00:01:29 Unstable branch on CRAWL.XTAHUA.COM updated to: 0.21-a0-625-g1119ec1 (34) 00:07:57 hm, why doesn't primary_mini_dummy place on Depths:1? 00:09:49 !vault primary_mini_dummy 00:09:49 1/1. https://github.com/crawl/crawl/blob/master/crawl-ref/source/dat/des/builder/dummy.des#L25 00:11:30 what does the 1-100 mean in DEPTH: ? 00:11:42 I guess that's D:1-100 ? 00:12:06 which, if it is, that's weird syntax 00:12:47 advil: yeah, not exactly sure why U:1 is excluded 00:13:00 maybe some look through the git logs will clarify 00:13:25 I was thinking maybe it's an attempt to make that level more interesting on average? 00:13:32 haven't tried to track down the history though 00:15:09 yeah that would be a reasonable guess, but it doesn't make too much sense 00:15:15 why that level in particular is so special 00:15:24 yeah, not sure 00:15:35 I know of some float vault that exclude U:1 and place in depths 00:15:39 I noticed because it doubles the rate of ghost vault placement on that level 00:15:47 and it's not uncommon for harder ones to do that (and sometimes unecessarilly in my view) 00:15:57 but that's something that tends to be done in the float vaults for every branch 00:16:09 yeah I'd say feel free to remove that 00:16:40 03advil02 07[spooky_vaults] * 0.21-a0-597-gb676c08: Tweak ghost vault generation rates 10(7 minutes ago, 1 file, 6+ 6-) 13https://github.com/crawl/crawl/commit/b676c088227d 00:16:40 03advil02 07[spooky_vaults] * 0.21-a0-598-g9c503fd: Don't spawn randomly-placed ghosts on level generation 10(5 minutes ago, 1 file, 0+ 8-) 13https://github.com/crawl/crawl/commit/9c503fd0f325 00:16:40 03advil02 07[spooky_vaults] * 0.21-a0-599-g053ff4d: Let ghosts use stairs 10(4 minutes ago, 1 file, 0+ 1-) 13https://github.com/crawl/crawl/commit/053ff4d4351e 00:16:43 but isn't the ghost vault tied to something like a CHANCE ? 00:16:56 I'd imagine it'd be nice to just have a flat chance of a ghost vault on each level 00:17:25 no, because I didn't know about CHANCE 00:17:37 yeah, see the des syntax file for details 00:18:22 just a way to do a specific roll on each level to place the vault, and you can have difference chance at different depths 00:25:11 does mapstat do something different with vaults with CHANCE? 00:25:18 it's not showing up in the % listing 00:29:33 hrm, I wonder if it would list those 00:29:54 they get listed in the aggregate numbers, it just doesn't appear to be including them in the % breakdown 00:30:06 oh, well, that would make sense 00:30:23 since that's a listing of (essentially) non-chance vaults 00:30:42 to show you what vault has what weight relative to other potential vaults 00:30:49 but for chance you know it's just whatever the chance says 00:31:00 I think it would be misleading to include chance in that listing 00:31:12 but you could have an additional listing for chance vaults I guess 00:31:56 I guess in theory you can divide number of placements by number of generated levels 00:32:07 and compare that to the CHANCE value to see if its similar 00:32:34 yeah, I just needed to double check that I did right 00:32:51 yeah, if the actual placement seems consistent with your chance (you probably need at least 100 iterations or so) 00:33:00 then that's perhaps all mapstat could tell you 00:33:08 that and if it's causing a lot of vetoes or whatnot 00:33:49 03advil02 07[spooky_vaults] * 0.21-a0-600-g7a4f597: Use CHANCE rather than WEIGHT for non-Vaults ghost vaults 10(5 minutes ago, 1 file, 4+ 4-) 13https://github.com/crawl/crawl/commit/7a4f5974e6cf 00:34:18 btw I thought a little about the mutation roulette thing, and tbf I haven't thought about this very much 00:34:41 but the 'purple stair' idea doesn't seem great to me since it's giving people who want to 'roll mutations' a conduct that they probably don't want 00:34:56 you could argue that the conduct is interesting on its own, but it doesn't really seem so to me 00:35:19 there's just not much logical connection between "you got a mutation" and "you ended up off stairs on a new level" 00:35:44 yes, I'm not sure it makes sense in regular crawl 00:35:44 it almost seems like something like a