01:31:38 -!- Kurper has quit [Remote host closed the connection] 02:09:45 -!- bhaak has quit [Ping timeout: 250 seconds] 02:15:59 -!- Zaba has quit [Ping timeout: 255 seconds] 02:22:27 -!- Zaba has joined ##crawl-dev 02:35:21 moin 02:38:26 Zaba! 02:38:40 Heavy armour feels very strong now 02:39:57 -!- galehar has joined ##crawl-dev 02:40:25 PIIIIIIIIIIIIIIIILLs. 02:40:28 Also, wrong channel, but hi anyway. 02:41:28 I just got a giant beetle boss in ZotDef that totally obliterated my defences :P 02:41:42 :D 02:41:50 Sweet! 02:42:02 has anyone other than me gotten a giant spore boss? 02:42:19 a ball lightning boss would be peculiar too 02:43:42 Damn imp bosses :P 02:44:26 imps are fucking nasty 02:44:28 what exactly is the purpose of those treasure chambers in zotdef? 02:45:03 To contain treasure, presumably 02:46:04 Damn, felled by a mere orc :P 02:51:05 -!- Textmode has quit [Ping timeout: 255 seconds] 03:24:43 http://forums.somethingawful.com/showthread.php?threadid=3332877&userid=0&perpage=40&pagenumber=136#post385742043 03:24:58 ghooostt mmmoootthhhh 03:27:22 This is even better, no idea it could work this well: http://forums.somethingawful.com/showthread.php?threadid=3332877&userid=0&perpage=40&pagenumber=96#post383554065 03:28:36 What vault is that loot-filled metal-walled room? 03:28:57 In http://forums.somethingawful.com/showthread.php?threadid=3332877&userid=0&perpage=40&pagenumber=136#post385740703 03:30:27 Also do we have a vault with a loot item at the end of a corridor protected by a shaft trap? 03:30:41 Because if not, we should. Ideally 2-3 shaft traps :P 03:30:47 I recall one 03:30:55 That vault is odd. 03:31:08 it may have been removed, though, because it's easy to detect and avoid with manual play, and screw over autoexplorers 03:31:16 screw/screws 03:32:30 : crawl.mpr("A basket of spiders falls from the above!") 03:32:36 That looks oddly phrased 03:33:23 remove the the? 03:33:50 Aye 03:34:45 that loot-filled vault is strange 03:34:55 I've seen it too 03:35:38 threw me off quite a bit, being free loot and all 03:36:13 03greensnark * r9ea01f5a050b 10/crawl-ref/source/dat/des/variable/traps.des: Fix trap trigger wording in spider basket trap. 03:42:40 !lg * map=~corexii 03:42:42 22. Icelos the Grave Robber (L3 DENe), slain by a hobgoblin (a -1,+1 orcish club) on D:3 (corexii shack village) on 2010-12-11, with 199 points after 2718 turns and 0:06:51. 03:42:48 !lg * map=~corexii s=map 03:42:49 22 games for * (map=~corexii): 8x corexii shantytown, 8x corexii shack village, 2x corexii checkpoint b, 1x corexii hall statues i, 1x corexii checkpoint a, 1x corexii leaking fountain, 1x corexii hall statues l 03:42:57 !lg * map=~shantytown -tv 03:42:58 8. Luquillo, XL8 HOCK, T:5237 requested for FooTV. 03:44:59 !lg * map!~entry map!= s=map 03:45:00 8283 games for * (map!~entry map!=): 841x uniq crazy yiuf cottage, 286x minmay sewer w, 257x sewer kobolds, 238x dpeg cavernosity, 236x shiori shiawase, 192x lemuel flame loot 1, 165x lemuel castle, 144x minmay sewer treatment, 134x tgw xom, 115x sewer frog island 02, 112x forest paths, 108x bailey polearm 2, 108x bailey polearm 5, 90x hall of Zot, 88x fountainhead, 86x vaults vault, 82x ossuary t... 03:47:32 in what des file would a vault consisting entirely of a net/zot trap in a chokepoint (to take advantage of their special properties) belong, or is that just too silly or unfriendly (just let the rng do it)? 04:04:29 mini_features.des 04:11:18 -!