=commitdiff;h=3a0f4b0643d3 18:31:40 <|amethyst> qoala: which stats? 18:31:56 -!- tolly has quit [Read error: Connection reset by peer] 18:32:42 <|amethyst> qoala: see {,un}marshallMonster 18:32:52 <|amethyst> qoala: ac, ev, speed, etc are marshalled 18:35:08 Looking at unmarshallMonster, it seems they're only handled if there's a ghost_demon? 18:35:30 <|amethyst> qoala: it's right at the very beginning 18:35:37 <|amethyst> qoala: m.ac = unmarshallByte(th) 18:35:49 <|amethyst> after mid and mname, before the coordinates 18:36:03 <|amethyst> kilobyte: should unmarshallCString unpack the length as unsigned, or should it assert on negatives? 18:36:38 ah, somehow ended up looking at the wrong function. 18:36:42 <|amethyst> kilobyte: I'm inclined towards the latter as it helps detect corruption (if the sign bit gets flipped) and we shouldn't need to marshall 32 K strings anyway 18:37:06 <|amethyst> kilobyte: (and marshallString would need to check for that case as well) 18:37:25 gah, I'm sorry I messed this up. 18:37:45 <|amethyst> qoala: it's okay, it's an easy one to miss 18:37:49 i didn't notice this at all either 18:37:56 too bad bytes with the high bit set are somewhat rare 18:38:05 <|amethyst> kilobyte: that's true 18:39:04 can spectral weapons even have a ghost.name? 18:39:18 no 18:39:38 <|amethyst> kilobyte: my question wasn't specifically about this problem 18:39:41 Well, it looks like if we find a way to detect when the ghost_demon is missing, then it should be able to create a new one based on what's already being loaded for the monster. 18:39:48 <|amethyst> kilobyte: so much as that we have a buffer overflow 18:40:27 oi guys 18:40:30 Bumblechubs: If polyefemus gets attacked by an invis person he should shout 'aaaa Nobody is killing me' 18:41:32 <|amethyst> kilobyte: (I realise trying to prevent arbitrary code execution from a specially crafted save is probably a sisyphean or at least augean task) 18:42:20 <|amethyst> Bloax: is that a reference to something? 18:43:05 <|amethyst> FR: if your char is named "Ulysses", polyphemus runs through the nearest wall, leaving a portal back to D:1 18:43:39 -!- orionstein is now known as orionstein_away 18:43:59 Might be a reference to Odyssey but I don't know. 18:44:27 But yeah, this would be a funny little thing. 18:44:38 Bloax: it is 18:45:08 <|amethyst> oh 18:45:11 <|amethyst> so it is 18:46:02 <|amethyst> then I approve 18:46:19 <|amethyst> maybe without the "aaaa", but "Help, nobody is killing me!" 18:46:29 -!- CeleryMan has quit [Remote host closed the connection] 18:48:47 hahaha 18:48:52 Maybe both. 18:50:24 <|amethyst> sadly I don't think it can be done without code 18:51:17 <|amethyst> since there's no prefix for "can't see player" 18:51:21 in old saves, constriction data means 16 zero bytes (SW can't either constrict or be constricted). In new ones, you have 11 zeroes, then a non-zero. 18:51:54 <|amethyst> kilobyte: can you read ahead that much and still be able to unpack? 18:52:15 <|amethyst> I haven't looked into the implementation that closely 18:52:28 when loading save chunks we always slurp everything into memory and read from it 18:53:53 Can the reader rewind? 18:55:07 enchanted lake (very early) prototype vault: http://pastebin.com/EMh2DJUB 18:55:24 Not sure about the best monster set, but the basic layout seems good enough 18:55:25 <|amethyst> qoala: not yet but I guess you'd just need to add a method 18:56:09 -!- myrmidette has quit [Ping timeout: 240 seconds] 18:56:22 somethink a little like those lake with a river leading into the mountain scenes in zelda games 18:56:52 maybe we need a zora monster to do this properly 18:57:08 <|amethyst> gammafunk: electric eel acts kind of like a zora 18:57:13 ah yeah, we would just be exposing the underlying fseek 18:57:16 |amethyst: haha, maybe that'd work 18:57:47 -!