00:01:08 -!- ZChris13 has quit [Quit: ChatZilla 0.9.88.1 [Firefox 11.0/20120312181643]] 00:01:15 Unstable branch on crawl.develz.org updated to: 0.11-a0-1007-ge138055 (32) 00:01:28 we should make leda's not cause fumbling, the spell would be much more widely useful if it didn't 00:01:36 right now it's basically ignored 00:06:55 -!- capablanca has joined ##crawl-dev 00:08:18 Windows builds of master branch on crawl.develz.org updated to: 0.11-a0-1007-ge138055 00:09:35 could you use flaoting to avoid fumbling? 00:11:45 you can't use flight while the spell is active or it cancels flight or something 00:17:26 -!- capablanca is now known as mikee_ 00:26:50 useful 00:27:35 Debian builds of master branch on crawl.develz.org updated to: 0.11-a0-1007-ge138055 00:34:37 -!- Mottie has joined ##crawl-dev 00:39:07 -!- MarvinPA has quit [Ping timeout: 264 seconds] 00:48:37 You know what would be amazing? Being able to export saves from the server to your computer (but not import because exploitz) 00:52:12 * due cocks head. 01:15:59 -!- mikee_ has quit [Ping timeout: 245 seconds] 01:20:47 -!- mikee_ has joined ##crawl-dev 02:19:28 -!- syllogism has joined ##crawl-dev 02:54:11 -!- heteroy has quit [Quit: ChatZilla 0.9.88.1 [Firefox 3.6.28/20120306064154]] 03:03:12 -!- dtsund has quit [Quit: dtsund] 03:09:06 -!- st_ has quit [] 03:27:00 -!- ToBeFree has joined ##crawl-dev 03:28:04 -!- alefury has joined ##crawl-dev 03:59:33 -!- edlothiol has joined ##crawl-dev 04:12:35 -!- blabber has joined ##crawl-dev 04:13:03 -!- monqy has quit [Quit: hello] 04:37:11 -!- blabber has quit [Quit: leaving] 05:00:06 OSX builds of master branch on crawl.develz.org updated to: 0.11-a0-1007-ge138055 05:04:27 -!- Ragdoll has joined ##crawl-dev 05:11:49 i am starting to really hate lua 05:12:08 or rather, crawls lack of documentation for its lua functions 05:13:01 -!- blabber has joined ##crawl-dev 05:25:25 -!- Mottie has quit [Ping timeout: 276 seconds] 05:37:46 -!- its_steve has joined ##crawl-dev 05:38:11 Howdy, the new trunk elven weapon graphics, are they meant to be in the lower left of the tile? 05:39:18 no 05:39:31 already on mantis, no idea if something is being done about it though 06:16:24 -!- lord-naughty has joined ##crawl-dev 06:16:53 -!- its_steve has quit [Quit: Page closed] 06:17:04 -!- lord-naughty has quit [Client Quit] 06:35:15 Phoniks (L17 DrEE) (Lair:5) 06:43:28 -!- ToBeFree has quit [Remote host closed the connection] 06:50:48 -!- elliptic has quit [Ping timeout: 265 seconds] 06:51:04 alefury: what do you mean lack of documentation? It's all there in l_*.cc! 06:51:06 ;) 06:53:14 -!- elliptic has joined ##crawl-dev 06:59:50 -!- Ganrao| has quit [] 07:00:37 I wonder what's the point of mon-place.cc:537 07:00:50 looks like its purpose is to give more crap monsters in the dungeon 07:05:44 -!- lord-naughty has joined ##crawl-dev 07:14:30 -!- MarvinPA has joined ##crawl-dev 07:17:16 -!- Vandal has joined ##crawl-dev 07:27:01 -!- Pacra has joined ##crawl-dev 07:30:11 galehar: i mean like a list of function declarations with a short comment what they do. also, tons of stuff is not in l_*.cc, but in dat/dlua 07:31:00 and theres no way to know what functions are available where. greping is hard because l_*.cc uses different names. 07:32:06 i already know exactly what i need to do to for the new portal timers, i just have no idea how any of the stuff i want to do works. i dont even begin to understand how the current portal timers work. 07:32:21 its a fucking mess 07:33:00 -!- lord-naughty has quit [Quit: Leaving.] 07:33:43 for example there are listeners (for entering the level, seeing the portal, and turns passing), but the logic is not anywhere in the same file or the files required by it. 07:34:03 talking about dat/dlua/lm_timed.lua btw 07:34:46 its not even in lm_trig.lua 07:34:52 or at least i havent found it yet 07:38:17 -!- lord-naughty has joined ##crawl-dev 07:51:44 there was a plan of using some doxygen thingies for documentation, but looks like that's dead 07:51:45 kilobyte: You have 4 messages. Use !messages to read them. 07:51:57 -!- lord-naughty has quit [Quit: Leaving.] 07:52:22 I'm quite tempted to just write some simple docs, that can be reworked later if someone has a fancier idea 07:52:47 especially clua and clua hooks -- these are supposed to be used for players, who rarely know how to RTFS 07:53:34 kilobyte: btw, are changes to all the lua crap regarding portal timers going to interfere with portal_branches? 07:54:25 Monster database of master branch on crawl.develz.org updated to: ERSION 07:54:48 uhm, almost :p 07:54:56 specifically, once i somehow learn how listeners work, ill have to make one that listens for tiles being mapped. then carpet all the entry vaults with a marker that checks for mapping and sight and starts the timers. 07:55:29 alefury: if they change portals themselves, I doubt so -- there'll be a minor conflict with portal properties, but that's all 07:55:41 ie, dst=, dstabrev=, etc 07:56:04 i dont think i need any of that 07:56:17 but i wouldnt know, i dont fucking get any of this :( 07:56:23 Monster database of master branch on crawl.develz.org updated to: 0.11-a0-1007-ge138055 07:56:31 that part is pretty clear, i just dont know where the fuck it gets evaluated 07:57:29 its like a wrapper within a wrapper within a wrapper, fed by arbitrary arguments in the .des files, eventually delegated to functions i cant find. 07:58:45 anyway, afk for now. 08:01:48 !log nht dsee d:7 08:01:49 1. nht, XL9 DsEE, T:10281: http://crawl.develz.org/morgues/trunk/nht/morgue-nht-20120407-112033.txt 08:01:52 oops, wrong channel 08:06:23 -!