02:47:39 Unstable branch on cbro.berotato.org updated to: 0.27-a0-79-gcaaa221d37 (34) 03:10:56 Irradiate description says "A bit loud" but in reality it's red noise. 13https://crawl.develz.org/mantis/view.php?id=12446 by Yermak 03:16:41 Bad Item 13https://crawl.develz.org/mantis/view.php?id=12447 by FakerFangirl 03:25:03 03wheals02 07* 0.27-a0-80-g58ad9d9: Quiet down Irradiate (#12446) 10(3 minutes ago, 2 files, 1+ 2-) 13https://github.com/crawl/crawl/commit/58ad9d9fa3b3 03:47:16 Stable (0.26) branch on cbro.berotato.org updated to: 0.26.0-11-g8f856b5395 03:50:27 Stable (0.26) branch on crawl.kelbi.org updated to: 0.26.0-11-g8f856b5395 03:52:45 Unstable branch on crawl.kelbi.org updated to: 0.27-a0-80-g58ad9d9fa3 (34) 03:54:27 Fork (bcrawl) on crawl.kelbi.org updated to: 0.23-a0-3840-g9efd504fd6 04:09:15 wheals, you made Irradiate even more op than it was already (: 04:24:12 Weird thing: I've been wearing Thief cloak through Vaults and Depths already and don't remember it leaking fog. In one of my previous pre-tourney games it leaked fog just fine. 04:25:02 <12e​bering> I had it in my BaNe game and it was releasing fog 04:25:13 <12e​bering> what's the exact version string of your current game? 04:25:19 <12e​bering> !lg . bane x=vlong 04:25:21 <04C​erebot> 1. [vlong=0.26-b1-32-g5d43845318] ebering the Tortoise (L27 BaNe of Kikubaaqudgha), escaped with the Orb and 3 runes on 2021-01-13 16:11:41, with 1463291 points after 92393 turns and 10:22:22. 04:26:02 0.26.0-10-g1f06aea511 04:26:43 Could it be related to me using mostly Statue Form? 04:29:08 <12e​bering> probably, since that gives you ac+gdr, the exact check is dam > you.hp_max / 10 && coinflip() 04:29:25 <12e​bering> where dam is how much your hp will be reduced by a single hit 04:30:12 hmm 04:37:35 oh yeah and stat form gives bonus AC 04:37:39 erm 04:37:40 max hp 04:37:55 so it's reducing dam and increasing you.hp_max to be in it 04:42:52 Yep, finally got fog: The crystal spear hits you!!! 04:46:28 ah yes, crystal spear is always good for a big hit 04:48:19 I forgot if flying creatures activated traps nowaways: red devil hopping backward and not activating teleport trap is bug or not? 04:50:19 <12e​bering> Flying creatures should, yes. Sounds bug to me 04:57:05 Red devil doesn't activate traps by hopping backward. 13https://crawl.develz.org/mantis/view.php?id=12448 by Yermak 05:20:12 Cigotuvi's Monster is not listed in the Vanquished Creatures section of the morgue 13https://crawl.develz.org/mantis/view.php?id=12449 by KamiKatze 06:57:12 Why doesn't status -Potion get dispelled when you are hit with dispelling bolt? 08:17:21 Yermak: seems a bug, perhaps it's just missing from the dispellable status list 08:40:10 yes, it doesn't have D_DISPELLABLE flag: https://github.com/crawl/crawl/blob/390cb15686453f144104744ad67521a19ac96d17/crawl-ref/source/duration-data.h#L458 08:41:04 may be intentional? 08:41:40 Icons for spells and abilities sometimes show up in the monster description and sometimes don't 13https://crawl.develz.org/mantis/view.php?id=12450 by Lici_the_Crawler 09:56:34 Autofight_throw with Minor Destruction quivered (range 5) fizzles at a monster out of range. 