00:11:11 <|amethyst> amalloy: there are a few different android version 00:11:20 <|amethyst> s 00:11:23 <|amethyst> amalloy: it's used for the SDL tiles versions 00:12:01 <|amethyst> s/s$// 00:12:07 <|amethyst> "it" being TOUCH_UI 00:12:19 <|amethyst> though it would be useful anywhere the keyboard is hard to use 00:12:25 <|amethyst> (so mostly mobile) 00:27:48 Unstable branch on CRAWL.XTAHUA.COM updated to: 0.21-a0-251-g5e8393f (34) 01:09:21 gammafunk: qw's winrate dropped by more in nospawn for HuFi in the tests I ran (22/2000 -> 12/2000) but low winrates are much harder to measure so it could just be variance 01:09:48 yeah, those are very small numbers 01:10:09 the difference between xp numbers at various stages of the game looks pretty similar to GrBe 01:11:23 well I'll hopefuly push this very soon, since I do want to work on some other things 01:12:24 I'm kind of curious if players will really notice the difference in difficulty; I'm sure they'll eventually notice no spawns from resting a long time and such 01:12:49 <|amethyst> and OODs can all be blamed on vaults then? 01:13:26 OODs can all be blamed on minmay 01:13:29 <|amethyst> or do you roll OODs for normal level spawns? 01:13:38 those do still get rolled, yeah 01:13:45 but there is no special "super OOD" roll 01:13:47 <|amethyst> but I guess not the super 01:13:48 <|amethyst> yeah 01:14:14 which was a quite low chance (1/5000) for absdepth >= 12 at level gen 01:14:45 but didn't seem worth preserving because vaults are good at rolling OODs 01:14:52 <|amethyst> yeah 01:15:30 <|amethyst> eventual replacement for level-gen monster spawning: 01:15:45 <|amethyst> have layouts place monsters 01:15:51 my guess is that the main noticeable thing is the wandering monsters right after level gen 01:16:18 and that this is also the main thing increasing difficulty 01:16:40 I might experiment a bit with removing that 1/8 chance and running qw more 01:17:26 i don't think i ever noticed that there are no wandering monsters when you enter a level. i wouldn't be surprised if most players don't notice that change 01:17:41 yeah amalloy, that was my reasoning 01:17:46 even if said monster kills them 01:17:48 it's also still possible that the wandering monsters are not a great idea, I'm not sure how they will play out 01:17:55 they would probably assume it's just woken from a shout 01:18:35 right, maybe there is some other solution, but at the same time winrate has been creeping up recently... 01:18:55 sure, making things more difficult isn't necessarily a problem 01:19:28 dang, nerf qazlal. someone on reddit just got their first win after ten years, after going qazlal, training a lot of invo, and never invoking anything 01:20:01 what combo was it 01:20:15 ...grfi 01:20:26 nerf grfi imo 01:20:31 true 01:20:35 <|amethyst> with a buckler and shields 27 01:20:39 haha 01:20:41 excellent 01:21:13 obviously, nerf bucklers 01:21:20 <|amethyst> (but did admit they don't really know much about skilling) 01:21:21 nerf the number 27 01:22:10 <|amethyst> don't allow training weapon/defensive/spell skills past a certain level determined by what you have in inventory/memorised 01:22:12 they've been playing so long they called their broad axe a bearded axe 01:22:34 they were called that in 0.13 or something, right? 01:22:36 <|amethyst> e.g. no training weapons above mindelay 01:22:43 0.13? nope 01:22:48 !lg amalloy 1 x=cv 01:22:50 1/418. [cv=0.15-a] amalloy the Chiller (L3 FeIE), slain by a worm on D:2 on 2014-08-08 02:41:12, with 51 points after 2342 turns and 0:11:24. 