01:11:21 -!- afd_ has joined ##crawl-dev 01:34:02 -!- Jordan7hm has quit [Ping timeout: 246 seconds] 01:40:44 -!- valrus has quit [Remote host closed the connection] 01:43:00 Another contributor may be the <1 centimeters (0.4 in) zombie worm (Osedax Latin for bone-devourer) Osedax inhabits whale falls in southern California, Monterey Bay, Sweden, and Japan. 01:43:12 there is a bone-devourer worm irl and it's not in crawl for some reason?? 01:47:51 see also: 01:47:54 Small decorator crabs, like this one, are often found wearing snipped off tips of stinging hydroids on their heads to fend off predators 02:08:33 -!- ekiM has joined ##crawl-dev 02:09:51 -!- eith has joined ##crawl-dev 02:18:46 -!- Kurper has quit [Remote host closed the connection] 03:23:35 -!- lorimer has quit [Ping timeout: 252 seconds] 03:25:32 -!- lorimer has joined ##crawl-dev 03:35:31 -!- Napkin has joined ##crawl-dev 03:35:39 morning! 03:35:39 Napkin: You have 1 message. Use !messages to read it. 03:35:59 -!- afd_ has quit [Remote host closed the connection] 03:45:36 -!- elliptic has quit [Ping timeout: 250 seconds] 04:15:49 -!- monqy has quit [Quit: hello] 04:57:38 -!- upsy has joined ##crawl-dev 05:07:51 -!- Textmode has quit [Ping timeout: 260 seconds] 05:21:47 -!- evilmike has quit [] 05:25:59 -!- dpeg has joined ##crawl-dev 05:44:14 Zaba: would you be ok if I added a fake feature for redefinition just the way we have >10 fake monsters? 05:44:57 hmm 05:45:11 kilobyte, what'd be involved? 05:45:27 for the user, "feature =" is so much cleaner than "cset_martian = item_detected {x1234}" "detected_item_colour = ultraviolet" 05:45:48 what I want right now is a way to highlight map edge 05:46:00 if it doesn't affect dungeon -generation- (and I don't see why it should), I'm cool with it 05:46:36 the default glyph, space, doesn't take colours well so I can't use mere emphasis like visited stairs 05:47:23 right, DNGN_ enums were purged from any fakes, so you could be annoyed by re-introducing them for some other use 05:47:56 there are still some reserved elements in it 05:48:21 okay, just one, actually 05:48:29 but no, it doesn't annoy me 05:52:37 I like Zaba's approach a lot :) 05:53:45 and the principal difference between those purged elements and ones that you might add is that the latter will actually be used for something :P 05:54:16 another option would be to sanitize our feature redefinition code 05:54:35 well, that'd probably be better in a long run, although I haven't looked at that code 05:55:53 Is the ice beast bug fixed? 05:57:03 dpeg: yes 05:57:09 cool, you rock 05:59:02 -!- Mu_ has joined ##crawl-dev 06:01:52 !seen galehar 06:01:53 I last saw galehar at Tue Mar 15 16:57:52 2011 UTC (18h 4m 1s ago) quitting with message Quit: Page closed. 06:04:29 03dpeg * r0f198270be03 10/crawl-ref/source/dat/descript/backgrounds.txt: Update Monk description. 06:05:37 hmmm. 06:14:53 what would be a nice name for "explore edge"? I mean, an unexplored square that is adjacent to an explored walkable/flyable one. 06:20:07 kilobyte: for code or in-game? Verb or noun? 06:24:14 just the config file, useable in "feature = explore edge {ยค, lightmagenta}" 06:24:56 "exploration edge"? "unexplored edge"? 06:27:11 exploration boundary? 06:27:13 -!- edlothiol has joined ##crawl-dev 06:27:14 not so easy :) 06:28:00 -!- casmith789 has quit [Remote host closed the connection] 06:28:23 -!- casmith789 has joined ##crawl-dev 06:39:08 horizon? 06:40:36 oh! 06:45:01 do you mean just "horizon"? Or *explor* horizon? 06:46:21 "explore horizon" might be best, yes 06:51:44 kilobyte: good enough? 06:52:00 it works... a very painful itch scratched. I wonder if most people would know about it if we just mentioned it in the docs. 06:52:47 on the other hand, it might be distracting for new players 07:02:22 03kilobyte * r4b33aec70b06 10/crawl-ref/source/wiz-you.cc: Update burden after wizmode species changes. 07:02:23 03kilobyte * r9adec8ab3baf 10/crawl-ref/source/ (directn.cc enum.h feature.cc terrain.cc viewmap.cc): "feature = explore horizon {xEA,,lightmagenta}" to make exploring Lair easier. 07:02:57 now a quick rebuild and I can return to the game :p 07:03:12 why exactly does it make exploring lair easier? 07:07:16 because it's a pain to search for every plant in a nook by hand 07:07:22 ah 07:07:28 Unstable branch on CDO updated to: 0.8.0-a0-5767-g9adec8a (32) 07:07:49 and there are several vaults with plenty of good stuff behind plants 07:27:10 -!- greensnark has joined ##crawl-dev 07:27:41 Boo 07:27:41 greensnark: You have 4 messages. Use !messages to read them. 07:27:59 waaaaa! 07:28:15 It's the un-scary-name guy :) 07:28:24 You have to admit bhaak sounds scary 07:28:34 !messages 07:28:35 (1/4) jpeg said (3w 18h 43m 18s ago): I know you've unsubscribed from c-r-d but could you send an email on your thoughts about release cycles to the mailing list? Or at least, the devteam members? FWIW, I fully agree. 07:28:49 Email is too much like work! 07:28:50 !messages 07:28:51 (1/3) elliptic said (2w 1d 17h 43m 19s ago): when=t10 is broken... only gets the second half of the tourney (try !lg * when=t10 1) 07:28:55 !messages 07:28:55 (1/2) cbus said (2w 1d 5h 51m 52s ago): snark snark! first day at work today :) 07:28:58 Ooh 07:29:09 !tell cbus Congrats, what's your job? 07:29:09 greensnark: OK, I'll let cbus know. 07:29:12 -!- felirx_ has joined ##crawl-dev 07:29:16 !messages 07:29:16 (1/1) casmith789 said (2d 22h 10m 28s ago): you are a legend 07:29:20 I know 07:29:39 -!- felirx has quit [Ping timeout: 240 seconds] 07:29:39 -!- joosa has quit [Ping timeout: 240 seconds] 07:29:39 -!- RjY has quit [Ping timeout: 240 seconds] 07:29:40 -!- joosa has joined ##crawl-dev 07:30:40 that's fine. bhaak should invoke the image of an evil sumerian god, someone who could kick cthulhu's ass. 07:32:41 greensnark: what a lot of noise 07:32:48 dpeg! 07:32:51 :) 07:33:06 So when's the release? :P 07:33:14 Or are we waiting for NH 4.0 before we do 0.8? :P 07:33:17 We're crawling towards release :) 07:33:28 We're crawling a tad faster than them. 07:33:37 make with NO_INLINE_DEPGEN dies because kilobyte has been naughty 07:34:00 elliptic and a few others of us were wondering of setting up a mini tournament when .8 is released 07:34:20 We had a funny bug where D:1 could be swarmed with ice beasts. Special "berserker fury and fiery elementalist" edition. 07:34:24 casmith789: You mean next year? :P 07:34:25 -!- RjY has joined ##crawl-dev 07:34:30 casmith789: yay! 07:34:37 please do 07:34:40 dpeg: Yeah, I noticed some excitement about the ice beasts on SA :) 07:34:51 greensnark: we are one step further away from using autotools, Zaba spoke up against it :) 07:35:09 bhaak: That just means Zaba secretly uses autotools 07:35:12 * dpeg patiently recommends using autofools instead. 07:35:27 Zaba also claims he does not use GNU screen and that he is not a billy goat 07:35:30 greensnark: I'm sure it'll be before august 1, anyway 07:35:34 Which just shows you how truthful he is 07:35:46 casmith789: I notice you didn't mention the year 07:36:31 Then again, Zaba has been looking deep into the dungeon builder. Everyone knows the dungeon builder is starting back, so we should be happy that Zaba still speaks comprehensibly. 07:36:58 staring back, I got my Nietzsche wrong 07:37:05 Billy goats are immune to abyssal effects 07:37:14 Something to do with their beards, I think 07:37:31 moin greensnark 07:37:32 Haha, b0rsuk 07:37:35 Heyy Zaba 07:37:53 I thought it was the effect of generations of them being used for sacrifices. You become more detached after a while, I'd think. 07:38:02 I see b0rsuk claims stone soup has been stealing his ideas without crediting him <3 07:38:26 greensnark: b0rsuk on tavern? Yes. I tried to soothe as good as I could (which isn't very good). 07:38:38 Just add him to credits 07:38:44 After all, he did do some work on that manual 07:38:57 he isn't in the credits? 07:39:05 I don't know about all the ideas we stole, but I can't recall many :) 07:39:15 dpeg: Is he? He thinks he isn't :) 07:39:17 probably because he refused to give his real name, claiming even the policeman calls him Mr Badger 07:39:38 Oh, he wants to be credited as Meles meles? 07:39:43 <3 07:39:54 But what about the okapi? 07:39:54 -!- valrus has joined ##crawl-dev 07:40:07 pointless_ loves b0rsuk quotes, he added half the learndb entries 07:40:11 ??b0rsuk 07:40:12 b0rsuk[1/7]: b0rsuk is a badger of such great power that he has mastered his own mortality. He is enormous, three-headed and capable of breathing fire. Born before the history of time, he has grown in might and magical mastery. Being immortal, there is nothing left to acquire but godhood. His claws are golden and enchanted. 07:40:20 Although the first entry is b0rsuk himself 07:40:35 I keep trying to spell it as b0rsulk 07:41:14 ??b0rsuk[7] 07:41:14 b0rsuk[7/7]: Finally, let it be known that my current nickname is 'b0rsuk', all lowercase in accordance with unix tradition. And "Borsuk" is not my name - "Borsuk Euroazjatycki" means "Eurasian Badger" (meles meles). 07:41:20 ^^ pointless quoting b0rsuk 07:41:45 He has some pretty good ideas. A slight problem is that he seems to think (a) no one understands and values his ideas, and (b) if they do, the ideas are stolen and miscredited. 07:42:00 Yeah, he has a bit of a persecution complex 07:42:14 But I'd say no harm in crediting him 07:42:22 After all, he may be nuts, but he has contributed 07:42:29 greensnark: I would blame his homecountry, hadn't we evidence to the contrary amidst the devteam. 07:42:40 greensnark: okay, will aproach him for real name. 07:42:45 Also Curio 07:42:52 tiles? 07:42:55 I'm surprised Curio isn't in credits already 07:43:00 Yes, he contributed a lot of tiles 07:43:14 I should ask Johanna to do that... I am missing all the tiles stuff. 07:44:22 -!- valrus has quit [Ping timeout: 250 seconds] 07:44:49 Yes, whoever's committing contributions should update credits accordingly 07:44:57 yes, that'd be best 07:45:08 greensnark: jpeg had a job interview where she talked about the orc god 07:45:15 Heh 07:45:21 not sure that'll help her getting the job, but it's sure unique 07:45:39 One of my job interviews someone googled me and found Crawl and then the guy turned out to be an Angband player 07:45:42 Hmm, I wonder if I talked about sewers and ice caves in mine. 07:45:43 "All my work in this company will be dedicated to Beogh!" 07:45:57 greensnark: was that good or bad? :) 07:45:58 greensnark: Was that good or bad? 07:46:07 * dpeg wins! 07:46:10 Well, I was surprised to find a RL player 07:46:18 But unimpressed with the Angband :P 07:46:35 Although he assured me he no longer played it 07:46:38 yes, but the tension between bandlike and hacklike is legend... he might decline alone for that :P 07:47:07 Everyone knows that only imbeciles and hapless morons play bandlikes. 07:48:24 03greensnark * re9e05cd52823 10/crawl-ref/source/makefile: Fix make failing with NO_INLINE_DEPGEN (needed for Mac universal binaries). 07:49:44 So when is the 0.8 release, for real? 07:51:18 there's no branch yet, but bugs are fixed (no triage, no release manager though) 07:51:38 No branch yet? So at least +3 weeks? 07:51:55 Too bad about that nice April 1 date 07:52:22 Why is there no release manager, anyway? Nobody has the time? 07:52:45 nobody spoke up :( 07:52:52 Let's get bhaak to do it :) 07:53:08 bhaak should do it, then we can finally get elbereth into crawl 07:53:13 Eek 07:53:23 I rescind my invitation 07:53:41 Heh, bhaak's actually playing Crawl 07:54:38 I think it's normal to gather intel from the enemy camp 07:55:09 i could do it, but are you really going to risk that crawl 0.8 will be really unnethack 3.5.4? 07:55:35 Are you calling that a risk? 07:55:44 a lot of patches on the last day :) 07:55:45 You should make it sound enormously desirable 07:57:24 I just need to put autoexplore on o and nobody would notice anything 07:57:47 bind o to autoexplore, all other keys to autofight 07:58:01 and nobody will care about the differences :P 07:58:14 So are we going to rename megabat before or after 0.8? :P 07:58:55 there's no need to rename megabats, we could just make the current 'dummy' bat monsters spawn instead of them :P 07:59:14 Megabats have got to the point where Crawl is an elaborate parody of itself 07:59:17 add an option to rename monsters 08:00:12 The closest thing to megabat is the new Ubuntu codename 08:00:28 "Oneiric Ocelot" <3 08:00:40 Way to go to make the release sound megadumb, Ubuntu 08:00:57 only because nobody told them about the okapi 08:01:43 dpeg: I can see you're going to name your next child "Okapi" 08:02:03 I think Keskitalo named his daughter after giant goldfish 08:02:33 And I see from the forum that someone named his son after felirx_ 08:05:45 -!- Twilight-1 has quit [Read error: Connection reset by peer] 08:07:08 Is "Faith worth it on a Zinner? 08:07:55 How's the new recite 08:08:16 Lair animals should congregate around pillars of salt to lick up the salt 08:08:16 A hill giant comes into view. It is wielding a giant spiked club. 08:08:16 _The hill giant shouts! 08:08:16 _There is a stone staircase leading up here. 08:08:16 The hill giant throws a throwing net. The throwing net hits you! 08:08:16 _You become entangled in the net! You stop ascending the stairs. 08:08:21 the new Recite is awesome 08:08:34 Have to give Eronarn a lot of credit for this 08:08:44 -!- ivan has quit [Quit: ERC Version 5.3 (IRC client for Emacs)] 08:08:46 So is Zin worth playing now? :) 08:08:55 yes! 08:09:00 Nice 08:09:05 -!- djinni has joined ##crawl-dev 08:09:05 -!- ivan has joined ##crawl-dev 08:09:06 -!- ivan has quit [Changing host] 08:09:06 -!- ivan has joined ##crawl-dev 08:09:20 I don't see Eronarn credited 08:09:21 -!- Twilight13 has joined ##crawl-dev 08:09:34 You know folks should really make a point of crediting contributors :P 08:09:56 Or maybe he's credited and I just don't recognise the name 08:12:09 -!- Cryp71c has joined ##crawl-dev 08:12:12 greensnark: I made a note. 08:12:39 Rar 08:13:22 !coffee Cryp71c 08:13:23 * Henzell hands Cryp71c a cup of irish coffee, brewed by Gastronok. 08:13:27 mmmm 08:13:43 Thanks snarky 08:16:41 There will be a time when kilobyte's console crawl will look quite alien to everyone else :) 08:16:56 Doesn't it already 08:18:21 hur? 08:19:01 Cryp71c: he is using utf glyphs most people haven't ever heard about 08:19:05 ooooo 08:19:06 I think kilobyte plays on a 1080p terminal using a whole bunch of Coptic characters 08:19:10 I must look 08:19:16 Cryp71c has not proved yet he can speak... ah :) 08:19:33 dpeg, proved he can speak? 08:19:45 rar, mmmm, hur, oooo :P 08:19:52 ah...lol 08:19:57 I said thanks earlier! 08:20:05 true! 08:20:21 When are we portal-vaultising Hive 08:20:28 unclear 08:20:47 Or add superior bees 08:20:48 there are folks who want Hive to be removed altogether, or turned into minihives 08:20:49 :( my terminal won't display his UTF stuff. 08:21:16 but I think the idea of a race against the bees is interesting 08:21:26 Reminds me: we should add a lua file that runs at game startup and decides what vaults it's going to place and where it wants them 08:21:29 races are generally good, even or especially if you can prepare 08:21:43 greensnark: what's the benefits? 08:21:46 That way we can do things like one vault requesting that the next two levels have vaults x & y 08:21:52 oh wow 08:21:56 We can do things like deciding what temple vault is used ahead of time 08:21:58 serial vaulting in two dimensions 08:22:11 Or let's say we want a vault ot be placed somewhere from D:2-7 fairly 08:22:23 No way to do that currently without ridiculous gymnastics or C++ code 08:22:39 greensnark: you can do that, right? In other words, your "we" means "I" :) 08:22:52 greensnark, would overflow temples be reimplemented using that? 08:22:53 dpeg, what setting controls the glyph set used, and how can I get my terminal to display Kilo's correctly, so I can see what it looks like? 08:22:53 dpeg: It's easy to do, don't know if I will :) 08:23:09 Zaba: Yes, I want to move temple handling as the first thing 08:23:15 that's be nice 08:23:23 Cryp71c: I have no idea, I am finding it depends on the whims of my linux and console 08:23:50 Cryp71c: You just have to tell your client to expect UTF-8 08:23:54 What is wrong with the current solution for temple and overflow altars? 08:24:18 Don't know what client you're using, if it's Linux you can just set your locale 08:24:19 dpeg, nothing exactly wrong, but it's hardcoded, and what greensnark proposes would be a more generic way to do it 08:24:37 dpeg: Hardcoded in C++, if any new vaults want to do something similar they need new C++ code 08:24:37 -!- felirx_ is now known as felirx 08:24:48 greensnark: ah, good 08:24:52 felirx: hi! 08:24:59 just lurking 08:25:05 felirx: Hey, someone named their son after you :) 08:25:10 lurkers don'T get coffee, of course 08:25:32 kilobyte's wall glyph is crazy :P 08:26:19 greensnark, hrm, my locale is UTF, and gnome-terminal's current encoding is set to UTF... 08:26:23 greensnark: I suspect it's not the real deal... aren't there more esoteric UTF pages we might see wrongly? 08:26:31 (UTF-8) 08:26:41 greensnark, why, in the current code, whenever C++ and lua need to exchange information, global variables are employed? Is there no better way to do that? 08:26:43 Cryp71c: Maybe your font is just missing those characters? 08:27:00 Zaba: Give some examples? 08:27:16 greensnark, map_anchor_points 08:28:23 It would be better to get rid of that, yeah 08:28:28 Nit O 08:28:31 Oops 08:28:38 and there are more: Old lua special rooms used a global variable, too, but I killed that altogether a while ago. 08:28:45 I'd also like to move more of the vault placement code to Lua 08:31:04 greensnark, what parts of it, exactly? 08:31:08 Hah, someone suggested "Ozzy Osbourne" as an alternative Ubuntu name 08:31:20 Zaba: Selecting where the vault is placed 08:31:29 Oh, I think we discussed this before, right? 08:31:44 greensnark, yeah, looks like the system "Monospace" font is missing some UTF chars -_- 08:31:48 yes, we did; I've also been pondering it recently, together with other differences between mini- and normal vaults 08:32:18 it'd be nice to have something like a POSITION: directive, which could take 'float', 'mini' or 'lua:function_that_returns_the_position', or so 08:32:24 Zaba: Yep! 08:32:59 but, I'm not sure how that'd interact with ORIENT 08:33:03 I'd also like serial vaults to be able to request that individual vaults be placed within a certain distance of some point, etc. 08:33:06 it would probably replace and expand upon ORIENT, actually 08:33:18 Well, ORIENT is totally a hangover of the 4.0 code 08:33:32 4.0 had NSEW and encompass orients already 08:33:53 POSITION: northwest, etc. instead of ORIENT could work, too 08:34:02 Aye 08:35:04 greensnark, one thing that should certainly change is how minivaults are told apart from normal vaults by the _absence_ of ORIENT 08:35:25 Could just add ORIENT: mini if that's confusing :) 08:35:31 or POSITION: 08:35:52 but minivaults are ORIENT: float plus a few obscure semantics 08:35:54 The dungeon builder should also catch more errors like branch entry vaults not placing the branch entrance 08:36:17 greensnark, yeah, that's actually quite easy to do in theory, but somehow I didn't get to it 08:36:31 I think there's also a big difference in how minivaults and regular vaults are retried 08:36:42 One of the vault types gets about 20x the number of retries 08:36:49 I think I got stuck trying to make _build_vault_impl properly return a vault_placement pointer without screwing up memory management :P 08:36:50 That should also be normalised 08:37:29 Memory management and iterator invalidation are the two great Fun things about C++ 08:37:36 Especially iterator invalidation 08:37:50 greensnark, _find_minivault_place tries to find a minivault place 600 times 08:37:51 Good old invalidation rules designed to bite you in the posterior 08:40:13 damn, 600 times is a lot, isn't it 08:40:32 Especially compared to floating vaults which get something like 50 tries :P 08:40:38 they get one try 08:41:01 I think the builder is also pretty dumb about trying to place things like entry vaults on encompass vault levels 08:41:14 i.e. instead of noticing it has no hope it tries the full X times before giving up 08:42:45 greensnark, write_vault retries 25 times, so a total number for minivault placement would then be.. 15000? 