- Silurio has quit [Ping timeout: 260 seconds] 18:57:48 <|amethyst> I always thought they were Zolas 18:58:00 <|amethyst> oh I see 18:58:02 <|amethyst> http://zeldawiki.org/River_Zora#Zora_or_Zola.3F 18:58:21 |amethyst: We just need an eel that fires some kind of splash damaging electricity bomb 18:58:28 <|amethyst> qoala: well, fseek and you'd need to change _read_offset for the in-memory buffer 18:58:50 <|amethyst> let em guess... Windows doesn't have mmap? 18:58:57 <|amethyst> s/em/me/ 18:59:27 <|amethyst> because I'm not sure why you'd want to read into a buffer instead of mmapping 18:59:40 |amethyst: no need to fseek 18:59:54 also, fseek won't work on fragmented zlibbed streams anyway 18:59:59 <|amethyst> kilobyte: are you sure? reader::readByte seems to use fgetc whnever _file is set 19:00:09 <|amethyst> kilobyte: or is it not set there? 19:00:14 <|amethyst> s/whnever/whenever/ 19:00:27 <|amethyst> oh, I forgot about compression 19:00:39 If it's in our buffer, can we just change position within the buffer? (Still trying to figure out all the branch-logic in the reader) 19:00:40 <|amethyst> which explains not mmapping it too :) 19:01:33 |amethyst: none of the above 19:01:37 -!- Mu_ has quit [Quit: Defecator, may everything turn out okay so that you can leave this place.] 19:02:30 |amethyst: _restore_tagged_chunk() calls tag_read(), which for whatever reason reads everything into memory before processing 19:02:35 -!- CKyle has quit [Read error: Connection reset by peer] 19:02:56 which, as I mentioned above, is lucky as it's the only easily rewindable stream :) 19:03:32 <|amethyst> kilobyte: oh... are the _file and _chunk cases in reader::read* ever used? 19:03:54 for reading files and save chunks 19:04:14 yeah, it's a mess 19:04:25 -!- CKyle has joined ##crawl-dev 19:04:29 most of use cases go through that code twice 19:05:01 <|amethyst> oh, I see 19:05:08 then there's a "tag" that's a degenerated protobuf 19:05:33 <|amethyst> I missed that it was using reader::read to load the buffer into memory 19:05:59 <|amethyst> so it's layered ish 19:07:48 -!- dtsund has quit [Quit: dtsund] 19:08:28 in fact, the old save format did use a kind of protobufs, but with pretty large individual fields ("tags") 19:08:46 like, "all monsters on the level" 19:10:09 the format of individual tags was supposed to be immutable, and you were supposed to add new tags without ever altering old ones 19:10:15 obviously, this did not work 19:10:55 <|amethyst> kilobyte: you said "11 zeroes, then a non-zero" 19:11:06 <|amethyst> kilobyte: why 11? 19:11:16 HD, saved as a short 19:12:23 -!- alefury has quit [] 19:12:28 <|amethyst> kilobyte: ? 19:12:46 name (string -> 2), species (short -> 2), job (2), religion (1), best_skill (2), best_skill_level (2) 19:12:46 <|amethyst> kilobyte: In the ghost_demon data? 19:12:51 these are all zeroes 19:13:06 then you get xl/HD, as a short, and it's non-zero 19:13:08 <|amethyst> oh, I thought it took the player's skills, okay 19:13:08 yes 19:13:30 the "best_skill" field is always zero 19:13:36 possible danger: old spectral weapon could have HD0 if the player had no skill in the weapon. I've since fixed this, but... 19:13:49 ... 19:14:00 might be unlikely enough to not be a problem, though. 19:14:11 qoala: you just stopped me from fucking it up, thanks 19:14:31 max_hp is > 0, right? 19:14:35 yes 19:15:04 minimum 10 19:15:13 -!- Silurio1 is now known as Silurio 19:15:24 up to 44 19:15:45 256 is not possible? 19:15:48 well, nonzero short at any rate 19:15:54 I'll just read it as short 19:16:00 256 should be impossible. 19:16:26 what byte order do we use? 19:16:40 <|amethyst> int16_t data = (b1 << 8) | (b2 & 0x00FF); 19:16:42 little endian 19:16:50 <|amethyst> huh? 19:16:53 eh? 19:16:54 <|amethyst> unmarshallShort uses nbo 19:17:04 ... and you're right 19:17:39 Well, it looks like {ghost_demon gh; set_ghost_demon(gh);} ought to work. None of the fields that are actually set by the new SW code are ever used again. I tested replacing the ghost loading code with this, and SW properly loaded. 19:18:10 and the new SW code handles a superset of that set by the old stuff. 19:18:43 I just turned it into MONS_GHOST, safer this way 19:19:07 or, should it go away somehow? 19:19:21 -!- ProzacElf has joined ##crawl-dev 19:19:26 <|amethyst> don't you have to do something to the player to expire the spell? 19:19:47 <|amethyst> or the caster, rather 19:19:51 if we can kill the spectral weapon safely, that might be best. there's a function in spl-summoning for cleanup 19:20:24 The player is already fully loaded by the time the weapon is being read, yes? (thank goodness no monster SW yet) 19:21:16 -!