01:22:54 wow 01:22:59 it's a classic death 01:23:15 i watched the replay a while ago. i actually correctly kite one worm 01:23:25 PleasingFungus used to talk about 0.12 as if it was linley crawl or something 01:23:26 but then i fight another and when i run out of mp i just tab it 01:23:44 <|amethyst> pre-0.8 is ancient to me 01:23:52 yeah, same 01:23:52 <|amethyst> since 0.8 is what I started with 01:24:06 i started offline when 0.14 was new 01:24:06 I started with 0.10, but I think I may have played a game of 0.9 01:24:27 !lg . 1 x=cv 01:24:28 1/247. [cv=0.11] johnstein the Grave Robber (L3 HaNe), slain by a kobold (a +0,+0 club) on D:2 on 2012-10-13 17:22:36, with 58 points after 2316 turns and 0:58:49. 01:24:40 <|amethyst> it was "broad axe" in initial revision 01:24:43 weird 01:24:44 nice, 0.11 also my first online version 01:24:59 although I think it was the alpha 01:24:59 !lg . 1 x=cv 01:24:59 <|amethyst> !lg . 1 x=cv 01:24:59 i wonder what game i'm thinking of that has/had bearded axes 01:24:59 1/777. [cv=0.10] gammafunk the Cudgeler (L6 HEFi), mangled by Sigmund (a +0,+0 scythe) on D:4 on 2012-04-26 22:42:48, with 466 points after 5210 turns and 0:28:36. 01:24:59 1/14542. [cv=0.8] Neil the Ruinous (L3 DECj), slain by a snake on D:3 on 2011-08-10 11:11:11, with 192 points after 2668 turns and 0:06:16. 01:25:02 oh wow, nm 01:25:36 amalloy: are you just thinking of lugafu or something? 01:25:50 no 01:25:52 which became trog which is sort of associated with axes (not really) 01:25:53 <|amethyst> !lg . 1 x=v 01:25:55 1/14542. [v=0.8.0] Neil the Ruinous (L3 DECj), slain by a snake on D:3 on 2011-08-10 11:11:11, with 192 points after 2668 turns and 0:06:16. 01:26:00 <|amethyst> and hair 01:26:04 <|amethyst> of which beards are made 01:26:28 probably hearing "bearded axe" made me remember seeing it somewhere, like in diablo or something, and it was a long time ago so figured it was an old version of crawl 01:26:46 it is a curious adjective to use for an axe 01:26:56 <|amethyst> it is a real kind of axe 01:27:08 <|amethyst> we should totally rename one of ours 01:27:11 <|amethyst> https://en.wikipedia.org/wiki/Bearded_axe 01:27:19 bearded scythe 01:27:19 broad axe. make my delusion a reality 01:27:34 almost entirely bearded axe 01:27:42 <|amethyst> looks more like a hand axe 01:27:42 like, a scythe could concievably trim a giant's beard maybe 01:28:07 ?/beard 01:28:08 Matching terms (2): beard_stats, bsdbeard; entries (6): beard_stats[1] | blasthardcheese[2] | bobbens[3] | desperate[29] | duhz[3] | tiles[16] 01:28:24 <|amethyst> You reap the beard! 01:31:18 do we have standards for how much indent we prefer after a line break e.g. if we're breaking in a function's args? 01:31:26 monster* mon = _place_monster_aux(mg, nullptr, place, force_pos, 01:31:26 dont_place); 01:31:31 in this case 01:31:37 <|amethyst> 8 seems to be pretty common, but 01:31:38 my vim wants to indent like that 01:31:51 <|amethyst> if you can line it up with the first argument IMO that's better 01:31:57 ok 01:31:59 <|amethyst> that doesn't always work 01:32:07 can I have vim try to do that by default? 01:32:16 <|amethyst> probably 01:32:20 I think emacs used to, but maybe I'm misremembering 01:32:25 I'll google it tomorrow 01:33:55 <|amethyst> :h cino-( 01:34:31 |amethyst, FYI, I took a stab at the Mara clone_mons bug with the OCS. no hurry or anything, but I wonder if you could give it a scan in the next week or so to see if it was what you were expecting/describing. 01:34:48 <|amethyst> which PR#? 01:34:53 https://github.com/crawl/crawl/pull/608 01:34:58 !bug 11171 01:34:58 https://crawl.develz.org/mantis/view.php?id=11171 01:37:34 <|amethyst> johnstein: I see two things in "Simplify handling..." 01:38:06 <|amethyst> johnstein: 1. you should leave the part of the comment about "If an enslaved caster...", since that explains why we don't just use caster->attitude 01:39:07 <|amethyst> johnstein: 2. you can combine the monster *clone clone = clone_mons and if (clone) lines all into one (well, probably wraps to two lines) 01:39:32 <|amethyst> if (monster *clone = clone_mons(....)) 