purple fountain that gives at least 1, maybe 1-3 mutations before drying up might be nice 00:35:54 here's a repost from earlier: after thinking about mutation roulette for a while, I think my current favorite is something like: new kind of mut potion that does 1-2 random mutations, and expire some reasonable # of turns after the player first notices them (a la blood) 00:35:54 basically chunks but generated as regular items, with a lower (and more even) frequency 00:36:29 fountains would be a bit more flavorful 00:36:43 yeah, I'm not sure the timer actually adds anything terribly interesting 00:36:57 I suppose there's the "well do this now or never" aspect 00:37:15 however that was not actually the main thing people liked about the conduct 00:37:31 if there's no time limit, probably you just hoard randmutation sources and use them preferentially when you have a bunch of bad mutations 00:38:05 sure, but you do have to live with a lot of bad mutations 00:38:15 for this item, it's not died to a monster drop 00:38:22 I guess this starts to expire only after the player sees it? 00:38:26 yeah 00:38:46 that seems maybe a little clunky 00:39:16 I do like the now or never aspect of chunks (not that I really ever ate them) 00:39:47 well I definitely ate them and just got a bunch of good mutations; I think at the core of things people like just trying to roll for a good thing and mostly losing but sometimes not 00:39:55 which obviously you can have with or without a timer 00:40:17 as for the hoarding aspect, I'm not actually sure if it's a problem; backtracking is kind of a problem, but we have things like shops 00:40:26 it's not really too much of an issue 00:40:47 this perishable item, it would just be a normally generating item? 00:40:56 that's what I had in mind 00:41:44 I don't have very strong opinions though 00:42:12 maybe I'll run these two ideas by the others who have opinions about purple chunks 00:42:44 it's sort of nice to use an item in that you can use item placement mechanics, to the extent that we really control those; a feature is arguably a little harder to control 00:42:51 but if we only place them through vaults it's fine 00:43:08 it improves the flavor behind dungeon features a little 00:43:14 or just adds some 00:43:24 it too could be timed based on the player seeing it 00:43:34 yeah, or leaving the level or something 00:43:35 but an item is a little cleaner if we're talking timed things, probably 00:43:43 right, could do that 00:43:49 so probably item vs feature and timed vs not-timed are orthogonal decisions 00:43:58 I guess we could prompt them before they left the level 00:44:25 or when they see it give a message saying it will run out 00:44:25 only thing is people who never want to use these getting a prompt like every other level 00:44:34 yeah, a simple timer 00:44:54 starting when they see the feature 00:45:04 and just give timing messages, working teh same way portals do 00:45:44 maybe a non-trivial issue is just how common these should be 00:46:18 I'm not what you would call this perishable item 00:46:32 for now it could be a food type 00:46:43 I have no sense for what a good number of mutation rolls might be 00:46:58 if it's an actual potion, for instance, it's a little confusing since he have something literally called a potion of mutation 00:47:08 yeah, that's true 00:47:14 ~1 per level didn't sound too crazy for me, but I really don't know 00:47:34 ~1 per level would be reasonably sane 00:47:38 ensure that 27 mut fountains generate each game 00:47:46 before you had like average of 40 chunks in dungeon alone 00:47:52 and like 300 across the game 00:48:00 vaults, depths, elf all had lots of purple chunks 00:48:03 wow 00:49:42 btw the ghost vault draft is basically done 00:50:06 I'm a bit worried about there being bugs to do with bones files on servers that I won't anticipate 00:50:47 well it would need to get merged fairly soon, but we certainly could 00:51:36 probably good to get some code review before merging at least 00:52:01 as much as I don't want to say this, it is feeling a bit rushed to me for 0.21 00:52:09 yeah 00:52:12 may need some