13https://crawl.develz.org/mantis/view.php?id=12451 by Yermak 10:29:43 Unstable branch on crawl.akrasiac.org updated to: 0.27-a0-80-g58ad9d9 (34) 10:49:21 has anyone ever had thoughts about having the webtiles server in its own repo somehow? It might take a lot of work to disentangle, so I'm not sure how viable it is 10:50:51 advil: I had mentioned that too; aidanh didn't like the idea because of the difficulty of keeping codebases properly integrated 10:51:13 I'm not paraphrasing his reasoning very well, so you might want to look up the ##crawl-dev chat if you want to see his points 10:51:55 but he is in favor more generally of keeping the codebases together in the same repo rather than splitting up the integral parts that need to talk to each other into separate repos 10:57:32 yeah the version-specific stuff would be impossible to move I think 10:58:35 I was wondering because at least in principle the non-versioned parts of the server could be more independent of the game repo 11:19:16 hm, I think the remaining blinking in console targeting might just be an artifact of sending a full screen redraw somehow 11:19:23 I can't get it to happen locally 11:28:16 I just hold out hope that someday we can make a libcrawl 11:29:25 can move stuff like objstat and mapstat to python 11:29:36 I guess you could reimplement them both now in lua using dlua, that doesn't sound very fun 11:31:42 in some ways the seed explorer is very similar to those, it's a lot slower though 11:35:24 is lua/dlua just that much slower, or is the seed explorer doing a lot more processing somehow? 11:35:57 objstat is counting basically everything on the level each iteration and doing some running total 11:36:00 *totals 11:36:02 objstat/mapstat do skip a lot of aspects of levelgen as I understand it 11:36:15 whereas the seed explorer is trying to exactly duplicate in-game levelgen 11:38:57 hrm, I wonder what those aspects are. Objstat is supposed to be counting data for actual game levels as well, and mapstat does track vetoes, so it's not simply choosing sets of maps 11:39:54 I'm familiar with some aspects like overflow selection being a bit different, but it's been a long time since I've looked at that and have forgotten the differences 11:40:57 but if objstat weren't generating levels equivalent to what you could see in-game, at least on-average, there'd be a problem with its data 11:41:09 I probably should look through its invocation of the builder sometime and figure out what differences might exist 11:41:45 even if it resulted in a slowdown, it'd be better if its data had as much fidelity as possible compared to what you can get in-game 11:42:11 it is balling `builder()` so everything involved in actual map generation should be the same 11:42:12 *calling 11:42:32 but it skips all the infrastructure surrounding the builder call 11:42:57 I also have no idea if the various between-level constraints are tracked? e.g. not reusing vaults and so on 11:43:12 I had guessed not but I haven't looked very closely at the code 11:44:11 it's possible it isn't that much faster in the end 11:45:40 yeah, I should probably take a look at its approach to dungeon generation, then; I need to run data for 0.25 and 0.26, so would be good time to make sure objstat is actually getting correct results 11:50:39 I would think it's still pretty accurate at bulk statistics even if it doesn't try to mimic all the exact details 11:52:28 yeah, that's my hope at least! but I'd like to know and at least document the differences, if not actually bring it closer to reality 11:52:28 one could probably the objstat/mapstat bulk data collection stuff so that it works with seed explorer though 11:52:32 one reason why this is important is we sometimes use objstat/mapstat results to make balancing decisions about item/monster/vault weights etc 11:52:35 one thing I know is that I did not do anything with branch/level generation order in the mapstat code 11:52:52 so unless someone else did, it doesn't match the pregen order 11:53:09 but if it's resetting data between levels (which I think it might?) then that might not matter anyways 11:54:27 yeah, obviously both are run only on subsets of levels in many cases, so there's not really a notion of pregen