01:24:19 -!- Paroid has quit [Ping timeout: 265 seconds] 01:27:14 -!- monky has quit [Read error: Connection reset by peer] 01:34:46 -!- monky has joined ##crawl-dev 01:36:19 -!- Textmode has quit [Ping timeout: 240 seconds] 01:37:04 -!- TGWi has quit [Read error: Connection reset by peer] 01:50:17 -!- Radoslav_ has quit [Ping timeout: 245 seconds] 01:59:01 moin 02:17:39 okay. I just implemented a Rune of Zot into 100 Rogues 02:18:03 it will randomly do one of two things: give you a new Skill Point (it's got a Diablo2-esque skill tree) OR reduce your maximum Energy Points (mana) 02:18:15 so, a sort of crapshoot 02:18:30 hope that sounds OK. Wanted it to be chaosey 02:18:50 something xom-related would work better for chaos, in my opinion 02:19:24 Isn't Zot a chaosey realm 02:20:32 -!- Noom has joined ##crawl-dev 02:21:06 zot isn't really well-defined, as far as I know 02:21:23 there are a few chaotic things there, but it's mostly draconians and dragons 02:22:22 it's really all over the place 02:22:27 yeah, zot isn't as much of a chaos realm as abyss, for example 02:25:26 ohhh right 02:25:30 I was thinking of the abyss actually heh 02:25:48 Well... What would you think consuming a "rune of zot" would do 02:25:49 to a person? 02:25:57 like "using" it 02:26:05 if it cast a spell on you, or something, what do you think it would do? 02:27:37 they actually don't do anything other than unlock the realm of zot 02:28:02 I'd advise copying a different item or making up an item that references or contains the properties of something crawlie 02:28:33 but don't do hand or tooth of xom because linley already used those in white butterfly and garden of coloured lights, respectively 02:30:43 -!- Pseudonut has quit [Remote host closed the connection] 02:34:56 zot is a crashed spaceshit 02:34:58 ship 02:40:39 and what's the orb of zot, then? 02:41:29 it could be its drive core, but then, it doesn't explode on any random occasion, so probably no 02:42:20 some piece of valuable cargo 02:42:29 you're a looter 02:43:05 hmm. 02:43:16 goddamn it 02:43:20 _builder_basic is so crappy 02:44:43 just _why_ does it build a bunch of rooms 02:45:05 a level built entirely with _make_trail already looks interesting 02:45:27 and there are more than enough other layouts that build bunches of bloody rooms 02:45:41 except for they don't do it on top of an existing good-looking level 02:47:07 -!- syllogism has joined ##crawl-dev 02:47:24 oh bloody hell. 02:47:43 just how many functions are there in dungeon.cc for building a 'room' 02:48:36 with each used in one or two places 02:48:46 -!- valrus has joined ##crawl-dev 02:53:12 -!- valrus has quit [Ping timeout: 245 seconds] 02:58:38 -!- Noom has quit [Ping timeout: 255 seconds] 03:00:37 I wonder.. can 4 + random2(5) + random2(6) be simplified somehow? 03:01:22 why would you need so many randoms in the first place augh 03:01:50 monky: for extr arandom 03:02:12 well, I wonder whether it can be reduced to a single random2 call 03:02:19 so it's nearly equivalent to the old version 03:02:19 I imagine it produces a desirable distribution, but I don't see what's so special about that 03:02:50 classical RPGs sometimes want more than one die 03:02:54 it's the same principle 03:02:54 I'd probably change it to 3 + 2d6 or 5 + 2d5, myself, but maybe I'm crazy 03:03:40 err 03:04:28 random2(n) is 1dn - 1, isn't it? 03:04:44 oh perhaps 03:04:56 it's random_range(0 to n). 03:05:12 random2(10) = returns a value of 0 through 9. 03:05:21 yeah 03:06:00 I entirely forgot about that 03:06:05 -!- monky has quit [Quit: hello] 03:06:09 int temp_rand = random2(750); int no_rooms = ((temp_rand > 63) ? (5 + random2avg(29, 2)) : (temp_rand > 14) ? 100 : 1); 03:06:14 could this perhaps be simplified somehow, too 03:06:18 because.. well, seriously. 