- dpeg has joined ##crawl-dev 04:11:23 hi ploog 04:11:47 dpeg, did you know that your lossy-artefact ridden relative graced us with her presence yesterday? 04:13:43 I had no idea! 04:14:03 But I did have an idea for Ashenzari... 04:14:49 Compress him with aforementioned file format? 04:15:13 The god should stay as passive as now, but we want to strenghten him in battle (as battle is what a large portion of crawling is about). So we already have monster detection, and this could be improved a little. But what if Ashenzari would sometimes foretell moves of opponents? 04:15:39 I am thinking of "Boris is about to use IOOD soon." rather than "Boris will move northeast." 04:16:02 !seen galehar 04:16:03 I last saw galehar at Wed Dec 15 08:39:57 2010 UTC (1h 36m 6s ago) joining the channel. 04:16:32 ooh, I like that. 04:16:48 It doesn't have to be programmatically passive though. 04:17:34 Boris is about to kill you. 04:17:59 sorear: <3 04:18:08 due: it is passive in the sense that no activity on behalf of the player is necessary. 04:18:14 sorear: could be actually used, yes. 04:19:02 Well, we could send the message "x is about to do y", and then set a flag meaning that they will do this. 04:19:10 Rather than making the monster AI predictable. 04:19:23 sorear: <3 also. 04:20:13 due: yes, that's what I meant. 04:20:45 We could restrict to spells, and Ash announcing a spell means that the monster will use that spell soon, or next time it sees the player (should he cowardly flee). 04:22:26 It would only apply to monsters with spells, too, yes? 04:22:30 So not to breath weapons, blah blah. 04:25:45 yes, for a start 04:25:59 not sure other events are attention-worthy enough 04:26:13 Using escape items such as potions? 04:27:44 not sure that's relevant enough 04:27:50 the player is not going to die from this :) 04:27:59 telling what items monsters have could be nice 04:28:13 watch out for that imp it has a wand of paralysis 04:28:26 monky: good idea, yes 04:29:55 hey dpeg 04:29:55 galehar: You have 1 message. Use !messages to read it. 04:30:00 !messages 04:30:00 (1/1) jpeg said (13h 2m 15s ago): Is it intentional that level 0 skills show their progress in the "show all" toggle? 04:30:49 good point 04:31:05 should I hide the progress of level 0 skills? 04:34:33 have ash brand all monsters that carry weapons of draining 04:34:54 distortion too 04:35:33 galehar: yes, hiding them is good. 04:37:51 -!- monky has quit [Quit: hello] 04:37:52 -!- Mu_ has joined ##crawl-dev 04:37:55 if you predict a move, set a flag and then force the monster to do it, then the monster behaviour is changed when you worship Ash. Even if only by a little bit, it might leads to problems. 04:38:05 I agree with galehar. 04:38:09 Also hi! 04:38:13 hi :) 04:38:28 detecting wands and weapon brand is good 04:39:14 galehar: hm, you misunderstand 04:39:30 galehar: set a flag saying "if you would normally cast this spell, then do so now". 04:39:39 galehar: and the code that determines which spell to use can simply be copied from the monster AI. 04:40:37 predicting spells is a huge can of worms 04:40:55 the monster's estimation is not going to stay put, too 04:41:35 if the monster would fireball your ass a round before, he won't do that if you stepped next to him or an ally came close to you 04:41:50 Hm. 04:41:53 That's true. 04:41:55 So in that case. 04:41:56 !tell jpeg good point. I'll fix it. 04:41:57 galehar: OK, I'll let jpeg know. 04:42:00 "The monster may shortly cast ". 04:42:42 but then it's useless 04:43:18 -!