order to begin with (and we do have to skip non-specified levels for performance reasons) 11:54:39 so you can e.g. run 100k iterations of d:1 and look at that, if you want 11:55:00 just get 1k iteration of e.g. swamp:4 or something, and expect it to run much more quickly 11:55:15 but for "3-rune" runs like I do, that could potentially use pregen order 11:55:25 running 1k of swamp:4 in the context of pregen would ... take some time :D 11:56:21 I did once generate like 1000 full seed descriptions, it wasn't completely horrendous to do 11:56:31 oh yeah, my 1k iterations of the 3 jobs to cover two 3 rune branch sets and one for extended takes like 8 hours utilizing 4 cores 11:57:06 well, 3 cores I guess, since it's only 3 jobs; oh actually no there are 4 jobs aren't there, so it is 4 11:57:21 but those are what I do for the final data and it takes a very long time 11:58:34 that is sort of interesting now that I think of it though, you can't really get perfect swamp:4 data unless you do run those previous levels and properly exclude vaults previously used for each iteration 11:58:51 but that shouldn't really matter much for statistical purposes like you said 11:59:42 hm yeah on a non-dgl build over ssh on CBR2 I'm still getting some blinking in targeting mode, I'm now thinking it is just the cost of a full screen redraw (which, I don't understand why it would be needed) 12:11:08 huh, this is not happening for me on cao 12:11:53 Different ncurses versions? 12:11:58 yeah I wonder 12:13:27 also fine on cko 12:27:11 cbr2 is nothing but problems! 12:44:47 I blame....ZureaL! 12:50:09 well, cbr2 is certainly using a newer version of ncurses than cao, unsurprisingly 12:50:09 from the docs it seems not entirely unexpected that a full clear happens on the next refresh() after a clear() call 12:51:42 wow, I misspoke yesterday. was thinking of clearok(curscr, TRUE) for immediate clear/redraw 12:52:18 but yes, clear() vs. erase() sets clearok(stdscr, TRUE) so a clear happens on the next refresh() 13:00:53 CURSES! Foiled again!! 13:43:49 03ebering02 07* 0.27-a0-81-g768c70b: Fix a teleport closet 10(7 hours ago, 1 file, 1+ 0-) 13https://github.com/crawl/crawl/commit/768c70b9d499 13:43:49 03ebering02 07* 0.27-a0-82-g0c344bb: Restore Zyme sickness 10(6 hours ago, 1 file, 1+ 1-) 13https://github.com/crawl/crawl/commit/0c344bb9d350 13:43:49 03ebering02 07* 0.27-a0-83-g80642e4: Don't offer conjured things god blessings 10(6 hours ago, 1 file, 4+ 1-) 13https://github.com/crawl/crawl/commit/80642e47ccf2 13:43:49 03ebering02 07* 0.27-a0-84-gb0dbf93: Simplify Tornado interactions with distance movement 10(5 hours ago, 7 files, 14+ 49-) 13https://github.com/crawl/crawl/commit/b0dbf93ca225 13:43:49 03ebering02 07* 0.27-a0-85-g28f6fc2: Anger monsters who block player attacks 10(3 hours ago, 2 files, 12+ 4-) 13https://github.com/crawl/crawl/commit/28f6fc227366 13:43:49 03ebering02 07* 0.27-a0-86-gf40771b: Remove an obsolte comment 10(3 hours ago, 1 file, 0+ 5-) 13https://github.com/crawl/crawl/commit/f40771be7fcf 13:43:49 03ebering02 07* 0.27-a0-87-g0fe236c: Adjust the charging of Majin-Bo's hp cost 10(3 hours ago, 1 file, 32+ 27-) 13https://github.com/crawl/crawl/commit/0fe236cf81fa 13:43:49 03ebering02 07* 0.27-a0-88-gea5d599: Fix a tension check for runic seals (8085, 10542) 10(82 minutes ago, 1 file, 10+ 3-) 13https://github.com/crawl/crawl/commit/ea5d59904d4e 13:43:49 03ebering02 07* 0.27-a0-89-g897d3da: Don't update constriction when testing door seals 10(59 minutes ago, 1 file, 4+ 4-) 13https://github.com/crawl/crawl/commit/897d3da516b7 13:43:49 03ebering02 07* 0.27-a0-90-g33242fe: Prevent trampling a constricted actor (12432) 10(37 minutes ago, 1 file, 15+ 5-) 