time to mature 00:52:14 in trunk 00:52:27 we don't normally have such big features coming in so close to release 00:53:10 I suppose we could branch without this and let it stay in 0.22-a 00:53:18 against our usual rules of putting in big features 00:53:25 it's sort of a special case in terms of features 00:53:40 and it does just need testing, which delaying it doesn't help 00:54:05 we had previously avoided putting any significant new features in trunk during freeze until after tourney 00:54:30 I wouldn't really mind leaving it in a branch until after tourney 00:54:59 ok, maybe we should make that the plan, then 00:55:12 it could even be useful to test it out in an experimental for a little while 00:55:13 I can still give it a look and maybe you can even get a patented amalloy code review 00:55:39 of course I shelled out thousands for the amalloy code review and he didn't even point out that constriction bug!!! 00:55:47 heh 00:56:30 hmm 00:56:40 my milestones seem to be slightly bugg 00:56:42 *buggy 00:56:56 if we don't merge it, for 0.21 I'd like to consider (i) disabling ghosts in portal branches, and (ii) maybe tweaking some speed/brand stuff that has been discussed before 00:57:06 I have a "Identified a Sif Muna gift" followed by "received a Sif Muna gift" 00:57:24 and a repeated "Identifed a Sif Muna gift" 00:58:13 I guess it has something to do with items being identified from the ground? 00:58:45 https://pastebin.com/Ugfn6J0b 00:58:45 https://www.reddit.com/r/dcss/comments/7ku5o7/sif_gift_missing/ maybe relevant 00:58:52 are ghosts in portals really that bad? I see how they'd be pretty tough to place 00:59:01 I mean presently, really 00:59:07 see turns 29870, 35370 00:59:17 in newghosts they'd be hard to use in a meaningful way 00:59:32 so I guess that's a great reason to disable them in portal branches 01:00:16 the problem with portals now is griefing 01:00:16 which I think is still happening to some degree 01:00:23 yeah, they won't generate in portal branches for newghosts (unless someone makes a map with one) 01:00:55 I wonder if you could make a spooky entry vault 01:00:58 hall_of_ghosts 01:01:10 for D:1 I mean, so an arrival vault I guess, using the right terminology 01:01:26 anything's possible! would sort of waste the bones entries though 01:01:40 yeah, you'd probably need something sort of special to do that 01:05:38 CanOfWorms: do you know how you got the delayed identification for the 29870 turn? in wizmode I just get them identified immediately (which mostly happened to you, too) 01:06:20 i.e. it seems to have shown up on 29866 and been identified on 29870 01:06:33 might have been trampled off it by an elephant 01:07:10 ah interesting 01:08:01 that's the only doubled one that I'm spotting, unless I'm missing something 01:08:25 the identified / received order is a bit weird too 01:08:35 CanOfWorms: also, this is not a big deal, but just to avoid confusion, what you're talking about is just a note, not a milestone 01:08:57 ah 01:20:35 Unstable branch on crawl.develz.org updated to: 0.21-a0-625-g1119ec1 (34) 02:00:07 Windows builds of master branch on crawl.develz.org updated to: 0.21-a0-625-g1119ec1 02:27:41 advil: what do you want a review of? i don't see a PR from you 02:28:30 spooky_vaults branch 02:28:43 dunno if he made a PR of it though 02:37:29 !tell advil if you'd like a review of spooky_vaults please make a PR for it. reviewing a PR is much more pleasant than reviewing a bare branch 02:37:30 amalloy: OK, I'll let advil know. 02:39:22 (plus without the PR i didn't realize it was waiting for review) 02:55:07 Monster database of master branch on crawl.develz.org updated to: 0.21-a0-625-g1119ec1 03:21:48 Unstable branch on crawl.beRotato.org updated to: 0.21-a0-625-g1119ec1 (34) 03:23:04 to PR, or not to PR, that is the diff 03:34:51 -!- Evablue_ is now known as Evablue 04:07:36 Evablue (L21 GnCj) (Vaults:1) 04:13:24 -!- amalloy is now known as amalloy_ 04:15:53 -!- amalloy_ is now known as amalloy 04:37:34 Evablue (L21 GnCj) (Vaults:2) 05:07:49 Unstable branch on crawl.akrasiac.org updated to: 0.21-a0-625-g1119ec1 (34) 05:56:42 -!