08:42:49 that's wicked 08:43:35 and just 25 for floating vaults (unless some other retries happen along the way and I'm not seeing them) 08:43:56 no wonder i hardly ever see my floating vaults :P 08:44:06 Mu_, that's not the reason 08:44:13 you hardly ever see floating vaults because they are rare 08:44:20 oh ok :P 08:45:00 Yeah, non-minivault dummies are pretty harsh in general 08:45:11 We should totally reduce the number of minivault retries 08:45:29 Mu_, 25 tries is plenty for any sane vault to be placed as secondary 08:45:44 Mu_, 1 try is always sufficient to successfully place a sane vault as primary :P 08:46:17 and floating vaults are placed as primary in D. Why the hell's the difference between D and other places, anyway? 08:46:22 greensnark: I cannot change the dummy weights. 08:46:44 Zaba: Should change it to always place them as secondary 08:46:51 dpeg: Why not :) 08:47:15 greensnark, hm. 08:47:27 greensnark: because I have no idea what I am doing :) 08:47:29 all those retries must be good for testing i guess? i've been making some serial vaults and i'm guaranteeing one on D:17 for testing with all of its subvaults and i get a noticable lag when i enter the level i guess since it's trying really hard to place them all? :P 08:47:35 greensnark, a while back, I made a change that made _all_ vaults be placed as secondary, and that broke D:1 and branch endings 08:47:38 among other things 08:47:46 Mu_: Yep 08:47:53 Mu_, yes 08:48:01 Especially if you're on a debug build, you can totally notice when minivaults aren't being placed :P 08:48:07 Because Crawl goes into a dreamy trance 08:48:21 could make the number of retries higher in development builds, with a warning when it overflows the number that's used in normal builds 08:48:23 Also minivault collision detection is horrifically inefficient 08:48:49 We should rewrite from scratch, when I listen to you guys :) 08:49:02 dpeg: Nah, refactoring is almost always better :) 08:49:06 Just look at Crawl itself! 08:49:11 that reminds me, i don't suppose there's a crafty way i can make D:17 always use the open or cross layout is there 08:49:19 Jesse Jones tried to rewrite it and call it Crawl 5.0 :) 08:49:38 Mu_: Not yet 08:50:14 But Zaba and I have plans to run a Lua function that decides what the dungeon will look like at game start 08:50:27 I'd like to use normal vault picking functions to choose layouts 08:50:33 but that'd require all layouts to be in some form of lua :P 08:50:47 I thought you'd already done that? :P 08:51:00 and reimplementing the existing C++ layouts in lua is ranging from painful to insane 08:51:17 well, aside from _bigger_room, maybe 08:51:17 Just add a Lua wrapper layout that calls the C++ function :P 08:51:31 but a large chunk of that is _builder_extras, which is bloody hell and insanity 08:51:52 greensnark, that's kind of ugly 08:51:59 -!- ais523 has joined ##crawl-dev 08:52:04 Zaba: Baby steps! 08:52:23 Your goal: select layouts using weight+chance 08:52:38 yeah 08:52:46 So first Lua-wrap the C++ layouts 08:52:52 And gradually convert them 08:53:42 I might do that.. 08:53:44 Incidentally, it would be nice if layouts took the build area as coordinates + size + mask instead of assuming they have the whole level to use 08:53:51 But again, that's long term :) 08:54:02 yeah, that'd be nice 08:54:22 could somehow employ subvaults there. But that wouldn't mix well with layouts that directly use grd instead of mapgrd. 08:54:28 So you could have a vault that can request a labyrinth as a subvault 08:54:43 and I believe you don't like using mapgrd for layouts :P 08:54:48 Or a vault that requests a swampy room 08:55:21 Zaba: mapgrd is better, but rewriting all the layouts to use it is a lot of work :P 08:55:25 mapgrd is more composable 08:55:50 oh, I also believe you don't like using the whole NAME/MAP/ENDMAP syntax for layouts :P 08:56:03 You seem to have an uncanny memory :P 08:56:21 I should be able to just call a layout function! 08:56:33 but what about weight/depth specification? 08:56:49 Zaba: First-class mapdef structures that can be manipulated from Lua :) 08:56:59 *that* would be nice 08:57:05 and we've also discussed it :P 08:57:06 * dpeg suspects that Zaba prints out everything greensnark utters here and nails it to his walls. 08:57:24 Yeah, this conversation is major deja vu 08:57:26 Everyone should do that actually. 08:57:33 I do it myself 08:57:42 My walls are a little crowded in consequence 08:58:39 oh man being able to designate a section of vault as swamp and so on would be amazing 08:59:15 Mu_, would need to rewrite the swamp dungeon build in lua as a subvault 08:59:34 greensnark, actually, it could work with subvaults for each layout + layout vaults that fill the whole level with the subvault glyph 08:59:54 that doesn't even immediately seem like an awful hack 09:01:17 Swamp and Shoals builders are pretty math intensive, they'd suffer from going to pure Lua 09:01:22 Unless we do Luajit 09:01:39 However, C++ code should still be usable there 09:01:48 could make them simple wrappers that call a few API functions (implemented in C++) to do all the heavy lifting 09:01:54 Right 09:12:06 -!- syllogism has joined ##crawl-dev 09:13:52 greensnark, damn.. need a default-tags: directive 09:18:48 and I need a faster rig, it seems 09:18:51 takes ages to start crawl 09:19:07 for me it takes ages the first time after a recompile, and is then very fast on other loads 09:20:47 spam on Mantis :( 09:20:49 -!- ais523 has quit [Read error: Connection reset by peer] 09:22:37 -!- ais523 has joined ##crawl-dev 09:24:22 hrm 09:25:13 greensnark, calling builder_basic from a vault causes crawl to get stuck in the call to it on startup, while reading .des files and executing lua chunks 09:27:28 greensnark, ah, it's stuck in _make_trail because that expects a level full of rock, but grd is probably not initialized at this point 09:27:30 how silly 09:27:41 it loops on while (grd(pos) != DNGN_ROCK_WALL && grd(pos) != DNGN_FLOOR); 09:28:41 I guess dgn_reset_level is not yet called at this point 09:31:54 greensnark, why is lua called when .des files are being read, anyway? 09:33:26 about starting Crawl: it takes so long because there's a million fsync()s while building the database 09:33:50 which are useless, especially that a partially built database should be thrown away (which we don't!) 09:34:02 kilobyte, well, in my case, it's because my code is obscurily buggy :P 09:35:16 kilobyte: Firefox had that problem; they found that using a different API to do the syncing did wonders on Linux (the one for syncing just part of a file) 09:35:23 but if you don't need to sync at all, that'd be better still 09:36:56 ais523: actually, the fsyncing problem in Firefox is bogus, too 09:37:17 somehow that doesn't surprise me 09:37:24 they added fsync() all around the place because NTFS likes to corrupt files during crash 09:37:55 well, Linux can use NTFS 09:38:26 even on non-journaled ext2 this type of corruption can't happen... so all you'd lose is a handful of seconds worth of browsing history 09:38:54 not sure if it's a problem with the filesystem or the implementation 09:40:29 new sqlite (which is what Firefox internally uses) has a nice fsync-less mode, too... it uses a huge (several megabytes big) journal, and transactions are not actually written into normal places until the journal is full 09:42:46 !learn add pearl_dragon < casmith789> pearl dragons are the sigmund of d:26 09:42:47 pearl dragon[2/2]: < casmith789> pearl dragons are the sigmund of d:26 09:44:26 well, at least the new holy monsters seem sufficiently deadly 09:44:49 although the fixed placement of the new vault is leading people to just skip past d:26 on the basis that it's so insanely hard compared to the levels around it 09:46:06 which is valuable balance feedback 09:46:19 it's not too hard to skip the holy vault 09:46:34 pearl dragons aren't that strong vs heavy armour chars anyway 09:46:37 unless they're undead 09:47:03 my MiGl took a max of 100 damage when he zerked it 09:54:14 was that Mi an AC char? 09:56:58 yes 09:58:02 he was doing 60 damage per turn at range + cloud damage against my ddne with ~20ac though 10:00:00 When wielding ranged, clicking adjacent ally incites an attack (https://crawl.develz.org/mantis/view.php?id=3596) by XuaXua 10:02:39 Zaba: .des file Lua has to be run when compiling the .des file to detect errors in the .des 10:03:02 After the first-time compile the only stuff that runs is Lua preludes 10:03:16 And the preludes are necessary to set up functions for markers, etc. 10:03:38 greensnark, well, I'm calling dgn.builder_basic, which works with grd that is not yet initialized, and that makes it go in an infininite loop 10:03:43 what'd be the best way to work around that? 10:04:01 Zaba: There's an api call to check whether you're being run at compile-time 10:04:27 crawl.game_started() will return true if the vault is being run for real 10:04:30 bhaak: NEVER go down stairs with Slow+Exh 10:04:35 oh, ok 10:04:38 thanks 10:06:36 I'M INVINCIBLE!!! 10:07:10 but I also was quite laggy, I almost wandered into an ogre :) 10:08:59 I wonder why stairs are such a big deal in so many roguelikes? 10:10:26 abrupt changes of the environment 10:10:30 ais523: well, it's very natural, isn't it? 10:10:40 perhaps 10:15:47 -!- eith has quit [Ping timeout: 252 seconds] 10:16:55 uh. 10:18:01 Ulch, cao is pretty low on disk space 10:18:15 Probably <2 weeks to disk full 10:18:33 it works 10:18:48 [TAG] Found map layout_basic tagged 'layout' 10:18:49 -!- elliptic has joined ##crawl-dev 10:20:28 bhaak: still exploring with Slow+Exh :O 10:20:29 greensnark: I like the way you used "Ulch", a delayed instadeath indicator in NetHack 10:20:46 Heh 10:20:56 I'll have to make some time to offload some ttyrecs :( 10:21:02 Not going to happen tonight though 10:21:16 would compressing them be enough? or are they compressed already? 10:21:23 They're already bzipped 10:21:40 rax said something about giving the VM more space, but I think she's busy 10:21:57 bhaak is unmoved :) 10:21:59 So for the nonce I think I'll backup the ttyrecs elsewhere to make space 10:22:37 hrm 10:22:39 it's broken, too 10:23:25 i'm Swiss, i'm being used to walk around slow+exhausted 10:24:37 greensnark, I think I misunderstand CHANCE 10:24:48 Join the club :P 10:25:02 specifically, the way it works when picking a vault by tag 10:25:30 I don't think CHANCE is applied at all for tags 10:25:33 Is it? 10:25:35 it is 10:25:43 if you give random_map_for_tag 'true' as a third argument 10:26:03 but, anyway 10:26:06 Aha 10:26:40 how do I do it so I always get _some_ vault from random_map_for_tag? Is CHANCE wrong for that? 10:27:16 You mean it doesn't always give you a vault? 10:27:20 That sounds pretty weird 10:27:22 I'm not sure 10:27:32 but could it be that the chance roll fails for all available vaults? 10:27:48 Yes, that can happen, but why would all vaults have a CHANCE: to start with? 10:27:54 a good question! 10:27:57 Shouldn't you be using just weights instead 10:28:05 bhaak: :) 10:28:19 CHANCE is really meant for things like: "I want all levels to have a 5% chance of a bazaar" 10:28:39 greensnark, yes, it probably should 10:28:47 just how to I convert percentages to weights :P 10:29:26 Zaba: Use a Lua callback in dungeon.cc that picks the layout to use :P 10:29:51 hmm? 10:30:06 Assuming you want arbitrary conditions to select the layout 10:30:24 How does it work now, anyway? 10% chance of X, 15% chance of Y, everything else: use random layout from this list? 10:30:34 well, no 10:30:45 It's just a weight already, no? 10:30:45 the distribution table for _builder_normal is roughly this: http://sprunge.us/BIQP 10:31:14 it's 'on arbitrary depth range, X chance for layout foo, Y chance for layout bar' repeated several times, and otherwise builder_basic :P 10:31:45 So use WEIGHT with depth ranges 10:31:51 that's what I'm trying to do 10:32:59 Nobody's likely to howl if you change the distribution a bit for that matter 10:33:43 Unless you make spotty or wide-open levels too common :P 10:34:46 The howling is for free. 10:34:59 greensnark, I can mix chance and weight, right? 10:35:09 Yeah, anything that goes wrong in the dungeon afterwards will be blamed on Zaba 10:35:15 because, for example, it makes sense for shoals/swamp/city to be CHANCE: 100% in shoals, swamp and dis, respectively 10:35:18 -!- valrus has joined ##crawl-dev 10:35:41 but city can also be WEIGHT: 1 in Vault and 10 elsewhere 10:35:42 Zaba: You can, but if you have both CHANCE+WEIGHT for the same level, the meaning is poorly defined. 10:35:52 As in: not defined at all 10:35:57 no, for different depth ranges 10:36:05 Different ranges, yeah 10:37:08 But you could just use DEPTHs instead 10:37:27 DEPTH: Shoals for the shoals builder 10:37:37 And when picking layout by tag, tell it to check the depth as well 10:37:53 well, for many layouts, I have DEPTH n-m, without the branch 10:37:59 DEPTH:* 10:38:32 Yeah, in those cases CHANCE: 100% for the exceptions should be fine 10:40:25 I've been playing with a treetop parser for Sequell queries 10:40:37 Pretty nice to work with, although error messages are disastrously bad 10:43:43 hrm. 10:44:49 how to make it so that branches like Vault don't get any layouts aside from two specific ones, and that those two are chosen with WEIGHT? 10:45:27 the problem is the DEPTHs specified as n-m present on other layouts, but I can't list all branches for all layouts, that's just mad 10:48:07 You could use CHANCE: with chance_vault_layout 10:48:31 So CHANCE: 100% on both layouts for the vaults, then include chance_vault_layout and weights 10:48:41 I'm not sure that it actually does what you want with the weights though :P 10:48:59 Alternative: use !Vault on all the layouts that shouldn't be used in vaults 10:49:16 can I have !Vault in default-depth? 10:49:32 Yeah, but it won't be inherited if the individual layout has its own DEPTH line 10:49:44 ah damn 10:49:58 Might be nice to have a DEPTH: variant that just adds on to the default depth instead of completely replacing it 10:50:13 yeah 10:52:26 okay, !Vault in individual layouts works 10:55:17 crash on reload (https://crawl.develz.org/mantis/view.php?id=3597) by Joellercoaster 10:56:01 kilobyte: When are you merging the unicode branch? 10:57:17 kilobyte pretends he isn't listening :) 10:57:33 greensnark: quite a bit of stuff to do, and it would need DGL changes first 10:58:18 Ah 10:58:29 So only 0.9? 10:59:07 kilobyte pretends he isn't listening :) 10:59:13 sorry, wrong console 10:59:17 Haha 11:02:51 ugh 11:05:30 -!- Kurper has joined ##crawl-dev 11:05:55 The Wise man's fear was highly disappointing :( 11:11:38 Really? 11:11:46 I got the Kindle book but I haven't dug into it yet 11:11:52 But the reviews seemed pretty positive 11:12:26 i haven't started it yet, either. 11:12:53 OTOH, i need to reread the first book again, 'cause i can't recall anything about it. 11:13:11 There was a pretty funny recap comic by Rothfuss, did you see it? 11:13:22 no 11:13:28 http://blog.patrickrothfuss.com/2011/02/our-story-thus-far/ 11:13:29 nope 11:14:01 oh god the felurian 11:15:08 he should hire a new editor :P 11:15:29 Damn, now I have to read it soon 11:15:41 oh, right. now i remember. i still need to reread it, to find out whether i like it more now :P 11:16:51 -!- jpeg has joined ##crawl-dev 11:17:33 I love the panel where they spoof Gandalf in Moria 11:18:07 -!- ais523 has quit [Remote host closed the connection] 11:19:24 Trying to add Curio to the credits... 11:19:56 "git diff" complains about "Terminal not fully functional" 11:20:25 I specifically had to install Strawberry Perl to get the perl script to run, damn unicode stuff 11:20:33 Ah, windows? 11:20:51 yep 11:20:53 git diff tries to run less if it's on a tty, so it's probably not happy with the windows console 11:21:20 Running it from the bash prompt ought to work, or you can use some option to tell it not to run a pager 11:21:22 not a problem normally, but I'm worried I screwed up something with the unicode 11:22:24 looks fine in Notepad++, but not having a diff to check makes me nervous 11:22:46 Can notepad++ tell you the encoding? 11:23:33 utf-8 I think 11:23:42 You can check the diff with gitk after you commit, I think 11:24:54 ah well, if I broke anything someone will revert :P 11:25:45 greensnark, I need to add a function to set the level layout type to the lua api---what library does it belong to? 11:26:07 Um, dgn? :) 11:26:24 well, there's l_dgn, l_dgnbld, ... 11:26:30 We should document the major Lua namespaces and a brief note on what goes where 11:26:37 and, don't those end up in metatables for various things? 11:27:02 I don't remember what goes where :) 11:27:49 I always have trouble with the l_foo files. I just list them all, look at the names that look promising, and add stuff to the file where it looks least out-of-place 11:27:59 I should patent my system 11:28:25 greensnark: hey, that's my system! :D 11:28:26 you should've done it earlier, before ibm patented patenting 11:28:38 03j-p-e-g * r8f521e6de0e0 10/crawl-ref/CREDITS.txt: Add Curio to the credits. 11:29:24 I'm curious to know what a person with a curious name like curio's real name is 11:30:06 Paul 11:30:14 Paul Svinarchuk sounds pretty close to Pavel Svinarhuk to my uneducated ear 11:30:19 Are they related? :) 11:31:25 greensnark: they're worlds apart 11:31:38 you mean he's already in there? 11:31:45 that'd be embarassing... 11:31:52 jpeg: No idea, but the names seemed rather similar, apart from Pavel/Paul 11:32:05 Pavel is the slavic form of Paul, so ... yeah 11:32:51 Svinarhuk might be a common name 11:33:25 I got the name from here: https://crawl.develz.org/mantis/view.php?id=3045 11:33:26 Sounds like the sort of thing Superman might do to conceal his secret identity 11:34:07 woot? he needs more concealment than glasses??? 11:34:24 Yeah, if the glasses don't fool them, adding one letter to the surname totally will 11:35:18 it's the same guy, apparently 11:35:41 my git log's broken now, so I've got trouble doing a search 11:36:02 but there's a mention of adding him to the credits 11:36:28 now, how do I unbreak git? 11:37:49 I was about to offer up as release manager but I really can't work this way... 11:37:52 * jpeg goes off to cry 11:38:11 git --no-pager log 11:38:25 jpeg: this is the curse of windows, you really need a linux 11:38:27 or alternatively install msys or cygwin 11:38:37 I am using msysgit 11:39:10 is nopager supposed to be uninterruptible? 11:39:44 -!- casmith_789 has joined ##crawl-dev 11:40:05 it doesn't call less so yes 11:40:32 you could try 'git --no-pager log | $pager_of_your_choice' 11:40:37 not very helpful then :( 11:40:50 -!- casmith789 has quit [Ping timeout: 255 seconds] 11:40:56 -!- casmith_789 is now known as casmith789 11:41:16 jpeg: would it help if you could log onto a linux machine? I am sure kilobyte would set up an account on dis for you. 11:41:28 This is how I do all my crawl work. 11:41:30 at least piping works 11:42:21 but it worked fine until the credits commit... and it was completely unnecessary! 11:42:43 ah, this way you can set your fav pager: git config --global core.pager 'less' 11:43:21 I think that's already my global setting 11:43:33 there's a bot account in mantis 327 11:43:40 yes 11:43:46 or I guess someone just got it 11:44:23 Blade looks so odd when rotated 11:51:51 yes 11:51:54 -!- eith has joined ##crawl-dev 11:51:55 flip it back imo 11:52:42 can't flip it 'back', it just can be randomly rotated now 11:52:58 could add no_rotate to it, though 11:57:36 oh, do that then 11:57:54 !tell galehar Would you lock the Og forum thread? 11:57:55 dpeg: OK, I'll let galehar know. 12:01:39 in case anyone is interested: "export TERM=linux" solved my git log troubles 12:01:47 yay 12:01:58 * dpeg appoints jpeg to release manager =) 12:02:24 happened before, and now I remember that strawberry perl overwrites the TERM variable, so yay! 12:02:52 dpeg: :) 12:03:31 -!- eith has quit [Ping timeout: 246 seconds] 12:03:36 okay: are there any reasons not to branch right away? 12:04:47 I'm sorta antsy about getting rid of scrolls of paper for reals :P 12:06:11 back later 12:06:18 -!