- HousePet has quit [Quit: Divide by cucumber error] 08:07:07 -!- lord-naughty has joined ##crawl-dev 08:08:50 Monster database of 0.10 branch on crawl.develz.org updated to: 08:10:06 Monster database of 0.10 branch on crawl.develz.org updated to: 0.10.1-54-ga41a5f4 08:12:32 |amethyst: there's a couple of commits to monster on CDO's repository 08:16:40 -!- lord-naughty has quit [Quit: Leaving.] 09:00:28 <|amethyst> kilobyte: thanks... I see just one per branch, is that right? 09:08:14 <|amethyst> @?-version 09:08:14 Monster stats Crawl version: 0.9.1-1-gec69250 09:08:17 <|amethyst> @??-version 09:08:17 Monster stats Crawl version: 0.11-a0-1007-ge138055 09:08:39 <|amethyst> So how do we actually run 0.10 monster on CDO? 09:08:49 <|amethyst> @???-version 09:08:49 Monster stats Crawl version: 0.11-a0-1007-ge138055 09:22:05 -!- Blade- has joined ##crawl-dev 09:25:23 -!- valrus has joined ##crawl-dev 09:38:29 -!- valrus has quit [Remote host closed the connection] 09:38:53 -!- valrus has joined ##crawl-dev 09:39:06 hm, whatever ended up being the plan for transformations? they cancel in one turn now but it looks like lichform is still inconsistent 09:41:20 well, I've gotten the impression that some people think it makes sense that you can't go directly from lichform to blade hands 09:41:44 personally I don't think there is much point in preventing this now that it isn't any faster than cancelling lichform 09:41:56 yeah 09:42:01 well 09:42:11 it would be actually, since cancelling is 15 aut 09:42:29 isn't casting a form 15 aut when you are in one? 09:42:43 aah 09:42:45 so it is 09:43:10 in that case yeah, not much point have lichform still be different i guess 09:43:21 -!- qqryq has joined ##crawl-dev 09:45:20 if nothing else, I think that if you can't go directly from lichform to blade hands then you shouldn't be able to go directly from blade hands to lichform 09:53:02 lol 09:54:27 what's hilarious in crawl's code, are all the occasions that some error condition which shouldn't even happen still has handling code 09:54:34 -!- ixtli has joined ##crawl-dev 09:54:43 -!- ixtli has quit [Client Quit] 09:55:27 mumra: so I heard that you're doing Vault overhaul and vaults need to be made for that...exactly what kind of vaults do you want? I may make some. 09:55:32 e.g. decks.cc 586 - 745 -- nearly 200 lines of code helping the player in case a deck count has gone wrong 09:56:43 <|amethyst> mumra: and when I was implementing deal four I triggered that code a few times :) 09:57:01 wait ... it's more than that ... 519 - 745 ... an entire function called "_check_buggy_deck" ;) 09:57:05 <|amethyst> mumra: "can't ever happen" really means "can't happen until some dev goes and breaks stuff" :) 09:57:26 |amethyst: yeah, that code is entirely redundant in a release 09:57:36 |amesthyst: it's the sort of thing that test suites are for 09:58:25 <|amethyst> I disagree... releases have bugs; ripping out detection of error states in releases would be bad 09:59:56 <|amethyst> we have certainly had players hitting assertions in stable releases 10:00:13 <|amethyst> _check_buggy_deck is a little unique because it actually tries to "fix" the problems 10:00:15 |amethyst: the problem is you can't ever catch all possible error states ... and assertions are one thing, this function takes it a bit further :) 10:00:40 <|amethyst> what would be gained by removing it for every release and re-adding it in every alpha? 10:01:05 |amethyst: nothing -- but in a TDD project, it wouldn't be part of the main codebase 10:01:06 <|amethyst> (where "removing" could mean just "not calling") 10:01:44 stuff like that should be in a test suite (crawl already even has a stress test) 10:01:48 <|amethyst> and when a player does get into a state that would trigger the check? 10:01:57 -- i'm not saying it's worth changing, i just find it occasionally hilarious 10:01:58 <|amethyst> you can't test every possible combination of doing things to a deck 10:02:19 |amethyst: a player wouldn't get into that state if tests had very good coverage 10:02:51 |amethyst: well true, but the fact that the code is even there indicates to me that the underlying system could be implemented in a less-easily-breakable-way 10:03:20 <|amethyst> I think in a complex game like crawl, most of the tests will not have good coverage 10:04:12 <|amethyst> A test isn't likely to find a bug that triggers, for example, when you start worshipping a god on the first level of pan, abandon the god on another level of pan, and then descend further 10:04:34 <|amethyst> I do agree that the deck stuff is kind of error-prone 10:05:06 well, you have to start from the ground up with tests anyway -- i'm not saying it's even possible ;) 10:06:34 Blade: re Vaults 10:06:53 Blade: I'm designing something where lots of small rooms are populated from minivaults 10:07:29 Blade: https://crawl.develz.org/wiki/doku.php?id=dcss:brainstorm:dungeon:branch:vaults 10:08:23 Blade: It's basically a combination of a lot of the ideas on that page -- big doors, and most monsters / loot / stairs are in the boxes (with patrolling bands in the corridors) 10:08:24 mumra: I think generating all monsters in rooms is a pretty bad idea... makes getting from room to room completely boring 10:08:44 elliptic: Yeah there are patrolling guards in the corridors 10:08:55 elliptic: i wanted to have bands of guards "stationed" outside some doors as well 10:09:16 ellptic: also, I'm going to have a trigger that, now and then, opens up a random door and wakes the monsters inside 10:10:06 -!