03:06:26 what the hell. 03:06:37 random_choose_weighted 03:07:08 what would the equivalent weights be? 03:07:22 according to comments, the odds are 91.47%, 6.53% and 2%, respectively 03:07:40 That probably came before randomm_choose_weighted existed. 03:07:42 so I guess it'd be something like 92, 6 and 2? 03:07:57 636, 49, and 15 03:10:01 okay 03:10:31 -!- Niccus has joined ##crawl-dev 03:11:19 -!- Niccus has quit [Read error: Connection reset by peer] 03:12:27 -!- Niccus has joined ##crawl-dev 03:12:45 -!- Niccus has quit [Read error: Connection reset by peer] 03:14:03 -!- Niccus has joined ##crawl-dev 03:14:15 -!- Niccus has quit [Read error: Connection reset by peer] 03:15:38 -!- Niccus has joined ##crawl-dev 03:15:48 -!- Niccus has quit [Read error: Connection reset by peer] 03:18:25 sorear, temp_rand = random2(1200); if (temp_rand < 417) drawing = DNGN_ROCK_WALL; else if (temp_rand < 156) drawing = DNGN_STONE_WALL; else if (temp_rand < 40) drawing = DNGN_METAL_WALL; /* How can one simplify this?.. */ 03:21:43 sorear, seems like the chance of drawing to be changed is 417/1200---could that be moved out into a separate one_chance_in(3)? 03:22:09 Zaba: that code is broken as pasted, the else ifs are unreachable 03:22:41 more like x_chance_in_y, unless you *want* to be rounding stuff 03:23:08 I wonder whether the difference between 417/1200 and 400/1200 is negligible 03:24:31 and well, the code appears in exactly the same way in _plan_4, at least as far as I can see 03:25:53 -!- elliptic has quit [Ping timeout: 264 seconds] 03:26:08 sorear, huh.. I wonder how to fix this. 03:26:09 a lot of distributions in Crawl are ludicrously overspecified 03:26:21 yeah. 03:26:50 6586 // NB: comparison reversal here - combined 03:26:53 * sorear thinks most everything should become one of random_range, random_normal, random_lognormal 03:28:15 sorear, what should be done to this specific case? As you've pointed out, the else ifs are indeed unreachable, so it is basically a something-in-1200 chance of drawing being set to DNGN_ROCK_WALL, right? 03:29:11 Zaba: i'd reverse the comparisons 03:29:18 longstanding bugs are still bugs 03:29:22 hm. 03:29:28 okay.. 03:29:38 it can be rebalanced later if it causes a problem 03:29:44 sorear, would you consider the difference between 417 in 1200 and 400 in 1200 negligible? 03:29:51 yes 03:30:19 okay.. 03:30:56 -!- Niccus has joined ##crawl-dev 03:30:59 -!- Niccus has quit [Read error: Connection reset by peer] 03:32:25 -!- Niccus has joined ##crawl-dev 03:32:28 -!- Niccus has quit [Read error: Connection reset by peer] 03:32:36 sorear, I wonder what's the best way to use random_choose_weighted for dungeon features 03:33:56 I guess I just need to cast its return value 03:34:21 static_cast(random_choose_weighted(blah)) 03:36:42 ok, seems to work 03:44:21 03zaba * r18c9eceaab22 10/crawl-ref/source/ (dungeon.cc dungeon.h): Remove unused spec_room::random_spot(). 03:44:22 03zaba * r11d4efe95fe7 10/crawl-ref/source/dungeon.cc: Make _builder_basic correctly place the second pair of stone stairs. 03:44:24 03zaba * r8e62dd779d2c 10/crawl-ref/source/dungeon.cc: Move room building from _builder_basic into a separate function. 03:44:25 03zaba * r6643f6037cc2 10/crawl-ref/source/dungeon.cc: Use random_choose_weighted in _plan_4 to simplify random choices. 03:52:38 g'morning! 03:52:56 hey duegod! did you finish merging zotdef? 03:53:17 -!- Pseudonut has joined ##crawl-dev 03:54:10 Napkin: kilobyte did it instead 03:54:14 -!- monky has joined ##crawl-dev 03:54:56 ah, kilobyte? did you finish merging zotdof? 03:55:16 i'd like to add it to CDO soon 03:55:17 yes 03:55:19 he did 03:55:49 so I can access it the same way as sprint? awesome :) 03:56:00 is zotdef better now? 03:56:25 Napkin: ./crawl -zotdef, I think. 