00:08:42 nikheizen: that was a fun time 02:35:04 if i wear fencer's gloves will i riposte with longbow bashes 02:42:10 minmay: yes, apparently 02:43:24 yes 02:43:28 from experience 02:46:41 so i'm in shoals and pressing f. a lot right 02:46:45 because i don't want deep water to eat my ammo 02:47:09 this also means that if i get caught in a net i'll waste a bunch of turns because 'f' prints a message instead of bringing up the targeting interface, and '.' waits instead of struggling against the net 02:52:02 Monster database of master branch on crawl.develz.org updated to: 0.21-a0-110-g8e1637b 02:56:18 okay 07:06:25 -!- amalloy is now known as amalloy_ 09:38:07 Anyone know the maximum size for vaults? And does ORIENT affect this (encompass vs float, etc)? 09:39:11 Also, what exactly does encompass do? 09:50:06 <|amethyst> encompass means the vault is the whole level 09:50:56 <|amethyst> meaning a max of 80 x 70 for encompass vaults 09:52:00 <|amethyst> other vaults should probably be smaller so there's room for the layout to place its stuff 09:53:07 <|amethyst> minivaults (without an ORIENT: line) should be smaller, since they're placed on top of an already-generated level (after the layout) 09:53:34 <|amethyst> there used to be a 12x12 limit for minivaults; that is gone now, but you should still try to keep them small 09:54:06 <|amethyst> otherwise the level generator might have problems finding a place to put them that doesn't cut off parts of the level 09:57:57 amethyst: thanks! do you know if there's something I can do to my vault to give it high priority (level generator places it first and builds everything else around it) 09:58:11 for reference, I'm making a river vault 10:01:00 rough draft: https://pastebin.com/t3nxACEP 10:01:20 I was trying to test it in wizmode, but it crashes my game half the time 10:03:55 <|amethyst> hm 10:03:58 <|amethyst> testing it how? 10:04:14 <|amethyst> &L or &P ? 10:04:23 &P, it was working sometimes when I used encompass 10:04:32 <|amethyst> but then it's your whole level 10:04:44 <|amethyst> hrm 10:04:44 yeah, guess that's a bad idea, lol 10:05:05 is there a different limit for float or the directions? 10:05:19 <|amethyst> the only real limit is 80x70 10:05:31 <|amethyst> might be useful to use a fulldebug build (make debug) to see what is causing it to fail 10:06:47 <|amethyst> hm 10:06:54 oh, I'm dumb. Could have tried the console version 10:07:05 yep, just gave me an error 10:07:44 map too large - 59-14 (max 53x46) 10:07:48 <|amethyst> oh 10:07:57 <|amethyst> I did not know about that limit 10:08:01 I always forget about console 10:08:16 sorry for wasting your time 10:08:51 but it should be fine if I leave it as float and just reduce the length to 53, you think? 10:09:03 <|amethyst> yeah 10:09:07 mmk 10:09:20 <|amethyst> it *might* help (result in fewer vetoes) to give it the "transparent" tag 10:09:56 ok will do :) 10:09:56 <|amethyst> if you don't, the whole thing will count as wall for the purpose of level-connectivity checks 10:10:59 <|amethyst> which is usually fine, but for a vault this big that could be a problem 10:11:00 <|amethyst> hmm 10:11:10 <|amethyst> so, since you don't have no_rotate 10:11:29 <|amethyst> you might want to restrict it to 46x46 10:11:41 oh 10:12:10 <|amethyst> otherwise it will be too big when it gets rotated, and it will retry, so you'd have the effect of no_rotate anyway, but more expensive 10:12:22 good point 10:13:21 <|amethyst> apparently the limit is: 10:13:49 <|amethyst> for float, centre, and corner orients, 2/3 in each direction of the level size, so 2/3*(80x70) = 53x46 10:15:08 ty, I'll just keep it as float then and reduce to 46 10:15:31 <|amethyst> for edge orients, 2/3 in the direction of the orient, the full level size in the direction parallel to the edge. So a map with ORIENT: east (after rotation) can be 53x70, and ORIENT: north 80x46 10:15:43 <|amethyst> not sure about minivaults 10:17:03 <|amethyst> (I don't remember if they get MAP_FLOAT orient, or MAP_NONE or something else 10:17:07 <|amethyst> ) 11:47:04 hm was it crashing because of the map too large error? 