- Pingas has joined ##crawl-dev 10:11:02 mumra: there are rooms everywhere in crawl, I don't see any real appeal to making vault levels just full of them... 10:11:19 IMO more ordered layouts would be cool, but you need more open space 10:11:47 ok, so you disagree with most of that wiki page? 10:11:56 mumra: it'd be cool to have a trap that opens a door and then sounds an alarm 10:12:16 mumra: no, I just prefer the layouts with more open space, like example 1 and 2 10:12:22 bringing corridor guards and also a room 10:12:43 elliptic: oh right, that's fine, the layout generator i've made can do either 10:12:44 <|amethyst> example 1 is similar to mumra's layout, isn't it? 10:13:03 |amethyst: note - I drew those examples 10:13:21 examples 3 and 4 seem too much like layout_roguey to me, is all I meant to say 10:13:32 <|amethyst> mumra: I only saw the ring layout in the version I was playing with yesterday 10:13:53 |amethyst: Example 1 is what I based the layout generator on, except the walkway width is randomised, and there are less rooms (but slightly larger), and the positioning of the rooms is randomsied 10:14:13 -!- vadatajs has quit [Ping timeout: 246 seconds] 10:14:25 open space is sort of vaults's thing, yeah 10:14:31 elliptic: yeah, a single rare encompass vault that looks that way is ok, having an entire branch with nothing but these would be terribly boring 10:14:44 so is the monster set going to change at all? 10:15:07 most of Vault's appeal is in the space between rooms 10:15:44 kilobyte: that's mainly because there's never anything interesting in the rooms 10:15:59 mostly V feels like boring free xp right now 10:16:00 |amethyst: about monster, your edits in dcss010 install "monster-stable" while Gretell expects "monster"; I just deleted the latter and symlinked 10:16:02 kilobyte: the wiki page is talking about giving meaning to the name "vaults" by actually having interesting stuff in the rooms ... 10:16:03 any change to that sounds good to me 10:16:04 @?cacodemon 10:16:04 cacodemon (082) | Speed: 10 | HD: 13 | Health: 78-116 | AC/EV: 11/10 | Damage: 22, 22 | Flags: 05demonic, see invisible, lev, regen, !sil | Res: 06magic(156), 11elec+++, 09poison+++, 04rot, 13neg+++, 13torm | Vul: 08holy++ | XP: 1895 | Sp: b.energy (3d20), slow, polymorph other, confuse. 10:16:37 <|amethyst> @?-version 10:16:38 Monster stats Crawl version: 0.10.1-54-ga41a5f4 10:16:47 <|amethyst> ah 10:16:54 Blade-: they're no different from deep[ish] D, except for the layout 10:17:15 D has more vaults, ironically enough 10:17:27 and it kinda gets away with things because it has so many different layouts 10:17:34 V feels like 7 levels of exactly the same thing 10:17:38 kilobyte: so the plan is to shrink V to 5 levels, so that means 4 levels before the end... I'd suggest having about half of them do something ordered like mumra is working on and half of them be current V layout (but with more vaults probably) 10:17:57 kilobyte: exactly, late D and V are pretty much exactly the same, it would be really nice to have a distinct feel to V 10:18:26 V already has a pretty distinct feel really, it is just seven levels long without much variety 10:18:30 elliptic: my layout will also do a "chaotic city" style like the current one, except using subvaults inside (some) rooms 10:18:54 perhaps we could raid, say, DoomRL, for city layout ideas? 10:19:00 elliptic: the weights of how many subvaults get used or how many rooms are left empty can be tweaked very flexibly 10:23:18 Unstable branch on crawl.akrasiac.org updated to: 0.11-a0-1007-ge138055 (32) 10:24:29 I think one issue with distinguishing D and V monster sets better is that we don't actually have that many high-level threats to pick from, and trying to use different monsters in the two of them means even less variety in each 10:25:26 what about qwarves? 10:25:40 it seems like they'd make a decent late-game threat 10:25:52 Eronarn: something like that trap idea - evilmike had an idea for stationary "vault sentinel" monsters that don't actually ever damage the player, but try to tip battles against you with effects like alarm, or buffing monsters, etc. 10:26:26 mumra: i like that idea... good opportunity for alliteration 10:26:48 'sentinel statue' that can come alive and become a 'shrieking sentinel', 'shielding sentinel', etc. 10:26:59 -!- vadatajs has joined ##crawl-dev 10:27:24 elliptic: I'm experimenting with trying to make combinations of lower-level monsters actually threatening by vault placement ... e.g. you come down a staircase and are surrounded by big kobolds with reaching weapons 10:27:45 elliptic: i think we need new high level monsters :) 10:27:55 big kobolds are popcorn by v 10:28:17 big kobold annihilator :p 10:28:19 (jk) 10:28:24 Blade: someone did mention dwarves already (it might be on the wiki page) - the problem with dwarves is that currently they're pretty uninspired 10:28:26 @??big kobold 10:28:26 big kobold (08K) | Speed: 10 | HD: 5 | Health: 17-38 | AC/EV: 3/12 | Damage: 7 | Res: 06magic(20) | Chunks: 09poisonous | XP: 97. 10:28:45 eronarn: yeah... currently we have dragons, giants, and sometimes *taurs (depending on character)... and uniques 10:29:07 vampire bands 10:29:07 elliptic: but a whole bunch of them, in combination with a sentinel statue that was hasting them, combined with boris just walking through the door ... 10:29:07 other stuff like orc packs or ogre packs are pretty weak by late D/V 10:29:18 mumra: adding in boris is cheating :P 10:29:34 heh :) 10:29:44 golems would be good if we could get them to be fun to fight 10:29:49 slime creatures can be deadly if you're inexperienced 10:30:09 Eronarn: what about a golem with perma-100% reflection? 