03:56:34 it's in the master branch now, casmith789 - so i trust it is :) 03:56:34 casmith789: I don't know, but it's in trunk so it getting better is inhernetly easier. 03:56:42 ah. 04:17:09 -!- Radoslav_ has joined ##crawl-dev 04:43:10 -!- casmith789 has quit [Ping timeout: 245 seconds] 04:55:14 -!- Mu_ has joined ##crawl-dev 04:56:47 Red drac breath does not prompt if you'll hit yourself. (https://crawl.develz.org/mantis/view.php?id=2917) by Pseudonut 05:04:11 -!- Keith_burgun has quit [Quit: Leaving] 05:14:08 -!- monky has quit [Quit: hello] 05:31:59 Large-sized Hydra tile (https://crawl.develz.org/mantis/view.php?id=2918) by Denzi 05:35:35 oo 05:35:54 ! 05:35:59 Denzi makes me happy. 05:48:56 -!- casmith789 has joined ##crawl-dev 06:22:24 -!- Pseudonut has quit [Remote host closed the connection] 07:54:53 -!- casmith_789 has joined ##crawl-dev 07:55:40 -!- casmith789 has quit [Ping timeout: 245 seconds] 08:12:53 Miasma is autoexcluded even when playing as undead (https://crawl.develz.org/mantis/view.php?id=2919) by minmay 08:14:23 -!- Radoslav_ has quit [Ping timeout: 240 seconds] 08:22:56 Hive entrance placed on top of another vault (https://crawl.develz.org/mantis/view.php?id=2920) by MarvintheParanoidAndroid 08:23:10 -!- syllogism- has joined ##crawl-dev 08:25:20 -!- syllogism has quit [Ping timeout: 255 seconds] 08:27:08 03j-p-e-g * rfd56415b0068 10/crawl-ref/docs/changelog.txt: Start on the 0.8 changelog. Just highlights for now. 08:33:40 -!- casmith_789 is now known as casmith789 08:45:09 03kilobyte * r7e2ae6061030 10/crawl-ref/docs/changelog.txt: Mention no setgid multiuser installs in the changelog. 08:59:16 re 09:07:15 hmm 09:10:12 Napkin, I would like to download http://crawl.develz.org/saves/dumps/MarvinPA-79ee111-101205-1415.tar.bz2 for debugging purposes, but it says authorization is required 09:12:47 03kilobyte * r74196f79007e 10/crawl-ref/docs/changelog.txt: Mention new saves in the changelog. 09:13:03 Zaba: if your mantis account is labelled as a dev, it probably already lets you get them 09:13:27 you just need to authenticate each time 09:13:46 (not sure about this, it might be that Napkin added that manually) 09:13:50 ah 09:13:56 you are right 09:14:29 !coffee Zaba 09:14:30 * Henzell hands Zaba a cup of café au lait, brewed by Crazy Yiuf. 09:14:37 greensnark, what up 09:15:08 It's been raining continuously for almost 24 hours 09:15:15 Traffic is all snarly 09:15:22 And when I say snarly, I mean it's faster to get down and walk 09:15:25 ouch. Well, here it almost doesn't stop snowing 09:15:55 Heh, I hear snow sucks 09:16:42 it's slightly better than mud 09:16:57 and that's what we get a lot of in the other seasons 09:19:20 03kilobyte * r25499eb83592 10/crawl-ref/source/ (4 files in 4 dirs): Hijack the unused "pyjama gnome" nisse tile for an "extinct species" vault gnome. 09:24:16 -!- dpeg has joined ##crawl-dev 09:25:18 !seen galehar 09:25:19 I last saw galehar at Sat Dec 4 00:09:15 2010 UTC (1d 15h 16m 4s ago) quitting with message Quit: ChatZilla 0.9.86 [Firefox 3.6.12/20101026210630]. 09:25:54 !tell galehar Hi Raphael! I suggest you add the uncontroversial bits to the coding conventions (i.e. all but the last one). That's still good progress. 09:25:54 dpeg: OK, I'll let galehar know. 09:27:55 !seen due 09:27:55 I last saw due at Sun Dec 5 11:35:59 2010 UTC (3h 51m 56s ago) saying Denzi makes me happy. on ##crawl-dev. 09:28:02 due: around? 09:28:06 moin dpeg 09:29:06 Hi Zaba! Great work on the dungeon builder. I am not the target audience, but it is always great to see somehow doing the dirty work. 09:29:46 dpeg, I just hope #2920 is not caused by me 09:30:24 tell me which one is that? 09:30:28 -!