- Nikolaos has quit [Quit: Wrong button bad with computers] 19:21:37 -!- myrmidette has quit [Ping timeout: 276 seconds] 19:21:37 {you.props.erase("spectral_weapon");} Though I think the code will clean this up on its first attempt to use the weapon, if you don't clear it. 19:22:12 yeah, if it can't find the weapon, the owner's prop is erased automatically. 19:22:29 <|amethyst> and that's by mid so no chance of a false positive 19:22:43 <|amethyst> so just nuking it should be fine? 19:23:01 yeah, nuking it shouldn't break the SW code. 19:23:11 <|amethyst> turning it into a ghost might be problematic because then it would find it 19:23:36 already handled 19:23:43 Is it safe to nuke a monster during file load? 19:23:43 <|amethyst> cool 19:24:23 qoala: I wouldn't risk it 19:25:11 So what does turning it into MONS_GHOST actually end up doing? 19:25:21 -!- ProzacElf has quit [Quit: Leaving] 19:26:06 it's there as I'm uncomfortable trying to kill the monster at that time 19:26:07 <|amethyst> qoala: exactly what it sounds like 19:26:11 ghost (15p) | Spd: 10 | HD: 1 | HP: 1 | AC/EV: 0/0 | 07undead, 10doors, evil, lev | Res: 09poison+++, 12drown, 04rot+++, 13neg+++, 13torm, 04napalm | Vul: 08holy++ | XP: 1 | Sz: Medium | Int: normal. 19:26:11 <|amethyst> %??ghost 19:26:24 we could set a flag then kill it later 19:27:18 https://dl.dropboxusercontent.com/u/63152810/Suaverick.png 19:27:28 is this an improvement or is it now 19:27:30 not 19:27:47 <|amethyst> kilobyte: I ended up doing something similar for the constriction changes (mindex to mid) 19:28:14 <|amethyst> kilobyte: but even then a ghost is better in 99% of cases 19:28:27 <|amethyst> s/is better/feels safer/ 19:29:08 Out of curiosity, in the #7276 save, was it a spectral weapon left behind on the previous floor? 19:31:04 <|amethyst> hm 19:31:12 <|amethyst> yes, but now that I look they never knew the spell 19:31:48 even worse, the ghostified monster knows primal wave 19:31:48 <|amethyst> so now I wonder if there's some other corruption 19:31:52 <|amethyst> hrm 19:32:25 wait. What's that about primal wave? 19:32:35 elemental wellspring (11E) | Spd: 10 | HD: 15 | HP: 73-107 | AC/EV: 8/8 | 11non-living, !sil | Res: 13magic(immune), 10elec, 03poison, 12drown, 04rot+++, 13neg+++, 13torm | XP: 1857 | Sp: primal wave (3d21), water elementals | Sz: Big | Int: plant. 19:32:35 <|amethyst> %??elemental wellspring 19:32:43 <|amethyst> (is the next enum) 19:34:01 0: Primal Wave (180) 1: Primal Wave (180) 3: Summon Water Elementals (162) 19:34:07 looks pretty similar :p 19:34:11 an auto-die flag would probably be safer >.> 19:36:39 <|amethyst> err 19:37:12 -!- ahahaha has quit [Quit: Page closed] 19:37:13 -!- Ajonos has quit [Ping timeout: 250 seconds] 19:37:16 <|amethyst> kilobyte: yeah, see 19:37:23 <|amethyst> %git 2fab1c 19:37:24 07mumra * 0.13-a0-1964-g2fab1c1: Post-merge Skald cleanup 10(7 days ago, 26 files, 169+ 158-) 13http://s-z.org/neil/git/?p=crawl.git;a=commitdiff;h=2fab1c1e1f4a 19:37:48 hrm, for new (but pre-canary) saves, fourth byte has a value of 100 19:37:52 <|amethyst> it stuck spectral weapon ahead of elemental wellspring 19:38:45 -!- evilmike has quit [Read error: Connection reset by peer] 19:39:24 oy vey 19:39:32 <|amethyst> oh 19:39:33 so there was an elemental wellspring on the floor, which is being unpacked as a broken spectral weapon? 19:39:37 <|amethyst> wellspring *and* polymoth 19:39:55 <|amethyst> yes, though the ghost thing is still an issue 19:42:44 hrm, so how come fourth byte is 100? (after three zeroes) 19:44:54 <|amethyst> in the ghost? 19:45:08 <|amethyst> or rather, not in the ghost 19:45:21 -!- Blazinghand has quit [Ping timeout: 248 seconds] 19:45:22 <|amethyst> in the constriction info? 19:46:34 <|amethyst> that would be held_type = 100 wouldn't it? 19:46:37 <|amethyst> hm 19:47:35 crap ... i see what i did there 19:50:20 -!- Nakatomy has quit [Remote host closed the connection] 19:51:32 elemental wellspring (11E) | Spd: 10 | HD: 15 | HP: 73-107 | AC/EV: 8/8 | 11non-living, !sil | Res: 13magic(immune), 10elec, 03poison, 12drown, 04rot+++, 13neg+++, 13torm | XP: 1857 | Sp: primal wave (3d21), water elementals | Sz: Big | Int: plant. 