- Mu_ has quit [Quit: Defecator, may everything turn out okay so that you can leave this place.] 04:43:49 interesting thing: the console version doesn't build, yet there were several commits by three different people since then. WTF? 04:45:11 Hah. 04:45:17 People aren't checking their commits! Naughty. 04:46:31 -!- Mu_ has joined ##crawl-dev 04:47:00 another idea: give away a part of the monster spell list 04:47:29 You can do that with spoilers. 04:47:50 sometimes they are chosen randomly from several 04:48:52 Ah, yeah. 04:57:06 kilobyte: Most folks don't try to recompile after rebasing :P 04:57:33 * greensnark only changed Lua and is therefore blameless and above censure and utterly magnificent. 04:58:29 well, all but the last 3 04:58:53 Why this parsimony. Would it kill you to concede at least, say, the last point? 04:59:37 Of course it wouldn't kill me. But I also do not like to give off an impression of untruth. 05:00:04 Good, so you do concede the last point. You could have just said "Ok" without all this extreme verbosity. 05:00:48 But that wouldn't have led to your misinterpretation of my words! 05:01:19 Words mean exactly what I want them to mean, nothing more and nothing less. 05:01:36 Especially other people's words 05:01:51 !coffee 05:01:51 * Henzell hands greensnark a mug of café au lait, brewed by Snorg. 05:02:07 I need to add more coffee types 05:02:11 And more brewers 05:06:58 kilobyte: I've just compiled the console version without problem (not including the very last trivial commit by Darshan) 05:07:40 galehar: tags.cc has tiles references in the console build 05:12:31 !coffee 05:12:31 * Henzell hands greensnark a cup of cappuccino, brewed by Makhleb. 05:12:35 There we go 05:14:17 makhleb?! 05:14:46 galehar: then where did your compiler take env.tile_bk_bg from? 05:16:36 I don't know 05:16:44 I can't look into it right now, I have to go 05:16:50 see u later 05:17:03 kilobyte: You fixing that? 05:19:25 due: Is rax around on ##c-o these days? 05:19:43 cao is running low on disk space again and there's very little free space to shuffle ttyrecs off to 05:20:19 Only 12G of space in the chroot, which means about 5-8 weeks to go 05:20:21 greensnark: Yes 05:20:26 Though not likely at 5am. 05:20:39 Are the ttyrecs bzipped? 05:20:43 Yes 05:20:53 Is the 7zip format any better for compressing them? 05:20:53 There's a cron job that bzips ttyrecs 05:21:07 I remember from experience that 7zip can be much, much ... zippier. 05:21:12 7z is better, yes 05:21:26 it compresses only slightly better 05:21:32 And recompressing all those ttyrecs will take up a lot of time and CPU 05:21:36 Yeah. 05:21:52 kilobyte: It's better for text, I at least seem to recall. 05:21:59 (I managed to get about 100mb of text down into 7mb.) 05:22:13 and there is no such format at "7zip". It is a program that bundles more than ten various compressors in one binary. 05:22:36 lzma :) 05:22:51 you can use lzma without 7zip 05:22:55 Yes, I know. 05:22:58 and you really should use xz instead 05:23:03 I was using 7zip as a synonym for lzma. 05:23:15 Well, the lzma family. 05:23:40 Most people say 7z when they mean lzma because it's the default 05:23:41 but, here's a comparison: xz compresses somewhat better than bzip2, compresses slower, decompresses A LOT faster, uses A LOT more memory 05:24:23 it is much better than bzip2 if you 1. don't care about memory, 2. compress once, decompress many times 05:24:27 Feel free to slap me a lot; I like it. 05:24:30 We could save a lot of space on cao by identifying startscummers and tossing those ttyrecs 05:24:40 !gamesby Meow 05:24:40 Meow has played 21597 games, between 20071031 and 20101016, won 2 (0.0%), high score 7203896, total score 20063257, total turns 6714101, total time 663:24:37. 