01:39:45 ah yea, good call 01:39:53 <|amethyst> johnstein: I'm not sure how I feel about ATT_SAME 01:42:49 |amethyst, yea, the more I thought about it the less it seemed to fit with the other monster_attitude_types 01:42:59 <|amethyst> johnstein: I kind of feel like there should be checks somewhere to make sure we never actually give that attitude to someone 01:43:13 <|amethyst> but it would be better not to have to have those checks 01:44:28 <|amethyst> hmm... you could have an overload instead of making that optional 01:44:56 <|amethyst> first function monster *clone_mons(const monster* orig, bool quiet = false, bool* obvious = nullptr) 01:45:22 <|amethyst> second function monster *clone_mons(const monster* orig, bool quiet, bool* obvious, mon_attitude_type mon_att) 01:45:50 <|amethyst> then have the first one use orig->attitude 01:45:54 <|amethyst> however, I think I see a problem 01:46:29 <|amethyst> what happens when an enslaved rakshasa self-clones? 01:47:47 <|amethyst> (that uses cast_phantom_mirror, which doesn't pass an explicit attitude) 01:48:34 <|amethyst> so I think it should be ->temp_attitude() instead? 01:49:11 <|amethyst> you might also be able to use caster->temp_attitude() to simplify the _mons_summon_monster_illusion case as well 01:50:08 <|amethyst> that would also better handle the purely hypothetical case where a permanent or summoned ally with such a spell gets hexed 01:51:03 <|amethyst> which AFAIK can only be done with control undead... I guess maybe it could happen with an yred-enslaved spectral rakshasa? 01:53:03 <|amethyst> also, totally pedantically—and it probably doesn't matter, since it likely to be squashed—but that should be "Change a docstring" instead of "Add a docstring" :) 01:53:57 yea (to the docstring), but really, I shouldn't have commited the docstring in the first commit :P 01:54:05 I almost went back to fix that 01:54:14 <|amethyst> I'd say go ahead 01:54:35 <|amethyst> squash and reorder things as much as you like, no one else is working on the branch 01:55:13 <|amethyst> (just try to make sure it builds at every commit) 01:56:16 when you said "so I think it should be ->temp_attitude()", what is 'it'? 01:56:45 <|amethyst> in the case that's currently ATT_SAME, instead of "orig->attitude" use "orig->temp_attitude()" 01:57:27 oh, temp_attitude is a Real Thing that exists Today 01:58:06 <|amethyst> oh, also, re removing that coord_def: 1. you can just write coord_def pos or coord_def pos(0,0); 2. the fact that it's 0,0 means the if (!in_bounds(pos)) is always true so you can pull out the body and cut the if 01:58:11 <|amethyst> yeah 02:00:09 thanks 02:01:08 <|amethyst> oh 02:01:46 <|amethyst> and then, because you'll still have the inner if (!in_bounds(pos)) return nullptr; (when you couldn't find a location), you can also remove the ASSERT_IN_BOUNDS 02:01:58 <|amethyst> since you just checked that 02:44:14 -!- amalloy is now known as amalloy_ 02:45:27 -!- amalloy_ is now known as amalloy 03:21:35 Unstable branch on crawl.beRotato.org updated to: 0.21-a0-251-g5e8393f (34) 03:33:09 -!- amalloy is now known as amalloy_ 03:34:27 -!- amalloy_ is now known as amalloy 04:10:08 -!- amalloy is now known as amalloy_ 06:19:07 !tell espais So far not a single email was returned - seems like they started using real email addresses or the registrants are all real (2-3 a day) 06:19:07 Napkin: You have 11 messages. Use !messages to read them. 06:19:08 Napkin: OK, I'll let espais know. 10:19:59 Do monsters still drop their unthrown stack of stones if you kill them before they exhaust them? Playing as an EE I seem to be running out of stones a lot more than I remember 10:19:59 espais: You have 1 message. Use !messages to read it. 10:20:02 !messages 10:20:03 (1/1) Napkin said (4h 55s ago): So far not a single email was returned - seems like they started using real email addresses or the registrants are all real (2-3 a day) 10:20:32 <|amethyst> espais: they do... are you remembering things from before Sandblast started always using stones? 