- amalloy is now known as amalloy_ 06:17:00 did deep dwarves ever get nerfed at all 06:38:38 yes 06:38:41 we removed wands of heal wounds and give them an ability that gives them overall less heal wounds 06:54:24 03Aidan Holm02 07https://github.com/crawl/crawl/pull/632 * 0.21-a0-626-g7094902: Remove menu.set_maxpagesize interface 10(9 weeks ago, 7 files, 3+ 38-) 13https://github.com/crawl/crawl/commit/7094902431da 06:54:24 03Aidan Holm02 07https://github.com/crawl/crawl/pull/632 * 0.21-a0-627-gd88edeb: Assert that menus have a valid selection flag 10(9 weeks ago, 2 files, 19+ 4-) 13https://github.com/crawl/crawl/commit/d88edeb20028 06:54:24 03Aidan Holm02 07https://github.com/crawl/crawl/pull/632 * 0.21-a0-628-g1f189c4: Explicitly set MF_SINGLESELECT on describe menu 10(9 weeks ago, 1 file, 1+ 0-) 13https://github.com/crawl/crawl/commit/1f189c4a3687 06:54:24 03Aidan Holm02 07https://github.com/crawl/crawl/pull/632 * 0.21-a0-629-gd2a446e: Remove unused menu.f_drawitem hook 10(9 weeks ago, 2 files, 2+ 7-) 13https://github.com/crawl/crawl/commit/d2a446ec42b6 06:54:24 03Aidan Holm02 07https://github.com/crawl/crawl/pull/632 * 0.21-a0-630-gc0777e5: Remove column_compoer set_pagesize() 10(9 weeks ago, 4 files, 2+ 20-) 13https://github.com/crawl/crawl/commit/c0777e5c2b6a 06:54:24 03Aidan Holm02 07https://github.com/crawl/crawl/pull/632 * 0.21-a0-631-g4ea9e22: Refactor Menu scrolling into MenuDisplay 10(9 weeks ago, 7 files, 408+ 318-) 13https://github.com/crawl/crawl/commit/4ea9e22959bd 06:54:24 03Aidan Holm02 07https://github.com/crawl/crawl/pull/632 * 0.21-a0-632-g6f054d7: Remove limit on lookup-help results length 10(9 weeks ago, 1 file, 8+ 17-) 13https://github.com/crawl/crawl/commit/6f054d7f4d22 06:54:24 03Aidan Holm02 07https://github.com/crawl/crawl/pull/632 * 0.21-a0-633-gd1a600a: 2-col tile menus: wrap horizontally and support line scrolling 10(11 days ago, 4 files, 259+ 177-) 13https://github.com/crawl/crawl/commit/d1a600a1aadd 06:54:24 03Aidan Holm02 07https://github.com/crawl/crawl/pull/632 * 0.21-a0-634-gc20e1d3: Remove Menu text-only option 10(23 hours ago, 10 files, 46+ 89-) 13https://github.com/crawl/crawl/commit/c20e1d311ac0 06:54:24 03Aidan Holm02 07https://github.com/crawl/crawl/pull/632 * 0.21-a0-635-gcc831ee: Move menu title rendering to mdisplay 10(20 hours ago, 8 files, 144+ 184-) 13https://github.com/crawl/crawl/commit/cc831ee4531c 07:06:01 gammafunk: do you have any instructions available on how to set up the beem-bot to watch all games on a server? 07:06:47 TZer0: The home page of the repo mentions what packages you have to install, and you mostly just have to edit the config file for your settings 07:06:53 TZer0: but you realize I run beem for CUE already? 07:06:56 or is this for someone else 07:07:02 ??beem[2 07:07:02 beem[2/6]: Use `!player-only on' in your own game's chat to have beem respond only to your queries. Use `!player-only off' to re-allow everyone's queries. 07:07:04 ??beem[3 07:07:04 beem[3/6]: Source https://github.com/gammafunk/beem ; (very simple) webtiles python library: https://github.com/gammafunk/webtiles 07:07:14 Yeah, but it only happens on some games, right? 07:07:17 The ones with the most people 07:07:40 not quite, it has an autowatch for one player with the most spectators who isn't subscribed already 07:07:44 and it has subscriptions 07:07:50 so you just have to subscribe to have it watch you 07:08:13 I didn't want it to just watch every game no matter what 07:08:24 ??beem 07:08:24 beem[1/6]: beem is a WebTiles chat bot that sends commands to the DCSS IRC knowledge bots. For details, see https://github.com/gammafunk/beem/blob/master/docs/commands.md ; if you see beem in chat, type the following to have it watch your games on that server: !subscribe 07:08:39 to subscribe, people just have to type !subscribe in any tiles chat where beem is present 07:08:58 not an ideal system necessarilly, and I might make a HTTP thing for easier subscriptions at some point 07:09:11 but it's been working ok for a couple years now 07:09:19 Ah, right 07:10:21 Thanks for clearing that up! 07:10:24 np 07:10:35 also I noticed that rebuild url doesn't seem to work for CUE 07:10:39 is there a new url? 07:10:43 Yeah, I need to fix that, sorry. 