05:24:44 But that's too much work 05:24:46 Meow has been eclipsed 05:24:51 Oh? 05:24:52 Meow was a bit player 05:24:55 !gamesby syban 05:24:56 syban has played 11930 games, between 20090920 and 20101215, won 11 (0.1%), high score 10870709, total score 55570192, total turns 33234357, total time 1164:58:21. 05:25:01 Not him, then 05:25:08 !gamesby Sebi 05:25:09 Sebi has played 70051 games, between 20080418 and 20101210, won 0, high score 197916, total score 4597448, total turns 3629059, total time 735:37:15. 05:25:12 Ah right 05:25:15 ... 70051, wtf? 05:25:28 !lg * ((quitting||leaving) s=name 05:25:28 Malformed argument: leaving) 05:25:35 !lg * ((quitting||leaving)) s=name 05:25:38 264109 games for * (((quitting || leaving))): 68317x Sebi, 20754x Meow, 10756x syban, 8228x adamzap, 7926x Ivo, 7264x TGW, 5737x Xiberia, 5604x Fieros, 4545x uru, 4432x Arved, 4233x Donatan, 4030x Eronarn, 3613x dirge23, 3155x Eir, 2914x DrPraetor, 1903x oxeimon, 1620x tcircuits, 1472x ion, 1461x b0lt, 1426x Gamerha, 1334x KiloByte, 1310x syllogism, 1267x chilliwack, 1121x Covenant, 1118x Yorikke,... 05:26:05 you might also want to delete certain mummies 05:26:17 Hmmm. 05:26:25 !lg * race=mu max=dur 05:26:25 52110. Cuckatoo the Farming Talismancer (L27 MuFi), worshipper of Trog, escaped with the Orb and 61 runes on 2007-08-01, with 2200969 points after 3190190 turns and 224:36:31. 05:26:31 !lg * race=mu max=turns 05:26:32 52110. Stabwound the Farming Skullcrusher (L27 MuSu), worshipper of Nemelex Xobeh, escaped with the Orb and 111 runes on 2008-03-09, with 9425283 points after 92568259 turns and 128:15:16. 05:26:34 !lg * race=mu max=turns -2 05:26:35 52109. Zornaq the Farming Annihilator (L27 MuRe), worshipper of Sif Muna, escaped with the Orb and 20 runes on 2009-09-26, with 2324410 points after 17521353 turns and 36:21:51. 05:26:39 !lg * race=mu max=turns -3 05:26:40 52108. Zornaq the Farming Annihilator (L27 MuNe), worshipper of Sif Muna, blasted by a smoke demon (divine providence) on Tomb:3 on 2009-09-14, with 1484457 points after 10481497 turns and 24:04:23. 05:26:42 !lg * race=mu max=turns -4 05:26:42 52107. KiloByte the Invulnerable Farmer (L27 MuFi), worshipper of Okawaru, escaped with the Orb and 16 runes on 2008-12-03, with 2213368 points after 8213459 turns and 32:23:58. 05:26:53 Could you get the relevant game_ids and dates from Sequell and then use them to determine which ttyrecs to be pruned? 05:27:02 !hs Sebi 05:27:02 70051. Sebi the Eclecticist (L18 DESt), worshipper of Sif Muna, mangled by a troll (led by an orc knight) on Vault:4 on 2010-08-12, with 197916 points after 77943 turns and 20:00:00. 05:27:07 !hs Meow 05:27:07 21597. Meow the Farming Swordmaster (L27 MfWz), worshipper of The Shining One, escaped with the Orb and 18 runes on 2009-10-12, with 7203896 points after 257090 turns and 27:42:52. 05:27:35 Easier to read the logfile directly 05:27:56 True, I suppose. 05:28:02 "Quit the game on turn 0", etc? 05:28:06 But I'd have to be really careful with auto-purge scripts so I'm not eager to dive in until space gets even more critical :P 05:28:43 How much space would be needed? I suppose we could throw together for offsite storaage... 