- jpeg has quit [] 12:06:22 greensnark, okay, I think it works more or less well now 12:09:57 committing it would be for the great annoyance of kilobyte, though :P 12:10:26 because it makes it necessary for all layouts to be at least wrapped in lua :P 12:11:14 dpeg: tiles don't work over X forwarding so jpeg would be able to use only text there... 12:11:48 kilobyte: yes, I know 12:12:06 well, at least I suspected that it'd be insanely cumbersome to do tiles over X 12:12:30 Also, I don't think anyone will shed a tear or two for the unknown scroll of paper. 12:12:30 Zaba: well... I've written a single PHP page in my life. Its text was 12:12:57 kilobyte, way to go :P 12:13:22 dpeg: it is _supposed_ to work, but doesn't 12:13:49 -!- monqy has joined ##crawl-dev 12:13:51 kilobyte, well, the existing C++ layouts would be called in a similar manner 12:13:55 dpeg: heck, I've played Quake 15 years ago... and four guys logged in from Irix boxes to a single Linux machine 12:14:02 i'm sad scrolls of paper are going because now they can't be used for ashenzari :( 12:14:20 kilobyte, the benefit of using DEPTH/WEIGHT/CHANCE for choosing the layouts rather than the _builder_normal insanity is quite large, I'd say :P 12:15:35 Eronarn: does it matter much if the Ash special scroll is Curse Foo or Paper? 12:15:38 dpeg: also, it seems to be caused by latency breaking something. I built crawl-tiles on my phone and sshed into it from a desktop computer, it worked fine. Yet when doing the same over a 25 ping 2Mbit link it just locks up. 12:16:08 dpeg: paper is more flavorful - ash is just telling you how to write down a curse, rather than modifying a scroll 12:16:43 Ash could affect scrolls of acquirement and fog, just as well. 12:17:30 Zaba: so go for it... even such a Lua-hater like me can see the benefit 12:17:59 (one other thing that could be considered: ash's curse seeps into you as you wander the world (gain piety, before the piety boost from wearing cursed stuff). thus curse scrolls can be a boost, but you cannot rely on finding enough in the game for regular swapping; and if you don't have one you must wear an uncursed item for a while until it becomes bonded to you) 12:18:38 okay :P 12:18:45 and when I'll get around to implement layout_delve or forest, doing this the way of my PHP page will work just fine :p 12:19:40 kilobyte, by that time, greensnark will come up with a sane API for writing level builders in lua and all will be well :P 12:20:27 kilobyte, with first-class mapdef structures in lua and such niceness :P 12:21:06 actually, lua is _less_ readable than the equivalent C++ code, so the main reason for using script languages doesn't apply here 12:21:19 it's DEPTH/WEIGHT/CHANCE what's a good thing 12:21:45 I need to fix up the weights for layouts first, though 12:21:46 or.. hm 12:22:13 I might just commit it as-is, but some layouts will be rarer, others will be more common 12:22:16 and I can't tell which :P 12:22:30 but maybe that's a good thing 12:23:14 dpeg: btw, had an idea to make MSH actually salvageable: it summons a dancing weapon silver hammer. a *GIGANTIC* hammer, titan-sized. knocks back enemies multiple tiles, based on their size, when it swings 12:23:50 Eronarn: so you'd salvage nothing but the name 12:24:03 kilobyte: the name is the only interesting thing about it 12:24:27 -!- eith has joined ##crawl-dev 12:25:28 going more literally -- giving the hammer a silver brand plus slaying could be better 12:26:54 why not just delete the spell? 12:27:05 kilobyte: nobody will use a spell that gives a hammer silver brand unless it's so much slaying that it's OP 12:27:19 -!- valrus has quit [Remote host closed the connection] 12:29:12 i also believe that a spell like that isn't great from a design POV; it has the problems of all the temp buff spells except it's limited in which weapons you can use it on making it even more narrow 12:29:57 that's why I didn't run to implement it, even though I like the idea 12:30:43 I believe silver brand can be useful early on. 12:30:58 but of course it is fringe 12:31:05 i don't think we need a fringe spell 12:31:17 it's also 100% a beatles reference if anyone's concerned about that sort of thing 12:31:26 if we're going to have silver brands early on - why not a zin thing? a weapon blessing, except one that happens at very low piety (say, 50) 12:31:45 would be an interesting twist on the weapon blessing mechanic which as of yet is max-piety-only 12:31:48 redundant with TSO 12:31:53 Eronarn: being playing exclusively Zin these days btw 12:31:56 !whereis Infest 12:32:05 No games for Infest. 12:32:07 grrr 12:32:11 kilobyte: TSO's is much better 12:32:13 @whereis Infest 12:32:14 Infest the Nimble (L12 DrPr), a worshipper of Zin, saved on Lair:3 on 2011-03-16 after 26974 turns. 12:32:37 dpeg: zin definitely feels better; there's still room for improvement but i think he is a viable god now 12:32:49 still very concerned about the orc thing thouhg 12:33:11 Eronarn: orc thing? 12:34:27 dpeg: beogh is counted as an evil god, and most orcs are worshipers of beogh. therefore, they get hit very hard by Zin's recite - regularly killing normal orcs 12:34:52 ah 12:34:54 not that I have any say but I think 0.8 should be released soon so we should stop adding features :) 12:35:13 I noted that orc priests are affected more strongly, and liked that. 12:35:13 i'm of the opinion that it makes no sense for beogh to be an evil god anyways, and changing that would fix the problem 12:35:19 casmith789: yesyes 12:35:29 we even almost have a release manager! 12:35:42 who is the release manager? 12:35:45 and his Lordship is visiting us regularly to inquire about release progress 12:35:46 or who almost is 12:35:50 sssh 12:36:01 is it MarvinPA 12:36:05 no 12:36:16 is it the CIA? 12:36:18 good, I need him in fighting form for the tourney 12:36:39 (priests would still be hit harder, fwiw. it goes: non-zin god, evil god, chaotic god, priest of a non-zin god - so right now orcs are 2, orc priests are 3. if it were changed, orcs would be 1 (and no longer get outright damage effects), orc priests would be 2) 12:36:59 casmith789: :) 12:37:15 Eronarn: yes, there's a comment on "heresy", I liked that a lot 12:39:36 Why is Beogh actually evil? 12:39:47 I am sure when I designed Beoghie, he wasn't yet. 12:41:01 jpeg thinks he's hitler 12:41:09 it's on mantis but I can't look now 12:41:14 dpeg: when arguments have been had in chat over it, the justification is that because he hates TSO, TSO designates him as evil 12:41:28 has anyone ever said that 12:41:31 yes 12:41:42 not TSO specifically, but 'the good gods' 12:41:50 the argument was that "pro-orc" is evil 12:41:59 as opposed to "kill everything" which is clearly neutral 12:41:59 (it's not) 12:42:12 I don't think that jpeg thinks Beogh is Hitler. 12:42:17 beogh wants you to kill orcs too 12:42:24 there's literally nothing to suggest beogh hates good gods 12:42:54 Why does Beogh hate TSO? I always thought he'd deny the existence of non-orcish gods :) 12:43:22 anyways i don't think it's fruitful to get into this discussion again 12:43:30 a better question would be: is there a compelling gameplay reason for beogh to be an evil god 12:43:31 Vehumet's status was also moved back and forth, iirc. 12:43:48 i say no: he doesn't do anything associated with the other evil gods (demon summoning, abyssal stuff, necromancy, undeath) 12:43:51 liking orcs is evil, summoning demons is neutral 12:44:09 Eronarn: okay, will ask in another mail 12:44:27 whereas having almost all orcs be evil has some negative gameplay effects - more pronounced now with the zin thing but it can come up in other cases i assume 12:45:00 hitler's in 2306 12:45:03 literally hitler 12:45:51 kilobyte brought it up though 12:46:29 dpeg: another thing to ask in mail would be about revising the zin rules for chaoticness/etc.; right now the piety gain and the recite versions are slightly askew 12:46:48 i think the recite versions are better, but either way, they should be brought in line before .8 12:47:36 Eronarn: I always left that to dolorous 12:48:02 although I think Keskitalo also has a hand in Zin rules 12:48:39 yeah, i don't recall asking him specifically about the piety thing though presumably he read my code when he reviewed it 12:49:10 ??new monsters 12:49:11 I don't have a page labeled new_monsters in my learndb. 12:49:13 ??new monster 12:49:14 I don't have a page labeled new_monster in my learndb. 12:49:16 hmmm 12:49:18 ??learndb 12:49:19 learndb[1/9]: A html page of learndb entries is at http://crawl.akrasiac.org/learndb.html. You can also access the learndb at http://crawl.develz.org/wordpress/bots. 12:51:24 !learn add todo_monsters decorator crabs, velvet worms, osedax worms (bone-worms), tardigrades, shrikes, bladderworts, sundews, rocket hawks, aerial jellyfish 12:51:26 todo monsters[3/3]: decorator crabs, velvet worms, osedax worms (bone-worms), tardigrades, shrikes, bladderworts, sundews, rocket hawks, aerial jellyfish 12:54:31 03zaba * r8181c35602f0 10/crawl-ref/source/ (dat/des/builder/layout.des dungeon.cc dungeon.h l_dgnbld.cc): Choose dungeon layouts by picking a vault with tag "layout". 12:54:43 03zaba * rdcdbbad24810 10/crawl-ref/source/dat/des/builder/layout.des: Remove redundant calls to fill_area in some layout vaults. 12:56:20 -!- Zaba has quit [Ping timeout: 252 seconds] 13:00:44 -!- Zaba has joined ##crawl-dev 13:01:40 -!- jpeg has joined ##crawl-dev 13:02:41 back, have we branched yet? 13:02:45 :P 13:02:58 everyone is awaiting your orders! 13:03:37 greensnark: why would we need to wait 3 weeks after branching? what's the difference if all commits get cherrypicked anyway and no new features get added? 13:03:47 !tell greensnark Why would we need to wait 3 weeks after branching? what's the difference if all commits get cherrypicked anyway and no new features get added? 13:03:47 jpeg: OK, I'll let greensnark know. 13:06:29 which features are still being worked on for 0.8 to a degree that we can expect several dozen commits? 13:07:12 in particular, how far along are we with the holy monsters? 13:07:25 i think they're still pretty unfinished 13:07:26 due's been busy 13:07:52 i'd like to have a monster version of new-recite working for the Silver Stars, but who knows if that will even be possible 13:08:23 no reason not to delay the unfinished ones, but I should probably ask due about his plans 13:09:00 Eronarn: what are the salt pillars? 13:09:18 are the monster petrified for good or is it a permanent effect? 13:09:22 for good 13:09:30 they crumble away in a short amount of time, like toadstools dying 13:09:41 ah... okay 13:09:49 might make a difference in the tiles design 13:10:03 but I guess that means the actual shape is not important, beyond flavour 13:10:15 also, are you in the credits yet? (came up earlier today) 13:10:46 -!- aze5 has joined ##crawl-dev 13:10:55 i envisioned it as being only a very rough outline, like here: http://www.thercg.org/youth/images/0504_lot_01_s.jpg 13:11:07 but a featureless blob could be fine too 13:11:11 and no 13:11:15 jpeg: We don't need to wait 3 weeks after branching, it's just that the lack of a branch suggests a lack of attention to stabilising the build, and I think 0.8 needs at least a few weeks of effort toward stability fixes with no major new changes 13:11:15 greensnark: You have 1 message. Use !messages to read it. 13:11:42 Creating a branch means new changes (to master) stop altering the landscape all the time 13:11:49 i see, so it's mostly about the message it sends within the team and to the public 13:11:52 isn't turning things into salt pretty unthematic for the purity-of-form god 13:11:58 been wondering about that 13:12:25 jpeg: Not just a message. The problem with attempting to stabilise master is that you can't stabilise master with new changes going in all the time 13:12:36 OG17: ah, but he only does it to impure things. and he turns them into a pillar of pure salt! 13:12:45 "Stabilising bleeding edge code" is a bit of a contradiction in terms, that kind of thing 13:12:58 OG17: zin is based on the OT god, as such he's quite hypocritical 13:13:02 The first main advantage of a branch is that new changes stop going into the branch 13:13:10 Unless they're intended to fix problems in the branch 13:13:18 jpeg: I am only aware of Ash and holyhaul. 13:13:33 there are still quite a few annoying bugs that really should be fixed before 0.8 release, btw 13:13:49 dpeg: trample, if we want to try getting that in - it is definitely the kind of change that can introduce crashes 13:13:53 greensnark: I meant that lately, changes have been either bug fixes or balancing changes to unfinished features needed for 0.8 13:13:55 since it reorders some code 13:14:10 but of course, that does put developers under a lot of pressure 13:14:31 elliptic: which ones, in particular? 13:15:15 the one that comes to mind first is the bug with colors on ctrl-F searches being wrong 13:15:36 Eronarn: under which name would you like to be credited? 13:15:36 second one that comes to mind is travel exclusions being funky 13:15:48 monqy: funky, how? 13:15:55 I'll try and find the report 13:15:57 jpeg: Eronarn Palazzo, the same as on git 13:16:00 Mutations can override racial mutations gained by leveling. (https://crawl.develz.org/mantis/view.php?id=3598) by TwilightPhoenix 13:16:29 https://crawl.develz.org/mantis/view.php?id=2930 13:16:35 is the stash color bug 13:17:07 there's also this: https://crawl.develz.org/mantis/view.php?id=3514 13:17:19 terpsichore is currently completely harmless 13:17:32 Eronarn: that's not a real name, is it? dpeg is quite adamant about that, though I disagree 13:17:50 jpeg: you add it, not I 13:18:00 Darshan didn't care either. 13:18:27 dpeg: the thing is we've got quite a number of aliases already 13:18:37 Real names are preferable, but if the contributor insists on a netname, they get it 13:18:56 And in elliptic's case I just didn't get around to asking for their real name :) 13:19:08 elliptic: You want your real name in credits? 13:19:13 it would also be nice to fix https://crawl.develz.org/mantis/view.php?id=3444, which is just an off-by-one error 13:19:21 greensnark: I'd prefer elliptic 13:19:30 i don't 'insist' on it, but that is the name i use online most everywhere 13:19:39 Yeah, good enough 13:19:43 the best way to handle things is with firstname 'alias' lastname 13:20:05 that's fine; my real name is James Meickle 13:21:34 ok I found the report I was looking for: https://crawl.develz.org/mantis/view.php?id=3303 13:22:20 oh right, that one is really annoying also 13:22:29 that one and the ctrl-F one show up constantly 13:22:45 Eronarn: so... Eronarn Palazzo or James 'Eronarn' Meickle - your choice :) 13:23:39 Did we figure out if Curio is really Pavel aka Paul Svinar[c]huk? 13:23:52 yes, he is 13:23:54 Aha 13:23:59 Jude added him under that name a while ago 13:24:20 as to which of them is "correct", no idea! 13:24:53 chuk sounds more Slavic, but I know about as much about Swahili as I do about such names 13:25:20 I think I'm basing that off "Ivanchuk" 13:25:27 So a sample size of 1 :) 13:26:55 I'll send him an email 13:27:10 jpeg: the latter is fine 13:31:57 now about b0rsuk... 13:32:47 jpeg: I can write a mail. Shall I? 13:33:02 ??b0rsuk[7] 13:33:03 b0rsuk[7/7]: Finally, let it be known that my current nickname is 'b0rsuk', all lowercase in accordance with unix tradition. And "Borsuk" is not my name - "Borsuk Euroazjatycki" means "Eurasian Badger" (meles meles). 13:33:15 You should just credit him as B0rsuk Eroazjatycki :) 13:33:45 greensnark: he will get mad, especially since you forgot the u 13:33:58 I think Eronarn has more cause to be mad :P 13:34:09 (for being associated with b0rsuk) :) 13:34:10 it's the name he listed on Mantis ;) 13:35:06 I'll wait for Curio to reply before I commit any further changes to the credits 13:35:16 once a day is quite enough :P 13:35:38 is the release documentation up-to-date? 13:37:01 jpeg: I went through the whole manual recently, although there are definitely omissions. 13:37:06 for example, new status lights 13:37:15 autofight not included 13:37:31 I'm talking about git quickstart.txt, the release process 13:37:36 how to set up a new branch etc. 13:37:52 without that, I wouldn't have a clue 13:38:29 let's get water status lights in .8 :D 13:38:36 speaking of things that should happen before the release, how about removing or at least redesigning the orb chamber subvault deal, especially now that orb apportation is harder? 13:39:22 weren't a couple of the orb vaults removed again? 13:39:34 no 13:39:50 there's no such thing as too many removed vaults, anyway :P 13:40:06 branching in 3... 2... 1... 13:40:09 any objections? 13:40:24 * greensnark has a fix for 34444 first 13:40:26 congratulation on your promotion :) 13:40:28 Well, with one less 4 13:40:47 -!- valrus has joined ##crawl-dev 13:40:48 greensnark: <3 13:40:59 jpeg: Yay, thanks for being release manager! 13:41:26 if I get the job (as I hope), someone else might have to take over 13:41:32 or we need to be quick about the release ;) 13:41:46 jpeg: time pressure? 13:41:51 I vote for being quick :) 13:41:54 Is that the Beogh job? 13:42:04 yeah 13:42:07 yay 13:42:16 Quick release is good 13:42:24 Of course, I work exclusively for Trog. 13:42:34 I explained about how I coded a new god that offers permanent allies ... and they were really impressed 13:42:47 "This research has been cosponsored by T.R.O.G." 13:42:48 I er... somehow forgot to mention that allies existed in Crawl before that 13:42:57 * jpeg whistles innocently 13:43:53 it only occured to me belatedly that they probably think I did it all from scratch 13:44:06 jpeg: Don't tell them now, okay? 13:44:29 dpeg: don't worry, I won't :) 13:44:30 -!- valrus has quit [Read error: Operation timed out] 13:44:36 03greensnark * raa4f317c0c08 10/crawl-ref/source/ (misc.cc player.cc): [3444] Fix off-by-one error in autopickup_no_burden checks. 13:44:39 jpeg: speaking of Beogh, did you read the awesome story on SA where someone's wife/girlfriend played Crawl for the first time, and got quite far with a Beoghite? 13:45:02 dpeg: yeah, awesome 13:45:18 newbies love Xom and Beogh both, even if powergamers don't 13:46:08 yes, they make good stories, which is what many of them like 13:49:13 Just give the reinforcements a little buff when near max XL and piety. :f 13:51:06 hmm 13:51:42 so, things that are missing after my change are the gateways to abyss on layout_big_octagon and occasional replacement of rock walls with stone or metal on _plan_* levels 13:51:54 both are recreatable with vault techniques in layout.des, though 13:52:00 but yawn 13:52:29 xom would be less annoying if he didn't statdrain so much, especially now that statdrain takes longer to heal 13:53:02 xom should always add the drained points to some other stat 13:53:45 because just draining is drama without enough lol, and xom likes them balanced 13:54:19 like temporary stat boosts, you mean? 13:54:33 also, that's treading in Jiyva's territory 13:54:53 +1 for what monqy said... playing xom currently feels like you have statrot 1 mutation 13:54:57 or maybe even statrot 2 13:55:27 yeah, Xom will get another overhaul in 0.9 or 1.0, depending on whether I have time or not 13:55:35 you often end up with stats drained by 8 or so for a substantial period of time 13:55:52 more good stuff during tension, etc. 13:55:54 (if you don't use a restore abilities potion or royal jelly) 13:56:14 on the upside, xom never kills you directly - unless bored, of course 13:56:50 jpeg: bbbut Gozag! 13:57:09 probably if xom gives you statrot mutation then it will kill you in combination with the xom stat drain 13:57:28 dpeg: Xom should be less work than Gozag, so Xom comes first 13:58:03 sniff 13:58:09 oh, and I pushed a new branch 13:58:13 what now? 13:58:36 dpeg: maybe Xom for 0.9 and Gozag for 1.0? 13:58:40 jpeg: I think we keep pushing to trunk and you have to cherrypick bugfixes into the new branch 13:58:48 jpeg: double sniff, but sure :) 13:59:05 jpeg: btw RangerC gave a sign of life 13:59:12 0.9 is August, which is practically around the corner 13:59:32 yes 13:59:56 gozag? 14:00:40 God of stones. 14:00:57 gozag ym sagoz, gozag the greedy, god of gold 14:01:12 I mostly love the name :) 14:05:58 hmm... what's the syntax for cherry-picking several commits? 