10:20:34 Hey espais :) I think it works nicely 10:23:14 so i'm running a game right now 10:23:20 a kobold threw 2 stones at me 10:23:21 i killed him 10:23:25 he only dropped his dagger 10:23:34 napkin: great! glad to have helped out 10:24:16 |amethyst: noticed similar behavior with goblins 10:24:31 i kind of assumed they'd have more in their packs to toss at me 10:25:03 thanks again for your help, espais! 10:25:14 <|amethyst> kobolds get 1d5 stones when they do get stones, so probably it just ran out 10:26:08 <|amethyst> goblins should only get stones when they have a sling, or when they pick them up off the ground 10:26:56 right, but they drop them on death, no? 10:26:58 or should i wait till they throw a bunch at me? 10:27:07 <|amethyst> they do drop them on death 10:27:49 <|amethyst> the only thing that should be destroying items on death are summoned monsters (and then the ammo would also disappear on use) 10:28:08 they might just have fairly small stacks 10:28:15 Kobolds can have as many as 1 10:29:12 i see 10:29:36 %git ef5153f041f09a8b719b4f545c09e6d7af0cb6f3 10:29:37 07Brannock02 * 0.20-a0-325-gef5153f: Rework and simplify Sandblast 10(9 months ago, 1 file, 17+ 12-) 13https://github.com/crawl/crawl/commit/ef5153f041f0 10:29:37 i've been adapting my strategy to kill the popcorn more with a weapon, but i do notice that i run out of stones more than i used to 10:30:09 before 9 months ago, you probably only used stones sometimes 10:30:24 <|amethyst> looks like goblins with launchers get 1d7 + 1d10 + 1d12 + 1d10 - 3 so also a (smaller) chance of only one or two 10:30:39 wtf is that formula 10:30:43 <|amethyst> s/goblins with launchers/monsters with slings/ 10:30:54 <|amethyst> _generate_missile_item 10:31:00 <|amethyst> I have no clue :) 10:31:15 !source _generate_missile_item 10:31:16 1/1. https://github.com/crawl/crawl/blob/master/crawl-ref/source/makeitem.cc#L649 10:31:23 tbh i've recently been trying to do more mage classes 10:31:30 so i didn't have much exp with EE short of a run here and there 10:31:34 that's some @crawlcode material there 10:31:52 mmmmmagic numbahs abound 10:32:12 oy, whoever made that vault with electric eels that spawns early is devilish 10:32:19 else if (get_ammo_brand(item) != SPMSL_NORMAL) 10:32:19 item.quantity = 1 + random2(7) + random2(10) + random2(10); 10:32:20 else 10:32:20 item.quantity = 1 + random2(7) + random2(10) + random2(10) + random2(12); 10:32:46 <|amethyst> oh hey 10:32:58 <|amethyst> that second case (the else if) can probably be removed 10:33:17 <|amethyst> oh, wait, no, because of needles of poison 10:33:27 <|amethyst> :( 10:33:40 lots of special things in there 10:33:58 relatively minor, but nice double negative on if (!no_brand) 10:34:09 <|amethyst> the || get_ammo_brand(item) == SPMSL_RETURNING could be removed though 10:34:26 <|amethyst> since only javelins and tomahawks can have that brand now, and they're already covered 10:34:44 <|amethyst> unless the off-brand ammo shop can still make large rocks of returning, I guess 10:34:47 that is a bizarre string of randoms 10:34:58 <|amethyst> oh, but large rocks were already handled above 10:35:16 %git 072319d150da380014ea476c734d2739393fe58e 10:35:16 07elliptic02 * 0.11-a0-1864-g072319d: Decrease the size of large missile stacks. 