19:51:32 <|amethyst> %??elemental wellspring 19:51:35 polymoth (12y) | Spd: 12 | HD: 10 | HP: 39-73 | AC/EV: 0/10 | Dam: 20 | fly | Res: 06magic(53) | Chunks: 07contam | XP: 975 | Sz: small | Int: insect. 19:51:35 <|amethyst> %??polymoth 19:51:42 spectral weapon (05() | Spd: 30 | HD: 27 | HP: 43-44 | AC/EV: 10/10 | Dam: 14 | 11non-living, lev | Res: 13magic(immune), 05fire, 02cold, 10elec, 03poison, 12drown, 04rot+++, 13neg+++, 13torm, 04napalm | XP: 5864 | Sz: Medium | Int: plant. 19:51:42 <|amethyst> %??spectral weapon ; scythe 19:51:56 <|amethyst> you can tell them apart by m.speed I guess? 19:52:39 we can check the minor version presumably? 19:53:18 <|amethyst> mumra: that commit didn't introduce a tag 19:53:33 <|amethyst> mumra: though this is only an issue between the two most recent tags 19:53:40 -!- Poncheis has quit [Quit: ChatZilla 0.9.90 [Firefox 21.0/20130511120803]] 19:54:19 <|amethyst> it's not so for *all* version between those two tags 19:54:27 <|amethyst> s/ion/ions/ 19:55:46 ah, i'd got it into my head there was a version tag but no there isn't 19:55:54 <|amethyst> btw, we're over halfway to the maximum number of monster types 19:56:08 <|amethyst> (int) MONS_POLYMOTH == 563 19:56:26 wow 19:56:27 <|amethyst> (int) MONS_NO_MONSTER == 1000 19:56:47 (How much does that go down for TAG_MAJOR_VERSION > 34?) 19:56:47 i don't think we're gonna have over 400 new ones any time soon 19:56:48 <|amethyst> some of those will go away on the next major bump 19:57:28 <|amethyst> Grunt: 10 it looks like? 19:58:32 <|amethyst> lab rat, war dog, ant larva, bumblebee, wood golem, monstrous item mimic, monstrous feature mimic, rock troll, blessed toe, arachnoid 19:58:43 -!- ackack has quit [Ping timeout: 264 seconds] 19:59:23 -!- riot has quit [Ping timeout: 260 seconds] 20:00:36 <|amethyst> kilobyte: now I wonder if Roshnak's save in 7270 is the same thing 20:00:43 <|amethyst> kilobyte: Upgraded the game from 0.13-a0-1887-g8bcbd07 to 0.13-a0-2010-g5476c2e 20:01:55 -!- rebthor has quit [Ping timeout: 246 seconds] 20:02:28 Oh yeah, on the off chance someone ends up contemplating a new major version before random gods go in, we need more DNGN_UNUSED_ALTAR spaces. We may be able to squeeze them without a bump, but if the version gets changed for something else, it'd make things easier. 20:03:07 <|amethyst> qoala: I'd rather they go in sooner than that :) 20:03:22 <|amethyst> qoala: random gods that is 20:04:03 yeah, I wasn't sure how likely it was or how soon it would be that something else required a major version. 20:04:32 <|amethyst> qoala: shouldn't be too hard to put them at the end and move them on a major bump 20:05:03 Well, altar code assumes that all altar features are in a contiguous block and in the same order as religion_type 20:05:12 -!- Arkaniad|Laptop has quit [Ping timeout: 252 seconds] 20:05:30 I hadn't gone through enough code to be sure how easy breaking such an assumption would be. 20:07:21 <|amethyst> qoala: hopefully not any time soon... there have been I think two since I started in 2011, and one of those (0.11 to 0.12 compat) was in effect rolled back with herculean effort on kilobyte's part. 20:07:53 <|amethyst> qoala: ah, usually stuff like that isn't too bad to fix 20:08:04 <|amethyst> qoala: and if it is that means there needed to be a function anyway 20:08:26 <|amethyst> qoala: and then you only have one place to put the ugly code 20:08:54 -!- bonghitz has quit [Remote host closed the connection] 20:09:10 <|amethyst> I'd probably even skip DNGN_UNUSED_ALTAR_1 20:09:33 <|amethyst> since that would just make the code more complicated 20:10:02 -!- rkd has quit [] 20:11:06 Yeah, if the assumption goes away, then the placeholder wouldn't be necessary anymore. 20:12:10 <|amethyst> well, it's necessary for other reasons, but could rename it to UNUSED_FEATURE_1 20:12:24 right 20:13:47 Do the servers have any convenient way to query and aggregate what version saves are currently at? If we're lucky, there might not be anyone in the space between 9e80fb2 and 6cfaa00. It's even less likely that such a save has an active spectral weapon. 