10:30:15 silme creatures generate pre-lair sometimes, we are talking about D:20+ 10:30:35 Eronarn: and immunity to refrig/fire/poison/, so clouds don't work too 10:30:38 kilobyte: i think they should all have some effect, but also they need to not be slow 10:31:16 !lg * killer=~slime creature place=vault 10:31:16 112. Lightli the Metallomancer (L17 DDEE), worshipper of Makhleb, mangled by a large slime creature on Vault:5 on 2012-04-01, with 132015 points after 41689 turns and 2:39:47. 10:31:18 kilobyte: by D:20+ I think monsters need to be ranged or >10 speed to be serious threats 10:31:43 Slime creatures are great ... I always see them as V monsters for some reason 10:31:57 (I didn't realise until looking at the code that D/V monster weights are actually identical) 10:32:07 there are other movement abilities... charges, teleports, things that can ranged immobilize you 10:32:15 has anyone considered representing hunger as a bar so you actually know how close you are to hitting the next stage 10:32:56 Eronarn: With the sentinel statues, you can buff golems, blink them closer, that kind of thing - make them more of a threat without directly giving them abilities 10:33:47 @??crystal golem 10:33:48 crystal golem (038) | Speed: 7 | HD: 13 | Health: 102-132 | AC/EV: 22/3 | Damage: 40 | Flags: 11non-living, see invisible | Res: 13magic(immune), 04fire+++, 12cold+++, 11elec+++, 09poison+++, 12drown, 04rot+++, 13neg+++, 13torm | XP: 615. 10:33:52 mumra: yeah, the totems in brogue work well for that 10:34:03 what about giving these such 100% reflection, in line with crystal walls? 10:34:07 also, can have the statues only come alive if things that help them are nearby 10:34:18 kilobyte: definitely in favor of crystal golems being reflective, yes 10:34:20 kilobyte: you'd just kite them with a polearm then 10:34:25 but they need to be speed 10 10:34:30 Eronarn: That could be good, make them indestructible until they're awake too 10:34:54 mumra: i wouldn't do indestructible... i'd just have them all over the place, so only people with unlimited disint could kill them all 10:35:18 (the non-living ones just disappear, the living ones wake up with a shout and take non-brittle statue damage from disint) 10:35:18 Eronarn: or summons, or swiftness and a polearm 10:35:46 (elliptic raised a good point about polearms here) 10:36:57 Eronarn: My overall idea was to have a very restricted monster set in the V corridors (golems, slime creatures, vault guards, little else), and more diverse groups of monsters locked away in the vault rooms; then sentinel statues both in the corridors and in the rooms to continually make everything more threatening 10:36:58 well crystal walls only reflect bolts of fire/cold/elec, which the golems are immune to anyway 10:37:38 so could make the golems reflect those just for flavour i guess 10:37:53 making all 4 levels of V golem-themed seems like a bad idea to me 10:37:54 mumra: that sounds really boring. imo spice up vault guards - like have caster ones - and give them better, group-based, defensive AI 10:38:03 they can be described as polished in addition to being crystal 10:38:16 and have whatever magic shields of reflection have 10:38:39 elliptic: there are lots of other monsters in the rooms - and since the rooms contain the stairs, you'll have to deal with those threats 10:38:54 <|amethyst> can' 10:38:59 mumra: nobody wants to fight multiple golems every level for four levels 10:39:27 <|amethyst> can't an ash worshipper or someone with m.mapping avoid most of the rooms in the layouts that have corridors? 10:39:47 making them harder to kill doesn't make a speed <=10 melee monster any more interesting 10:39:47 |amethyst: sure, I don't see a problem with that 10:39:48 |amethyst: yes, just like on any level of the dungeon 10:39:53 so it doesn't seem very worthwhile to me 10:40:01 MarvinPA: they should be speed 10 :) 10:40:02 MarvinPA: yeah 10:40:08 <|amethyst> in most of the dungeon the things outside of rooms are the same as the things in rooms 10:40:12 Eronarn: 10 is less than or equal to 10 10:40:13 :P 10:40:16 elliptic: so it's not a problem with using golems, it's the fact that they're not interesting, right ...? 10:40:19 10 is much much better than 9 at least 10:40:36 |amethyst: well the idea is to make Vaults unique 10:41:04 <|amethyst> I guess the loot is an encouragement not to dive 10:41:08 |amethyst: just like polished branches like Spider or Slime have focussed monster sets and unique layouts 10:41:32 |amethyst: loot and XP is the reason why most players clear every level 10:41:44 |amethyst: elliptic is not the 99% ;) 10:41:56 mumra: partly it is that they are one of the most boring monster types in the game, and partly it is that they annoy certain playstyles far more than others 10:42:46 elliptic: what playstyles do they annoy sorry? (i've never been annoyed by golems but i've only ever gotten casters as far as V) 10:42:49 most conjurers: can't kill them with spells or need to kite them forever doing very low damage with IMB or throw icicle 10:43:05 i've never had a problem killing them with spells 10:43:09 most melee chars: just kill them 10:43:24 (especially if they have polearms) 10:43:45 <|amethyst> D&D golems are supposed to be anti-caster, though 10:43:55 <|amethyst> not that crawl needs to follow D&D 10:43:57 mumra: they are immune or nearly immune to 90% of attack spells 10:45:59 <|amethyst> fast generate_awake golem with M_MAINTAIN_RANGE, so everyone is equally screwed 10:46:02 <|amethyst> :P 10:46:33 elliptic: well, their frequency can be kept low (and >50% of monsters on a level will be inside the rooms anyway) 10:47:10 but what do they add? 10:47:42 are you arguing they should be taken out of the game altogether? 