10(5 years ago, 1 file, 3+ 3-) 13https://github.com/crawl/crawl/commit/072319d150da 10:35:31 elliptic last edited it so maybe there's a rational explanation for the randoms 10:35:32 ohhh those are 1d rolls 10:35:47 yes, more or less 10:35:48 <|amethyst> espais: random2(x) is 1dx - 1 10:39:41 makes sense then 10:39:51 i was looking at that and my brain hurt trying to understand why you'd do that 10:40:01 die rolls though...that is a bit clearer :) 10:40:06 advil: nah, I just didn't care about them 10:41:05 I'm pretty tolerant of silly formulas like that 10:42:38 changing it to 1 + random2avg(blah,blah) would be sensible 10:47:48 well, the formula turns out to be pre-dcss, so I can't really blame you 10:48:36 upside of the crazy formula is that searching for "random2(15)" is likely to find it in old version where the structure is very different 10:48:48 shouldn't those numbers all be defines? 10:49:56 <|amethyst> only if they appear in more than one place IMO, and even then I'd prefer a const or a constexpr function 10:50:24 I would say, only if the appear in one place AND have some meaningful semantics 10:50:28 *they 10:50:28 <|amethyst> yeah 10:50:37 <|amethyst> I mean, #define ITEMGEN_NUMBER_AMMO_1 7 #define ITEMGEN_NUMBER_AMMO_2 10 etc doesn't really do anything 10:50:47 <|amethyst> s/do anything/make anything easier to understand/ 10:51:15 <|amethyst> but ITEMGEN_AVG_AMMO_PILE_SIZE might, if it actually meant that 10:51:52 <|amethyst> but even with that, ITEMGEN_AVG_STONE_PILE_SIZE ITEMGEN_AVG_NEEDLE_PILE_SIZE ITEMGEN_AVG_JAVELIN_PILE_SIZE seems like a bit much 10:52:03 <|amethyst> when that is clear from the context in which it is used 10:52:21 <|amethyst> (again, if the numbers only appear once with that meating) 10:52:26 <|amethyst> s/meat/mean/ 10:53:25 to me i'd think all those should be replaced with a get_die_roll(int mod) function or something like that 10:53:38 that way you could easily swap out RNG mechanics if you wanted different rand distributions 10:53:48 <|amethyst> what would "mod" be? 10:53:50 I think that's more or less what elliptic had in mind with his suggestion 10:54:01 oh 10:54:03 I misread 10:54:16 they already are that 10:54:19 the second half of the 1dX 10:54:31 <|amethyst> but 10:54:33 oh is that what random2 does 10:54:46 well 1dX-1 10:54:51 <|amethyst> under what circumstances would you want to, say, change all the die rolls in the game to have a lower variance 10:55:07 not necessarily that, but there are different random distributions to draw from 10:55:25 <|amethyst> right, but do you want to affect every die roll in the game at the same time? 10:55:32 !source random.cc 10:55:32 https://github.com/crawl/crawl/blob/master/crawl-ref/source/random.cc 10:55:37 <|amethyst> item gen, monster gen, damage, AC, ... 10:56:12 if you wanted, say, a mersenne twister vs poisson 10:56:16 i dunno 10:56:16 <|amethyst> oh 10:56:21 just tossing it out there 10:56:36 <|amethyst> we already have one place where you can change generators 10:56:58 <|amethyst> but, again, I have a hard time seing a situation where you'd want to change all uniform distributions into poisson at the same time 10:57:27 wasn't sure if it would throw people for a loop if the RNG changed its behavior suddenly 10:58:11 <|amethyst> it certainly would if it changed distributions, especially since we use the uniform distribution to build more complicated ones 10:58:21 <|amethyst> if you changed the distribution from random2, for example, then binomial would no longer give a binomial distribution 10:58:42 <|amethyst> it shouldn't break anything (other than reproducibility of seeded tests) to change the generator though 10:58:53 <|amethyst> we've changed the generator several times in fact 11:00:11 i see 11:00:14 not a big issue 11:00:24 i was just thinking that it