- Radoslav has joined ##crawl-dev 09:30:34 "Hive entrance placed on top of another vault" 09:31:08 no no 09:31:19 I thank vaults could be placed overlapping since forever. 09:31:26 There is a tag to prevent that. 09:31:41 hm.. really? 09:31:53 But I think it'd be better to have non-overlapping as default and add a tag that says "you can put another vault on me, if you like". 09:32:01 Zaba: never really sure with me... sorry. 09:32:13 But there have been complaints about that in the past, and I've seen it too. 09:32:43 from what I can see in the code, it shouldn't really happen 09:34:18 aha 09:34:26 Zaba: vaults are generally protected? 09:34:30 this might be new 09:34:59 they're always marked with MMT_VAULT, and that is generally checked for when placing new vaults 09:36:54 entrance vaults are makred, you mean? 09:37:01 all vaults are 09:37:25 03j-p-e-g * r343c64c023ea 10/crawl-ref/source/rltiles/item/scroll/i-amnesia.png: New tile for the scroll of amnesia. 09:37:26 03j-p-e-g * r42c7d3773c85 10/crawl-ref/source/rltiles/ (UNUSED/monsters/sea_snake.png dc-mon/animals/sea_snake.png): Use minmay's sea snake tile. 09:37:27 03j-p-e-g * rd02296d59bf4 10/crawl-ref/source/rltiles/dc-mon/statues/training_dummy.png: Per Denzi's request replace training dummy tile with another one. 09:38:11 -!- Radoslav has quit [Ping timeout: 240 seconds] 09:38:20 well 09:38:25 it's kind of complicated, but generally they are 09:39:04 aha, that's news to me (good news) 09:44:05 -!- TGWi has joined ##crawl-dev 09:50:15 mornafternoon 09:50:26 Hi! 09:58:22 children! 10:01:53 -!- Zaba has quit [Ping timeout: 264 seconds] 10:10:15 -!- Zaba has joined ##crawl-dev 10:15:15 -!- Radoslav has joined ##crawl-dev 10:22:08 03kilobyte * r6033f7cc289b 10/crawl-ref/source/ (rltiles/dc-corpse.txt tilemcache.cc tilepick.cc): Fix weapons and corpses on monster kenku (entry vault). 10:31:57 thought for earth temp-brand spell: it could be no damage, but a knockback effect 10:33:52 (i actually kind of have an idea that maybe we should have three brand spells: bludgeoning = knockback; cutting = spread to enemies adjacent to both you and the target; piercing = enemies on other side of target from you) 10:34:09 would be more interesting than slashing-vorpal-spell, bludgeoning-vorpal-spell 10:34:49 the idea of perm-silver sounds more fitting for the spell's name 10:35:19 as someone proposed, it could both give a silver brand (perhaps permanently) and give a damage bonus (temporarily) 10:36:16 kilobyte: well, i'm not really working from the name here, more the current mechanical effect - i think there's room for a silver-related spell but i'm not sure i like the proposals to make MSH just 'brand spell but silver' 10:36:53 the name should change, even if it ends up as a silver brand spell 10:37:58 huh.. 10:39:14 since the current name is a blatant pop culture reference 10:39:36 what? no, it's a reference to 10:39:40 ??maxwell's patent armor 10:39:40 I don't have a page labeled maxwells_patent_armor in my learndb. 10:39:41 -!- Radoslav has quit [Ping timeout: 264 seconds] 10:39:43 ??maxwell's patent armour 10:39:43 maxwells patent armour[1/1]: +10 plate mail, -tele, -cast. No longer called Edison's. 10:39:47 yeah you're wrong 10:40:14 TGWi: 'pop culture' definitely doesn't fit in the 300 year rule, so it wouldn't make sense for it to have been added. qed 10:40:20 why are there so many bad unrandarts :( 10:41:09 MarvinPA: a lot of them are pretty flavorful/cool, they just were clearly designed by people who weren't actually good at crawl 10:41:26 yeah 10:41:44 i will get round to looking at the code for them at some point 10:42:22 imo get a big list of all of them, then run through them one by one and triage the easily fixable ones from the really awful ones 10:42:41 like wucad mu 10:43:28 Its power varies in proportion to its wielder's intelligence. 