10:48:15 if they're broken, maybe they can be fixed 10:48:17 aside from electric golems, yes 10:48:34 D tends to be about fighting monsters in rooms, in corridors, or in semi-open areas after coming from behind a corner 10:48:45 having V be the same looks like a big step backwards to me 10:49:36 improving golems would be great, but until that happens I don't see them improving the V monster set 10:50:25 kilobyte: have you read the vaults wiki page out of interest? the "rooms" idea had good feedback from evilmike and lemuel 10:52:11 kilobyte: Actually it *is* more about open space. D has a lot of corridor fighting. My new V layout has *no* 1-tile-wide corridors. So you are always fighting in open space. 10:53:09 kilobyte: It's just that as you do so, you're cracking open these ancient vaults (with big doors), and stuff spilling out of them (the only time you'll really fight *in* a room is when you go downstairs, or teleport) 10:53:13 mumra: it'd be cool to have a trap where walls change into grates in a 1-wide corridor, exposing two boxes with reaching enemies inside 10:53:24 but only when you're in the middle of it 10:53:37 reaching doesn't go through grates 10:53:51 or one where the corridor sides seal up, but the boxes on either side open 10:53:59 I thought that was going to be change though? 10:54:38 mumra: lemuel's only comment there is that "it would reward stealthy play" -- but I fail to see how stealth would be different than in D, or even current V 10:56:04 kilobyte: well lemuel started with "I like the general approach." -- but you're right, I don't quite see his point about stealth 10:58:36 kilobyte: anyway, there was no *bad* feedback for the overall idea, i'd just be interested to know what you think would be good to make vaults more interesting 10:59:42 I'm afraid I have no good ideas right now, I just think this one is worse than status quo. 11:00:29 actually, I do have an idea -- have different branches in rotation with V; but that's a rather big project, and I have my plate full 11:00:35 I think making the rooms in V interesting (they can contain all the items, stairs, etc, and some can have minivaults) is good, and making the layout more ordered sometimes is also good, but I just don't see any real advantage to limiting the monsters that are outside the rooms... you should have to face a full assortment of enemies to get from one room to the next 11:02:59 mumra: by the way, one thing to think about while working on layouts -- currently the most interesting rooms in V are those with multiple doors 11:04:20 kilobyte: you realise that if a new branch were designed to replace V, it would be much better than V, because new branches have much higher standards 11:04:35 since then there is the option of retreating through a different door from where you entered, and also you can't guard multiple doors at once 11:07:50 mumra: do they? there have been exactly two new branches in the last few years... Shoals took a rather long time to develop and Spider is still in an extremely rough state 11:08:47 spider has been getting very good feedback to say it's an extremely rough state ... 11:09:07 but these branches are both very unique with an interesting monster set and unique level design 11:09:15 where Vaults has: city layout, D monster set 11:09:22 (and V:8) 11:09:37 spider is shaping up well, but it needs a lot more work 11:09:54 my only point is that there isn't much history with adding new branches to modern crawl 11:10:07 V is also a far more important branch than shoals/spider 11:10:13 fair points 11:10:38 V is, challenge games aside, a mandatory branch 11:10:47 every game gets V and V contains a huge amount of XP and subbranches 11:10:54 I think that's why people think it should be far more interesting - for something so important it has surprisingly little content 11:11:29 mumra: yeah, and that's why ideas for improvements are needed 11:12:19 this doesn't mean any random idea is good -- remember Sturgeon's law: "90% of everything is crap" 11:12:43 rooms with empty corridors would make a good encompass vault, though 11:12:56 kilobyte: yeah i know, but a whole lot of people seemed to like the rooms idea (wiki page, tavern thread, last night on here) 11:13:05 something as specialized as shoals (incredibly painful to explore, even without levitation) or spider (ghost moths from the first level, no non-animals) would make an awful V 11:13:32 mumra: I like the rooms idea too! but not all of it 11:14:25 er, that should have been "even with levitation" :P 11:15:12 elliptic: what is it you didn't like, was it basically stuff about corridors? we got sidetracked by golems which are basically unrelated ... 11:16:16 elliptic: you mentioned multiple doors, that can be achieved also, in a couple of ways: the city layouts can have rooms with multiple doors, and in the more closed layouts can have rooms joined onto other rooms 11:16:47 elliptic: my original thoughts were "no corridors", it was only evilmike who objected to that 11:16:56 what's the current upside of V? It has plenty of open space, but is not entirely open space (like the "all floor with some water" or big_octagon layouts, which are boring) 11:16:57 mumra: my main criticism was restricting the monster set outside of rooms 11:17:00 (I mean, no 1-tile-wide corridors) 11:17:24 I'd go the other way: no monsters in rooms 11:17:45 -!- Wensley has quit [Ping timeout: 246 seconds] 11:18:24 elliptic: it's easy to tweak the monster set anyway (although I think that D/V monster sets should be differentiated) 11:18:48 -!