would be cool if... 11:00:25 then 11:00:33 i come to find out others already had that thought ;) 11:00:49 <|amethyst> if you're interested in investigating more, 11:01:01 man crawl-ancient in github kills my browser 11:01:18 <|amethyst> the RNG itself is in pcg.cc, where PcgRNG::get_uint32 is the generator 11:01:36 (also i have some projects i work on that involve playing with the rand distribution) 11:01:37 <|amethyst> the RNG support functions like random2 are in random.cc 11:04:16 if (many_many >= 75 && many_many < 80) 11:04:16 return; // Dis - Is this even possible? 11:04:20 those were the days 11:07:29 so, the stone generation core magic formula appears to come from the 2000jun19 version, and was constructed by taking a much more insane item generation function and collapsing a lot of cases that were previously summed across a bunch of steps 11:07:50 <|amethyst> see? progress! 11:08:36 <|amethyst> that was, what, 3.something ? 11:08:47 I'm not entirely sure how the versioning worked back then 11:09:05 <|amethyst> oh, looks like it was between 3.4 and 4.0 11:09:27 oh no 11:09:38 I'm wrong, it goes back earlier, I was looking at the wrong code 11:10:28 amusingly, I think I mistook dungeon generation code for item generation code 11:10:35 <|amethyst> it may be a Linley original, though "much more insane...summed across a bunch of steps" sounds a little more like Linley :) 11:10:40 which may say something about me but definitely says something about how this worked 11:11:19 <|amethyst> at least there were enums or defines for monster and item types by that point 11:11:30 <|amethyst> instead of numeric codes in the source that are sometimes commented 11:11:35 heh 11:11:57 I really should go to campus instead of following blames for the last half hour 11:14:09 probably I should learn how to do git blame diving from the command line, it would go a lot faster 11:16:41 <|amethyst> it's still a pain 11:16:48 <|amethyst> would be nice to have an interface for it 11:17:26 <|amethyst> where you could click on a line and automatically switch to a blame of the version that line came from and a diff with the previous 11:17:39 github seems to have that now, actually 11:17:44 <|amethyst> ah, cool 11:17:46 but I get a lot of beachballing on large files 11:18:00 there's a thing just to the right of the age in the blame view you can click on 11:18:01 <|amethyst> I'd rather have a terminal interface myself, but 11:18:18 <|amethyst> my usual workflow for blaming is git blame file 11:18:32 <|amethyst> then I see the version hash on the line in question, let's say deadbeef 11:19:02 <|amethyst> so I do git show deadbeef to see where that line came from 11:19:14 <|amethyst> then when I know where it came from, which could have been another file, I do 11:19:21 so here's a bit of an oddball question 11:19:30 <|amethyst> git blame deadbeef^ otherfile 11:19:40 <|amethyst> and repeat until I have tracked down the true source 11:19:48 i just got stair-pounded by a hydra and cyclops...do you think it would be helpful to have a 'listen downstairs' type spell/item/ability? 11:20:20 <|amethyst> ignoring game design and balance issues, that would be terribly difficult to implement 11:20:28 <|amethyst> because we don't generate the level until you enter it 11:20:49 <|amethyst> (well, it could be implemented for levels you've already visited) 11:25:48 what would stop level generation from happening earlier? 