05:29:05 But then the actual transfer would take up enough IO etc to be not worthwhile. 05:29:18 do we have an estimate how big part of ttyrecs are due to startscumming? 05:29:36 rax mentioned giving the VM more space which would be really nice 05:29:37 I'd say this might be a significant _number_ of files, but not _size_ 05:29:42 True 05:30:11 They're just the easiest thing to go after if we're deleting stuff 05:30:13 !lg * s=xl 05:30:14 869312 games for *: 403640x 1, 97552x 3, 96487x 2, 72728x 4, 46462x 5, 31700x 6, 25245x 7, 21144x 8, 18545x 9, 14935x 10, 10745x 11, 6402x 12, 4287x 13, 3795x 27, 3471x 14, 2704x 15, 2081x 16, 1440x 17, 1099x 18, 882x 19, 711x 20, 602x 26, 559x 25, 533x 21, 529x 24, 518x 23, 510x 22, 6x 0 05:30:32 Quantity over quality, though. 05:30:36 !lg Sebi s=xl 05:30:36 70051 games for Sebi: 69121x 1, 465x 2, 191x 3, 100x 4, 43x 5, 38x 6, 28x 7, 16x 8, 16x 9, 12x 10, 9x 11, 5x 12, 2x 13, 2x 17, 2x 14, 1x 18 05:31:02 !lg Meow s=xl 05:31:02 21597 games for Meow: 20319x 1, 709x 2, 187x 3, 96x 4, 46x 5, 36x 7, 32x 10, 31x 6, 30x 9, 27x 8, 26x 11, 11x 12, 8x 13, 7x 16, 6x 27, 4x 19, 4x 15, 4x 14, 4x 20, 3x 17, 3x 18, 2x 21, 1x 24, 1x 25 05:31:13 If you have thousands of small ttyrecs, it would add up. 05:31:23 !ttyrec Sebi ((quit||leaving)) xl=1 05:31:36 Oh right, that's going to hit the http listing wall 05:32:02 68257. Sebi, XL1 DEAr, T:1: http://crawl.develz.org/ttyrecs/Sebi/ 2010-12-10.11:52:22.ttyrec.bz2 2010-12-10.11:52:24.ttyrec.bz2 05:32:10 Really now 05:32:21 How did he manage two ttyrecs on a 1 turn quit 05:32:30 Maybe it's a mistake :P 05:33:56 !lg Sebi ((quit||leaving)) xl=1 x=cv 05:33:57 68257. [cv=0.8-a] Sebi the Charlatan (L1 DEAr), got out of the dungeon alive. on 2010-12-10, with 20 points after 1 turn and 0:00:01. 05:34:15 waaait... what's the point of scumming a DEAr in 0.8? 05:34:28 These startscummers are not exactly models of rational thinking 05:34:44 for a +8 ring of slaying in sight 05:34:56 it did give a significant advantage when it actually worked. But today... 05:35:11 casmith789: you don't know the ring's plusses 05:35:21 Maybe they don't know that stats are fixed now and assume they're just having a really weird long streak of constant stats 05:35:22 you also have to scum for a scroll of ID in sight then 05:36:02 maybe scumming for specific entry vaults? 05:36:14 Possible 05:36:14 elliptic: The orb of fire one, say? 05:36:32 I just came across one which was full of stones 05:36:42 so you could scum for that one as an EE 05:36:47 I think it honestly doesn't matter 05:36:58 there is one with a lugonu altar 05:37:04 Well, it doesn't matter to the point of changing the game. 05:37:11 oh my god -- it's full of stones! 05:37:21 greensnark: Can I have your star babies? 05:37:48 * greensnark doesn't do stars. 05:37:49 maybe he's scumming for a rod as a Ar 05:38:04 given that rods are ridiculously rare and are the most fun way of playing evocations 05:38:29 I think all these theories make the mistake of assuming Sebi has a good reason for startscumming 05:38:56 when greensnark startscums, he has a reason! 05:39:06 :( 05:39:48 incidentally, zotdef really encourages startscumming for particular layouts 05:40:19 elliptic: I've noticed that, not that I had much time to do anything more than die a few times. 05:40:23 obviously this is fairly mild startscumming because there aren't many options, but it is still sort of strange 05:40:44 I startscum Sprint 1 to get Ijyb without a nasty weapon. 