07:10:45 ??rebuild[2 07:10:45 rebuild[2/2]: Trunk update start times in UTC (second number during summer time): CAO: 1700; CDO: 0600/0500; CUE: 2300/2200; CBRO: 0800/0700; CXC: 0500/0400; CPO: every 15 minutes; CJR: every hour 07:10:47 oh ok, cool 07:10:54 no rush or anything 07:11:09 It is like one of the last two points remaining from the server-move 07:11:29 yeah everything else seems to be working great for CUE 07:11:37 *newCUE 07:12:08 It was a bit of work to get it set up. 07:12:18 And there's still a hack in there 07:12:47 for accessing files between host and VM? 07:13:06 Nah, the redirects into the webtiles-server and ssh is a bit hackish right now 07:13:13 It works... but I'm not 100% happy about it. 08:08:24 03Aidan Holm02 07https://github.com/crawl/crawl/pull/632 * 0.21-a0-635-g13ce829: Move menu title rendering to mdisplay 10(21 hours ago, 8 files, 144+ 185-) 13https://github.com/crawl/crawl/commit/13ce829408a7 08:13:16 <|amethyst> aidanh: I see you just committed something, are you here? 08:18:59 03Aidan Holm02 07https://github.com/crawl/crawl/pull/632 * 0.21-a0-636-g7684dd0: Centre the CRT display on local tiles 10(3 minutes ago, 2 files, 10+ 2-) 13https://github.com/crawl/crawl/commit/7684dd06d9b0 09:20:23 |amethyst: yep 09:20:42 sorry; irc client was silently disconnected 09:23:57 btw, if anyone would like to review #632, that'd be a huge help 09:24:13 as I'm getting to the point where progress depends on both #632 and #642, and they're both rather big 09:37:38 aidanh: I'm not sure how invasive those PRs will be; I was planning to do feature freeze on this Friday or Sunday at the latest 09:38:22 if you think it will take some time to iron out major bugs, we could try to merge them right after tournament 09:38:50 but if you think they can work fine in that time frame, we could try to merge them 09:39:02 For those interested, I updated this: https://github.com/crawl/crawl/wiki/0.21-Development-and-Release-Plan 09:39:02 when's the tournament? 09:39:06 ??tornament 09:39:06 tornament ~ tournament[1/3]: The 0.21 tournament will run from 20:00 UTC Friday 5 Jan 2018 to 20:00 UTC Sunday 21 Jan 2018. 09:39:34 Feature freeze probably this coming friday, possibly as late as Sunday 09:40:37 I think they're mostly bug free; what I mostly need feedback on for #632 is the changes to two-column menu scrolling 09:41:25 the other one is a UI framework, so there's no point merging it without it being used somewhere, and the highscores rework was just a proof-of-concept really 09:41:58 ziggurnaut (L8 DgAE) (D:6) 09:44:52 most of the bug fix PRs should be good to go though 11:26:19 03gammafunk02 07* 0.21-a0-626-g79970e1: Show constriction and held statuses show properly (Pekkekke) 10(13 minutes ago, 2 files, 3+ 3-) 13https://github.com/crawl/crawl/commit/79970e119989 11:51:10 gammafunk: did you see the discussion about BVC on stationary monsters? 11:51:10 advil: You have 1 message. Use !messages to read it. 11:51:30 it seems maybe a bit strong in that case 11:56:23 advil: too strong in what way? 11:59:53 in my testing it lasted 14 turns at the lowest spellpower where I had it castable 12:00:31 and they can't escape except via the timer 12:01:13 Yeah, but stationary monsters can't escape any spell over any number of turns if you can hit them with it 12:01:25 they're stationary, so things like airstrike etc can be used against them repeatedly 12:01:42 yes, but bvc has no cost after the initial cast, right? 