- edlothiol has quit [Ping timeout: 245 seconds] 11:19:05 I tend to agree with kilobyte that the monsters are more interesting outside of the rooms, though I think some "rooms" should really be minivaults (with monsters and/or loot as usual) 11:19:10 kilobyte: you're suggesting basically a bunch of loot rooms guarded by monsters? could use the sentinels idea and have alarms inside the rooms to stop people just ctele-ing from one room to the next 11:20:06 mumra: I was mostly joking -- rooms with no monsters have little purpose; but there's no need for many rooms 11:20:15 -!- lord-naughty has joined ##crawl-dev 11:20:37 kilobyte: the actual layout I've designed has much less rooms, it's a range of about 8-15 11:20:45 ie, in the current V layout, having no rooms but solid boxes would be an improvement 11:21:03 kilobyte: I noted on the wiki those examples had way too many rooms. I was going to draw up some better ones but I just wrote some LUA instead :) 11:22:05 i don't really understand why monsters are necessarily more interesting in the open ... that's no different to opening a room and monsters spill out, suddenly you're in the open with them ... 11:23:35 -!- edlothiol has joined ##crawl-dev 11:24:02 mumra: two important things: 1) they don't spill out instantly and 2) in open space, they'll be at edge of LOS when you first see each other 11:24:56 I can't recall more than one or two interesting situations created by the open space in V, for what it's worth 11:25:06 we aren't talking about completely open space either (that's for the octagon levels), just mostly open space 11:25:13 the only times I've really felt threatened there are in moments of extreme stupidity 11:25:31 the open space doesn't do much to make it interesting, especially not 7 times in a row 11:25:52 Blade-: V is being shortened to five levels, so s/7/4 11:26:08 okay cool 11:26:11 elliptic: monsters at the edge of LOS is usually pretty easy for casters to deal with, potentially much harder for melee; so you surely want a mix 11:26:18 that helps the problem but doesn't yet remove it 11:26:50 mumra: not really... lots of ranged stuff in V, casters often have difficulty just trading attacks with them 11:27:11 is the xp going to be redistributed to anywhere or is that considered unnecessary 11:27:17 (there is too much xp in midgame anyway) 11:27:42 there was the idea of giving xp when you pick up a rune 11:27:53 that sounds cool 11:28:32 mumra: also, many spells aren't full LoS range 11:28:48 <|amethyst> is the intent to add new content to replace the various cut levels? 11:28:54 <|amethyst> or to make the game shorter? 11:29:02 either sounds good 11:29:06 shorter 11:29:06 or both 11:29:21 |amethyst: I think shorter, though there was also talk of adding in a one-level dwarf branch somewhere 11:29:27 new content tends to be put into rotation rather than being an addition 11:29:33 elliptic: a melee character has to close the entire distance to the threat and then kill it 11:30:02 elliptic: and surely it's boring if everything you ever meet is at full LOS range? 11:30:03 mumra: and usually has much better defenses, and can dodge around a building to create a not-very-good corner 11:30:28 my melee dudes are almost always less threatened by ranged threats than my ranged dudes rae 11:30:29 are 11:30:30 mumra: again, we aren't talking about completely open octagon levels... 11:32:09 <|amethyst> does shorter game also mean things like god wrath are shortened? 11:32:55 making god wrath not be 50k turns long for some gods would be good 11:33:27 elliptic: of course, to keep our rules, it'd be packing the same amount of wrath into a shorter time :p 11:33:40 -!- DaneiTWO has joined ##crawl-dev 11:33:44 -!- Danei has quit [Ping timeout: 252 seconds] 11:34:05 god wrath right now is incredibly tedious 11:34:19 it is less tedious if you don't wait it out in temple 11:34:25 -!- lord-naughty has quit [Quit: Leaving.] 11:34:38 I've never done that 11:34:54 it's just boring to have to deal with wrath effects once every long time for a really long time 11:35:17 elliptic: basically i think there can be a balance of tactical situations with the rooms - sometimes you have to deal with things up close, sometimes you're meeting things at longer distance (corridor guards, or spontaneously opening rooms) 11:36:00 mumra: the rest of crawl is full of situations where you open a door into a room and there are scary monsters there 11:36:37 elliptic: i just think "vaults" should imply, well, "vaults" -- i.e. ancient storage rooms containing horrors or wonders 11:37:15 the open chaotic levels sort of imply the security has got a bit lax over the years :) 11:37:15 -!- Mottie has joined ##crawl-dev 11:37:40 it'd be good to have vaults that have been forcibly busted open, and then something really powerful nearby 11:37:43 sure, some of the rooms can be interesting 11:38:04 but don't let that detract from putting interesting stuff (monsters) outside of them too :) 11:39:32 elliptic: sure, there has to be a decent threat in the corridors (that was part of the plan, the "no monsters at all outside the boxes" idea is clearly bad) 11:40:42 more than a "decent threat" IMO... things that aren't downright murderous are a waste of time :P 11:41:17 elliptic: What I'm talking about is a designated "vault security force" that isn't just "everything from D" 11:41:25 e.g. yaktaurs 11:41:47 mumra: using a really limited monster set usually doesn't make something more interesting 11:42:17 along with other stuff 11:42:44 yaktaurs seem like a weird thing to pick on since they work better in V than in D, too 11:43:07 basically I think