14:06:21 just list their hashes one after another? 14:06:26 imo implement gozag as an empty altar 14:06:27 git cherry-pick commit1 commit2 commit3 ... commitn, I guess 14:06:39 who cares about mechanics, he's gozag 14:06:40 also apparently the -x option is useful, anything else? 14:06:59 -!- ais523 has joined ##crawl-dev 14:09:57 TGWi: you can read about the gold god on the wiki, no need to explain it here 14:14:09 Zaba: re blade getting rotated, i thought it looked okay so i removed no_rotate when i made it encompass, if others disagree then feel free to put it back though :P 14:14:14 -!- MarvinPA_ is now known as MarvinPA 14:14:42 dpeg: I'm not mocking 14:14:44 it's a cool name 14:14:45 :( 14:15:01 oh and other stuff for 0.8: was someone working on monster spines for porcupines? 14:15:24 TGWi: it's also a pretty cool god :) 14:15:34 MarvinPA: Cryp71c? 14:15:36 Cryp71c: ^ ? 14:15:55 spiny frogs, porcupines and perhaps someone else should get the spines effect 14:16:18 03dpeg * r8e8335248872 10/crawl-ref/source/dat/des/variable/mini_monsters.des: A few changes to Shiori's sword in stone vault. 14:16:45 could give it to a demon 14:17:02 dpeg, yep, I'm here. was just talking earlier about porcupines having spiny. 14:17:07 MarvinPA: I was working on it but then I gave up and gave my progress to Eronarn 14:17:14 If we give spiny to spiny frogs, they're going to be quite a bit more formidable than they are presently. 14:17:15 um... builder/layout.des:74: attempt to call global 'layout_type' (a nil value) 14:17:21 ah cool 14:17:23 Cryp71c: have you been coding that? 14:17:25 and yeah I'd be careful buffing spiny frogs 14:17:31 dpeg, no, I havn't. 14:17:33 ok 14:17:38 monqy, you've been working on spiny for monsters? 14:17:42 they're a flagship threat of lair 14:17:45 Cryp71c: I was 14:17:54 TGWi: will be more flagshippy with spines 14:17:58 :P 14:18:06 dpeg, lair will be universally hated. 14:18:16 @??Spiny frog 14:18:17 spiny frog (08F) | Speed: 12 (swim: 60%) | HD: 7 | Health: 24-52 | AC/EV: 6/9 | Damage: 2604(medium poison) | Flags: amphibious, cold-blooded | Res: 06magic(28), 03poison, 12drown | Chunks: 09poisonous | XP: 408. 14:18:21 Cryp71c: right after Slime and Tomb, you mean? 14:18:23 we could have them appear later 14:18:24 Cryp71c: I generalized the spiny effect but had a bit of trouble getting it to trigger properly, what with fight.cc being structured the way it is 14:18:25 well they could just have their hp reduced a little or something if they get it 14:18:26 crawl will fall apart and literally refuse to compile 14:18:28 or move them deeper, yeah 14:18:33 spiny frogs, hydras, elephants, sound like a cool combo 14:18:33 the earth will catch on fire 14:18:55 MarvinPA: don't listen to them, they want to soften Crawl at the core. The spiny core, I might add. 14:19:04 heh 14:19:18 dpeg, lol, I'm just not prone to buffing creatures that are already fairly buffy. 14:19:31 And not buffy in a hot vampire killing way. 14:19:41 dpeg: everyone likes tomb 14:19:43 (spiny frogs kill vampires too though) 14:19:46 (or at least I do) 14:19:56 even if it was made a lot harder in .8 14:20:02 monqy, has Eron done any more work on it? 14:20:10 !tell Zaba dat/des/builder/layout.des:74: attempt to call global 'layout_type' (a nil value) 14:20:11 jpeg: OK, I'll let Zaba know. 14:20:19 Cryp71c: no idea 14:20:27 Eronarn, yo? 14:20:30 !poke poke 14:22:13 !tell Eronarn I was told Monqy passed along Spiny monsters stuff to you, any progress on its implementation? 14:22:13 Cryp71c: OK, I'll let Eronarn know. 14:23:49 Hopefully I'll be back to do some more dev :) Work and baby have been ~_~ 14:23:57 Cryp71c: how old? 14:24:04 dpeg, 13 month 14:24:17 great, everyone is sound? 14:24:39 Yeah, pretty well. Wife's working from home, though she's talking about quitting her job, my life may go back to being ~_~ if she's not working :) 14:24:40 TGW: having spiny frogs get spines is clearly desirable; the frogs can be nerfed in a number of other ways to compensate if we think that's needed 14:24:49 Cryp71c: I can relate 14:24:52 dpeg: yes 14:24:54 boy or girl? 14:24:58 dpeg, yeah, they should def get spiny. A 10% drop in hp should be sufficient. 14:25:02 Girl, she's just starting to walk. 14:25:40 maybe a bit more hp reduction, we'll see. If monster spiny is anything like player spiny, it can be pretty powerful. 14:25:48 evolution of homo sapiens offspring: walk, talk, crawl 14:26:07 I presume thats in <--- this direction and not----> ? 14:26:17 no no :) 14:26:26 Ahhh. 14:26:30 Crawl as in DCSS 14:26:38 Cryp71c: well, our choice what level we give the monster... porcupines should have weaker spines than spiny frogs, who should have weaker spines than some appropriate demon 14:27:31 dpeg, agreed. Do monsters have mutations, just as players do? If so, a cleverly crafted universal mathematical equation would work for both players and monsters, to minimize redundant code. 14:28:25 0.9 should see more rockin' TGW animal suggestions implemented imo 14:28:30 could just base the damage on monster hd 14:28:55 guard llamas dropbears coconut crabs anglerfish snapping shrimp skunk 14:28:56 MarvinPA, that would probably be the next best implementation. 14:29:05 and chameleon instead of gila monsters 14:29:12 monster mutations and monster demonspawn 14:29:18 mmm monster demonspawn 14:29:34 porcupines probably shouldn't be lair-depth 14:29:41 or maybe they should show up in packs in lair and singularly in D 14:29:52 TGWi, earlier than lair, IMO. 14:29:55 Not packs. 14:29:55 esp. if spines are HD-based, they're not really a threat 14:29:59 i sort of thought about starting on monster demonspawn but then i realised i had no plan on how i'd do mutations and stuff 14:30:05 TGWi: solution: increase hd 14:30:16 megaporcupine 14:30:29 We're already going to have spiny frogs in lair. Might as well have porcupines show up earlier in dungeon with their current HD. 14:30:37 TGWi, I refuse your rediculousness :) 14:30:42 MarvinPA, can monsters have mutations? 14:30:46 oh, we need warthogs too 14:31:00 not as far as i know 14:31:06 <3 warthogs, 2-square knockback on a warthog charge. 14:31:13 so it'd require some new fancy way of implementing it 14:31:35 MarvinPA, perhaps we have good cause to implement it with spiny and monster demonspawn. 14:31:53 Might even allow room for improved code on other mechanics that stem from mutations that monsters mirror. 14:32:00 perhaps, although also most of the player demonspawn mutations wouldn't actually work very well on monsters i think 14:32:12 so the monster ones would have to end up quite different anyway 14:32:15 No, most of them aren't appropriate. 14:33:14 i guess the most basic thing would be some giving them some random resistances and attack types 14:33:32 But scales, the resistance trees, breath facets, spiny, body-slots, and a few others would apply. 14:34:55 As far as implementation goes, what about something like how enchantments are implemented? It's been a while since I've been in the code but I recall monsters having an enchantment implementation that is smaller than the player enchantment implementation. 14:35:10 s/smaller/smaller in memory size 14:35:30 -!- jpeg has quit [] 14:38:09 Cryp71c: i didn't know that i was doing it! i have nothing against doing it, i just didn't know 14:38:09 Eronarn: You have 6 messages. Use !messages to read them. 14:38:25 !tell eronarn check your messages dude 14:38:26 TGWi: OK, I'll let eronarn know. 14:39:21 Eronarn, lol, <3 your messages. I don't mind doing it, I implemented player Spiny, and monster spiny was on my to-do list before I went away for a while. 14:39:44 I was just told monqy had passed part of his implementation along to you, which is why I targetted you specifically. 14:39:49 monqy, is your code in trunk? 14:40:57 spiny code? not that I know of 14:41:57 monqy, do you still have it? 14:42:04 probably not 14:42:19 Ah ok, well if it turns up I'd take a look at it. But don't stress over it. 14:42:55 Eronarn, so nevermind, ignore my message and ignore monster spiny :) I'll do it. 14:43:24 ;D 14:43:30 :D* 14:44:05 -!- eith has quit [Quit: sleep] 14:44:16 if whatever pastey service I used didn't remove it, there should be a link in the channel logs somewhere, but I don't remember the date 14:44:48 MarvinPA, its been ages since I was deep in the code, and I was never great at understanding the big picture behind how actors / monsters / players were related and whatnot, so if it ends up that implementing mutations for monsters is the direction we go, I won't be able to do it in any timely fashion on my own. 14:45:28 Cryp71c: specifically, what i wanted to do was tackle stuff-happens-on-damage effects in a vaguely unified way 14:45:29 Eronarn: You have 7 messages. Use !messages to read them. 14:45:43 for spiny, acid blood, monsters that shock you when you hit them, whatever 14:46:19 Eronarn, I would love to see that as well, lots of fight-related stuff could benefit from a vaguely unified implementation. 14:46:24 yes 14:46:28 Eronarn: I was going to do it like that but then I ran into troubles with things triggering from different things, like spiny is supposed to trigger on misses 14:46:52 monqy: right, probably the best way would be to use a series of hooks 14:47:20 something like: tried to attack; hit shield; hit; damaged; killed 14:47:41 that's 5 separate points at which an on-hit effect could reasonably want to trigger at 14:48:17 possibly one for actually-missed, too, for stuff that only matters if you try to hit and miss completely 14:48:27 It might be a bit more granular than that, imo, to cover all the bases for a unified implementation, but that sounds like a good summary of the organization. 14:48:50 Cryp71c: you don't want to get too granular at the start or it will be so complicated that it never happens 14:48:54 5 or 6 code points is doable 14:49:32 Ofc, but it seems counterproductive to change certain effects (like, say...fumbling) because our plan from the get-go was not well designed. 14:49:52 add an enum for these 'attack phases'; new function do_hits(enum); do_hits calls relevant other functions 14:49:56 or at least change those effects until some time down the road when this new "unified implementation" is corrected to properly address all attack possibilities. 14:50:32 you could have multiple functions, rather than a function that looks at an enum, but imo it would be a lot cleaner to have all of these effects in one place rather than across 6 different functions 14:50:37 Eronarn, are you talking about these attack phases existing for the monster-attack, player-attack, or both? 14:50:41 both 14:50:56 they already do, to some extent, because the game prints messages at those phases 14:51:05 Eronarn, 1-function implementation (1 for monsters, 1 for players) seems more appropriate. 14:51:11 right 14:51:37 Separate functions seems a bit too functionally-oriented anal-retentive. A practical implementation. 14:51:47 a practical implementation instead* 14:51:53 Cryp71c: monster vs. player should take place at the called-function level. the code that calls those functions should be as agnostic as possible 14:52:39 probably it'd work something like spells - some of them need separate functions, some of them don't 14:53:23 spines were actually really easy to generalize, but triggering them was kind of nightmarish 14:53:54 yeah, i'm most concerned about making stuff trigger at the right times... it's easy enough to add new code to do something 14:53:58 Eronarn, so the do_hits(enum) would have a switch on enum to handle each way the function's actions could split. But on a specific effect (say, spiny) the effective code for calculating spiny damage and/or applying it should be agnostic, potentially lending to 1 function for the spiny effect for both players and monsters to share. 14:54:03 it doesn't help that monster and player stuff is even structured differently 14:54:03 the issue is doing it in a way that makes adding stuff later easier 14:54:40 Cryp71c: right; there might need to be storage of which effects have happened, also 14:55:03 for example, if you have a forcefield effect that blocks an attack outright, spines shouldn't trigger 14:55:09 Cryp71c: 1 function for both monster and player spiny effect worked really well when I tried it 14:55:14 Cryp71c: I mean shared, of course 14:55:50 i can probably help code this some time this week; i don't know if you intend to do it from scratch or want a hand with it 14:55:52 Eronarn, how agnostic is the attacker / defender logic? I remember it seeming like - if used correctly - you could use those paradigms and have a single function for use by both attacking players and attacking monsters, but in other circumstances I saw implementations that were clearly for one or the other (and even had hard-coded checks to see if attack was player, for example). 14:56:23 monqy, what did you base your effect on? Players use mutation levels, did you have monsters using HD? 14:56:40 Cryp71c: there's a lot of duplicated logic between player/monster; and then at a certain point the player moves over to using the monster code path 14:56:51 Cryp71c: I used a fixed value for monsters and left a comment saying that people with some sense of balance should edit it 14:56:59 so they have differences, but also a lot of similarities 14:57:25 using a hook system means that to some extent this can be ignored 14:57:39 Eronarn, that's a shame. It seemed like the way monsters and players inherit from actor class, there was lots of room for character (whether player or monster) agnostic code...and yet that potential was never really capitalized on. 14:57:40 both code paths have the points for the hooks to be attached 14:57:52 even if the code around those hooks is slightly different for player vs. monster 14:58:16 Eronarn, and by hooks, you're referring to calls to this do_hits() function, with various enum values? 14:59:39 Cryp71c: on agnostic code, there's a lot of code in fight.cc that is monster/player-specific for no reason at all, like before I modified it the spiny code converted actors to monsters to call member functions that all actors have 15:00:08 Cryp71c: yes 15:00:17 03greensnark * r253c36b556fb 10/crawl-ref/source/ (menu.cc menu.h stash.cc stash.h): [2930] Fix stash search results not being coloured correctly when matching item heaps. 15:07:10 Do you guys recommend an IDE that can auto-detect available functions and all that when working with the DCSS code? 15:08:25 i tried eclipse, it was pretty slow because of the size of the codebase 15:08:54 Eronarn, you just use a text editor w/ coloring, now? 15:10:16 03galehar * r3e8b1d49f20f 10/crawl-ref/source/tilereg-dgn.cc: Don't fire at friendly monsters when clicking on them. 15:15:06 -!- valrus has joined ##crawl-dev 15:22:14 Cryp71c: coloring is optional 15:22:24 well, ofc. :) I like working with it. 15:22:35 i mostly use nano; i sometimes use gedit 15:23:22 i have syntax highlighting in nano on on my desktop, not on on my server 15:42:05 -!- evilmike has joined ##crawl-dev 15:42:26 -!- Textmode has joined ##crawl-dev 15:43:26 -!- bhaak has quit [Ping timeout: 240 seconds] 15:43:42 -!- syllogism has quit [] 15:43:52 -!- xyblor is now known as battlemage 15:44:11 Eronarn, still there? 15:44:30 -!- Jordan7hm has joined ##crawl-dev 15:45:58 Well, anyone can answer this potentially, why do we constrain ourselves to 80-column lines of code, sometimes sacrificing readability for it? 15:47:49 it lets you open two pieces of code side by side on a typical screen at a sensible font size 15:53:32 ais523, seems odd, my resolution isn't great but I can fit two and have another half to 3/4 of a 3rd editor in extra space. 15:53:50 Was just curious, seemed like a severe restriction that yields fewer benefits than it lends. 15:54:12 I use a netbook 15:54:22 anyway, I'm not responsible for the restriction, just explaining its most common practical effect 15:55:06 I understand, thanks. 15:55:27 ais523, can you fit two pieces of code side by side on your netbook? 15:56:09 yep, I can fit in around 170 or so columns on the screen at my usual programming font size 15:58:57 it's not a "real" netbook, I get the impression it was meant to be a netbook but then beefed up so that it could manage to run Windows 7, just about 15:59:19 although I use Linux on it, which it can handle really easily by comparison (it visibly and audibly struggles when it has to run Windows) 16:01:02 03Cryp71c * r928664955e14 10/crawl-ref/source/fight.cc: Code cleanup 16:01:19 There are still many devices around that are limited to 80 character lines; plus, limiting windows to 80 characters makes it possible to have several windows side-by-side. The default wrapping on such devices disrupts the visual structure of the code, making it more difficult to understand. 16:01:34 So are we developing on gitorious now, instead of sourceforge? 16:01:51 Also, if you have a standard then you can have a single window set-up for all editing. 16:02:38 Cryp71c: yes 16:03:00 dpeg, so is that commit to sourceforge going to make its way to gitorious, or do I need to recommit it tomorrow after I get gitorious setup? 16:03:36 if gitorious isn't pulling from sourceforge, someone, either you or someone else, will have to push it 16:04:14 well, if I don't see it there tomorrow, I'll just push it myslef. 16:04:17 G'day! 16:04:18 -!- Cryp71c has quit [Quit: Leaving] 16:33:52 -!- pointless_ has joined ##crawl-dev 16:37:42 -!- battlemage is now known as xyblor 16:41:12 -!- galehar has joined ##crawl-dev 16:42:35 hi 16:42:35 galehar: You have 1 message. Use !messages to read it. 16:42:40 !messages 16:42:41 (1/1) dpeg said (4h 44m 45s ago): Would you lock the Og forum thread? 16:42:54 done :) 16:44:25 yes, thank you! 16:44:31 it was getting silly, wasn't it? 16:46:47 indeed 16:49:06 -!- golgepapaz has joined ##crawl-dev 16:53:33 this reminds me 16:53:35 !messages 16:53:35 (1/7) greensnark said (3w 1d 5h 10m 31s ago): Is the misc items page up-to-date or do you have ideas you haven't added there? 16:56:18 -!- golgepapaz_ has joined ##crawl-dev 16:57:11 -!- edlothiol has quit [Ping timeout: 255 seconds] 16:58:02 -!- golgepapaz has quit [Ping timeout: 240 seconds] 16:58:14 -!- golgepapaz_ is now known as golgepapaz 17:09:59 hii 17:14:32 03galehar * r48aafb78fca1 10/crawl-ref/source/ (mon-pathfind.cc mon-stuff.cc): Monster use pathfinding to go around clouds. 17:17:30 -!- golgepapaz has quit [Quit: ChatZilla 0.9.86 [Firefox 3.6.15/20110303024726]] 17:22:34 !tell greensnark Yes, the misc item page is good. I just went through it again. 17:22:34 dpeg: OK, I'll let greensnark know. 17:22:37 due: hi! 17:22:38 !messages 17:22:38 (1/6) LordSloth said (3w 10h 40m 6s ago): oh, two other things on the difficulty front. One, my perspective is skewed by slaying rings. And two, one factor that makes the early game seem more difficult is the extra scroll types. Some of the old reliables like teleport seem harder to come by 17:22:47 -!- golgepapaz has joined ##crawl-dev 17:22:47 !messages 17:22:48 (1/5) LordSloth said (3w 10h 37m 59s ago): and my old 2-stack identification strategy tends to result in seemingly fewer useful scrolls for early confrontations. The potion game is probably the same as always, though I do see more magic and agility potions than I'd like for heal wounds. Though I'd change my tune as anything other than MDHU. 17:23:02 !messages 17:23:02 (1/4) GreyKnight said (2w 4d 4h 19m 14s ago): hi 17:23:05 !messages 17:23:06 (1/3) kilobyte said (1w 4d 11h 57m 34s ago): XP is assigned proportionally to damage dealt. Piety is not (yet?), gains some monsters might get for killing neither. 17:23:36 kilobyte: around? 17:27:31 !messages 17:27:31 (1/2) kilobyte said (1w 4d 12h 56s ago): both Ce and Na bardings are capped at +4 enchantment, yeah. 17:27:35 !messages 17:27:35 (1/1) galehar said (1d 13h 28m 21s ago): We can add flags for forcing or preventing mimics in vaults. However, isn't it very spoilery? Should be used with care. 17:28:57 galehar: I was concerned with the currently existing fake naga guardian vaults. They look like ordinary treasure vaults, but part (or sometimes all) of their items are mimics. That would still work? 17:29:47 yes sure, we can keep it 17:30:33 kilobyte think it wouldn't be good to change item mimics (he commented on the wiki) 17:30:39 bedtime for me 17:30:40 bye 17:32:56 -!