20:13:53 |amethyst: we might be trying too hard, though 20:14:15 though fixing wellspring-->spectral_weapon would still need to be done. 20:15:54 -!- kekekela has quit [Quit: Page closed] 20:15:59 just got a deep scratch across the face. I got tired and dared to stop a 15 minutes petting session without permission. 20:16:43 <|amethyst> ah, cats 20:16:59 <|amethyst> only one of our (6) does that 20:17:19 <|amethyst> he also tries to eat and/or claw off glasses even if you are petting him 20:18:16 this one uses his claws only if I pet him not the way he wants 20:18:51 -!- ProzacElf has joined ##crawl-dev 20:18:51 <|amethyst> My main concern with breaking save compat on the servers is that people with old games won't get bugfixes 20:19:00 <|amethyst> meaning we could leave their game unfixable 20:19:55 <|amethyst> not that we can guarantee that that doesn't happen anyway, but it does mean (to me) that a compat break should be as last resort as possible 20:20:14 <|amethyst> s/last/last a/ 20:21:15 <|amethyst> "as possible" of course being flexible 20:21:25 wow, how is there no tag at all for new skald? why on earth didn't i put one in? i'm sure i was even thinking about that at one point 20:21:43 <|amethyst> mumra: was there anything that needed it? 20:22:08 no 20:22:35 <|amethyst> usually we don't put in a tag unless we have to check it 20:23:18 i think i went through this process though, and in fact thought "well it doesn't specifically need one, but it's probably safer to have a marker" and then forgot to do it ;) 20:23:36 <|amethyst> that might not be the best policy in the world, since it does mean you can screw up saves by going backwards (e.g. local console + tiles sharing a dir but they're not exactly the same version) 20:24:05 <|amethyst> the ghost thing did need a tag though 20:24:33 -!- CeleryMan has quit [Remote host closed the connection] 20:24:51 yeah, I was surprised to find that all monsters could contain a ghost_demon, but failed to check if such had branch-logic in tags.cc 20:25:55 hopefully I'll have a better grasp of save compat by the time random gods are ready for merging. 20:26:36 yeah it's not too hard once you wrap your head around it 20:27:01 this problem has possibly saved me from making the same mistake upgrading chimera to use ghost_demon 20:27:30 -!- Zermako has quit [] 20:28:30 <|amethyst> qoala: have you read the save compat doc? docs/develop/save_compatibility 20:28:36 <|amethyst> .txt 20:28:57 Yes, though I'm not fully certain of everything that gets saved, and where. 20:29:02 <|amethyst> yeah 20:29:45 <|amethyst> tracing through a level being loaded (maybe not single stepping, but at least making a call graph) is a good exercise 20:30:27 <|amethyst> also means you'll see a lot of save compat code :) 20:30:29 I know what to do when adding/removing fields, but don't know all the flags that cause differences in the number of things being marshalled. 20:30:55 <|amethyst> I probably would have missed this one too 20:31:11 <|amethyst> certainly I didn't notice it when I skimmed over the commit 20:31:59 <|amethyst> I only noticed it when tracing ttf's game 20:32:04 How would you recommend making a call graph? I've picked up some experience using gdb to step through and examine things, as well as some intuition for where things can go wrong with code I'm familiar with, but am still new to debugging tools. 20:32:37 * Grunt read "step through time" in there somehow <_< 20:32:51 <|amethyst> qoala: I was thinking by hand, but I think doxygen can do it? 20:33:00 (Cheibriados says, "Take your debugging easy.") 20:33:24 <|amethyst> Grunt: well, gdb does have reverse-step 20:33:59 -!- mumra_ has joined ##crawl-dev 20:35:51 -!- mumra has quit [Ping timeout: 246 seconds] 20:39:10 -!- mumra_ is now known as mumra 20:40:41 New branch created: recover-sw (1 commit) 20:40:41 03kilobyte 07[recover-sw] * 0.13-a0-2174-g9a0535e: Recover saves with bad spectral weapons/wellsprings. 10(14 minutes ago, 2 files, 33+ 2-) 13http://s-z.org/neil/git/?p=crawl.git;a=commitdiff;h=9a0535eb52de 20:40:47 -!- MaxFrost has quit [Ping timeout: 256 seconds] 20:40:56 could someone take a look at this commit? 20:41:04 does it appear to be right? 