12:02:01 so I don't think this is so much a thing about stationary monsters 12:02:26 you just cast it and do whatever else you want as long as the statue remains in los 12:02:47 right, my point is that this is not some fundamental issue 12:02:54 otherwise cloud spells are a problem 12:03:05 or rather this isn't anything related to stationary monsters 12:03:12 clouds dissipate a lot faster than bvc constrict 12:03:40 I don't think there's a statue in the game that can survive one cast of BVC 12:03:45 maybe roxanne, I didn't test 12:03:54 yeah, it's not an issue for statues is my point 12:04:08 so it would be ok to lower the max duration of the enchant, but "damages monsters over time" is exactly what these spells do 12:04:31 cloud spells, poison spells, this spell 12:04:47 so you could lower the roll for the enchant so that it tops out lower 12:05:16 Unstable branch on crawl.akrasiac.org updated to: 0.21-a0-626-g79970e1 (34) 12:05:53 yes, that's my point 12:05:56 14 turns seems a bit long 12:06:08 and statue monsters are basically designed to be like this; they don't regenerate but they have very odd/powerful attacks 12:06:23 so they're meant to be taken out with some kind of specialized spell 12:06:39 sure, it seems reasonable that this would be strong against them 12:06:46 but it's sort of a 5MP auto-kill right now 12:07:12 sure, but try using LRD against a statue 12:07:18 it's a 5mp auto-kill, basically 12:07:27 anyhow you could probably halve the enchant roll 12:07:46 anyways, I don't think players will complain if you *don't* change it 12:07:56 btw if you do change something that report was from Lightli IIRC 12:09:34 Unstable branch on crawl.jorgrun.rocks updated to: 0.21-a0-626-g79970e1199 (34) 12:12:09 another statue auto-kill spell: fireball 12:12:15 another statue auto-kill spell: exclusion 12:12:26 we should make exclusion properly into a spell 12:13:37 There is an actual concern mibe reported to me, which was that monsters in constriction near harmful clouds won't try to move 12:14:22 So I might try to update their AI to consider that as something similar/analogous to being in a harmful cloud 12:14:37 but I'd have to look that portion of the monster action code 12:17:35 ok, so why even have a timer on BVC constrict? 12:19:52 sorry? is that relevant to the cloud thing? 12:20:39 the timer was in general because 1) it's an enchantment, those are duration-base and 2) it gives some spell-power influence and a way to limit the constriction 12:20:52 it doesn't seem to limit it at all? 12:21:38 well how much it limits it will depend on the numbers you use 12:21:48 I'm not sure I really understand what you're asking though 12:22:05 The duration formula could probably use some adjustment, as I said above 12:22:22 ok 12:22:25 I didn't catch that 12:22:32 that's what I'm suggesting 12:23:27 It's just not a "statue monster problem" nor is this the only spell that destroys statues very easilly 12:24:03 fireball a statue out of los, lrd a statue (very often one-shots them) 12:24:23 it's just a case of "does this spell have all the right knobs balanced all the right way given its power level" 12:24:45 so for duration we probably want something less like infestation (which I think I used the same formula there) and something a lot more aggressive 12:24:59 another approach could be to not have a spell-power influenced escape chance 12:25:37 but that's a bit awkward maybe, I'd have to think about that a bit 12:26:02 the point about statues and this spell is that they can't take action to escape constriction, the point Lightli made (after playing a full game with the spell) is just that it currently feels very op against stationary monsters for exactly this reason 12:26:06 which may be fine 12:26:30 advil: what action to statues take when you fireball them out of LOS? or when you use LRD on them? 12:26:51 it's just a general issue with the spell (does it balance it's duration of constriction) 12:27:02 s/it's/its/ 12:27:43 or does it have the right way of limiting constriction, for that matter 12:28:08 those are one turn actions... 