05:40:54 But it's not really scumming because I can't help but die to him. 05:46:19 I startscum games to get ones when I can get alive past D:1 05:46:30 heck, I startscum games to get ones when I can win 05:46:51 kilobyte: Did you commit that console compile fix yet? :P 05:48:25 03kilobyte * r86e40da49d07 10/crawl-ref/source/tags.cc: Fix debug non-tile builds. 05:48:45 Thanks 05:48:59 Maybe galehar builds non-debug builds 05:49:32 Btw, did someone change the random name generator? It seems to produce much better names than it used to 05:50:58 No, we upgraded your firmware. 05:51:32 Ok, I have had it with the centaur bosses at XL 4 :P 05:52:15 \o/ 05:52:22 :D 05:53:27 re 05:55:39 hi zaba 05:55:52 !coffee Zaba 05:55:52 * Henzell hands Zaba a mug of black coffee, brewed by Trog. 05:58:59 what up 05:59:40 Trog is such a racist. 06:03:01 -!- syllogism has joined ##crawl-dev 06:20:48 TROG IS A RACIST HOG! :D I'd say pig but it doesn't rhyme. 06:21:53 !coffee due 06:21:53 * Henzell hands due a cup of cappuccino, brewed by Ijyb. 06:27:41 -!- Hehfiel has quit [Remote host closed the connection] 06:29:35 :D 06:36:05 -!- Hehfiel has joined ##crawl-dev 06:39:03 The new crosstraining highlight on the skills screen is nice 06:40:20 greensnark: yes, galehar's work is impressive 06:40:44 The screen now explains things many players were not aware of, I think. 06:41:38 There's definitely a slow period in ZotDef between say XL 5 - 13ish 06:41:53 When your defences easily keep up with the attack waves 06:42:22 Things only get lively if you get a wave you're completely unprepared for like a jelly wave 06:42:31 Otherwise that early period is a yawn 06:42:50 greensnark: yes, the game could be sped up- 06:43:21 Would simply scaling up wave threats do the trick? 06:43:26 I think so 06:43:52 Still haven't played enough to make recommendations though :) 06:44:14 But that middle comfort zone feels pretty grindy :P 06:44:20 the game up to xl3-4 is basically start-scumming: there's a small chance the monsters will hit you just rarely and weakly enough to survive 06:44:36 Yeah, very early game is tricky unless you have a very strong character 06:44:45 trolls are likely to pass through it, but other races, not really 06:44:51 And when you get a good boss early on like a centaur :) 06:45:12 s/tricky/random/ -- there is little strategy you can do, it's mostly pure luck 06:45:35 Luck of the wave, yeah, because you can't menuever 06:45:39 Er, maneuver 06:45:42 kilobyte: I believe the early game can (and should be) addressed after the threat curve is better established. 06:45:58 But I don't mind early game lethality as much as the XL 5 - 13 lull 06:46:03 yes 06:46:21 -!- Vandal| has quit [] 06:46:21 once you get a couple of oklobs, you are mostly good to go 06:46:29 The good thing is that once past the lull things start ramping up nicely again 06:47:03 an interesting term for "1-2 fire giants per turn" 06:47:09 :D 06:47:11 maybe it's not just about the threat, but also about the available abilities? 06:52:24 -!- Vandal| has joined ##crawl-dev 06:53:47 you get a vital ability, oklobs, at xl7, and then nothing useful till the last few levels 06:53:55 !lg * zotdef s=xl 06:53:55 3069 games for *: 1541x 1, 578x 2, 302x 3, 163x 4, 127x 5, 105x 6, 58x 7, 56x 8, 31x 10, 28x 9, 14x 12, 14x 11, 11x 14, 10x 13, 8x 15, 5x 16, 4x 19, 4x 18, 3x 17, 2x 27, 2x 22, 1x 26, 1x 23, 1x 25 06:54:15 and about no one gets there 07:00:22 !lg * zotdef win 07:00:23 2. Ereshkigal the Demonologist (L27 MuSu), worshipper of Vehumet, escaped with the Orb and 15 runes on 2010-12-12, with 67988863 points after 21083 turns and 4:39:33. 