13https://github.com/crawl/crawl/commit/33242fe616db 13:43:49 03ebering02 07* 0.27-a0-91-g4f573a8: Fix a teleport closet (MakMorn) 10(6 minutes ago, 1 file, 1+ 0-) 13https://github.com/crawl/crawl/commit/4f573a835dff 13:46:44 advil: fixing teleport closets will touch the des cache, do you know if it impacts seed stability 13:47:31 it will 13:47:54 dang rip 13:48:14 hm, which things to backport during t, always a fraught question 13:49:05 I guess these are all "quirks" and its better to notionally have people play the same version 13:49:10 possibly none? 13:49:19 the game breakers are the teleport closets but those hit seed stability 13:50:11 the tele closet is the only one that actually does look like something I would backport, but yeah 13:50:23 maybe the zyme one if I'm understanding it right and zymes are currently broken 13:50:37 <12e​bering> yes 13:50:42 <12e​bering> currently zymes are melee only 13:51:34 Unstable branch on crawl.kelbi.org updated to: 0.27-a0-91-g4f573a835d (34) 14:11:03 New branch created: ncurses-erase (1 commit) 13https://github.com/crawl/crawl/tree/ncurses-erase 14:11:03 03advil02 07[ncurses-erase] * 0.27-a0-80-gbfda08c: Use erase() for some ncurses screen clears 10(4 minutes ago, 11 files, 38+ 15-) 13https://github.com/crawl/crawl/commit/bfda08c6d931 14:14:27 ebering I think both of those closet fixes should be fine to merge actually 14:14:38 afaik just adding no_tele_into won't change seeds 14:15:09 and maybe removing closets >> seed stability anyways, not sure 14:23:04 hm, I suppose people must still use console spectating since we got reports about it 14:23:04 but this ttyplay stuff is janky nonsense 14:26:08 04Build failed for 08ncurses-erase @ bfda08c6 06https://github.com/crawl/crawl/actions/runs/486182471 14:39:27 -!- aidanh_ is now known as aidanh 14:44:47 03advil02 07[ncurses-erase] * 0.27-a0-80-gc1b49d3: Use erase() for some ncurses screen clears 10(38 minutes ago, 12 files, 39+ 16-) 13https://github.com/crawl/crawl/commit/c1b49d350d1e 14:47:02 well, that commit does fix blinking in targeting mode on cbr2, still not sure I understand why 14:48:12 well, I understand every part of why except why it's not noticeably blinking elsewhere 14:49:09 I do wonder if somehow, "padding" is getting added to the clear code, but I have not been able to ascertain how to debug that 14:49:21 terminfo looks fine 15:29:52 03advil02 07[ncurses-erase] * 0.27-a0-80-g4201424: Use erase() for some ncurses screen clears 10(83 minutes ago, 12 files, 29+ 20-) 13https://github.com/crawl/crawl/commit/4201424199f6 16:00:00 How should I go about reporting a crash on a 0.26 game that just happened in CDO? It crashed when I was entering Slime, and I got the URL from the backup save game feature, but I don't think I have anything else useful to say about the bug... 16:02:24 (and I didn't see fresh crash reports in Mantis nor anything that looks like a crash bugfix in git log after the 0.26 release) 16:05:02 <12e​bering> what's your username 16:05:02 <12e​bering> on cdo 16:05:34 !crashlog 16:05:35 23312. OrcKing, XL11 MiFi, T:9524 (milestone): https://crawl.project357.org/morgue/OrcKing/crash-OrcKing-20210113-232542.txt 16:05:45 !lm * cdo crash 16:05:50 2743. [2020-07-23 18:37:53] hayenne the Grappler (L8 PaCK of Xom) Crash caused by signal #15: Terminated (D:5) 16:05:55 hm 16:06:00 Discord|: Rjs 16:06:12 !lm * cdo 16:06:13 3067052. [2021-01-14 20:36:13] Rjs the Unseen (L13 KoAK of Lugonu) reached level 6 of the Lair of Beasts on turn 16118. (Lair:6) 16:06:25 Yes that one 16:06:25 !lm * cdo x=cv 16:06:26 3067052. [2021-01-14 20:36:13] [cv=0.26-a] Rjs the Unseen (L13 KoAK of Lugonu) reached level 6 of the Lair of Beasts on turn 16118. (Lair:6) 16:06:58 crash reports