12:28:17 seems to me if it were something like "max of 5-7 turns at low spellpower" it would at least feel better 12:28:29 yes, that's true I think 12:28:32 I don't know why that's an important distinction 12:29:00 to put it another way, this change really won't have any real, practical effect on how easilly this spell kills statues 12:29:06 it's not really a statue killing issue 12:29:58 I think it probably does matter for how the spell feels...in any case I don't care, I like the spell, I'm just passing on a report (that I think is pretty valid), do what you want with it 12:30:27 it's a corner case for sure 12:36:05 don't worry, I know how to process lightli reports 12:36:41 to lightli, things exist in a binary statue of "broken" or "useless" 12:37:11 thankfully this still is still in one of the two binary states 12:37:18 *this spell 12:38:11 and s/binary statue/one of two binary states/ 12:38:27 was this posted on reddit or tavern, I wonder 12:39:27 must have been on SA, rip 12:46:42 it was in this channel 12:47:32 yesterday some time 12:48:24 yeah, I see; I'm going to adjust down the duration just on the principle that it should matter if it exists, but 12:48:35 need to think about what are all the knobs this thing should have 12:49:10 it can just have a pretty aggressively low duration and that should work ok for now 12:49:40 maybe about half of what it is now could work? 12:50:00 half average probably, yeah, but I'll do some testing 12:50:18 the constrict->cloud issue I'm not sure, need to look at the AI and see if that's hard for some reason 12:50:42 oh apparently there's a bug, too 12:51:08 where monsters getting "directly" (so, normally) constricted 12:51:08 ah the constrict displacement thing? 12:51:08 can get swapped with other hostiles 12:51:08 yes 12:51:08 I saw that in discord 12:51:14 it does sound like something that's probably always been there 12:51:24 oh, I was thinking very likely a bug I introduced if that's a thing 12:51:30 it seems it'd have been seen/reported 12:51:36 but I could be wrong about that 12:52:13 I didn't think very hard about it, so it could be new 12:52:20 yeah, maybe someone would've noticed on an op 12:52:53 well this was not based on player constriction in that sense 12:53:05 anaconda constricting a player, naga ritualist displaced it 12:53:14 ah, I see 12:53:51 yeah, that would be noticeable 12:54:27 probably a constriction check related to monster movement was not specific enough 12:55:38 03advil02 07* 0.21-a0-627-g87fa41b: Don't generate ghosts in portal branches 10(76 minutes ago, 3 files, 20+ 13-) 13https://github.com/crawl/crawl/commit/87fa41b98fa8 12:55:38 03advil02 07* 0.21-a0-628-g72aa5ef: Normalize ghost min speed and prevent more ghost brands 10(60 minutes ago, 1 file, 6+ 5-) 13https://github.com/crawl/crawl/commit/72aa5ef99f8e 12:55:57 happy for those to be tweaked, but I wanted to get something in 12:58:54 why ban those brands? 13:00:14 I think the idea was to prevent the worst kinds of potential relatively griefing and horrible-to-ever-fight ghosts from being a thing 13:00:23 *relatively early 13:00:40 chaos because of griefing, though I'd take the point that it's not so bad, distortion yeah for ^ 13:01:07 holy was already out 13:01:07 so is this making all ghosts have slow movement? 13:01:07 i mean they're no worse than those weapons just generating on early monsters especially 13:01:40 gammafunk: yes, but I'm not sure 12 is the right number, happy to change 13:02:04 well making a lot more slow monsters, this I don't like; monsters can still be killed with stairs 13:02:48 yeah, true...fwiw I got 12 from a suggestion of Lasty's in a much earlier discussion 13:02:53 what about 10? 13:03:01 yeah 10 seems more reasonable to me 13:03:07 I think MarvinPA is right that the the brand solution doesn't exactly make sense if we allow normal monsters to have those 13:03:13 chaos in particular is extremely rare unless people intentionally make CK ghosts 13:03:26 what if we instead (and for now) simply lowered the chance that the brands stuck? 13:03:26 which they do, or at least have done 13:03:32 oh making them all slow also sounds very bad, yeah 13:05:05 MarvinPA: one thing about ghost monsters is that players have an incentive to make e.g. CK ghosts, so unlike dungeon/vault monsters, they have an "outside" weight on generation 13:05:19 ok, I'll change it to 10 13:05:42 how often they get chaos depends on how often someone decides to run a bunch of CK 13:06:24 so we could just be pretty aggressive with a weighting against keeping certain brands as a reasonable stop-gap thing before the ghost vault branch gets merged in 0.22-a 13:06:34 03advil02 07* 0.21-a0-629-ga19c9dc: Tweak min ghost speed to 10 after further discussion 