11:26:03 say it is triggered once you hit the 'listen' ability 11:27:07 i'm not familiar enough with the code 11:27:13 but say i wanted to go downstairs 11:27:35 i've never visited it before, so once i trigger my ability/scry spell i generate the level if not generated 11:27:46 then perhaps for balance issues it is limited to the surrounding 8 squares 11:28:17 it doesn't happen to me often, but it's frustrating to take something squishy post-lair and die immediately to a group of enemies 11:28:33 <|amethyst> that would be possible, though I'd be wary of side effects 11:28:42 <|amethyst> things like starting portal timers 11:28:44 such as? 11:28:48 hm 11:28:51 well 11:28:54 <|amethyst> which could be dealt with 11:29:01 <|amethyst> but the problem would be identifying all of those cases 11:29:27 true 11:29:27 so say you do the scry 11:29:39 you start the timer because you generated the level, and you hear a whisper that a portal is open 11:29:59 or maybe the timer gets a slight delay and you get an 'opening soon' message instead 11:31:46 <|amethyst> yeah, that (the former) would be a reasonable solution to that specific problem 11:32:09 <|amethyst> design-wise, I'm not sure it's wise to make stairs safer 11:32:35 <|amethyst> but others can evaluate that better than I, I'm not much of a designer or balancer 11:32:43 <|amethyst> s/,/:/ 11:33:02 well 11:33:06 to me it'd be a quality of life balance 11:33:16 so here was the issue 11:33:23 ok, the basic magic formula for missile generation comes from 3.02, where it was complicated for no obvious reason from "quant = random3(9) + random3 (9) + 1;// + random3 (10) + 1;" 11:33:29 i was looking for orc because lair6 was too dangerous for my DeEE 11:33:40 i was going to d13 to get there 11:33:48 all 3 stairs going down were grouped next to each other 11:33:48 looking at item() in 3.01 I'm honestly shocked that modern dcss came to exist 11:34:05 i went down one, woke up a hydra and cyclops and a few other things 11:34:08 went immediately back up 11:34:15 https://github.com/crawl/crawl-ancient/blob/67903ca8be5cf57b8e43f236b80b07c04b7ac6f5/dungeon.cc#L3009 11:34:24 went back to my stash, freed up some inventory, hoped the monsters would wander away 11:34:25 ^ link to that version of item() 11:34:39 went back to a different stair and they were waiting for me 11:34:39 instant-death, no turns available 11:34:52 figuring out which staircase i could go down to at least TP/blink away would've been helpful 11:36:28 if something like the listening thing you've been discussing were implemented, it would be better if it happens automatically, because it would always be optimal to do anyways 11:37:16 would that be a balance issue though? 11:37:24 melee tanks can typically weather a beating like that 11:40:22 I think it might be better if it were something that went off memory, rather than trying to tap into current state on the level 11:40:45 you can actually check that with X> over stairs 11:42:00 so if you've seen the stairs you're going down you can already use X> to see that it's close to where you last were 11:42:21 something that simplified that process seems good to me 11:45:04 (X> shows monsters' last position, too) 11:47:49 <|amethyst> X] you mean? 11:48:23 did not know about that tbh 11:48:39 <|amethyst> this didn't work quite as well until a version or two ago 11:48:56 <|amethyst> it used to be that you only knew which stair went where if you had taken it 11:49:02 <|amethyst> now you just have to have seen both ends 11:53:16 if i get time i'll poke around at it 12:44:27 -!- amalloy_ is now known as amalloy 12:45:09 also until a few versions ago XV didn't work 12:45:43 so if you were playing console you just knew there was some g there, you had to remember whether it had been a goblin or a gnoll 12:46:05 Xv i guess 13:22:51 Unstable branch on crawl.akrasiac.org updated to: 0.21-a0-251-g5e8393f (34) 14:28:58 -!