10:43:28 Using it can be a bit risky. 10:43:35 hahahaha 10:45:28 also, now that we have res acid on the screen as a three dot resistance, why not just put res poison/elec as one, too 10:46:06 oh huh yeah 10:46:13 does that mean rCorr and preservation do stack now? 10:46:22 MarvinPA: yellow scales 10:46:32 yellow scales isn't actual rcorr though 10:46:38 no, but it's damage reduction 10:46:39 just racid 10:46:57 yeah, now there's 10:47:02 Res.Acid. : . . . Res.Corr. : . 10:47:09 you can have one without having the other 10:47:26 ok so the cloak and amulet stack, but nobody except yellow dracs/DS can get rAcid+++ 10:47:34 jiyvaites 10:47:39 oh yeah 10:47:44 also i don't think the cloak actually provides rAcid+ 10:47:50 i think it is just the normal cloak acid damage reduction 10:47:55 which is for some reason not rAcid+ 10:48:03 Fix the cloak of preservation giving rAcid++. 10:48:03 The cloak and the amulet add up. Wearing both gives rAcid++. 10:48:07 oh 10:48:09 so i think it does 10:48:09 that's dumb 10:48:14 it makes the cloak even better than it already is 10:48:22 and it's already one of the best items in the game 10:48:43 conservations, rcorr, racid+, and cloak reduction to acid damage 10:49:29 i'd say only yellow scales/jiyva should be able to get you more than rAcid+ 10:52:31 -!- sorear has quit [Ping timeout: 265 seconds] 10:55:34 also, going back to unrands: singing sword, zonguldruk, bloodbane, jihad, plutonium sword 10:56:00 5 long blades with bad base types and terrible extra downsides to make up for their bad base type 10:56:06 yay :P 10:56:33 Jihad is not that bad 10:56:39 the others, I agree 10:56:46 MarvinPA: i want zonguldrok to be more badass, but haunt you with spectral versions of uniques you've killed 10:56:55 it's still just a long blade of holy wrath and *rage 10:57:01 especially since lots of work went into singing sword's drivel 10:57:05 what is jihad? 10:57:07 the least bad of the bunch, probably, but still not great 10:57:26 _find_level_feature, count_feature_in_box, _find_in_area all overlap in functionality 10:57:27 +12+10 long sword of holy wrath, *rage, MR, -stlth 10:58:00 thanks. Yeah, doesn't sound that great. I've never found it, anyway 10:58:13 singing sword could be a shortblade, and be very likely to distract intelligent enemies 10:58:41 MarvinPA: and rN+ EV+3 10:58:44 ah yeah 10:58:46 zonguldruk is scummy rather than bad 10:59:13 most of the polearm/mace unrands are quite cool but should just be better base types (instead of spears/morningstars) 10:59:43 I wonder about adding an "eveningstar of disruption" 11:00:18 with an even bigger bonus than the old brand 11:00:22 what about changing jihad's enchantment to holy rN+++ 11:00:23 necromantic rather than holy 11:01:11 kilobyte: eveningstar of dispelling; 300% damage to undead, and 1% chance per turn of a word in output being typoed 11:01:16 for Jihad, rN+++ is redundant with TSO... rN+ is not since spamming daevas may get you below near-max piety 11:01:35 Eronarn: <3 11:01:42 why would you use it at all with TSO when you can bless a good base type? 11:01:58 why does it need to not be redundant with tso? 11:02:05 and get a trishula/scourge/blessed katana or something 11:02:35 yeah, i say remove the +EV and make it rn+++ 11:02:41 TSO wouldn't care if it were redundant 11:03:11 FR: include demon cock as a new staff, and let TSO enchant it 11:03:15 that's like four birds with one stone 11:03:25 hm. 11:04:09 TGWi: we could use more staff types in general 11:04:14 i think 'sceptre' would be a good one 11:04:18 Eronarn: that's one bird 11:04:35 TGWi: well, i mean add a new base type, then a demon version of it, then a TSO version of it 11:04:52 also fixes tso inconsistency, and relates the demon cock splash screen to the game 11:05:07 it's perfect 11:05:16 hahaha 11:07:22 we could add in sceptres, longstaves; demon sceptres; maybe one more 11:08:32 ooh, sodegarami 11:08:51 http://en.wikipedia.org/wiki/Sodegarami 11:10:28 other obscure weapons we don't have yet: hook sword, jian, katar, kukri, main-gauche, kusari, tetsubo, urumi 11:11:33 03kilobyte * r75874ee8246b 10/crawl-ref/source/ (5 files in 3 dirs): Add psyshvl's demigod tile, fix weapons and corpses for DG, DS, Ha. 11:16:19 -!