that the monsters inside the rooms and out in what you are calling the "corridors" (which I think should be much more open than that word suggests) should be drawn from the same set 11:43:25 MarvinPA: It was a suggestion on the wiki page that I didn't really agree with tbh 11:44:18 use redone vault guards instead of yaktaurs 11:44:20 In the layout I called them "walkways" not corridors, yes corridors is the wrong word 11:44:47 if we want to distinguish V and D monster sets somehow, that's great, but I don't really believe the layout needs to be coordinated with this 11:45:13 <|amethyst> the "rooms", being subvaults, would have mostly hand-picked monsters, wouldn't they? 11:45:18 also? a guardian spirit monster sounds like it could be a good add for vaults 11:45:32 |amethyst: some of them could, but plenty of vaults don't handpick monsters too 11:45:46 <|amethyst> true 11:46:05 Eronarn: does this have anything to do with amulet of guardian spirit or are you just trying to be confusing ;P 11:46:27 elliptic: i think it should, yes - have it take damage instead of other monsters 11:46:30 <|amethyst> when it takes damage, a percentage of that is subtracted from your MP 11:46:31 elliptic: Actually currently things are exactly how you're saying - subvaults can define their own monsters if they want, but a lot of them are just empty "furniture" vaults and they get whatever the overall V monster set is 11:46:36 "The guardian spirit shields the foo!" 11:46:43 mumra: yes, I know... 11:47:04 elliptic: No, I'm saying this is how I've set things up with the layout. 11:47:41 was there anything to set up? that's how vaults work in general... 11:48:01 anyway, what I'm saying is that non-rooms should have monster variety also 11:48:25 rather than being 12350897 *taurs and vault guards and golems, which is what you seem to want 11:48:42 elliptic: We're talking at slightly cross purposes 11:49:36 eronarn: could be interesting, though you'd also need to figure out how the guardian spirit attacks you itself 11:50:54 elliptic: Sure the native monster set can be diverse, I just think it should be differentiated from D *somehow*, there was an idea on the tavern for "artifical life" which I liked how it suited the idea of an ancient magical security system for an underground vault 11:51:08 elliptic: I am not saying this idea is the be all and end all of my layout 11:51:40 mumra: well, I see monster set improvements as independent from layout improvements 11:52:00 <|amethyst> what about more demons in vaults? 11:52:07 noooo 11:52:13 elliptic: only if there's some rough agreement on long vs close range 11:52:17 |amethyst: wouldn't they make more sense in D, where the hell/pan/abyss portals are? 11:52:24 kilobyte: what do you mean? 11:52:29 (I'm strongly for long range, mumra seems want to have close only) 11:52:46 kilobyte: no, definitely a mix 11:53:20 mumra: rooms only means monsters never get to fight you while being out of range of most effects 11:54:09 yaktaurs, crossbows, stone giants... all have full LOS range, while most good spells are far shorter 11:54:26 kilobyte: That's why monsters are needed in the corridors 11:54:36 <|amethyst> walkways 11:54:43 kilobyte: I'd say that like most places, V should have a mix of ranged and nonranged monsters 11:54:47 -!- vadatajs has quit [Ping timeout: 260 seconds] 11:54:50 kilobyte: and also, another room could have opened across the map, and monsters will wander into your LOS from a different direction 11:54:57 |amethyst: thanks ;) 11:55:04 walkways still sound really narrow to me 11:55:13 kilobyte: Also, the doors aren't soundproof 11:55:14 I want open space :P 11:55:25 <|amethyst> how open? 11:55:28 like current V 11:55:49 elliptic: my point is, with the rooms proposal, there's never a situation the yaktaur is at the edge of LOS while you have to get closer 11:56:08 current V layout I think has a good mixture of walls and open space... it is just the same for seven levels, and also the same layout that appears occasionally elsewhere 11:56:12 <|amethyst> elliptic: the ring layout seems to have larger open spaces that V 11:56:19 kilobyte: If there are yaktaurs in the corridors then this isn't a problem 11:56:20 <|amethyst> s/that/than current/ 11:56:39 |amethyst: not if all or most monsters are in rooms 11:56:50 <|amethyst> no one is putting all the monsters in rooms 11:57:02 elliptic: I can make the layouts have comparable amounts of open space to current V (but it's possible to have more variance in *either* direction) 11:57:03 (wimpying out, brother with his wife and stuff...) 11:57:04 <|amethyst> and the idea is that distance rooms open 11:57:05 |amethyst: not the example on wiki 11:57:23 example on wiki has a single width 6 ring corridor 11:57:25 why not just code it and try it out 11:57:33 Eronarn: I've already coded it 11:57:43 mumra: i mean with the monsters and stuff too 11:57:47 Eronarn: Yes 11:58:04 <|amethyst> elliptic: I just generated a vaults level and nowhere do I have anything near 40x6 of empty space 11:58:07 oh, okay 11:58:15 |amethyst: the 40x is irrelevant 11:58:42 <|amethyst> elliptic: so more like a 9-wide corridor? 11:58:51 vaults has lots of 10x10 and 8x15 and such 11:58:52 Eronarn: Well, it's just an experiment to test the layout, I don't have a lot of subvaults, and I only tweaked the monster set to see how that part of the code worked 11:59:32 elliptic: 6 width is quite open really - as open as 50% of spaces in current V 11:59:45 |amethyst: yeah... also V has many different open spaces as opposed to one predictable open space 12:00:13 elliptic: The rooms are also much bigger in the implementation 12:00:23 anyway 6 width isn't terrible, and that's much more than the example titled "walkways" on the wiki (width *3*) 12:00:42 which is why the term "walkway" scares me 12:00:48 -!