11:47:12 seems like that ought to be checked somewhere else 11:48:46 yeah, although this was in wizmode using &P 11:49:17 interestingly those savefiles I was using are bugged now, they crash if I try to load them 11:52:55 ok, that error will show up when it tries to regenerate the des files too (and cause crawl to fail to start) 11:53:23 that's more the behavior I was expecting 11:53:47 not sure if that's specific to a debug build or what 12:06:27 <|amethyst> advil: normally it would try again, and this time would pick a different map so not screw up 12:06:40 <|amethyst> but with &P it has no choice as to the map 12:07:27 <|amethyst> advil: oh, it does fail at mapgen? interesting 12:08:25 <|amethyst> ah, I see, this code is in validate_map_def, not when it is placed 12:08:39 <|amethyst> which makes the situation with rotation a little weird... 12:09:29 <|amethyst> ah, it gets checked again in can_dock which would happen at placement time 12:09:54 <|amethyst> hm, no, that check is > 13:30:49 probably a dumb question, but in a vault... can I not place items in shallow water? 13:33:44 KITEM isn't working for me here. The human skeletons are dropping 100% (should only be 50%) and none of the other items are dropping. https://pastebin.com/nt9GeBX8 15:02:08 <|amethyst> Wander: delayed_decay does KITEM: ? = human skeleton 15:04:34 <|amethyst> Wander: were it not for the 50% you could do: 15:04:56 <|amethyst> Wander: : dgn.delayed_decay_extra(_G, '?', 'human skeleton', 'any good_item / any superb_item / nothing w:20') 15:05:19 <|amethyst> Wander: but I don't think that would work with 'human skeleton / nothing' because it would try to add "never_decay" to the "nothing" 15:05:52 <|amethyst> Wander: you could look at the implementation of delayed_decay and use something like that 15:05:57 <|amethyst> Wander: in dat/dlua/lm_items.lua 15:06:13 <|amethyst> Wander: something like KITEM: ? = human skeleton never_decay / nothing, blah, blah 15:07:05 <|amethyst> Wander: then : _G.lua_marker('?', function () return function_in_los("dgn.activate_item_decay", nil, false, nil) end) 15:07:22 <|amethyst> Wander: but I might have some of that wrong, so you'll have to play around 15:51:07 |amethyst : thank you very much! I didn't get the fancy thing to work, so I just ditched the 50% and used your first suggestion. Works just fine! 16:40:53 -!- frd is now known as Wieselflink 16:49:39 10 New Vaults 13https://crawl.develz.org/mantis/view.php?id=11140 by WanderingBlade 17:10:25 "Attack new target" order behaves weirdly if there is disturbance nearby 13https://crawl.develz.org/mantis/view.php?id=11141 by Lavandula 17:23:18 !tv nago kaux~~something 17:23:19 1. nago, XL27 GrMo, T:82496 requested for FooTV: telnet://termcast.develz.org or http://termcast.develz.org. 17:26:05 <|amethyst> Wander: double check that they do in fact rot 17:26:31 <|amethyst> Wander: if the marker gets screwed up for whatever reason, the corpses will last forever 17:26:50 <|amethyst> Wander: (the marker makes them start rotting when they enter LOS) 18:22:12 |amethyst : yeah, I just checked and they rot. Thanks for the reminder, though! It made me realize that there were a couple unrelated errors, which I fixed. I think they came from last-minute tweaks to the loot in a couple of the vaults. 18:22:43 I uploaded a new file with the fixes 19:03:37 -!- amalloy_ is now known as amalloy 19:22:32 -!- crate_ is now known as crate 19:23:08 has there been any discussion/progress on merging/partial merging/rejecting the gnoll change PR? 19:27:00 yeah, sorry I haven't done anything with that 19:28:18 There wasn't a lot of excitement about gnolls from the brief discussions I had with other devs, and I didn't want to take ownership of that project given that I'm working on other things right now 19:29:52 The only opinion I heard was from MPA who said that while he wasn't really excited about Gnolls, the stat locking version is certainly better than the dynamic apt version, and that he leaned weakly towards not giving them a stat choice 19:37:01 well, I could be wrong from my own playtesting, so I would be okay with trying it without stat choice or some other stat growth at the moment 19:37:14 and it (or the base stat line) could be changed later 19:44:12 yeah, we should probably just merge your stat lock patch for the time being 19:44:44 I'm not going to be able to do that merge today, but I'll try to get to it tomorrow 19:44:48 if no one else does, that is 19:45:00 alright, that's cool 19:45:50 I was mainly checking up on it to see if I needed to start working on another implementation for gnolls, or if I could mess around with some major extended reform changes I've been brainstorming 19:57:12 ??tomb 19:57:12 tomb[1/4]: Accessed from the Crypt on the third level. Three levels deep, with a golden rune at the bottom. 19:58:12 ??stone stairs 19:58:12 I don't have a page labeled stone_stairs in my learndb. 