- syllogism- has quit [*.net *.split] 11:16:20 -!- CIA-6 has quit [*.net *.split] 11:16:34 -!- CIA-1 has joined ##crawl-dev 11:20:39 sceptres should be M&F 11:20:55 bloody, bloody hell 11:20:58 but looking around, I see triple staves as something with a potential 11:21:41 -!- syllogism has joined ##crawl-dev 11:21:45 while somewhat resembling a flail (of the non-morningstar type), you fight with it like a staff not a mace 11:22:12 I have a desire to kill _builder_basic with fire, but unfortunately it's one of the most commonly used level generators 11:22:22 as in, you can't hold it by one end and swing around, even if there are swinging moves 11:22:36 kilobyte: some sceptres are short rods, others are long staves 11:22:46 Zaba: because of the maps it generates or because of unholy code? 11:22:52 kilobyte, both! 11:22:57 but yeah, a segmented staff could work 11:23:02 kilobyte, but primarily the latter 11:23:42 kilobyte, I don't particulary like what it generates, either, but then, I don't really play crawl enough to be bothered by that... 11:24:56 speaking of triple stuff, it would be really, really nice to get rid of that Star Wars abomination 11:25:21 you mean double/triple swords? 11:25:27 yeah 11:25:34 the code is just.. well.. godawful, really. I don't know how to reasonably untangle it. 11:25:46 kilobyte: double sword isn't a sword with a blade on each end 11:26:01 _builder_basic itself is ok, but it calls other stuff... which uses spec_room in an obscure manner, and such 11:26:30 there's a possible avenue: the word "double sword" had two meanings, one being an European-type sword (double edged) and the other is a bastard sword 11:26:34 and all the one-shot utility functions that have overlapping functionality with other such functions 11:26:37 double sword: http://www.fantasyblades.net/images/products/an%20430%20double%20bld.jpg 11:26:44 it's like a razor 11:27:09 heh... someone actually made that thing 11:27:13 and the fact nothing can agree whether it should work with grd indices starting with 0, or 1, or X_BOUND_1, and whether it should end with GXM/GYM or GXM - 1/GYM - 1 or what 11:27:36 it doesn't make any sense to use it... less control, more weight, less damage 11:28:26 the triple sword always reminds me of this film http://www.imdb.com/title/tt0084749/ 11:28:32 the word "triple sword" is sometimes used for a type of claymore, with a crossguard that resembles short swords at an angle 11:28:35 kind of like long blades 11:28:52 kilobyte: it's a *magical* sword 11:29:01 like a quick blade is innately magical even with no brand 11:29:12 don't be ridiculous 11:29:14 crawl doesn't have magic 11:29:38 -!- sorear has joined ##crawl-dev 11:30:14 Zaba: during your refactoring, are you using any tools to find orphaned code? 11:30:44 sorear: running IRC as root? I wouldn't suspect *you* of doing that. 11:31:05 Fangorn_, no 11:31:16 Fangorn_, I just find awful bits of code and try to make them less awful, recursively. 11:31:50 kilobyte, why do you trust the information sorear's client sends about the username? 