- galehar has quit [Remote host closed the connection] 17:34:35 hrm, he left 17:35:08 items have a problem of 90% of them being useless 17:36:48 yet it would be beneficial to poke every of them -- for xp and safety 17:37:03 kilobyte: yes, I agree 17:37:09 let's discuss that tomorrow 17:37:27 I think that item mimics should just not react until they can bite you 17:38:06 kilobyte: but I queried you for the ally kills: you write that xp is proportionally assigned... does that mean 0% player xp if the ally does all the work? 17:38:51 50%, as before 17:39:37 ok, so it interpolates between 50% and a little less than 100% then? that's good 17:40:08 right... unless there was some collateral damage from enemies, and then it can be less than 50% 17:40:19 ah! 17:40:57 -!- pointless_ has quit [Quit: ChatZilla 0.9.86 [Firefox 3.6.7/20100713130626]] 17:48:23 03greensnark 07stone_soup-0.8 * rbdfb3c9c657b 10/crawl-ref/source/ (menu.cc menu.h stash.cc stash.h): [2930] Fix stash search results not being coloured correctly when matching item heaps. 17:48:24 03galehar 07stone_soup-0.8 * r5a5a2aaa422a 10/crawl-ref/source/tilereg-dgn.cc: Don't fire at friendly monsters when clicking on them. 17:48:25 03dpeg 07stone_soup-0.8 * r43b1431fd19d 10/crawl-ref/source/dat/des/variable/mini_monsters.des: A few changes to Shiori's sword in stone vault. 17:48:49 greensnark ftw 17:49:02 also it got branched, 0.8 by april 1 right? 17:51:12 casmith789: the branching is jpeg's 17:51:34 but greensnark fixed that stash colour bug apparently 17:51:37 yes 17:52:11 uhm... branching right now? 17:52:14 not a good idea 17:52:47 there's a load of things not in order 17:53:09 and you can't even install on a number of locales without stupid tricks in trunk 18:12:07 !lm * type=crash start>=2011-03-01 18:12:08 37. [2011-03-15] evilmike the Pacifier (L5 OgHe) ASSERT(y >= 1 && y <= sz.y) in 'libutil.cc' at line 801 failed on turn 3904. (D:4) 18:12:12 !lm * type=crash start>=2011-02-01 18:12:12 112. [2011-03-15] evilmike the Pacifier (L5 OgHe) ASSERT(y >= 1 && y <= sz.y) in 'libutil.cc' at line 801 failed on turn 3904. (D:4) 18:13:07 more than two crashes a day 18:13:24 this is console, from what I see on Mantis in tiles things are far worse 18:14:01 oh, both of those crashes were the ely description thing 18:14:53 kilobyte: well, greensnark lobbied for branching 18:15:03 I didn't use s=, it mentions just the very last crash 18:15:08 kilobyte: what is wrong with it? jpeg can just cherrypick all the fixes. 18:16:01 if I finish Unicode (which includes all the fixes for installing on Russian/Vietnamese/Korean), it's not really cherry-pickable without LOTS of merging work 18:17:43 telling people to manually unpack a zip file to a directory in the root of a partition is not really a good idea 18:18:44 on XP this was ugly but used to work... on Windows Vista/7 the system will prevent you from doing that unless you disable security features 18:21:30 kilobyte: hm, what now? 18:21:41 so far the branch only has three commits 18:21:53 you can send a mail to c-r-d an explain what's going on 18:22:03 ah, you did :) 18:22:07 -!- jpeg has joined ##crawl-dev 18:22:24 I could rush it, but we'd have to disable CDO updates unless I get DGL support soon enough 18:22:35 kilobyte: wait, didn't you say that you were going to postpone the unicode thing? 18:23:02 postpone until close to release since it needs only bug testing, not balance testing 18:23:33 oooh... a misunderstanding then 18:23:57 then again, we didn't have it in 0.7 - what changed? 18:24:15 and balance of new 0.8 stuff is not in a good shape, with all that "you don't need to train Hexes at all", Tornado being sometimes bad sometimes insane, Troves being an insult in most cases 18:24:45 telling people to manually unpack a zip file to a directory in the root of a partition is not really a good idea 18:24:51 on XP this was ugly but used to work... on Windows Vista/7 the system will prevent you from doing that unless you disable security features 18:25:30 (ie, the workaround is not really fine by no fault on our side) 18:26:55 er, no longer fine 18:27:13 troves are in a better state than for 0.7, nothing to delay release for 18:27:19 trunk users are mostly technically minded 18:28:25 kilobyte: I have been sending the 0.8 mails around... we mentioned troves and checked them off. We thought that Tornado balancing is a matter of number tweaking and that you have an idea about it. 18:28:28 troves: not a reason to delay a release for, but all troves I've seen recently are worth far less than the price 18:28:45 one thing that'd be nice: if tornado gave you DMsl while it was up 18:28:47 -!- upsy has quit [Read error: Connection reset by peer] 18:28:55 (but it should give everything in the area DMSl, too!) 18:32:10 -!- upsy has joined ##crawl-dev 18:32:17 For me, I saw the manual, baileys and ashenzari as my homework, and I tried to tick them off (baileys still not done, will come). What now? Revert Ench split? 18:33:09 hmm, is it any more unbalanced now than it was before? "Hexes are useless" vs. "Enchantments is too good" seem equally large balance problems 18:33:20 (also, is Charms currently a balanced school?) 18:33:33 ais523: strictly weaker than Enchantments were 18:33:37 ais523: no, but a bit weaker than before it is 18:33:51 charms/hexes is better than enchantments; the problem at this point is mostly some horrible school assignments 18:33:57 and the starting books 18:34:01 with the Haste nerf, removal of Extension and so on, I'd say they're fine 18:34:08 also some issues with overall playstyle capability of going hexes or charms 18:34:10 the main problems are the Cr/AM books and the horrible awful victory dancing En book, i'd say 18:34:14 oh right, books are bad too 18:34:38 MarvinPA: do you have an idea how to improve that? (I don't.) 18:34:45 MarvinPA: well Cr is kind of horrible awful victory dancing too now that you're working for silence 18:35:01 hmm, for En i'd say make invis pure hex i guess 18:35:08 invis should definitely be pure hex 18:35:25 reverting Ench split would be too much work, status quo is bad but not terrible 18:35:47 MarvinPA: sounds good to me 18:35:52 I'd disable buggy stuff like Tukima's Ball (and perhaps Terpsichore too) 18:35:52 (flavour and gameplay) 18:35:54 invis -> hex, brands -> charms 18:35:56 imo 18:35:57 for Cr brands could be charms but then yeah, that causes problems with silence 18:36:05 TGWi: and then fix AM? 18:36:05 silence -> haste again 18:36:09 invis should possibly be renamed to aura of blindness or something 18:36:12 yeah, agreeing with TGW here 18:36:15 there's not really any split that salvages AM 18:36:17 which comes to the same thing, but is more clearly flavourwise a hex 18:36:26 i have ideas for redoing AM, they can wait until .9 though 18:36:31 you could do some last-minute book repair 18:36:32 idk 18:36:39 we can disable AM for 0.8, no big deal 18:36:40 not many people do AM anyways so it's not a big loss to disable them for a version 18:36:41 silence is a fitting spell for crusaders though, and i like them not having haste in their starting book 18:36:51 MarvinPA: yes 18:37:04 disablinging Tukima's is fine, Terpsichore probably follows from that 18:37:10 it could be charms/air i suppose, since you cast it on yourself 18:37:15 ugh 18:37:15 no 18:37:16 bad 18:37:25 jpeg: what about this: could you please drop the branch for now, I'll rush and we'd branch, say, late this weekend? 18:37:31 what's in maledictions, now? 18:37:32 just have it be haste, they've had haste forever 18:37:41 remove haste 18:37:42 nobody else even gets haste in their starting book 18:37:49 we'd go on a disable-fest then 18:37:58 monqy: same minus silence iirc 18:38:08 which is good, it was too packed anyways 18:38:47 I'm not mega-opposed to silence in war chants, but 1) haste in war chants is not the end of the world 2) no silence in war chants is not the end of the world 3) charm brands is pretty important imo 18:38:57 you could replace it with some other spell 18:39:03 i think invis -> hex, silence and brands -> charms, temporarily disable AM would be fine, anyway 18:39:20 surely there are other mid-level charms 18:39:23 MarvinPA: you have free reign 18:39:28 i still say split the brands; distortion/pain/draining brand aren't flavorwise like fire/ice 18:39:30 discuss here and then decide, please 18:39:41 kilobyte: sure, it's not _that_ urgent 18:39:42 if you *really really* don't want haste in war chants, at least 18:39:43 brands don't fit among Charms the slightest 18:39:48 kilobyte: is this still about the unicode branch? 18:39:57 kilobyte: that's only because the flavour changed so that brands could go in hexes 18:40:09 jpeg: mostly... other fixes are minor 18:40:20 MarvinPA: the reasoning behind brands -> charms is about the starting background? 18:40:31 it's absurd to say that self-enhancements don't fit in the self-enhancement school "the slightest" 18:40:33 "charms affect yourself, hexes affect other things" is the flavour that seems to fit the current split best 18:40:38 like the new targetter, it has wrong messages in most places 18:40:44 ais523: yeah, currently 18:40:52 but you can easily change it to "charms beneficial, hexes detrimental", which moves mostly brands and nothing else 18:40:53 ais523: except that 'flavor', if you can even call it that, sucks and is pointlessly constraining on spell effects 18:40:56 but "positive effects vs negative effects" is almost identical 18:40:58 dpeg: mostly as if would fix crusaders, they were charms in the initial split too and the flavour fit just fine i think 18:41:00 target a wall? "Out of range". etc... 18:41:01 except for charms 18:41:03 it would fix* 18:41:04 Eronarn: the point is that you can fit flavour round most things 18:41:06 kilobyte: I'm worried about the large number of (crash) bugs, too, but we can't postpone forever 18:41:07 ais523: yeah, basically exactly what you said 18:41:25 I think the CDO crashes are all #1184 18:41:30 I've been trying to retroactively flavour some of the more bizarre bugs in NetHack 18:41:32 crashers are _typically_ easy to cherry-pick 18:41:34 ais523 for president 18:41:39 crusaders are named crusaders because they're charming and well-liked, just like paladins 18:41:48 Eronarn: :P 18:41:55 MarvinPA: I believe that the spell schools are more important than the background. I'd rather sacrifice Cr and keep Hex/Charms good for now (and currently, brands seem to fit Hexes better, speaking only about power here) 18:41:56 !lm * type=crash 18:41:57 674. [2011-03-15] tartakower the Prestidigitator (L16 MuNe) ASSERT(in bounds(mons->target) || mons->target.origin()) in 'mon-act.cc' at line 2377 failed on turn 56597. (Lair:8) 18:41:59 !lm * type=crash -2 18:41:59 673. [2011-03-15] evilmike the Pacifier (L5 OgHe) ASSERT(y >= 1 && y <= sz.y) in 'libutil.cc' at line 801 failed on turn 3904. (D:4) 18:42:00 !lm * type=crash -3 18:42:00 672. [2011-03-14] megabat the Spry (L21 DDNe) ASSERT(!crawl state.is repeating cmd()) in 'main.cc' at line 4139 failed on turn 72144. (Crypt:5) 18:42:01 things like garter snakes hiding under their own corpses are easily flavoured as "you see a garter snake playing dead", for instance 18:42:08 dpeg: the brand spells aren't powerful except for Cr though 18:42:14 and they're almost all low level 18:42:21 dpeg: brands aren't a major part of why charms are good 18:42:27 (okay, Cr and AM but we're probably losing the latter) 18:42:28 I've found brand spells useful as an early pure-spellcaster trying to learn to fight 18:42:29 yes, but a small one 18:42:30 but they *are* very important to cr 18:42:55 e.g. my current Ne was using drain brand or whatever it's actually called to beat uniques 18:42:58 We should rename Cr anyway... not for 0.8, we can just disable along with AM. After that, we build a new Hex background, complete with new name and brands. 18:42:58 a better target would be invis to pure hex, because it's fitting in terms of playstyle 18:43:00 kilobyte: ah, haven't seen those ones yet 18:43:08 dpeg: as for power: hexes *needs* partial resist on MR IMO 18:43:11 dpeg: "witch" for a hexer 18:43:15 I'm strongly opposed to ditching Cr 18:43:22 the all-or-nothing on most of its spells is killing it 18:43:27 TGWi: I ignore lines with "strongly" opposed :) 18:43:43 dpeg: invis, and also rmsl/dmsl 18:43:44 Eronarn: yes, we can move the old Ench->MR effect to Hexes 18:43:50 a bit of crutch, but what the heck 18:43:52 which are both non-playstyle spells but strong 18:43:55 dpeg: no... that's not what i mean 18:43:55 MarvinPA: still with us? 18:44:02 Eronarn: something I was playing with mentally was "repeating a hex gives it a progressively better chance of piercing MR"; I think it's an interesting idea but probably wouldn't fit Crawl 18:44:03 dpeg: Ench->MR is no more 18:44:13 kilobyte: I know, I thought Eronarn wanted it back on Hexes 18:44:15 kilobyte: I think dpeg meant resurrecting it linked to Hexes 18:44:17 ah 18:44:23 ok, would make sense 18:44:26 dpeg: i mean monsters shrugging off effects totally on their MR check. For example: Confusion = confuse on pierce their MR, daze if you come close to piercing their MR, otherwise nothing 18:44:27 which would certainly make flavour sense and also buff it 18:44:46 ais523: someone on the forums suggested something along those lines too 18:44:47 but there's a difference between buffing something weak, and buffing it in a sane way 18:44:51 right now, confuse is either render them totally helpless, or it does nothing 18:44:54 not sure if it has potential, but maybe 18:44:56 I think that might count as sane, but I'm not sure 18:44:59 I don't like tying extra stuff to skills but maybe I'm crazy 18:45:07 monqy: i agree, i'm against hexes giving +MR 18:45:18 Eronarn: I'd like to see a sort of "slightly confused" status, like NetHack confusion (Crawl confusion = NetHack stunning) 18:45:38 ais523: i added a Daze status, it makes you lose 1 in 5 actions (but not attack self, move around, etc.) 18:45:45 Eronarn: yes, that should come, but we won't even touch that for 0.8 18:45:48 That's dazing? 18:45:52 note that player confusion is helluva milder than monster confusion 18:46:06 kilobyte: mostly because players own !oHealing 18:46:19 the effects are actually quite similar if the players are as stupid as the monsters, just they aren't 18:46:25 ais523: also self-attacking, walking into lava 18:46:31 yep 18:46:32 dpeg: i don't think we're going to get hexes working well without it. before, if you were a pure enchanter, you would be ok because as your spells started to peter out you'd get haste 18:46:36 not only... monsters move totally randomly, players have only a small (1/3?) chance of that 18:46:37 also things that players don't do when confused 18:46:38 and so it was worth training 18:46:39 Eronarn: of course we do 18:46:41 now, this is not the case 18:46:55 Hexes are not garbage, they're just not as powerful as Charms 18:47:05 eronarn: uh, enchanters are still good 18:47:06 stay on the carpet, folks (that's a Germanism) 18:47:08 self-attacking could be removed outright, I don't get the point in it really 18:47:08 if you stop training hexes when hexes gets worse, hexes gets a *lot* worse than if you, say, stopped training conjurations 18:47:31 kilobyte: galehar likes it :) 18:47:37 the problem with hexes is simply that most chars aren't interested in them 18:47:39 (assuming you talk about the monsters) 18:47:47 not that chars who like hexes are too weak 18:47:57 hexes are fine but only stabbers really want them, whereas most casters want some charms skill, yeah 18:47:58 elliptic: yes, so the school is more backwater for now... but other Crawl versions had other backwater schools 18:48:05 it is not a big a deal as claimed here 18:48:13 dpeg: yeah, I don't think it is a huge problem either 18:48:30 it wouldn't destroy the game even if there was a spell school that was completely unusable, except as a trap for newbies 18:48:31 not every caster wants tmut skill either but some do, adding more varied hexes later on will help a lot too 18:48:39 Divinations? 18:48:42 MarvinPA: okay, let's make a plan: Invis goes Hex, brands stay Hex. Your choice whether that invalidates Cr (disable then). Disable AM in any case. Okay? 18:48:44 MarvinPA: except remove darkness because ughh 18:49:11 dpeg: that sounds good to me, sure 18:49:12 yeah, hexes isn't that dissimilar to tmut 18:49:25 dpeg: imo fire/ice brand to charms, other brands stay hexes (because they use necromantic or evil or poison magic) 18:49:41 eronarn: you are crazy, all brands should be the same school 18:49:42 MarvinPA: Eronarn's proposal is also fine with me, your choice. 18:49:47 nec/hexes is way more flavorful sounding than nec/charms 18:49:55 and the charms don't have to be together any more than anything else does 18:49:57 could save Cr, but think for yourself =) 18:50:08 they function similarly mechanically, but they aren't even found in the same book with each other 18:50:35 kilobyte: okay, Tornado next. Do you have a plan or should we collaborate? 18:50:44 any thoughts on sure blade? pure hex if the brands stay as hexes? 18:50:46 is there a balance reason why fire/ice brand affect the weapon rather than the player? 18:50:54 uhm, one brand being in Charms and another in Hexes would drop any sense from the split 18:50:57 (I understand the flavour reasons) 18:51:03 kilobyte: no it wouldn't 18:51:12 do you think wizards go around talking about 'brand' spells 18:51:25 dpeg: a good part of the problem is that you can still cast other spells in that time 18:51:27 crawl didn't fall apart when we had airstrike in Air instead of Air/Conj 18:51:27 a buff that makes your attacks catch fire is different in nature from a poisoning spell that's delivered via weapon 18:51:41 those are just two entirely different flavourings of brand spells 18:51:46 you could even rename them to draw the distinction 18:51:54 Weapon brands already exist and make sense to people, brand spells imitate them. 18:51:56 the point is that you can match flavour to the mechanics 18:52:00 -!- edlothiol has joined ##crawl-dev 18:52:01 other _air_ spells are damn weak or don't work with Tornado right, but a Ziggurat char can pump other skills 18:52:02 dpeg, kilobyte: I don't think there's anything terribly wrong with tornado as is... very strong, but not necessarily better than fire storm, say 18:52:05 so getting the mechanics right is more important 18:52:09 elliptic: the schools are wrong 18:52:11 people aren't using it that much 18:52:11 and, say, firestorm in the meantime 18:52:15 kilobyte: dmsl? that's air 18:53:00 ais523: it doesn't do any damage... sorry for being unclear 18:53:05 kilobyte: I value elliptic's opinion highly, if you want a nerf, you could say that Tornado hampers all elements. 18:53:06 ah 18:53:16 ais523: mechanically: it works really well to have fire/ice in Charms for both crusader and the other few books they're in (also founnd with fire/ice charms) 18:53:26 dpeg: as in, a negative spell enhancer? 18:53:41 sounds like a good idea 18:53:42 well, it could globally affect magic 18:53:47 I don't see people using tornado with other elemental spells at the same time 18:53:48 you have to shout over the storm 18:53:54 that soudns like the sort of thing that's dangerously reverse-abusable 18:54:01 unless they are highly scummed-out zig chars 18:54:09 and zig balance shouldn't be a big deal 18:54:12 kilobyte: yes, just a wild idea 18:54:13 not to underthink this but what about removing conj and lowering the damage? 18:54:18 if you want to worry about tornado, worry about how good it is on zot:5 18:54:32 TGWi: replace conj with hex 18:54:36 TGWi: reskilling, victory dancing :( 18:54:40 TGWi: that's what i'm in favor of. though i'd also like to see it DMsl vs. physical projectiles for everything in the area 18:54:58 kilobyte: you can reskill and victory dance two schools, too 18:55:04 it just takes longer 18:55:26 it's pretty silly to say that we can't have tornado a L9 pure air when we just moved shatter to L9 pure earth 18:55:29 -!