19:59:32 ??stair dancing 19:59:33 stair dancing[1/1]: The process of dragging enemies up stairs so as to kill them in smaller groups. 20:03:26 !learn set tomb[1] Accessed from the Crypt on the third level. Three levels deep, with a golden rune at the bottom. In 0.20, Tomb has its stairs replaced with one-way staircases that are like escape hatches that lead to a pre-determined destinations, thus preventing {stair dancing}. A one-way staircase that returns to the previous level can be found near by (but out of LOS) of each stair case's 20:03:27 tomb[1/4]: Accessed from the Crypt on the third level. Three levels deep, with a golden rune at the bottom. In 0.20, Tomb has its stairs replaced with one-way staircases that are like escape hatches that lead to a pre-determined destinations, thus preventing {stair dancing}. A one-way staircase that returns to the previous level can be found near by (but out of LOS) of each stair case's 20:03:32 destination. 20:03:44 oops 20:03:52 let's try that again 20:05:01 !learn set tomb[1] Accessed from the Crypt on the third level. Three levels deep, with a golden rune at the bottom. In 0.20, Tomb has its stairs replaced with one-way staircases that are escape hatches leading pre-determined destinations, thus preventing {stair dancing}. A returning one-way staircase can be found nearby (but out of LOS) of each staircase's destination. 20:05:01 tomb[1/4]: Accessed from the Crypt on the third level. Three levels deep, with a golden rune at the bottom. In 0.20, Tomb has its stairs replaced with one-way staircases that are escape hatches leading pre-determined destinations, thus preventing {stair dancing}. A returning one-way staircase can be found nearby (but out of LOS) of each staircase's destination. 20:05:10 ??tomb[2 20:05:10 tomb[2/4]: Need to go down to go up to go down. You can avoid this by random teleporting around on Tomb:2 until you land in the correct region. 20:05:57 Floodkiller: not sure if writing up your ideas somewhere (e.g. on the Doku wiki) first would be helpful or not 20:06:23 but it might be a good idea to get some kind of feedback on your full ideas before you start coding 20:06:33 yeah, I planned on it 20:06:38 cool 20:06:44 notes/brainstorming isn't quite finished yet though 20:27:44 !lm cao cv=0.20 x=vlong 20:27:44 No milestones for cao (cv=0.20). 20:27:44 !lg * cao cv=0.20 x=vlong 20:27:44 42648. [vlong=0.20.0-15-g7f486f2] Phibes the Grave Robber (L1 MuNe), slain by a jackal on D:1 (mainiacjoe_arrival_reflecting_pool) on 2017-07-10 00:27:22, with 1 point after 65 turns and 0:01:23. 20:27:58 johnstein: for some reason, cao seems to also have not updated to 0.20.1, can you look into that? 20:28:34 |amethyst: pinging you as well, if you have time to check 20:44:24 <|amethyst> gammafunk: running now (and I modified the update script that runs nightly) 20:51:12 awesome, thanks 21:24:15 -!- mngr__ is now known as daiy 22:09:40 |amethyst, what was the problem? 22:09:58 <|amethyst> johnstein: it was never added to ~crawl-dev/bin/do-update 22:10:09 <|amethyst> johnstein: (which is what gets run from the crontab) 22:10:54 doh 22:21:12 I was watching a TV and saw an old version of my vault 22:24:50 I always love reading the cwz trunk summaries through google translate 22:24:59 * Bulk bolt fixes. 22:25:03 # All gammafunk works. Too many, so omit. 22:25:31 I think it was asciiphilia who explained to me once that the words "vault" and "bolt" are the same in Korean 22:26:06 so that leads to a lot of cwz players making jokes along those lines 22:26:53 heh 22:27:20 checkers: cwz player ranting about your vault change: What? Are you going to die mana fast and angered dragon? ㅋㅋ 22:27:43 I think they're saying you're crazy since people will die to mana drain + zerked dragons 22:41:31 the korean metagame is incomprehensible 23:21:48 gammafunk: that is indeed the point 23:26:33 -!- GiantOwl is now known as Kalir 23:30:07 you'll have to explain that to them 23:30:43 recommend you type a response into google translate and post 23:30:46 what could go wrong? 23:40:20 gammafunk: it's probably also convenient that the best korean spelling for gammafunk is gammapunk. good source of jokes 23:44:14 google translate one of my dcss related dreams imo 23:49:49 call yourself checkers-san and post in mangled japanese 23:49:52 that'll show 'em 23:54:53 amalloy: nice try "ah"-malloy, but ascii told me they already have a special name for me 23:55:10 it's like....gamfunk...or something 23:58:05 besides it looks like they're leaving my name untranslated 23:58:09 # 전부 gammafunk 작품. 너무 많으므로 생략.