11:32:22 since dungeon.cc is a huge file that uses little of external stuff, it may be enough to make most stuff static 11:32:37 kilobyte, most stuff is already static 11:33:06 unfortunately, code rarely becomes orphaned in it 11:33:59 because many code paths reinvent all the little utility functions with interfaces different just enough to prevent all of them from being easily replaced with one 11:36:22 03greensnark * r20c306a60f10 10/crawl-ref/ (13 files in 4 dirs): Add support for Lua hooks triggered during vault generation (dpeg) 11:36:25 03greensnark * r99398b9f31ca 10/crawl-ref/source/main.cc: Fix warning noises when sizeof(int) != sizeof(long) 11:37:22 non-static functions used by no other file: dgn_excavate, dgn_place_feature_at_random_floor_square, dgn_postprocess_level, dgn_random_direction, dgn_random_point_in_bounds, dgn_square_is_passable, dgn_unregister_vault, dgn_vault_excavatable_feat, map_feature_at, remember_vault_placement, smear_feature, spotty_level 11:37:44 kilobyte, umm. 11:37:59 should I turn them static? 11:38:05 I guess. 11:38:22 hmm... I'd swear spotty_level is used from lua, but it isn't 11:38:37 yeah, could have sworn some of those are interfaced from lua, too 11:38:43 but apparently those are independent implementations 11:39:03 yay 11:39:14 bloody, bloody hell 11:39:15 what a mess 11:39:36 What's a mess? 11:40:08 greensnark, dungeon.cc and related stuff 11:40:15 Ah, is that all 11:40:25 I really try not to think of other parts of crawl these days. 11:40:51 Monster movement AI code will make dungeon.cc look good 11:41:03 Just take a squint at it when you feel dungeon.cc's getting you down 11:41:15 wait... is vehu extending sandblast or does it hit 2 range without stones normally? 11:41:29 greensnark, in a nutshell, what does that lua hook stuff let one do? 11:41:33 Also, someone let me know if that vault change commit I made breaks anything. Just leave me a !tell. 11:41:56 Zaba: The commit message gives an example -- dpeg wanted one vault to be able to request that a bunch of others be placed on the same level 11:42:06 greensnark, does it help in any way to bring closer the moment when all level generation can be done in lua? 11:42:39 Moving all the levelgen logic to Lua can only happen after you finish that dungeon.cc cleanup :P 11:42:47 greensnark: about that commit with int lengths... is there a compiler which does give a warning there? 11:42:52 greensnark, well.. not really. 11:42:56 kilobyte: My Mac laptop 11:43:04 greensnark, all I do is bottom-up cleanup, it won't bring any major changes 11:43:12 i686-apple-darwin10-gcc-4.2.1 (GCC) 4.2.1 (Apple Inc. build 5664) 11:43:13 I checked clang too 11:43:45 hmm, perhaps old gcc might fail to upgrade types when needed 11:44:03 Yeah, it said it was treating that check as always-false 11:44:05 of course, your commit is correct, I just wondered where could you get the warning 11:44:18 Er, always-true 11:44:28 _correctly_ 11:45:14 Ok, I'm off, drop me a !tell if vaultgen explodes after 20c306a60f10 11:46:31 hmm, not actually... while it is impossible for it to reach that value, it's only you.elapsed_time that is capped, not you.num_turns 11:48:17 03zaba * r6848e169a5c5 10/crawl-ref/source/dungeon.cc: Simplify the collision check in _build_rooms by using count_antifeature_in_box. 11:48:25 03zaba * r232f44b3850f 10/crawl-ref/source/dungeon.cc: Simplify _branch_entrances_are_connected by using rectangle_iterator. 11:48:36 03zaba * rbed5ca280dfd 10/crawl-ref/source/dungeon.cc: Use the coord_def variant of dgn_replace_area in _build_rooms. 