- jpeg has quit [] 18:55:37 zot:5 is very open so tornado is reliably really strong there if you get it castable in time 18:55:39 Eronarn: one of my ideas is to shift all targetting randomly by +/- 30โฐ 18:55:40 I would like to say that the storms aren't pinnacles of balance either... we don't have to be more critical than there 18:55:50 elliptic: just like with storms, right? 18:56:05 Eronarn: Shatter is weeeak 18:56:10 kilobyte: that'd be interesting... though it shouldn't affect rays and bolts and such (so you can fire lightning bolts out of a tornado) 18:56:17 dpeg: well, it is stronger than the storms in really open spaces like zot:5 18:56:20 kilobyte: we can make tornado weak, too 18:56:25 but yeah, fire storm is basically easymode for the rest of the game 18:56:29 weak, pure air tornado >>>>>> strong, air/conj tornado 18:56:29 in any sort of space 18:56:34 elliptic: Air is supposed to be the school for open spaces, so that's good 18:56:47 it's also the school of closed spaces 18:56:54 if you wanted a tornado nerf, make it check AC somehow 18:57:01 at least conj/air is 18:57:07 it doesn't already? 18:57:09 isn't air the school of map-layout-is-relevant? 18:57:17 kilobyte: I believe that since Air is the level 9 spell, it's just not that crucial. 18:57:35 It should be good enough to see use. Easier to get feedback for overpowered features and then to nerg. 18:57:38 *nerf 18:57:46 ais523: seems like it 18:57:47 airstrike checks AC, and tornado doesn't seem like it should be so different... I understand that the damage mechanics makes this a little tricky to implement, but AC should matter for it IMO 18:58:00 elliptic: being tossed around in armour is _worse_ so AC should increase the damage you get :p (we could decrease base in that case) 18:58:06 yeah if tornado isn't checking AC that is wrong imo 18:58:17 kilobyte: why is being tossed around in armour worse? 18:58:19 kilobyte: that argument would work better if tornado actually moved monsters around :P 18:58:39 also, I'm thinking of tossing orbs of fire around 18:58:54 Eronarn: depends on armour type, I guess. Leather would help, plate would make things worse. 18:58:59 we don't want elaborate changes for 0.8, we want balancing tweaks 18:59:03 keep it simple, folks 18:59:10 currently tornado ignores the 20/20 AC/EV that orbs of fire has, and all of their resistances 18:59:23 elliptic: yeah, that's something it is supposed to do but I never got around to finishing that yet :( 18:59:32 just make the damage work like airstrike, which is already a known to be fine spell 18:59:52 it seems like it could check something there and not just ignore every possible type of defensive attribute 18:59:59 idea: we can nerf Tornado for 0.8 and apply the elaborate ways for 0.9 19:00:07 kilobyte: yes! 19:00:18 so it'd be a weak spell for 0.8, a fun one later 19:00:34 it won't see much use in 0.8 if it is a weak L9 spell 19:00:36 tornado => air, checks AC, maybe tone down the damage a bit 19:00:39 see: shatter 19:00:45 (before the recent buff) 19:00:52 air with an AC check seems good 19:01:01 shatter is utterly meh 19:01:21 kilobyte: but what was it before boost? :) 19:01:24 shatter is great against non-flyers 19:01:25 shatter at L9 earth is reasonably balanced 19:01:32 I wouldn't call it meh now 19:01:56 admittedly that excludes fiends and lots of pan lords but considering that you can get it castable for zot or v:8 quite easily, it's pretty nice 19:01:57 it doesn't win the game for you like firestorm or tornado, but it is also much easier to get 19:02:04 that reminds me, should tornado work against incorporeals? 19:02:08 and it is a *very* fun spell 19:02:20 Eronarn: you'd expect it to bash them against the ceiling or something 19:02:27 and it's up to flavour whether that hurts them or not 19:02:48 ais523: i actually wouldn't expect a ghost to get blown around by wind much at all 19:03:14 I would 19:03:33 hmm, that's the sort of question which would be well asked on the Tavern 19:03:40 "flavourwise, would you expect a ghost to be blown around by wind?" 19:03:48 sounds like a bad idea, getting tavern input 19:03:55 empty question, so we can make up a solution 19:04:02 anyways, just idle curiosity 19:04:09 important part: tornado to pure air, make it check AC 19:04:27 the dmsl would be nice but it could probably wait 19:04:32 kilobyte: fine with you? 19:04:40 fancy stuff like moving monsters around for later 19:07:45 -!- casmith789 has quit [Remote host closed the connection] 19:08:06 -!- aze5 has quit [Quit: ChatZilla 0.9.86 [Firefox 3.6.15/20110303170030]] 19:08:18 not fine with checking AC, pure Air and (for now) weaker damage is cool 19:08:30 trivializing high-EV low-hp monsters is not good 19:08:37 -!- ais523 has quit [Remote host closed the connection] 19:08:46 kilobyte: okay! 19:09:03 should it be a negative spell enhancer, too? 19:09:13 so we have settled Hexes, Tornado... I will revert the trove shrinking commit, once I find out how to do that 19:09:21 (we'd replace it with targetting troubles later) 19:09:31 kilobyte: I have no idea -- your choice! 19:09:35 why not have it check AC, kilo? 19:10:54 kilobyte: will you tell jpeg once branching is acceptable again? The hexes, tornado, trove changes will be small enough to be cherrypickable 19:12:55 I don't think it needs three nerfs 19:13:26 ekiM: three? pure air is a buff 19:13:46 I heard spells, Ac check, and lower damage. 19:13:56 dpeg: sure... I ponder even asking Napking to block CDO updates, or blocking them myself unless I get charset conversions in DGL. 19:14:01 I think pure air and lower damage is more than enough. 19:23:37 hrm, needed three casts to kill a shedu 19:24:00 (with damage reduced to 72%) 19:28:34 03kilobyte * r7c86c88bf02b 10/crawl-ref/source/ (4 files in 2 dirs): Tornado changes for 0.8: pure Air, 80% damage, negative spell enhancer. 19:30:00 what about Ogres? We could do some simple changes to not have another release with a bad joke race... We can then spend another year or two arguing how to fix them better. 19:31:19 is TGWi's hp formula any good 19:31:22 ogres seem fine to me, fighting gain depending on racial hp would distinguish them more too 19:31:29 it rxo 19:31:30 roxc 19:31:31 rox 19:31:36 sorry about that 19:31:38 that might be too much of a change for 0.8 though, dunno 19:31:40 it doesn't rxo or roxc 19:32:05 it would improve trolls, who don't need a buff 19:32:17 make them even worse at fighting then 19:32:54 would def not recommend it for 0.8 though 19:33:05 18 as denominator might not be the correct value 19:33:19 TGWi: this is what I mean, let's touch ogres only for now 19:33:54 aren't ogres pretty fine 19:34:14 i'd say so 19:34:24 they're extremely weak 19:34:27 if that's what you meant 19:34:32 no immediate changes needed, I meant 19:34:38 question is whether that's alright 19:34:48 they're perfectly good pure casters 19:35:03 I hear they're good berserkers too, though I haven't tried it 19:35:18 MarvinPA: what are you going to do with cr brands? 19:35:24 everything is a good berserker 19:35:27 that's because berserker is good 19:35:53 monqy: ogres are good berserkers if you s/Og/Tr/ 19:36:04 early game is mostly about your class, and early game is the most lethal part 19:36:09 so yes berserkers are good 19:36:49 trolls are nearly strictly better than ogres for anything but conjuring... and neither race is any good at that 19:36:59 ogres are fine at magic 19:37:17 admittedly, I prefer them as summoners and transmuters 19:37:44 ogres are fine 19:38:01 slightly worse apts? No problem. Trolls do enough damage, both AC and EV is negligible so it doesn't matter if it's 0.01 * (troll skill) or 0.01 * (ogre skill), less Fighting doesn't matter if it's more than offset by better base hp. 19:38:42 kilobyte: nobody is arguing that TrBe is better than OgBe afaict 19:38:57 elliptic: so let's fix that! 19:39:16 what's wrong with TrBe being better than OgBe? 19:39:17 why shouldn't TrBe be better than OgBe? i'm confused 19:39:21 yeah 19:39:37 !apt tr 19:39:38 Tr: Air: -4*, Armour: -2, Axes: -2, Bows: -4*, Charms: -4*, Conj: -3*, Xbows: -4*, Dodge: -2, Earth: -1, Evo: -3*, Exp: 150, Fighting: -2*, Fire: -3*, Hexes: -4*, Ice: -3*, Inv: -2*, Long: -2, Maces: -1, Nec: -2, Poison: -3*, Polearms: -2, Shields: -2, Short: -2, Slings: -4*, Splcast: -4*, Stab: -2, Staves: -2, Stealth: -5*, Summ: -3*, Throw: -1, Tloc: -3*, Tmut: -3, Traps: -4*, Unarmed: 0 19:39:39 !apt og 19:39:39 Og: Air: -3, Armour: -2, Axes: -3*, Bows: -3, Charms: -3, Conj: -3*, Xbows: -3, Dodge: -1, Earth: -3*, Evo: -2, Exp: 110, Fighting: 2!, Fire: -3*, Hexes: -3, Ice: -3*, Inv: 0, Long: -3*, Maces: 1, Nec: -3*, Poison: -3*, Polearms: 0, Shields: -1, Short: -4*, Slings: -3, Splcast: 2, Stab: -2, Staves: -1, Stealth: -2, Summ: -3*, Throw: 1, Tloc: -3*, Tmut: -3, Traps: -2, Unarmed: -1 19:40:14 heck, if you don't agree with bumping base hp, even Fighting 5 Maces 5 wouldn't be out of whack 19:40:36 read as "humping base hp" 19:40:38 was very confused 19:40:42 just to be clear, you want to buff ogre melee? 19:41:07 nothing wrong with that, though +5 apt sounds excessive 19:41:17 how good are giant spiked clubs 19:41:34 slightly better than dwhips with more hands 19:41:37 I'd want to buff ogres wholesale, but if you disagree with a general bump, it's not the time for controversial changes 19:42:38 22 damage 7 delay vs 12 damage 5 delay... not "slightly" IMO, unless your skill is low 19:43:26 no brands unless you do Trog... in which case you get a vamp/speed spiked club by Lair end 19:43:48 (oh sorry, vamp is for Tr not Og again...) 19:44:04 (Tr doesn't use maces) 19:44:05 tr is better off unarmed anyway (because UC damage is insane) 19:44:27 an even siller thing: it's ogres not trolls who have food issues 19:44:59 ogres don't have food issues really because they can eat rotten chunks 19:45:15 that compensates for the fast metab 19:45:33 more food issues than Tr but still nothing significant 19:46:15 you'd have to calculate your nutrition level, and berserkers get hungry in most inopportune moments 19:46:22 -!- Mu_ has quit [Quit: Defecator, may everything turn out okay so that you can leave this place.] 19:46:28 kilobyte: mainly I guess I just don't think ogres are that weak... a couple minor apt buffs sounds fine but also I don't see any need for them 19:46:39 kilobyte: most ogres aren't berserkers 19:46:40 berserk and food is insane... 19:47:21 being merely "angry" will berserk you at starving, yet drugs, spells or divine powers are blocked at hungry 19:47:57 the spell isn't blocked at hungry 19:48:02 i don't think potions are either? 19:48:04 idea: what if we allowed berserk at anything better than "very hungry"? 19:48:25 kilobyte: just give enough room so that people can't kill themselves by berserk hunger 19:48:30 I don't know how much that is 19:48:41 also this would be a trog buff and he really doesn't need it 19:49:22 I agree that trog refusing to berserk you at hungry is strange given that the other effects do, but there are a lot of balance ramifications 19:50:08 MarvinPA: just checked: spell works, potions don't 19:50:16 ah okay 19:50:19 yeah, it is frightfully inconsistent 19:50:22 potions should prolly 19:50:32 either way some more consistency would probably be a good thing, agreed 19:51:08 but instakilling people with starvation is bad too... 19:51:17 elliptic: it's not about power, it's about annoyances like "plenty of food, but I have to wait until this specific moment to engage so I get hungry just at this point" 19:52:38 -!- Mu_ has joined ##crawl-dev 19:52:54 well, your change would be a clear buff for most berserkers early-game 19:53:02 maybe it doesn't matter because their early game is already easy 19:53:25 please leave ogres untouched for 0.8 19:53:48 we should not let loud complainers dictate our actions 19:53:54 can be trivially offset by making berserk speedup x1.8 or x1.9 19:54:43 well, that would be a clear nerf for berserkers late-game :P 19:55:29 sorry :p 19:56:02 current state of affairs may be good or bad, depending on person, but there's no bug 19:56:33 dpeg: I seriously ponder reverting Heroism for 0.8. The Unarmed Combat formula is so broken the ability boosts it a lot (while being a nerf as intended for other weapons). 19:56:48 kilobyte: reverting to Might? 19:57:00 and we don't want everyone unarmed to go Oka 19:57:02 yeah 19:57:08 of course, I never play Okawaru, so I have to rely on hearsay 19:57:14 heroism with unarmed is a bit overrated 19:57:32 don't revert heroism 19:57:38 imo 19:57:48 not as good as it was with +10 skill, but since +10 skill was still a nerf for weapons... 19:57:51 people say "wow 5 extra base damage!!!" but don't realize that unarmed base damage is very different from weapon base damage 19:57:53 it's very unlikely "everyone" will go UC 19:58:02 might with unarmed was very good also 19:58:18 kilobyte: one reason for not reverting is that we will get more feedback and incentive to do something UC (that's a meta reason, but still) 19:58:25 TGWi: sure, what I meant is most UC chars seem to prefer Okawaru 19:58:34 kilobyte: they already did 19:58:35 which is likely a fad 19:58:43 might + auxes was really good also 19:58:49 meh, nothing wrong with makhleb or whoever for UC 19:58:54 kilobyte: yes, the groupthink is insane, probably because there's only ##crawl 19:59:11 nothing is broken imo, just try to fix UC in 0.9 19:59:14 if you want 19:59:23 TGWi: there isn't even a plan, afaik 19:59:26 if you want to do something about heroism, you could make the piety cost a little higher 19:59:54 TGWi: right, UC formula is definitely not 0.8 material, I'm just afraid of UC+Oka 20:00:05 dpeg: there are some ideas I think but not concrete formulas 20:00:12 eronarn has something I'm sure 20:00:24 Heroism takes 2 piety, Finesse 5. 20:00:37 grappling/throws for UC 20:01:03 grappling is planned anyway, at least 20:01:06 I thought heroism was 1d2, the same that might was? 20:01:17 are manouevres (sp?) planned for 0.9? 20:01:34 maybe it was 1d3 and you mean the average rather than the minimum 20:01:35 manoeuvres 20:01:42 -!- golgepapaz has quit [Quit: ChatZilla 0.9.86 [Firefox 3.6.15/20110303024726]] 20:03:47 elliptic: already bumped to 2 20:03:57 oh, good 20:05:09 TGWi: I'd hate anything that makes combat less straightforward. Fun pseudo-brands: yeah. Advanced positioning: detrimental. 20:05:59 how is less-straightforward combat detrimental 20:06:02 i have no ideas, never, never ever 20:06:04 I don't get this 20:06:10 Eronarn: sorry I confused you with someone else 20:06:33 kilobyte: would you object to the sidestep mechanic? 20:06:52 dpeg: not sure which of many proposals you mean 20:06:54 kilobyte: grappling is possible to make pretty simple 20:06:58 advanced positioning is fine as long as you don't have to do it to be good with a certain weapon type 20:07:10 there should always be a less-involved and more-involved playstyle 20:07:19 like, lock things in place when you attack sometimes 20:07:28 and automatically break if you do something else 20:08:03 TGWi: that's what I call a "pseudo-brand": you hit things and get benefits from that, but don't have to think of details 20:08:08 throws could be as simple as "things are occasionally delayed" 20:08:13 like, reaching is tedious 20:08:20 yeah def. not reaching 20:08:34 Chance of (Dex-14)/Dex that attacks are targeted at the square you occupied last turn, instead of this turn. 20:08:56 kilobyte: but only because of the interface 20:09:04 sure, semi-paralysing an opponent for a (fraction of a) turn is fine 20:09:20 I'm thinking like 20:09:29 one turn or so 20:09:29 I have proposed a better interface for reaching. The brand itself is interesting. 20:09:43 and grappling prevents them from moving and hampers their attacks or something 20:09:49 dpeg: this is sadly something that would make combat tedious :( 20:09:55 the new built-in autofight helps reaching, i guess! 20:09:58 even though realistic and interesting 20:10:13 MarvinPA: hell yeah... I about never use reaching without it! 20:10:16 what's the interface proposal? 20:10:26 it's a cool idea in theory and it would be nice if it worked well 20:10:34 (reaching) 20:11:06 reaching is cool in practice too 20:11:22 my idea was to use Shift-direction for the main (eight) directions 20:11:30 but autofight will also help 20:11:38 that could help yeah 20:12:07 you could change it so that regular attacks are identical a lot of the time 20:12:10 with the mouse it is trivial anyway 20:12:20 TGWi: yes 20:12:26 like, instead of "reaching past" you get a chance to hit things behind other things 20:12:35 whether you v or just attack the adjacent guy 20:12:35 reaching has balance problems too: water/lava scum and swiftness/butterfly kiting 20:12:52 kilobyte: not more problems than a spell 20:13:02 ban all ranged attacks!! 20:13:03 like airstrike 20:13:05 dpeg: hmm, good point 20:13:45 all spells (but freeze/new sticky flame) allow kiting 20:13:52 -!- upsy has quit [Read error: Connection reset by peer] 20:13:53 water/lava can be taken care of with AI (if you reach and they cannot attack you, they should retreat) 20:13:55 new sticky flame allows kiting 20:14:01 passive "piercing" or something for polearms would reduce the need for explicit evocation of reaching 20:14:28 TGWi: the fact that you can attack over an ally is important 20:14:34 hmm 20:14:36 Eronarn: I meant doing it with lava... of course every FE backpedals after casting most of the time 20:14:48 TGWi: one nice thing about our weapon system is that polearms includes everything from spears to giant choppy things 20:14:52 also, if there's only one enemy near the direction you shift-press, dpeg, it should choose automatically 20:14:56 so there'll be weapon style versatility 20:14:59 TGWi: yes 20:15:07 -!- upsy has joined ##crawl-dev 20:15:13 so it can attack things a knight-jump away if it's clear what you mean 20:15:19 also, fr: knights 20:15:21 and unicorns 20:15:34 TGWi: I have a plan for unicorns 20:15:38 a two-press thing might be interesting 20:15:41 kilobyte: for dinner? 20:15:50 two presses could also work 20:16:02 first one moves two spaces, second one moves one 20:16:15 like: ctrl, -> = hits any of the three reachable squares on that side (defaults center one if there's more than one valid) 20:16:49 dpeg: no. With blood-shot eyes, bloodstained horns... oh, and demonic holiness, too. Not named "unicorn" but "demon horse" too but otherwise identical. Ridden by hell knights the way of spriggan firefly riders. 20:17:15 kilobyte: mountain branch 20:17:15 do it 20:17:17 mounted charge, that kind of stuff 20:17:18 kilobyte: ah, good. No virgins. 20:17:40 TGWi: forest, dwarf, hive. A new branch is a lot of work. 20:17:57 :( 20:18:09 It's a fact. Check the Shoal's history. 20:18:14 we should have had employed huldras as foocubi :( 20:18:25 no sex please 20:18:34 only rape 20:18:40 gore, madness, self-deprecation = good 20:18:40 unicorn => stripeless monozebra 20:18:42 and only in demonology description 20:18:50 Eronarn: lol 20:18:51 !learn add dpeg <@dpeg> no sex please 20:18:51 dpeg[4/4]: <@dpeg> no sex please 20:18:54 TGWi: no sex for the player, exactly 20:18:55 :D 20:19:16 Eronarn: megaquagga! 20:22:20 i want to add tardigrades to crawl 20:22:21 cyan U 20:22:23 03MarvinPA * r8f58253b615a 10/crawl-ref/source/ (ng-setup.cc spl-data.h): Ench split: more spell school changes 20:22:32 ooh, someone link to commit 20:22:33 03MarvinPA * r911022f1a0cf 10/crawl-ref/source/ng-setup.cc: Don't give Halfling hunters a buckler for free 20:22:34 03MarvinPA * r13482d21ae5d 10/crawl-ref/source/ (book-data.h initfile.cc newgame.cc spl-book.cc): Ench split: temporarily disable Arcane Marksmen 20:22:58 http://gitorious.org/crawl/crawl/commit/8f58253b615a4a67f236942fa44bc163e1873320 20:23:19 giant trilobites, too 20:23:34 oh sorry, due would have my ass for breakfast... megatrilobites 20:24:26 it's weird no games have them as monster... pretty cool creatures 20:24:34 ??todo monsters 20:24:35 todo monsters[1/3]: sloths, cassowaries, moas, hoatzin, octopus, locusts, paper wasps, bullet ants, mantis shrimp 20:24:39 ??todo monsters[2] 20:24:40 todo monsters[2/3]: anglerfish chameleons pistol shrimp anomalacaris trilobytes skunk coconut crab drop bears guard llamas 20:25:19 kilobyte: yes, but aquatic, that lessens the appeal 20:25:24 silence charms/air? boo :( 20:25:31 also boo @ nec brands in charms 20:28:07 MarvinPA made a decision (not quite what I proposed either), and I trust him. 20:28:37 dpeg: no obvious attack is a worse problem 20:28:54 kilobyte: they are psychic trilobites, of course 20:28:57 Wikipedia says many species were predators... I don't get how they could anything though 20:29:03 Eronarn: doh! 20:29:05 MarvinPA: didn't crusaders used to have 3 charms when they had no hexes? could be wrong 20:29:20 -!