07:00:27 Still no third win 07:00:56 greensnark: I would have won but bugs ate my runes 07:01:16 !lg . zotdef 07:01:16 117. hyperbolic the Englaciator (L26 DEIE), worshipper of Vehumet, got out of the dungeon with 12 runes on 2010-12-15, with 491450 points after 22434 turns and 7:37:01. 07:01:17 I believe the powers can be balanced out. 07:01:17 Missing towers bug or did runes go missing? 07:01:32 runes went missing and/or never generated 07:01:51 For example, traps could make more damage, the more you place of a type. 07:01:57 Interesting 07:02:00 this had something to do with the fact that the game is unplayable without nemelex or jiyva to control the item count 07:02:18 for the last several thousand turns, I was constantly spammed with messages about having reached the item limit 07:02:52 right, no items in LOS ever get culled 07:03:22 kilobyte: oh, is that true? that might have helped had I known it 07:04:25 -!- casmith789 has quit [Ping timeout: 245 seconds] 07:10:18 do we have a save from an advanced ZotDef game somewhere? 07:17:45 Don't think so 07:18:01 I'm sure Mu would be happy to provide us with one 07:19:40 re 07:20:08 galehar: Do you generally build non-debug builds? 07:20:23 no 07:20:35 I usually build either debug tiles or non-debug console 07:20:46 because I don't know how to debug console 07:21:00 with either gdb or ddd, it doesn't work well :( 07:22:55 galehar: You can start Crawl and attach gdb from another terminal 07:23:03 Oh, are you on windows or something? 07:23:14 I think Cygwin gdb behaved oddly if you tried that 07:23:40 I used to dev on windows, but now I'm on linux (except when at work) 07:24:02 I'll try attaching, next time I have to debug console 07:24:06 I got vampiricplant (xl 13), pointless (xl 14), dpeg (xl 10) 07:24:23 kilobyte: Oh, I thought you meant xl>20 07:24:38 yeah, but there are none :( 07:24:53 I keep getting bear waves and amazingly they don't totally suck :P 07:25:00 A polar bear took out an ice status :'( 07:25:21 yay incompatible lua in the saves :( 07:25:51 You can probably ignore it safely, no? 07:26:04 I don't think zotdef does anything with Lua 07:26:19 I wonder why we don't use system lua on cdo -- Windows is the only platform that uses contrib 07:26:21 We need to fix that sometime 07:26:34 Saving Lua bytecode in saves, I mean 07:26:42 It's not really necessary 07:26:43 and our change to Lua is to be compatible over endianness, right? 07:26:57 I don't know if that was actually done 07:27:00 re: Startscumming Zotdef, can we get sprint-like map selection? That way we are free to devise maps of different difficulties, and other creative stuff.. 07:27:24 if that's the case, it's pointless since Windows doesn't exactly have variable endianness 07:28:48 Stupid polar bears keep attacking my expensive ice statues :P 07:29:00 greensnark: you need some burning bushes in the mix 07:29:09 I have them at the other end of the hall 07:29:12 speaking about pointless, his save has mounds of armour and glowing daggers 07:29:20 Keskitalo: I support this idea. 07:29:23 I also support bed, night. 07:29:24 uhm, you need to mix 07:29:28 due: bie! 07:29:33 night! 