20:41:07 <|amethyst> I can in about 10 minutes 20:42:30 I'm afraid I'm too tired 20:43:23 3:42... 20:43:35 <|amethyst> I can merge/cherry pick it to trunk 20:43:50 if something is wrong, please amend it 20:43:54 <|amethyst> cool 20:43:56 <|amethyst> will do 20:43:57 -!- SudoNinja17 has quit [Ping timeout: 250 seconds] 20:44:02 great 20:44:03 <|amethyst> thanks 20:44:04 bye 20:44:09 <|amethyst> night 20:44:16 <|amethyst> and wash that scratch if you haven't yet :) 20:44:41 yeah 20:47:18 good night, and thanks for helping fix my mistake. 20:48:38 Can a wellspring be constricted? 20:49:04 <|amethyst> polymoth -> wellspring is also an issue 20:49:36 didn't polymoth get merged after newskald? 20:49:37 <|amethyst> I think I'll first fix up the type based on the speed for those three 20:49:51 <|amethyst> oh, hm 20:50:03 <|amethyst> not merged, but enabled 20:50:10 <|amethyst> so I guess it won't actually show up 20:50:40 I thought it was not enabled, but merged 20:50:50 <|amethyst> err, that's what I mean 20:51:10 <|amethyst> exactly the opposite of what I say 20:51:12 ;-) 20:51:44 Speakin of speed, the change of SW speed to 30 is actually in the same chain of commits as making it a ghost. 20:51:55 I thought it was in a different set. 20:52:14 But it looks like we could probably differentiate purely on speed, if I'm reading the commit log correctly. 20:52:51 <|amethyst> hm 20:52:55 <|amethyst> what speed was it before? 20:53:16 25 20:53:24 <|amethyst> then we check for either 25 or 30 :) 20:54:30 unknown monster: "spectral weapon speed:25" 20:54:30 %??spectral weapon speed:25 20:54:39 Yeah. Whoops. I could've sworn I made that change at a different time, but those commits are right next to each other. 20:54:53 spectral weapon (11() | Spd: 30 | HD: 27 | HP: 43-44 | AC/EV: 10/10 | Dam: 10 | 11non-living, lev | Res: 13magic(immune), 05fire, 02cold, 10elec, 03poison, 12drown, 04rot+++, 13neg+++, 13torm, 04napalm | XP: 5864 | Sz: Medium | Int: plant. 20:54:53 %??spectral weapon 20:54:57 <|amethyst> well, we'd have to handle older saves anyway 20:55:03 <|amethyst> thanks for reminding me the speed changed 20:55:10 But complicated rewinding isn't necessary any more. 20:56:16 particularly since I'm unsure if the byte-wise detection would fail on a constricted wellspring. 20:56:43 <|amethyst> qoala: if you fix up the monster type before fixing up the ghost structure that's not an issue 20:57:34 But if the wellspring is constricted, wouldn't the constriction struct have nonzero entries, making the code think it's a new spectral weapon? 20:57:52 (no longer an issue if we detect on speed alone) 20:57:59 <|amethyst> qoala: I mean, use speed to detect the type 20:58:01 yeah 20:58:07 <|amethyst> qoala: then fix up the ghost thing separately 20:58:35 <|amethyst> at which point you know which type of monster it is and will only do the constriction check for real SWs 20:59:06 okay, I keep misplacing half-completed thoughts that aren't needed anymore. And it's not even late yet for me. ;p 21:00:41 Since the speed and ghost commits are adjacent, were applied from a large format-patch of multiple commits, and no server or precompiled binary would've been generated in the middle, the constriction check might not ever be necessary. 21:01:08 Note that the speed change occured first, so it'll be speed 30 SW that need the constriction check if you keep it around to be safe. 21:01:25 <|amethyst> qoala: not necessarily 21:01:35 <|amethyst> qoala: we don't transfer through all intermediate versions 21:01:53 really? 21:01:57 <|amethyst> really 21:02:08 <|amethyst> we don't even compile all intermediate versions 21:02:38 <|amethyst> you have to be able to load saves from any time since the last version bump 21:02:57 <|amethyst> occasionally there are commits in one batch that break saves, which is mostly ok but can cause problems for git bisect 21:03:15 but at least those problems are localized 21:03:32 to the system of the one doing the bisect 21:03:33 <|amethyst> yeah, chances are they don't affect the particular thing you're bisecting so it's okay 21:03:43 <|amethyst> and if they do, you probably were bisecting for that anyway :) 21:03:46 and in particular to whatever saves they messed with 21:04:09 I'm possibly confused about which side I'm arguing for. My understanding is that no one should have a binary between the speed and ghost changes, with the exception of a bisect. 