on github or mantis are fine, wondering why it didn't give us a crash milestone 16:07:08 !lm Rjs cdo -2 16:07:09 1334/1335. [2021-01-14 20:31:55] Rjs the Unseen (L13 KoAK of Lugonu) killed Nessos on turn 15225. (Lair:4) 16:07:11 !lm Rjs cdo -3 16:07:12 1333/1335. [2021-01-14 20:27:24] Rjs the Covert (L13 KoAK of Lugonu) left the Swamp on turn 15042. (Swamp:1) 16:07:58 mantis is less used these days because it doesn't allow new account creation 16:08:42 I went from L:6 back to L:5 where the Slime pit entrance was, and then entered Slime for the first time. (So maybe it was creating the Slime levels when it crashed?) 16:09:13 <12e​bering> !crashlog Rjs 16:09:14 I don't think I have an account on Mantis but I do have one on github. Or I could maybe just paste the URL that I got on here? 16:09:14 <04C​erebot> 1. Rjs, XL3 FeSu, T:3444 (milestone): http://crawl.develz.org/morgues/0.10/Rjs/crash-Rjs-20120704-170543.txt 16:09:35 <12e​bering> that's not the correct game 16:09:59 <12e​bering> there's no crashlog in the cdo morgue directory 16:10:02 <12e​bering> for 0.26 16:10:47 you can paste it here but it might vanish in the sands of time if no one fixes it immediately :-) 16:10:47 Weird... This is what I got from the Backup save game thing: http://crawl.develz.org/saves/dumps/Rjs-fce45f26fa-210114-2050.cs 16:11:13 what was the message with the crash? 16:11:17 if you know 16:12:29 Unfortunately I didn't see any, probably because I had ssh in a terminal that auto-exited when the connection closed... (I wonder if the tty recording is somewhere?) 16:13:31 was there any rollback when you logged back in? 16:13:51 I haven't tried - should I try? 16:14:08 <12e​bering> yes 16:14:19 Ok I'll try 16:14:54 it looks like that save is in lair so maybe 16:16:32 Apparently I got back in L:6 without any special messages (at least I _think_ I had returned to L:5 from there before I found the Slime entrance) 16:16:32 I only see a slime entrance in lair 6 fwiw 16:16:32 Hmm maybe it was L:6 then, I'm not quite sure.. I suppose I could try to find it again and see what happens then 16:17:02 Oh yes, it was in L:6. I'll try to get there again 16:17:07 yeah 16:18:14 Hmm so now I got to Slime without any errors so it didn't reproduce :( 16:19:46 yeah, I can't reproduce it either 16:20:10 I wonder if this was a connection drop that had some weird properties, rather than a crash per se? 16:20:21 though your exploration of lair 6 ought to have been saved 16:20:56 I guess it's possible though I haven't had any other connection issues with CDO 16:21:31 well, I think you're safe to keep playing in any case 16:22:00 New branch created: pull/1687 (1 commit) 13https://github.com/crawl/crawl/pull/1687 16:22:00 03Hatsuma02 07https://github.com/crawl/crawl/pull/1687 * 0.27-a0-92-ge4a9b34: donger sprites 10(11 minutes ago, 6 files, 0+ 0-) 13https://github.com/crawl/crawl/commit/e4a9b34f9a05 16:22:28 Ok... I'll also change my terminal to stay alive after the connection closes so I might be able to see some message if it happens again. (though if it was related to generating the dungeon, I suppose it's very unlikely that I would see it again) 16:22:34 hm there is a core file with the right timing: Jan 14 21:44 crawl-0.26-fce4-1610657075.core 16:28:04 does anyone have the wizmode bit set on cdo? would be useful to force a crash to see if it is behaving right 16:28:04 I don't think I do 18:25:42 Unstable branch on underhound.eu updated to: 0.27-a0-91-g4f573a835d (34) 19:20:14 Enslave Soul gave me hostile Rupert. 13https://crawl.develz.org/mantis/view.php?id=12452 by Yermak 20:07:06 -!- jfcaron_ is now known as jfcaron 20:52:12 Stable (0.26) branch on underhound.eu updated to: 0.26.0-11-g8f856b5395