10(31 seconds ago, 1 file, 1+ 1-) 13https://github.com/crawl/crawl/commit/a19c9dc5aa9a 13:06:58 I actually don't feel incredibly strongly that chaos ego ghosts are that pressing to actually fix in 0.21, but maybe advil and others do 13:07:48 I think there's still more ghost griefing going on than we have been quite aware of, though I don't know how much exactly 13:08:10 so as long as CK starts with chaos brand I'd rather remove one of the knobs they have 13:08:43 I do actually agree that chaos brand isn't so bad, but the main target of this griefing seems to be new/weaker players 13:09:01 or one main target, at least 13:09:02 who it really messes with 13:09:35 Unstable branch on crawl.jorgrun.rocks updated to: 0.21-a0-628-g72aa5ef99f (34) 13:09:40 yeah, there's a flip side to that, which is that said players learn not to mess with ghosts, but I don't have strong feelings either way, and will leave it to those who do 13:09:59 we're going to fix it more comprehensively, that's what I most care about 13:10:04 yeah 13:10:06 this is all a stopgap 13:10:16 in fact I plan on reverting these changes in the spooky_vaults branch 13:11:36 the only occasion i've noticed it being an explicit problem was when it was just people like doesnty being deliberately obnoxious a while back, but yeah from what i saw the branch looks like a good approach anyway 13:12:04 there've been several other players, and I've seen scattered reports that suggest it's still happening in portals 13:13:16 (one yesterday, which is a thing that prompted me to do it, though it's not a definitive report and this stuff is hard to track down) 13:13:39 probably the portal branch change is the most important for this 13:17:21 anyways, MarvinPA: do you think ghosts should get disto? 13:17:45 i don't see why not, yeah 13:18:20 that's the change here that's least connected to griefing, so I'm not super committed to it on those grounds 13:19:26 it's probably not a ton worse than elec to bump into very early (and much rarer) 13:26:38 I guess my thought would be that it's a lot more annoying, though as you say maybe not a lot more annoying than any distortion monster 13:26:56 in combination with ghost randomness it can be though 13:28:56 well the only case I really see as to why ghosts are different (that's logical from a game design standpoint) 13:29:18 is that players are making ghosts with certain egos substantially more common in practice 13:29:44 by why ghosts are different I mean why ghosts should have their egos handled differently 13:30:23 it's not a problem that a ghost has one of these egos, since monsters have these egos generally, but if they're very common that's bad 13:30:43 that's why I suggested maybe just aggressively down-weighting their retention of those egos 13:32:34 now that their max speed is move speed is no better than normal move speed, they're going to be at least much easier to escape 13:32:34 I may be wrong, but I think the distribution of distortion weapons (especially early) tends towards monsters that are very unlike player ghosts 13:33:15 just because of the stats on what monsters can have weapons at all + what uniques get disto 13:33:52 I've got to go do some errands, bbl, if someone wants to change it back or further I'm fine with that :) 13:35:02 Yeah, but non-ghosts also follow players through stairs, so they're harder to ignore in many ways 13:35:09 Yeah I'll leave those changes to those who most want to see 0.21 tweaked 13:35:21 we can just be glad that at least we won't have to worry about this in the same way in 0.22 14:09:34 Unstable branch on crawl.jorgrun.rocks updated to: 0.21-a0-629-ga19c9dc5aa (34) 14:10:14 oh also re: speed, it seems like just fixing ghost speed at 10 would make more sense than the current version 14:10:42 yeah 14:14:09 -!- amalloy_ is now known as amalloy 14:16:33 03MarvinPA02 {GitHub} 07* 0.21-a0-630-g344d351: Further ghost adjustments 10(7 seconds ago, 1 file, 2+ 25-) 13https://github.com/crawl/crawl/commit/344d3514eb71 14:17:12 another FIXME fixed the only way i know how 14:19:10 heh 14:19:25 yeah, just setting speed to 10 may be better 15:09:38