- valrus has quit [Remote host closed the connection] 20:30:02 ps, is anyone working on adding hypnotoads yet 20:30:09 i have a hilarious idea for them: give them monster Enslave 20:30:14 nope, looks like they had 3 fighting though 20:30:28 aha 20:30:30 hypnotoad becomes friendly and you can't bring yourself to target anything that would hit it; all of your allies become hostile 20:30:47 as a cute little anti-summoner gimmick 20:31:32 fighters and berserkers themselves only get 3 fighting so leaving crusaders at 2 is probably fine 20:33:04 sure 20:34:44 Are there any decent ideas for renaming the "Crusader" background to something better? 20:35:26 -!- edlothiol has quit [Ping timeout: 255 seconds] 20:35:34 Antimagician 20:37:36 no idea, but +1 to the notion that the current name is abysmal 20:37:46 i actually really like the current name 20:37:51 it's unique 20:38:03 if you have read a history book or two, you may find it strange 20:38:12 what're the odds of that? 20:38:17 Eronarn: then why it isn't "Mujaheddin" ("jihader")? 20:38:20 depends on the player 20:38:36 I have a hunch that Paladin is better than Crusader but of course we won't do that 20:38:51 kilobyte: presumably because all the crawl role names are in english 20:39:43 "crusader" has strong religious overtones... while our crusaders are not religious 20:40:34 i was serious about antimagician btw. make 'em anti-mage mages 20:40:56 they're on a crusade against bad wizards! 20:48:51 -!- Twilight13 has quit [Read error: Connection reset by peer] 20:51:29 idea: what if we hard-coded Heroism to give less than 5 UC, before the skill is fixed? 20:52:41 -!- Twilight has joined ##crawl-dev 20:53:07 -!- Twilight is now known as Guest79869 20:54:51 kilobyte: yes, wondered about that myself. If you think it's necessary, do it. But there have been voices that it may not be. 21:00:24 ok, it's you who's the biggest Okawaru hater, so I'd ask you to flip the coin or something 21:01:36 leave it in 21:01:40 crap... I didn't manage to tie his and Trog's gifts to monster tiers... Too late though, it would require some balance testing. 21:01:57 to monster tiers? 21:02:02 kilobyte: so you can get the gift for killing a rat? :) 21:02:11 currently you can 21:02:39 "Here, noble warrior, take this vorpal battle axe for your valiant efforts... of beheading a rat." 21:02:45 10 rats give 15 times as many chances as 1 ancient lich 21:02:58 (15 since half of them will drop a corpse too) 21:03:21 kilobyte: ouch :) 21:03:28 same for Trog... you do the Hive for a crapload of trivial kills 21:03:45 I believe that piety gain should be linked to monster tiers as well, but that's not something we can add on a whim. 21:03:59 Of course, the real solution is to just get rid of those puny monsters in the first place. 21:04:06 yeah, I wanted to go with gifts first 21:06:09 if you remember me arguing loudly against removal of Hive -- I'd like to stress that I meant having a non-random source of food 21:06:25 the critters are not something that's worth keeping 21:06:36 Orc is the same... 21:08:24 kilobyte: I know. My proposal would keep the monsters but make the whole thing more relevant because you have to kill them fast for optimal results (and only one attempt, too). 21:08:28 placing ten food shops, or having a pasture with yaks and a shed with loads of cheese would serve all the good purpose Hive has with none of its downsides 21:08:52 portal vault isn't fun 21:09:16 kilobyte: true, but Orc can be (a) scaled, in contrast to Hive, and (b) is a resource (orcs, armours, gold, shops), and (c) much shallower than Hive, so has more reason to be like it is 21:09:30 kilobyte: why not fun? 21:09:41 you often find Hive entrance earlier than Lair, and many builds can't do it outright without rPois, too 21:09:42 it would not use the current layout 21:10:46 for Orc, vanilla orcies provide just piety and possibly branded weapons 21:10:58 (mostly short blades) 21:11:04 yes, but it is our choice how many plain orcs we want 21:11:26 we can reduce their number, or replace them by better orcs -- no such option for hive 21:12:02 -!- st_ has quit [Ping timeout: 240 seconds] 21:13:05 kilobyte: Hive is currently trivialise because you can take it slow (rest up etc.)... if there is only one attempt and you're trying to be quick, it will be much more interesting 21:13:31 essentially, it will lead to the question: how many consumables are you willing to spend for how much food? 21:13:43 for a reward that is not necessary to survive, it would be fine 21:13:52 hive food isn't necessary to survive 21:13:56 not at all 21:14:06 yeah, most chars have no need for hive food 21:14:12 Of course, (assuming that food plays a role in your game at all), you can delay that until later (when you'll need less consumables for the same gain), but then you'll have spent more permafood 21:14:16 at least, that's the plan :) 21:14:24 blocking a troll from a full game or a spriggan from casting big spells just because they rolled poorly on jewelry lottery isn't good 21:14:27 Eronarn: it depends on the character 21:14:36 Eronarn: for _most_ chars, yeah 21:14:37 'blocking a troll from a full game' is not the same thing as 'necessary to survive' 21:15:04 i doubt even an unlucky conj troll would run out of food on a 3 runer 21:15:20 well, there is too much food in the game 21:15:23 uhm, so you really say that "not everyone is supposed to be able to win a non-basic game no matter their skills"? 21:15:29 our food clock is unsatisfyingly lax 21:15:46 you have to play Ce or a Zinner to get it more tight 21:15:53 dpeg: I agree, even if this might sound self-contradictory 21:15:58 kilobyte: hehe 21:16:02 -!- st_ has joined ##crawl-dev 21:16:07 kilobyte: that argument might make sense if there were a Demon Hive in hell with hellbees that you had to do to have enough food to do pan 21:16:13 but... that's not the case 21:16:14 I have problems with randomness, not the average 21:16:18 congnitive dissonance (I learned this term today) 21:16:32 it's a normal hive, with normal bees, that will have been trivial for a dozen levels by the time you 'need' the food to do an all-rune game 21:16:43 kilobyte: but we're also fine with the fact that sometimes you just lose games, no matter the skill 21:16:50 Hive as a portal vault is especially bad since you'd punish people who already rolled badly 21:16:52 (and even then you only 'need' the food for certain characters) 21:17:02 kilobyte: no, I don't think so 21:17:30 kilobyte: the portal aspect is important because it puts a lot more emphasis on resources (spending !resistance, or !might etc.) 21:17:45 bees are just not interesting enough when you can take them on slowly 21:17:50 dpeg: resources you might not have 21:18:16 kilobyte: people do ice caves with ice fiends in them without rC 21:18:28 dpeg: you'd have them 90% of the time, but not all 21:18:29 kilobyte: hey, if you roll poorly, you may have found four scrolls by D:20, which turned out to be random uselessness, curse armour, curse weapon and curse jewellery 21:18:32 if people are too cowardly to fight some bees, why should we hand them a giant pile of food 21:18:36 sure it can happen, but we believe in great numbers 21:18:48 Eronarn: no rC is just some more damage -- rest often and avoid clouds 21:18:56 wow 21:19:07 kilobyte: you can kill bees effectively with immolation, too :) 21:19:12 "just some more damage" is sort of a big thing vs an ice fiend 21:19:18 cannot believe that you are actually making this argument 21:19:28 bees without rpois: more dangerous than ice fiends without rC+ 21:19:42 This is not an opportune moment, but I am pretty sure that if we want to make hive meaningful, Just One Attempt (i.e. portal vault) is a good step. 21:20:05 ice fiends torment and what not 21:20:24 Eronarn: it depends on the time of the game. You don't go into ice caves pre-Lair. 21:20:40 kilobyte: I will get in touch with you before Hive is touched okay? 21:20:50 hive could be moved to post-lair depth 21:20:55 dpeg: what about making hive a guaranteed portal vault on lair 1-4, and you can't avoid LOS of it to not trigger the timer? 21:21:21 weighting it against the XP from lair, which is crucial for most characters 21:21:23 MarvinPA: spoily, affected by shafts 21:21:39 Eronarn: Lair:1-4 sounds good. I didn't think of a timed vault, I believe players should be able to choose when to enter. 21:21:52 dpeg: choosing when to enter just means that they'll come back at L25 or whatever 21:21:53 mm i guess, although i'm not convinced that there'd be any problem with making it a portal vault either anyway 21:22:06 you really have to force people to go in early if you want fighting bees to be fun 21:22:25 Eronarn: my proposal is a bit more conservative than that 21:22:47 a bit too conservative :) 21:22:58 may not even touch hive at all if it's just made 'you can only enter it once' 21:23:08 Eronarn: I think that is not true 21:23:28 if there are no choke points, it won't necessarily be a cake walk 21:23:31 this screws over people who HAVE to enter hive early on weak characters (a very, very small amount of players but it'll be immensely frustrating) and just makes it more of an annoyance for everyone else 21:23:34 Eronarn: bees are either pretty hard, or trivial. A single resistance makes a huge difference; it might be offset by general power of your character then. 21:23:36 if you can come back whenever 21:23:44 if you make it a timed portal vault, you turn Hive into a lottery 21:24:06 I am fine with this lottery for baileys and volcanoes, but I am more conservative when it comes to Hive/food 21:24:27 dpeg: if it's not a cakewalk for xl25 players, how is it supposed to be reasonable for everyone else 21:24:38 dpeg: i'm not talking a random one; i mean a guaranteed, once-per game one that is timed 21:24:49 i don't think i've done hive since i played cepr of zin and even that would've had no problems just waiting for hive to become trivial and then clearing it in one go 21:24:56 placed in such a way that either you do hive early, or do lair late 21:24:59 monqy: it is a cakewalk for then, I agree. But our foodclock should be tightened so that food matters before XL 25. 21:25:18 MarvinPA: yes, I also skip it. But the reason is overabundance of food. 21:25:40 the reason I skip it is because it's really annoying 21:25:42 Eronarn: I understand what you mean, it is a valid idea. I just approach the thing less radically. 21:25:47 monqy: of course 21:25:47 so the question then loops back to: why are we making this change when food will still be abundant without hive 21:26:03 Eronarn: I think everyone agrees that food should be more scarce 21:26:03 tightening the food clock can't be currently done because most of the game is spent in places without food. Which makes the food balance for minimal games terrible. 21:26:22 if by 'most' you mean 'not most' 21:27:05 it's places without corpses, but even they have some, and can have food inside them too 21:27:18 there aren't many places in the game at all that have both no food and no corpses 21:27:30 I would start by reducing number of chunks. 21:27:42 ie: huge piles of food for a 3-runer, roughly enough (for trolls with a modicum of conservation) in a full game 21:27:58 dpeg: oooh... now that's hitting the right place 21:28:08 dpeg: making each chunk give less nutrition might also be an option 21:28:30 yeah 21:28:36 this makes eating chunks less time-efficient, and makes them less calorie-dense as a food 21:28:37 I believe that food clock is too lax, but we don't have to spill out the kid with the water (another Germanism). We can slowly nerf until we feel the food again. 21:28:43 Eronarn: yes. 21:28:44 which may have more desirable effects than reducing the chunk frequency 21:29:10 "throwing out the baby with the bathwater" is an english turn of phrase, too :P 21:29:16 Eronarn: having 8 chunks from a single corpse just means that gourmands can stuff themselves from Hungry to Engorged. Not good. 21:29:18 dpeg: both perishable and permanent food come early 21:29:22 MarvinPA: damn! 21:29:25 :) 21:29:46 so you have to stockpile, and that stockpile has to last enough 21:29:51 of course, before any of that we should implement vintermann's food proposal, it is very good 21:30:03 yeah, i agree on the upper end the chunk # is excessive 21:30:09 kilobyte: what do you mean by "come early"? 21:30:14 though i think stuff should have a more consistent chunk number too 21:30:25 it varies by... a lot 21:30:40 dpeg: so berserkers can't even eat before "very hungry" while they have to not be "hungry" to zerk? 21:30:54 kilobyte: sure they can! 21:31:04 have to find clean chunks 21:31:30 eh, that sounds really annoying 21:31:39 I hope we can get vintermann's proposal (and ideally his patch :) for 0.9. I promise to redo prices for 0.9; among other things, food should be a lot more expensive. 21:31:51 Eronarn: at very hungry, contaminated will do 21:32:07 it is the best idea I've ever seen to make chunk properties relevant 21:32:17 yes, i mean that it sounds very annoying that you could be very hungry, eat a chunk, and wind up hungry and unable to eat enough to be able to berserk 21:32:28 dpeg: you get perishable food in living branches, and permafood in the Hive and all around the game except for Hells... which means most of what you collect came from living branches again 21:32:30 that is a problem with berserk, not the food proposal 21:32:43 kilobyte: yes 21:32:59 you don't get clean chunks in Orc, Elf, or often for long periods of time elsewhere 21:33:00 well, the proposal itself is fine 21:33:14 just that it would be incredibly annoying for stuff that requires you to be satiated or higher 21:33:30 Eronarn: is there anything apart from rage? 21:33:31 which is just berserk really 21:33:33 dpeg: btw, we should give high elves reverse saprovore - higher chance to be sick from contam meat 21:33:38 vamp weapons 21:33:45 Eronarn: yes, those elitist snobs 21:33:59 not use to dungeon cuisine 21:34:51 kilobyte: otoh, it might not be as bad as you think (with new food rules and berserk still only at satiated): players would think harder about raging, which is probably good 21:35:07 but I am also fine with relaxing the rage, especially since casters can do it anyway 21:35:33 the problem would be dropping from satiated => hungry right before a fight 21:35:38 and only having contam chunks in inventory 21:35:48 what if we auto-stopped berserk at Starving? 21:35:50 you can't really judge when that will happen, and you'd have to blow permafood, or leave the level and come back later 21:35:57 kilobyte: that sounds best 21:36:12 I believe the rage rule was not made for tactical reasons, but only for safety 21:36:19 Eronarn: yes 21:36:26 kilobyte: so at starving berserk ends and you are slowed as usual? that sounds good to me 21:36:34 kilobyte: hmm, here's a thought... berserk = per-turn-of-berserk food cost, but the cost in food you've paid doesn't show up until the end of the berserk, plus the drop out of it at starving 21:36:44 rather than cost-at-the-end 21:36:50 and perhaps give a warning & prompt when starting to berserk at Very Hungry 21:37:16 Eronarn: actually, we might just show the food cost immediately 21:37:24 because you will get only very little berserk then 21:38:01 @??spriggan berserker 21:38:02 spriggan berserker (04i) | Speed: 16 | HD: 12 | Health: 26-44 | AC/EV: 2/25 | Damage: 21 | Flags: see invisible | Res: 06magic(112) | Chunks: 07contaminated | XP: 1094 | Sp: brothers in arms, hand, berserker rage. 21:38:03 dpeg: no, I mean berserk would be allowed at all times otherwise 21:38:08 ^^very little berserk^^ 21:38:30 Monster icons drag in movement (https://crawl.develz.org/mantis/view.php?id=3599) by XuaXua 21:38:34 kilobyte: I understand, and I approve. I suggest to help players prevent erroneous rage when it wouldn't last long. 21:38:43 ah, cool 21:39:18 BTW, I played several HaBe and KoBe games recently to compare them to each other 21:39:28 kilobyte: Very Hungry -> Starving is short, if you're unlucky just one or two turns of rage. So a prompt is in order. If you start at Hungry, no warning should be necessary. 21:39:40 kilobyte: what are your findings? 21:39:52 these races are identical save for halflings having food problems 21:40:02 in that role, yeah 21:40:07 they do have some differences otherwise 21:40:09 let me guess: the next step is to remove Ha for 0.8? :) 21:40:19 e.g., now ko are good at crossbows 21:40:25 disable for 0.8, remove for 0.9 21:40:35 monqy: disable? why the half-measure? 21:40:39 Eronarn: you don't get a crossbow unlesss Hu 21:40:44 I cannot say I have a soft spot for halflings. 21:40:49 kilobyte: you can pick one up though 21:41:05 But I know that Eino has. 21:41:08 * Eronarn wants to add 'kobold snipers' early on with crossbows, with about the rarity of blowgun ones 21:42:34 Eronarn: I seem to find more lajatangs than crossbows 21:42:44 they're fairly rare until yaktaurs show up 21:42:51 but you can run across a bailey or the like 21:42:56 (save for Trog, who gives them like slings) 21:42:59 mostly we just need to spread it out 21:43:05 too many xbows late, not enough early 21:43:08 yeah 21:43:21 i love yaktaurs but there are just toooo many of them in the game 21:43:24 same with centaur packs 21:43:35 this is a problem for weapon types with a single item 21:43:45 hand crossbows 21:43:50 arbalests 21:43:57 repeating crossbow 21:44:01 Eronarn: elephants and vampire mosquitoes now 21:44:03 heck, you could have a Ballista 21:44:11 for large races 21:44:11 like Lenin in a tin 21:44:36 ogre ballistechnician 21:44:45 fires javelins 21:44:46 kilobyte: Ha and Ko are more different if you aren't just playing berserkers 21:44:51 Eronarn: ballistae <3 21:44:52 any two berserkers are rather similar 21:45:16 (seriously though, bring back hand crossbows :( ) 21:45:26 !apt DE 21:45:27 DE: Air: 1, Armour: -2, Axes: -2, Bows: 1, Charms: 4!, Conj: 1, Xbows: -1, Dodge: 2, Earth: 0, Evo: 1, Exp: 140, Fighting: -2*, Fire: 1, Hexes: 3, Ice: 1, Inv: 0, Long: -1, Maces: -3*, Nec: 2!, Poison: 1, Polearms: -3*, Shields: -2, Short: 0, Slings: -2, Splcast: 4!, Stab: 1, Staves: 0, Stealth: 2, Summ: 1, Throw: 1, Tloc: 1, Tmut: 1, Traps: 0, Unarmed: -2* 21:45:29 way too little ammo 21:45:33 oh good they did get their xbow apt nerfed 21:46:01 kilobyte: darts are alllllll over the place, or you could just make them fire bolts 21:46:30 you rarely get >10 javelins by Lair end, and often none at all 21:46:55 I meant ballistas 21:46:57 oh. yeah it'd need to only drop if it came with ammo 21:47:01 or it could fire spears too 21:47:04 kilobyte: you're serious about Ha axing? 