- evan is now known as Guest82756 16:55:39 03gammafunk02 07* 0.21-a0-252-geb74d03: Replace some 0 pointers with nullptr and simplify 10(5 days ago, 1 file, 8+ 10-) 13https://github.com/crawl/crawl/commit/eb74d0312ede 16:55:39 03gammafunk02 07* 0.21-a0-253-g6b34986: Don't try to spawn monsters from stairs 10(5 days ago, 2 files, 9+ 243-) 13https://github.com/crawl/crawl/commit/6b349860cb1b 16:55:39 03gammafunk02 07* 0.21-a0-254-g1d58f9a: Remove monster spawns over time in most branches 10(5 days ago, 1 file, 29+ 120-) 13https://github.com/crawl/crawl/commit/1d58f9a54f8a 16:55:39 03gammafunk02 07* 0.21-a0-255-g5a3d65e: Increase numbers of level generation monsters in most branches 10(5 days ago, 2 files, 16+ 16-) 13https://github.com/crawl/crawl/commit/5a3d65ea255c 16:55:39 03gammafunk02 07* 0.21-a0-256-gb9be33d: Add a monster placement proximity type for generating away from stairs 10(4 days ago, 2 files, 8+ 0-) 13https://github.com/crawl/crawl/commit/b9be33d25565 16:55:39 03gammafunk02 07* 0.21-a0-257-gbb444f2: Generate some monsters awake at level creation time on most levels 10(4 days ago, 1 file, 13+ 3-) 13https://github.com/crawl/crawl/commit/bb444f2580aa 17:09:06 Unstable branch on crawl.jorgrun.rocks updated to: 0.21-a0-257-gbb444f2 (34) 17:55:35 -!- Chase is now known as ChaseSP 18:14:38 Unstable branch on underhound.eu updated to: 0.21-a0-257-gbb444f2 (34) 18:22:48 ??rebuild 18:22:49 rebuild[1/2]: http://crawl.akrasiac.org/rebuild/ http://underhound.eu:81/rebuild/ http://crawl.berotato.org/crawl/rebuild/ http://crawl.xtahua.com/rebuild/ https://crawl.jorgrun.rocks/rebuild/ Bug |amethyst or Nap.Kin for CDO. Use your powers wisely. 18:34:23 Unstable branch on crawl.beRotato.org updated to: 0.21-a0-257-gbb444f2 (34) 19:03:30 -!- PsyMar2 is now known as PsyMar 19:59:14 %git 19:59:14 07gammafunk02 * 0.21-a0-257-gbb444f2: Generate some monsters awake at level creation time on most levels 10(4 days ago, 1 file, 13+ 3-) 13https://github.com/crawl/crawl/commit/bb444f2580aa 19:59:32 &versions 19:59:44 CAO: 0.21-a0-251-g5e8393f, CBRO: 0.21-a0-257-gbb444f2, CDO: 0.21-a0-2-ge494091, CJR: 0.21-a0-257-gbb444f2, CPO: 0.21-a0-257-gbb444f2, CUE: 0.21-a0-244-gd4adca8, CWZ: 0.21-a0-216-g6d3fd91, CXC: 0.21-a0-251-g5e8393f, LLD: 0.21-a0-142-gdd972a7a06 20:00:18 ??rebuild 20:00:18 rebuild[1/2]: http://crawl.akrasiac.org/rebuild/ http://underhound.eu:81/rebuild/ http://crawl.berotato.org/crawl/rebuild/ http://crawl.xtahua.com/rebuild/ https://crawl.jorgrun.rocks/rebuild/ Bug |amethyst or Nap.Kin for CDO. Use your powers wisely. 20:13:52 !kw nospawn vlong>=0.21-a0-257-gbb444f2 20:13:52 Defined keyword: nospawn => vlong>=0.21-a0-257-gbb444f2 20:13:57 !lg * nospawn 20:13:58 65. fish the Sharpshooter (L21 CeHu of Fedhas), blasted by a deep elf mage (bolt of cold) on Vaults:5 (vaults_vault; vaults_end_minmay_fences) on 2017-09-06 00:12:09, with 350126 points after 59100 turns and 2:27:54. 20:17:36 Unstable branch on crawl.akrasiac.org updated to: 0.21-a0-257-gbb444f2 (34) 20:18:49 -!- GiantOwl is now known as Kalir 20:37:02 gammafunk: idea: spawn all monsters in certain vaults awake / asleep. Like say everything asleep in crypt/tomb 20:37:08 s/vaults/branches/ 20:37:11 ~flavour~ 20:50:21 Connector placement failed 13https://crawl.develz.org/mantis/view.php?id=11213 by yesno 20:56:23 -!- amalloy is now known as amalloy_ 21:15:25 -!- amalloy_ is now known as amalloy 21:41:27 -!- amalloy is now known as amalloy_ 22:20:47 -!- Mayoi is now known as erkin 23:17:51 -!- Mayoi is now known as erkin 23:31:53 03johnstein02 07https://github.com/crawl/crawl/pull/604 * 0.21-a0-224-g4c24e63: Better handle invis prevention messages 10(5 minutes ago, 1 file, 11+ 10-) 13https://github.com/crawl/crawl/commit/4c24e63dc387 23:35:59 03gammafunk02 07* 0.21-a0-258-g4c0eb85: Fix a SUBST that put too many apocalypse crabs in a vault 10(3 minutes ago, 1 file, 1+ 1-) 13https://github.com/crawl/crawl/commit/4c0eb85f4910