21:04:19 so it could make them somewhat sadfaced if they didn't back up a save before they started something ... 21:04:28 <|amethyst> qoala: which would mean "don't check for 30, just 25" 21:04:48 <|amethyst> qoala: oh 21:04:57 <|amethyst> qoala: you're saying if it's 30 it's probably safe 21:04:58 -!- pelotron has quit [Ping timeout: 276 seconds] 21:05:02 yeah 21:05:27 <|amethyst> hmm 21:05:54 Given the commit order, there's no commit at which a hypothetical binary could've had a speed 25 ghost SW. And no-one should've built a binary with a speed 30 non-ghost SW. 21:06:07 <|amethyst> now to make an old save with an actual SW instead of a wellspring for testing 21:08:19 -!- kryft has quit [Ping timeout: 264 seconds] 21:09:22 -!- hhkb has quit [Read error: Operation timed out] 21:09:22 yeah, testing is good. I should probably do save migration tests whenever I make large changes to monsters, etc. (especially if I don't put in explicit migration code) 21:10:13 -!- Utis has quit [Remote host closed the connection] 21:11:01 qoala: I'd skip that "especially"; what might be true, though, is that you'll need to try extra-hard to REMEMBER in that case 21:11:18 mhm 21:11:45 -!- nooodl_ has quit [Ping timeout: 248 seconds] 21:12:18 I mean you could screw up the migration code too right? 21:12:46 well, yes. I don't need to remind myself to test new code paths that I know touched. Usually. 21:12:48 and I guess you probably know that so you'd probably try to test it ;-) 21:12:49 :p 21:13:15 yeah 21:13:50 Should save migration go in the same commit that changes the save? (when it's actually planned for) 21:13:57 -!- stabwound has quit [Read error: Connection reset by peer] 21:13:59 -!- mason- has quit [Quit: Leaving] 21:14:03 or is it usually the following commit? 21:14:08 same one is best 21:14:21 IMO 21:14:24 <|amethyst> yeah 21:14:46 -!- stabwound has joined ##crawl-dev 21:14:50 so some schmuck like me doesn't do a git pull smack in between them :) 21:15:57 <|amethyst> I'd even go further and say if it's a branch, you should in the last step rebase it onto master instead of merging so you can squash the save fix commits into the right places 21:16:13 <|amethyst> but that's not as critical since no one who counts has a save from the branch 21:16:44 What if it's a public branch? Rebase anyways because people will be referring to the commits in trunk differently? 21:17:06 (or rather can refer, not 'will' refer) 21:18:18 I don't undestand what |amethyst was suggesting, and I would definitely suggest you avoid following suggestions to rebase unless you understand those suggestions 21:19:37 -!- Snarwin has quit [Read error: Connection reset by peer] 21:19:40 I understood it but question its necessity (as per |amethyst's following comment) 21:19:51 -!- CeleryMan has quit [Remote host closed the connection] 21:20:15 Yeah, I was hoping for some explanation, because I understand some of the caution necessary around rebases. Was it meant to refer to *non-public* branches only? 21:23:53 I presume I should merge upstream down into random gods when I get back to it, so that I can add canaries around the random god data. 21:24:30 -!- Fhqwhgads_ has quit [Remote host closed the connection] 21:24:57 -!- hhkb has joined ##crawl-dev 21:26:41 Is it possible to skip to just after a loop is exitted in gdb without setting and clearing a new breakpoint? 21:26:51 -!- Roarke has quit [Ping timeout: 250 seconds] 21:27:47 hmm 21:28:10 -!- mizu_no_oto has quit [Quit: ["Textual IRC Client: www.textualapp.com"]] 21:28:55 http://stackoverflow.com/questions/14651073/is-there-a-gdb-command-to-finish-a-loop-construct 21:29:58 ah, thanks 21:31:14 what does that say, use "until"? 21:31:32 yes 21:32:05 yeah I was trying to find that command in the manual because I couldn't remember its name 21:32:27 -!- raskol has quit [Quit: Konversation terminated!] 21:33:45 -!- ProzacElf has quit [Ping timeout: 252 seconds] 21:35:20 -!