21:47:39 dpeg: not for 0.8 of course, and I tested only Be 21:47:49 dpeg: they just recetly got a differentiation patch 21:47:50 !apt ha 21:47:51 Ha: Air: 1, Armour: -2, Axes: -1, Bows: 2, Charms: 1, Conj: -2, Xbows: -1, Dodge: 2, Earth: 0, Evo: 1, Exp: 100!, Fighting: -1, Fire: 0, Hexes: -1, Ice: 0, Inv: 0, Long: 0, Maces: -2, Nec: -2, Poison: -1, Polearms: -3*, Shields: 1, Short: 3!, Slings: 4!, Splcast: -2, Stab: 0, Staves: -2, Stealth: 3, Summ: -1, Throw: 3!, Tloc: 1, Tmut: -4*, Traps: 0, Unarmed: -2* 21:47:52 !apt ko 21:47:53 Ko: Air: 0, Armour: -2, Axes: -1, Bows: -1, Charms: -2, Conj: -1, Xbows: 2!, Dodge: 2, Earth: 0, Evo: 2, Exp: 100!, Fighting: 1, Fire: 0, Hexes: 0, Ice: 0, Inv: 0, Long: -2, Maces: 0, Nec: 0, Poison: 0, Polearms: -2, Shields: -2, Short: 3!, Slings: 2, Splcast: 0, Stab: 2, Staves: -1, Stealth: 3, Summ: 0, Throw: 3!, Tloc: 0, Tmut: -1, Traps: 0, Unarmed: 0 21:47:54 most berserkers feel fairly similar 21:48:15 Eronarn: this patch is what I'm talking about 21:48:26 kilobyte: as pointed out, be is not a good way to test it 21:48:33 Too many --more-- delays - XP out of sequence with movement (https://crawl.develz.org/mantis/view.php?id=3600) by XuaXua 21:48:34 Be is an awful way to test it 21:48:36 to me the difference is really only the choice between whether you focus on ev and ac, and whether you decide to train up ranged skills 21:48:40 I can't think of a worse way to test it 21:48:47 elliptic: summ 21:49:31 if you play ar you will hate both races and want to remove them 21:49:43 Ha = worse casters, food issues. diff = slings vs crossbows 21:50:02 Ha also has intrinsic partial rMut 21:50:20 and Ha has more HP and better shields 21:50:29 ha are a lot better at charms 21:50:36 dpeg: it's not just Ha vs Ko too... MD vs Mi are mostly in the same standing 21:50:53 md and mi aren't even close 21:50:56 MD and HO are even more similar 21:50:59 md are *good* at some magic 21:51:01 !apt md 21:51:02 MD: Air: -2, Armour: 3!, Axes: 2!, Bows: -2, Charms: -2, Conj: -1, Xbows: 1, Dodge: -1, Earth: 2, Evo: 1, Exp: 130, Fighting: 2!, Fire: 2!, Hexes: -2, Ice: -2, Inv: 0, Long: 0, Maces: 2!, Nec: -3*, Poison: -2, Polearms: -1, Shields: 2!, Short: 1, Slings: -1, Splcast: -3, Stab: -2, Staves: -1, Stealth: -3, Summ: -2, Throw: -1, Tloc: -2, Tmut: -1, Traps: 1, Unarmed: 0 21:51:14 and Ko is better for conjurer-types, especially with sif, but Ha is better for crusadery builds 21:51:32 yeah, MD v Mi v HO are more similar 21:51:32 Eronarn: yeah, they're mostly strictly better than MD 21:51:40 HO has beogh 21:51:57 MD has earth and fire skills 21:52:03 elliptic: my point is, no different enough 21:52:05 and Mi has... horns? 21:52:07 making HO have equal weapon apts would be a good differentiation from HD 21:52:09 !apt HO 21:52:10 HO: Air: -2, Armour: 1, Axes: 2!, Bows: -1, Charms: -1, Conj: 0, Xbows: -1, Dodge: -2, Earth: 0, Evo: 0, Exp: 100!, Fighting: 2!, Fire: 0, Hexes: 0, Ice: 0, Inv: 0, Long: 1, Maces: 1, Nec: 0, Poison: -1, Polearms: 1, Shields: 1, Short: 0, Slings: -1, Splcast: -3, Stab: 2, Staves: -1, Stealth: -2, Summ: 0, Throw: 0, Tloc: -2, Tmut: -3, Traps: 0, Unarmed: 1! 21:52:21 kilobyte: I agree with you about MD and Mi 21:52:23 no reason for them to have axes 2! short 0 21:52:28 most orcs show up with short, even 21:52:39 kilobyte: to me, there are reasons to keep each of Mi (labyrinths), HO (orc branch and Beogh), MD (dwarf dichotomy). It is the absence of something like for Ha which makes them vulnerable (that's a pretty skewed scale, I know). 21:52:58 labyrinths are a pretty weak defense for Mi 21:53:04 mi are fine 21:53:05 elliptic: but not for me 21:53:07 reverse kenku 21:53:08 you could have labyrinths without Mi being a playable race 21:53:14 without any trouble at all 21:53:17 of course 21:53:18 (ok not really, but) 21:53:20 remove labs too 21:53:24 Mi have equal weapon apts while MD are weak at some, but you never switch weapon types anyway 21:53:25 -!- monqy was kicked from ##crawl-dev by dpeg [monqy] 21:53:32 -!- monqy has joined ##crawl-dev 21:53:37 sorry 21:53:37 kilobyte: I switch weapon types a fair amount 21:53:53 it'd be nice if crosstraining boosts atually applied to skill too 21:54:12 elliptic: sure, for Mi I also need that I like having players versions of monsters. But Mi is the weakest out of the three. 21:54:12 you mean, with more than 3 skill? 21:54:12 get maces 10 and you can fight as if you have axes 5 21:54:19 or whatever 21:54:37 this would of course be modified by aptitude 21:54:40 the crosstraining is very good already, people are just shy to use it 21:55:04 kilobyte: a recent MD of mine actually switched *away* from an exec axe of speed with 20ish skill in favor of a demon trident of speed + shield 21:55:16 dpeg: crosstraining is good for training, but no actual bonus to using it, which is why 21:55:47 weapon effects would be one way to encourage crosstraining 21:56:03 someone with spear + staff skill could use the effects of one while equipped with the other 21:56:19 so there's an actual incentive to learn two skills 21:57:23 -!- Mu_ has quit [Quit: Defecator, may everything turn out okay so that you can leave this place.] 21:57:44 hmm... actually, that'd work real nice with Oka heroism, too 21:57:54 definitely worth considering 21:58:01 Eronarn: are you aware of the proposal that high end weapons use two skills instead of one? 21:58:05 I think that has a lot of potential. 21:58:26 dpeg: no... i don't think that'd be good if it were the way spell skills work 21:58:32 if it were for weapon auxes only that might be cool 21:58:52 most weapons would pretty cleanly map to 2-3 applicable skills 21:58:55 elliptic: wait, wait... how do you switch AWAY from a freaking exec axe of speed? Was it a -8 randart with MUT+ *Rage *TELE -CAST? 21:59:07 like: bardiche = spear, staff, axe 21:59:34 kilobyte: demon trident of speed is really that good 21:59:39 stuff like whips is where there'd be weirdness, primarily 22:00:02 but you can address that by making weird weapons have very different sets of auxes, possibly unique to those weapons (like disarm on whips) 22:00:20 min delay 3 rather than 4, can use a shield also, much better accuracy (the exec axe of speed was -2,+9 and I was missing a ton) 22:00:35 a 20 base damage weapon with same brand and skill you have... 22:00:36 and I had some slaying and berserk to even out the damage a bit 22:01:15 it definitely felt like the right choice once I got the dtrident to min delay (which was very quick with crosstraining) 22:01:44 20 base damage vs 14 base damage isn't as big a deal when you have about 5 slaying and berserk 22:02:07 3 delay vs 4 delay is a nice 33% boost 22:02:15 was it a 3 or 15 runer? 22:02:15 and it meant I got a shield 22:02:35 3-runer, though I'm sure I could have 15-runed with that char without any trouble 22:02:51 I made the change in crypt 22:03:29 anyway, just a random example... obviously switching to the dtrident would have been even more clearly the right decision if I hadn't had such a good axe already :) 22:04:30 it's also an obvious argument for nerfing them 22:04:56 and the speed brand... rarity != balancing 22:05:02 preventing hell knights from getting speed weapons would be good 22:05:03 should move to a speed system where it's attacks-per-turn rather than turns-per-attack imo 22:05:05 Eronarn: weapons getting manoeuvres from cross-training skills sounds pretty cool 22:05:34 the speed brand is not a pinnacle in Crawl's design 22:05:37 kilobyte: well, demon trident of speed doesn't generate except as a randart or as a hell knight/margery/whoever weapon 22:05:43 TGWi: yeah... could do some neat stuff working non-weapon-skills into that, maybe 22:06:01 demon trident itself is slightly weaker than a katana 22:06:03 Eronarn: and with >1 attacks per turn would you "wind up for the attack" or whatever on lost turns 22:06:26 I'd return speed to long blades, give staves that dual-hand thingy planned but not implemented in 4.1, make speed brand *1.5. 22:07:03 elliptic: randarts would still be overpowered 22:07:04 monqy: the difference is that instead of stuff like 'min delay 4', where going from 4 to 3 is a huge deal, you have speed whatever, and then get to attack 30% faster than that speed or what have you 22:07:16 indeed 22:07:26 btw, melee attack delay should use div_rand_round 22:07:43 if you express it as turns-per-attack you have to give smaller and smaller boosts in terms of real numbers 22:07:46 kilobyte: that sounds good aside from nerfing sabre of speed 22:07:58 oh I meant <1 not >1 22:08:02 I'm curious about <1 22:08:03 which is currently reasonably balanced 22:08:12 mistyped there 22:08:13 monqy: it would work out exactly the same interface-wise 22:08:14 elliptic: would make quick blades a no-brainer 22:08:19 it would just be changing the underlying math 22:08:26 to use 'miles per gallon' rather than 'gallons per mile', so to speak 22:08:42 they are inferior unless you get a top brand currently 22:08:54 Eronarn: oh I was hoping for multiple attacks per keypress 22:09:04 less pain for fast weapon users 22:09:17 monqy: well, that could be done... it might actually be interesting 22:09:23 monqy: good idea! 22:09:27 especially if it applied to all sorts of fast attacks 22:09:46 elec, pain... 22:09:51 distortion 22:10:00 chaos 22:10:25 draining's hd-reduction? or does that check damage dealt 22:10:44 it's flat I think 22:10:48 the damage isn't flat 22:10:50 the hd reduction is 22:10:53 the hd is though yeah 22:10:58 that too but no one uses draining ever :p 22:11:05 maybe draining shouldn't tank your exp 22:11:08 people might use it 22:11:13 draining should be a stat boost imo 22:11:21 like it doesn't actually trigger often 22:11:23 you hurt their str or int or dex and buff yours 22:11:30 kilobyte: if you support removal of speed ego, I support the axing of Ha. 22:11:32 but the effect on xp gain is really substantial if it does 22:11:40 dpeg: keep speed for sbl imo 22:11:51 TGWi: it's bad because of kitting, anyhow 22:11:52 and staves? 22:12:05 dpeg: axing or differentiating more 22:12:11 imo keep speed as a randart-only 'chase' brand 22:12:19 kilobyte: it is a very plain ego, right there with vorpal 22:12:19 it's good to have rare stuff people are excited to find 22:12:22 draining is good, though 22:12:50 dpeg: I had a proposal a while ago regarding brand balance though 22:13:05 nerf speed like haste was nerfed, and make freezing and flaming damage bonuses apply before AC reduction 22:13:16 so speed is less better than it is now and it's even less better against AC 22:14:03 and vorpal is even more useless 22:14:12 vorpal can go back to 20 or so 22:14:17 not my fault it's useless :P 22:14:26 imo make vorpal a damage boost that also buffs weapon effects 22:14:30 there might just not be room for it 22:14:37 vorpal boosting weapon effects would rock 22:14:37 it already shows up as 'slicing', 'piercing', 'crushing', etc. 22:14:42 eronarn should get a medal 22:14:43 so it's keyed to the weapon 22:14:43 for rocking 22:15:00 are weapon effects slated for 0.9? 22:15:03 no 22:15:05 TGWi: it's your fault! the japanese tsunami is your fault too! 22:15:07 :( 22:15:11 vorpal weapons just become 'Better At Being Weapons', in such a system 22:15:38 TGWi: there is no overarching framework, just some ideas 22:15:47 dpeg: you know what there is some ideas for? lava orcs. 22:15:48 if there were solid proposals, though? 22:15:54 eronarn should get two medals 22:15:56 for double rocking 22:16:09 .9: octopodes, lava orcs, weapon effects, alchemy 22:16:18 dpeg: speed is better than vorpal, you actually notice it more than most other brands too 22:16:27 given that this is ##crawl-dev, it is plain to see why progress is so slow :P 22:16:32 Dungeon Crawl Stone Soup 0.9 Acid Lava Octopus Flensing Edition 22:16:40 you feel you get many weak blows 22:17:03 dungeon crawl stone soup 0.8: limned ________ 22:17:15 what's new in 0.8 anyway 22:17:20 everything 22:17:34 hmm... and what if instead of delay/1.5 it kept /2 but reduced damage (inc. slaying)? 22:17:41 ps TGWi want to hear the best idea? 22:17:44 kilobyte: you are talking about good/bad on a power scale? I am not denying that speed is good power-wise. It's just very bland. (Yes I know about the difference between "+40% damage" and "attack twice". I just think that distortion, reaching, draining, electrocution etc. are all a lot more interesting.) 22:17:46 Eronarn: i already did 22:17:48 it's' called lava orcs 22:18:12 speed is more interesting than vorpal 22:18:19 I think this is what kilobyte was saying 22:18:22 dpeg: no, I mean being less bland 22:18:28 lava orc pirate unique, on a ship of toenails, which bursts through the lava and spews out its damned crew (of lava orcs) 22:18:31 yes, I agree with that 22:18:45 but on the bland-ness scale, it comes second right after vorpal :) 22:18:46 it feels different rather than just giving more damage 22:19:18 are there other proportional damage boost brands beyond vorp, freeze, flame, drain, and speed? 22:19:43 holy I guess 22:19:45 and slaying 22:19:56 here's a zany idea for speed: if you attack with it you get +swiftness -attackspeed, and if you move you get +attackspeed -swiftness (neither ever becoming a penalty though) 22:20:31 might work better as a spell though 22:20:33 that's zany 22:20:44 it'd be fun though 22:21:16 Eronarn: "Ah, there's nothing like hot winds of hell blowing in your face." 22:22:45 can't get new pirates past LeDpeg, though 22:23:09 (let's hide Murray's profession when he lived) 22:23:41 kilobyte: no, have to stage a coup first 22:25:21 Eronarn: do you know how to build a ship out of toenails? 22:25:40 this always puzzled me in the Norse myth 22:26:25 very large toenails 22:26:33 also, in your case... toenails+lava = terrible toxic smell 22:26:33 it's actually just like two or three of them but they're big 22:26:50 kilobyte: the dead know 22:27:07 also, they're lava orc toenails 22:27:10 and thus immune to lava 22:27:58 orc high priests immune to fire... pretty buggy to me 22:28:10 yes, they should have lost that versions ago 22:28:15 unless they're lava orc spies 22:28:24 sorcerers have rF rC I think 22:28:27 @??orc sorcerer 22:28:28 orc sorcerer (13o) | Speed: 10 | HD: 9 | Health: 22-41 | AC/EV: 5/12 | Damage: 7 | Flags: evil, spellcaster, see invisible | Res: 06magic(36), 05fire | Vul: 08holy | Chunks: 07contaminated | XP: 627 | Sp: b.fire (3d17), b.draining (3d14), demon, paralyse, animate dead, teleport self. 22:28:33 nope just rF 22:28:36 @??psyche 22:28:37 Psyche (13@) | Speed: 13 | HD: 5 | Health: 39 | AC/EV: 0/12 | Damage: 7 | Flags: spellcaster | Res: 06magic(26) | Chunks: 07contaminated | XP: 371 | Sp: throw frost (3d6), cantrip, haste, throw flame (3d6), magic dart (3d4), invisibility. 22:28:37 still weird 22:28:38 still speed 13 22:28:49 make psyche a felid 22:29:00 a felid with chaos paws 22:29:01 psyche meows, "YOU ARE VIOLATING AREA SECURITY!!!" 22:29:53 Eronarn: how do you justify like axes getting pierce 22:30:07 or staves 22:30:13 or whatever polearms get 22:30:24 -!- eith has joined ##crawl-dev 22:30:31 hullo eith 22:30:40 sup 22:31:13 TGWi: axes can have points or hooks, and they can do hooks with their blade 22:31:52 probably the best way to do this would be a general rule and special-case where necessary with "rules" of increasing specificity 22:31:53 orc priests are immune to fire b/c of lack in the AI department 22:32:15 Eronarn: impede for polearms seems cool too 22:32:29 enemies moving toward you lose time 22:32:42 would make more sense for sharing 22:32:53 come to think of it that might be problematic with kiting 22:33:12 TGWi: it only works if you are 'set' (haven't moved) 22:33:27 perhaps even require doing a . to be set against a charge 22:33:55 impede would be one on staves, too - historically they've been used by a lone combatant to keep a group away 22:34:08 staves get block imo 22:34:23 you'd want multiple possible effects per weapon 22:34:28 fair enough 22:34:51 but pla sharing impede with staves doesn't make much sense if staves already has impede 22:35:18 a good starting list: impede, block/parry, pin, sidestep, disarm, trip, grapple, pin, pain, clobber, cleave, sunder, knockback, charge, overrun 22:35:51 TGWi: being good at both polearms and staves = more chance of it kicking in on either 22:36:07 choppy polearms may also, say, be able to cleave 22:36:08 what are pin, pain, clobber, sunder, and overrun? 22:36:11 hmm yeah 22:36:33 pin is an extension of a grapple (see also locks and holds and throws) 22:36:47 clobber is like with a clubstab, whacking someone and doing some mental status effect 22:37:00 ah yes 22:37:06 pain is with a whip - lacerate someone, doing less lasting physical damage but a lot more pain 22:37:15 sunder is breaking through a shield or armour 22:37:33 overrun is some variant on a trample (you move), whereas knockback = just they move 22:37:52 -!- valrus has joined ##crawl-dev 22:38:18 last two would warrant an evocation like reaching 22:38:25 there's also piercing through enemies, making enemies hit each other instead of you, blinding 22:38:48 which implementation of sidestep do you mean 22:38:59 there are lots of ways to implement sidesteps 22:39:08 the most fruitful way is to probably think of, what interface inputs do we have available? 22:39:17 I like the "enemies fire where you last were" thing but also I had an "enemies get distracted if you walk around them" thing 22:39:35 let's say: move before an attack, explicitly . and then not move after that, ctrl-move - as the three basics 22:39:52 we'll call those 'charge', 'ready', 'special' 22:40:23 if we break it down to those three, we can think about how to allocate possible effects between them, such that any weapon category has at least one of each 22:40:35 because some people will hate any 2, but it will be rare that someone hates all 3 22:40:44 replace special with passive, bro 22:41:09 er, well, yeah the initial discussion was ctrl-move but then it became ctrl-move to do a normal attack 22:41:12 oops 22:41:38 I don't like requiring ctrl-move to do things because it already does a lot of things 22:41:45 (plus it doesn't work with my rc) 22:41:51 anyway though 22:42:30 kilobyte: speaking of orc resistances, nergalle's rF+ rC+ should be removed 22:42:56 since there is no reason for it, especially now that her spellset has nothing to do with elements 22:43:00 so an example might be spears: stabby polearms get piercethrough off of polearms, choppy ones get cleave off of axes+polearms. lighter ones get impede off of staffs+polearms; heavier ones just off of polearms 22:43:12 something like that 22:43:57 the best way to allocate this is to have to say as few exceptions as possible 22:45:04 staffs should help with all impeding, easier to parse 22:45:31 flavour but whatever 22:46:56 each weapon desc would need to have some info 22:47:39 'this weapon can impede advancing enemies and, with skill in axes, cleave through foes' 22:47:42 preferably with as much repetition as ossible 22:48:05 should only vary by weapon type and maybe attack mode in cases where it's needed 22:48:06 like 'this is a balanced polearm that benefits from Staff skill' 22:48:15 vs. 'This is an unbalanced polearm that benefits from axes skill' 22:48:22 Hive wiki page redone. 22:49:04 TGWi: weapon type, handedness, damage type should get *most* stuff 22:49:10 there will only be a handful of exceptions 22:49:14 like whips 22:49:25 whips have a different damage type than other maces though 22:49:26 maybe scythes 22:49:38 piercing/crushing for most, and whips are slashing 22:49:39 so yeah 22:49:49 TGWi: right but they won't have sword-like effects even though they're slashing 22:50:01 what I mean is you should be able to say "slashing polearms let you __________" 22:50:07 -!- blackpenguin has quit [Quit: WeeChat 0.3.4] 22:50:09 right, yeah 22:50:15 that will generally work, with most of the exceptions in maces 22:50:24 ankuses let you tame elephants 22:50:35 fr: remove ankus because serously why bother 22:53:14 i'm gonna watch a movie but if you set up a page with the weapons with name/type/handedness/damagetype that'd probably be enough to work with 22:53:29 i actually have a bit of time until next week 22:53:47 can I make a new page for that 22:53:47 ? 22:53:49 should be a lot more free for .9 though 22:53:56 i think there is an exsiting one but you could make a new one sure it'd be neater 22:54:03 yeah that's what I'm thinking 22:54:09 there's an existing one but it's unorganized 22:54:14 most of the existing one was crap anyways iirc 22:54:14 some good info though 22:59:04 is there already some plan for a way to mitigate/resist/whatever holy damage? if not, is my idea awful? (I'll restate it here) 22:59:07 it'd be cool if instead of a "resist holy" amulet or ring or whatever, miasma would make a holy monster's attacks non-holy (aka they've been defiled so they can't channel their holy god's energy into their attacks until they're cleansed or something) 22:59:45 that's kind of cool 22:59:55 not sure how you would communicate that though 23:03:45 that's cool/flavorful but i'm not sure nec needs the buff 23:03:56 would be awesome against players, though 23:08:52 hex spell that blocks off god abilities (including your own) + holy attacks for everything in sight, like some sort of holy silence 23:10:00 Profane Word 23:11:19 also the spell should let you break god conducts without being caught when it's active... 23:13:37 -!- blackpenguin has joined ##crawl-dev 23:14:26 bad idea 23:17:51 -!- dpeg has quit [Quit: Lost terminal] 23:41:09 -!- upsy has quit [Quit: Leaving]