- lord-naughty has joined ##crawl-dev 12:00:49 elliptic: I have 9x9 rooms at the moment, but I'm trying to make it use any size of minivault - you could have a few really large rooms 12:01:11 mumra: I'm talking about non-room open spaces 12:01:59 yep, I was just pointing out that the rooms in those examples were way too small to be interested 12:02:02 s/ed/ing 12:02:49 6x6 is large enough for a room... some could be larger of course, but I don't see anything wrong with 6x6 12:03:38 basically there's this issue I'm having in Lua with vault maps, it's preventing me from using variable sizes of room. 12:03:56 Can anyone help interpret a stack trace? 12:04:18 <|amethyst> mumra: probably 12:05:50 http://pastebin.com/FgeVfEzr 12:06:21 That's the stack trace, when I try to call reuse_map 12:06:39 This is after I added that flag we talked about 12:07:03 <|amethyst> sadly, that's not a very useful stack trace 12:07:19 Does anyone online today understand the Ziggurat lua code, specifically dgn.resolve_map and dgn.reuse_map? 12:08:21 I'm trying to use them to handle determining the subvault size and placing it in a room, but I'm getting lockups or crashes whatever I try, even though I'm apparently only doing exactly what the Ziggurat code is doing 12:08:25 -!- vadatajs has joined ##crawl-dev 12:09:35 <|amethyst> wait 12:09:45 <|amethyst> what's this crap in dgn_map? 12:10:04 <|amethyst> char buf[2] = ""; buf[0] = gly; lua_pushstring(ls, buf); return (1); 12:10:26 <|amethyst> am I mistaken, or is that just asking for a segfault? 12:10:44 hmm, my crash was a segfault, right? 12:10:52 <|amethyst> yeah, from lua_pushstring 12:11:03 <|amethyst> can't tell whether you were actually in dgn_map though 12:11:22 <|amethyst> because most of your stack frames are unannotated, because of optimisation 12:11:41 <|amethyst> mumra: sec 12:12:04 |amethyst: Should I do a debug compile instead or something? 12:13:15 <|amethyst> mumra: try changing the two lines char buf[2] = ""; buf[0] = gly; at l.432 to one line char buf[2] = { gly, '\0' }; 12:14:06 <|amethyst> mumra: and see if that fixes it; also, yes, a debug compile would be useful 12:15:22 which file is it? there isn't an l_dgn_map 12:15:51 <|amethyst> sorry, in l_dgn.cc, function dgn_map() 12:22:18 -!- lord-naughty1 has joined ##crawl-dev 12:23:54 -!- lord-naughty has quit [Ping timeout: 246 seconds] 12:24:48 |amethyst: Didn't fix it, now I'm compiling with FULLDEBUG but it'll take a while. 12:26:15 <|amethyst> hm 12:27:27 elliptic: so, for the "square" layout, it already has a control for the width of walkways .. so i can set this as e.g. random_range(5,10) 12:28:08 mumra: The layout looks very nice, thanks for working on it! 12:28:30 mumra: sounds good :) 12:28:31 re: enemies, I like the artificial guardians idea, but of course that means lots of work making the set interesting 12:29:04 <|amethyst> hm... I think I'm still going to commit that fix... looks like we never saw it triggered because no one uses the three-argument version of dgn.map() 12:29:20 close range vs long range, I tried out a long ranged -only set, and it was terribly exhausting to play.. a mix is definitely good, and what I think works in open spaces is *groups* of monsters 12:29:50 Keskitalo: thanks 12:30:58 Keskitalo: yes, I think bands of monsters are best, this is probably why some people suggested Yaktaurs, although I agree with (was it kilobyte?) that they should be D, and V should have ... *something* else 12:31:26 -!- mikee_ has quit [Quit: I love my HydraIRC -> http://www.hydrairc.com <-] 12:31:35 bands of monsters don't need to all be the same type 12:31:38 well, I see it the other way around, yaktaurs for V and not for D 12:31:48 elliptic: thats what I though 12:31:58 elliptic: That's wasn't my implication :) 12:32:01 this is something that could be good elsewhere too... clump monsters betters 12:32:06 *better 12:32:15 elliptic: yeah, that would be great 12:32:32 elliptic: In part that was the idea of the rooms - it makes it really easy to add new subvaults containing hand-picked *groups* of monsters 12:32:48 elliptic: It's what I really liked about spider, how different combinations of monsters worked together 12:33:24 well, I don't think hand-picking specific combinations is necessary (or necessarily a good idea... makes them more predictable) 12:33:31 -!- ais523 has joined ##crawl-dev 12:34:49 mumra: one random idea for your layouts for monsters outside the rooms: put clumps of randomly picked monsters at intersections/corners of walkways 12:35:52 |amethyst: http://pastebin.com/H6TD4wiK 12:35:58 that's the crash after a FULLDEBUG compile 12:36:28 elliptic: I wanted to have groups guarding outside the doors; intersections are possible too 12:36:46 -!- dtsund has joined ##crawl-dev 12:37:20 that's possible too, yeah... especially for rooms with more treasure inside 12:38:13 <|amethyst> mumra: hm 12:38:53 <|amethyst> mumra: the stack frames still seem to be mangled :( 12:39:16 <|amethyst> mumra: how did you build? 12:39:49 elliptic: by "hand-picked" I mean some overall themes are hand-picked (kind of like Zig floors), then you, say, select 2-3 random monsters from that group, and generate a room with a random quantity of each of those monsters. 12:40:18 |amethyst: make FULLDEBUG=y 12:40:23 <|amethyst> mumra: oh 12:40:26 <|amethyst> mumra: try make debug 12:40:40 <|amethyst> mumra: that includes FULLDEBUG and also NO_OPTIMIZE 12:40:43 ah 12:40:55