- sbanwart has quit [Ping timeout: 256 seconds] 21:35:56 -!- dtsund has joined ##crawl-dev 21:38:58 -!- Notorion has quit [] 21:39:08 -!- ProzacElf has joined ##crawl-dev 21:39:09 -!- MarvinPA has quit [Ping timeout: 256 seconds] 21:42:47 -!- Pacra has quit [Quit: Pacra] 21:43:24 -!- Arkaniad|Laptop has quit [Ping timeout: 246 seconds] 21:44:05 Debian builds of the trunk branch on crawl.develz.org updated to: 0.13-a0-2174-g3a0f4b0 21:44:06 -!- nickajeglin has quit [Ping timeout: 246 seconds] 21:44:23 -!- Kalir has quit [Changing host] 21:46:36 -!- Bloax has quit [Quit: Error: Bloax not found] 21:57:34 -!- pelotr0n has quit [Ping timeout: 240 seconds] 22:01:24 -!- ShadeTornado has quit [Quit: Leaving] 22:08:23 -!- ussdefiant has joined ##crawl-dev 22:13:15 -!- ackack has quit [Ping timeout: 240 seconds] 22:17:19 -!- Infinite_Monkeys has quit [Read error: Connection reset by peer] 22:18:45 -!- MaxFrosty has quit [Ping timeout: 246 seconds] 22:27:04 <|amethyst> oh, monster_info needs to be fixed too 22:33:39 -!- Grivan has quit [Ping timeout: 252 seconds] 22:34:30 -!- Flex has quit [Ping timeout: 246 seconds] 22:41:59 <|amethyst> aha, that was Roshnak's problem too 22:43:09 <|amethyst> (really a wellspring) 22:44:34 (A tiles player is asking me what elemental wellsprings will look like in tiles. <_<) 22:45:14 -!- mong has quit [Quit: Leaving] 22:46:35 -!- Lightli has quit [Ping timeout: 250 seconds] 22:47:39 make something up 22:47:51 say they'll look like elemental wellsprings maybe 22:48:31 They just logged out, so can't really do that now :b 22:49:42 you should have told them the question mark was the permanent tile =P 22:50:35 say "an E" 22:52:44 lightcyan E 22:53:18 § 22:54:29 -!- CryptoCactus has quit [Quit: Leaving] 22:55:12 -!- Datul has quit [Quit: Page closed] 22:58:12 <|amethyst> could recolour fire vortex as a placeholder 23:01:37 -!- CKyle has quit [Quit: CKyle] 23:05:33 -!- ZebTM has quit [Ping timeout: 252 seconds] 23:17:24 -!- Arendeth has joined ##crawl-dev 23:17:54 How do I get sound enabled in crawl? 23:18:09 i uncommented the wind_sound option 23:19:44 nm seems when I redownloaded it i forgot to uncomment the line 23:19:48 -!- Arendeth has quit [Client Quit] 23:19:58 -!- ussdefiant has quit [Remote host closed the connection] 23:20:17 -!- elliptic has quit [Quit: Leaving] 23:21:38 -!- elliptic has joined ##crawl-dev 23:25:01 -!- Rebthor has quit [Quit: ChatZilla 0.9.90 [Firefox 21.0/20130511120803]] 23:27:16 (Silly idea-from-tilesland of the moment: having Dazzling Spray deal extra damage or something to eyeballs and great orbs of eyes) 23:29:52 The great orb of eyes squinches its eyes shut and vibrates in agony! 23:30:17 and i defy anyone who tells me "squinch" isn't a word 23:30:26 <|amethyst> okay, this turned out much uglier than I was expecting because I forgot slow/haste actually changed monster.speed 23:30:47 |amethyst: ? 23:30:48 <|amethyst> fortunately, none of 10, 12, 25, 30 is 2/3 or 3/2 any others 23:31:15 -!- six40sword has quit [Quit: ThrashIRC v2.9 sic populo comunicated] 23:31:25 <|amethyst> Grunt: trying to restore the spectral weapon/wellspring/polymoth breakage 23:31:54 <|amethyst> I got the idea to use m.speed to tell which it's supposed to be 23:33:53 -!- bonghitz has quit [Remote host closed the connection] 23:37:56 -!- Sizz has quit [Quit: Page closed] 23:39:35 -!- bmfx has quit [Read error: Connection reset by peer] 23:43:19 <|amethyst> okay, this is ugly but appears to work with hasted/slowed/whatever wellsprings, polymoths, and weapons from all relevant points of time 23:43:24 <|amethyst> s/of t/in t/ 23:43:49 <|amethyst> not ones with the wretched enchantment, but not worth handling imo 23:44:08 (The ugly code basks in the mutagenic energy of its kin and changes!) 23:44:41 Unstable branch on crawl.lantea.net updated to: 0.13-a0-2174-g3a0f4b0 (34) 23:45:17 03|amethyst 07* 0.13-a0-2175-ga079a5c: Recover saves with bad spectral weapons/wellsprings (kilobyte). 10(17 minutes ago, 1 file, 83+ 2-) 13http://s-z.org/neil/git/?p=crawl.git;a=commitdiff;h=a079a5cb2e01 23:45:41 <|amethyst> kilobyte will probably object to having his name associated with something so ugly 23:53:29 -!- minqmay has quit [Quit: Leaving]