11:50:01 Eronarn: veh extends sandblast 11:50:24 why? it's not a conjuration 11:50:37 i don't care if he does, but if he extends non-conj, the religion text is wrong 11:50:40 he boosts directly damaging spells 11:50:55 ??vehumet[3] 11:50:55 vehumet[3/4]: Vehumet's wizardry effect kicks in at piety 50. Reduces spell failure chance by 1/3 for the following spells: any Conjuration, any Summoning, Airstrike, Bone Shards, Ignite Poison, Ozocubu's Refrigeration, Sandblast, LRD, and Shatter 11:50:59 those ones (plus tornado) 11:51:13 MarvinPA: right but the range boost specifies 'extending the range of your conjurations' 11:51:19 rather than 'destructive spells' 11:51:24 ah right 11:54:05 Vehumet religious text is misleading (https://crawl.develz.org/mantis/view.php?id=2921) by Eronarn 11:55:03 03j-p-e-g * ra77d2aa694c0 10/crawl-ref/source/tilereg-inv.cc: Add missing spaces to secondary actions' wield tooltips. 11:55:06 03j-p-e-g * r8f1bec83804b 10/crawl-ref/source/ (7 files): Ctrl-mouseclick asks for quantity for drop/pickup. 11:55:08 03j-p-e-g * r8eb0c05ed5ff 10/crawl-ref/source/initfile.cc: Use #ifdef on a function only used for UNIX. 11:55:17 okay, spell formula time 11:55:19 03j-p-e-g * r4e29b8c0ab21 10/crawl-ref/source/ (8 files): Rename debug_prompt_for_item() -> prompt_for_item(). 12:05:24 greensnark: around? 12:11:43 ah, I see he left... busy backlog 12:12:01 btw, I am against three levels for rElec and rPois 12:12:15 this comes up from time to time, but is probably just power-spiralling 12:12:22 Erik was pretty adamant on this one 12:18:03 yeah, for a good reason 12:18:15 dpeg: i don't mean being able to wear more than one level; it already exists - player undead can't be poisoned with poison arrow for instance 12:18:37 -!- elliptic has joined ##crawl-dev 12:19:00 though i don't know if rElec for players can go above + 12:19:24 Eronarn: I see, but that's really different from rF and rC. 12:20:52 ??sandblast 12:20:52 sandblast[1/1]: Wield stones to do more damage and increase range from 1 to 2 with this spell. Sandblast with stones maxes out at 2d20 damage, which is really good for a level 1 spell, although damage reduction from enemy AC is applied twice instead of once. You can use large rocks, but damage won't increase beyond regular stones. 12:21:00 is the part about double AC correct? 12:21:07 think so 12:21:11 wow 12:21:25 the point is that sandblast is really bad 12:21:41 TGWi: it's really good when you consider that the alternative damage source for an EE is stone arrow 12:21:45 it's a cool idea but not when it's a new caster's starting spell 12:21:45 or LRD 12:21:55 I'll mention it on the wiki 12:23:04 these formulas: so bad 12:24:28 sandblast can be clearly made to work 12:25:35 -!- Noom has joined ##crawl-dev 12:26:43 sandblast still only has 1 diagonal range :( 12:26:53 even with stones, that is 12:27:32 it is clearly not perfect, but a spell for which stones are valuable ammunition is good starting stuff 12:28:57 yeah (although it's fun when you get one of the couple of starting vaults with about 100 stones in it as an EE) 12:36:32 it could use a damage boost when using it without a stone 12:37:09 maybe making it much harder to miss since it's a big cloud of sand 12:49:05 ping dpeg 12:53:19 -!- sorear has quit [Ping timeout: 260 seconds] 12:55:38 bhaak: pong 13:01:06 @??two-headed ogre 13:01:07 two-headed ogre (04O) | Speed: 10 | HD: 6 | Health: 21-45 | AC/EV: 1/4 | Damage: 17, 13 | Res: 06magic(32) | Chunks: 07contaminated | XP: 280. 13:01:10 @??ettin 13:01:10 ettin (07C) | Speed: 10 | HD: 7 | Health: 25-51 | AC/EV: 3/4 | Damage: 18, 12 | Res: 06magic(28) | Chunks: 07contaminated | XP: 284. 13:01:20 what's the point in having both? 13:01:29 there isn't one; remove two-headed ogres 13:01:51 that way we won't get people asking to play as them 13:02:29 it's on the wiki 13:02:39 I was thinking about removing ettins instead, since they hardly ever spawn and 2h-ogres are common, especially in packs... but you may have a point. 13:02:41 though if you drop one it's obviously ettins, ogres are in mage packs 13:03:44 i'd rather see dropping two-headed ogres and then adding a new kind of ogre 13:03:48