07:29:56 there's no benefit in having identical turrets next to each other, but a serious downside 07:30:49 (I mean, oklobs as the bulk of your force will be next to each other, but it's about not having ONLY oklobs) 07:33:40 Huh, Sigmund showing up pretty late :P 07:33:56 well, still on D:1 :p 07:38:22 Oka haste was already considered to be very expensive. With the haste nerf, shouldn't we reduce the price? 07:40:10 not sure, but he can take it 07:40:41 with the AC boost, we can disable Okawaru 07:41:38 Lets implement all the Okawaru proposals, and generate 2-3 of them per game 07:41:46 re Ash: what if instead of announcing a monster's move we have the player react to it (positively). "Boris shoots an IOOD. You saw that coming." (and evasion) ? 07:42:40 We have killed species. We have killed backgrounds. Now is the time to kill a god. 07:42:49 Chei 07:42:51 Zin 07:43:19 -!- dpeg has quit [Quit: leaving] 07:43:26 heh 07:45:37 I just checked: it's the endianness patch that causes save incompatibility 07:48:30 Stupid skeletal warrior wave is causing zombie issues :P 07:48:38 At least I haven't hit a jelly wave yet 07:49:49 Zin fix seemed easy, people seemed to enjoy stabber Zinnites when That One Bug was in master. 07:50:23 I know Eronarn is working on a more comprehensive Recite overhaul, but I'll take a look at that.. 07:50:53 And Chei is awesome. :P 07:51:02 greensnark: can you learn basic necromancy? 07:51:07 or no books? 07:51:13 No books 07:51:45 But my greater servants took them out 07:55:47 -!- MarvinPA has quit [] 08:05:21 kilobyte: http://crawl.develz.org/saves/dumps/greensnark-zotdef-09dc7e2-101215-1405.tar.bz2 if you want another ZotDef save 08:05:29 * greensnark finally going back to work. 08:05:40 @whereis 08:05:40 greensnark the Covered (L1 HuFi) saved on D:1 on 2010-12-08 after 13 turns. 08:05:53 thanks, but I already know what's going on 08:06:01 armour, glowing daggers, glowing clubs... 08:06:04 What's the bug? 08:06:19 Oh, item cull fail? 08:06:38 1. we need a better definition of junk, 2. you spend most of the time in view of all items 08:06:59 but for the lua issue: the patch neunon applied is buggy 08:07:43 what about just ignoring the endianness issue and reverting the patch? 08:07:55 since I can't find an obvious fix 08:09:09 Sounds fine to me 08:09:28 We should anyway stop squishing Lua bytecode into saves at some point 08:14:01 -!- Lollipop has joined ##crawl-dev 08:14:27 -!- Pialein has quit [Ping timeout: 240 seconds] 08:35:10 Morning 09:16:53 -!- Hehfiel has quit [Remote host closed the connection] 09:19:33 -!- Hehfiel has joined ##crawl-dev 09:29:16 -!- Hehfiel has quit [Ping timeout: 276 seconds] 09:29:24 -!- Hehfiel has joined ##crawl-dev 09:34:47 -!- Hehfiel has quit [Ping timeout: 250 seconds] 09:36:20 -!- Hehfiel has joined ##crawl-dev 09:40:29 -!- Hehfiel has quit [Ping timeout: 240 seconds] 09:42:33 -!- Hehfiel has joined ##crawl-dev 09:54:23 squish! squish! 09:54:26 yay squishing! 09:54:27 -!- lorimer_ is now known as lorimer 09:54:49 remember, it's always ok to cram all kinda crap into savefiles. who'll ever look in there? :D 10:00:43 Hey, it's lorimer in a tutu 10:01:33 greensnark, what is lua bytecode saved for, anyway? 10:01:44 No good reason 10:01:52 well, it's used for *something*, right? 10:02:08 The functions used to reload Lua markers are saved into the save file as bytecode 10:02:26 Also currently map Lua is saved, but it is not used for anything at all 10:02:27 what would be the proper way of doing it? 10:02:42 Register marker unmarshall functions by name and save the name of the unmarshall function 10:03:15 hm 10:04:34 greensnark: this is more of a four-four 10:16:57