01:35:41 ChrisM345 (L17 FoAK) Crash caused by signal #11: Segmentation fault (Shoals:1) 01:40:39 03Alex Jurkiewicz02 07https://github.com/crawl/crawl/pull/1253 * 0.25-a0-835-gc13bfda: Enhancements to webtiles game definitions 10(3 weeks ago, 7 files, 395+ 53-) 13https://github.com/crawl/crawl/commit/c13bfda030d8 01:40:39 03Alex Jurkiewicz02 07https://github.com/crawl/crawl/pull/1253 * 0.25-a0-836-g7de1179: Reduce log spam 10(3 weeks ago, 1 file, 2+ 2-) 13https://github.com/crawl/crawl/commit/7de1179e7ae5 01:40:39 03Alex Jurkiewicz02 07https://github.com/crawl/crawl/pull/1253 * 0.25-a0-837-g713cdae: Convert README to markdown and update 10(3 weeks ago, 2 files, 87+ 49-) 13https://github.com/crawl/crawl/commit/713cdae23492 01:40:39 03Alex Jurkiewicz02 07https://github.com/crawl/crawl/pull/1253 * 0.25-a0-838-g2004899: Remove version pin from Tornado dependency (advil) 10(3 weeks ago, 5 files, 21+ 18-) 13https://github.com/crawl/crawl/commit/200489945a45 01:40:39 03Alex Jurkiewicz02 07https://github.com/crawl/crawl/pull/1253 * 0.25-a0-839-g7f6036b: Improve some webtiles log messages 10(3 weeks ago, 2 files, 6+ 3-) 13https://github.com/crawl/crawl/commit/7f6036b9db42 01:40:39 03Alex Jurkiewicz02 07https://github.com/crawl/crawl/pull/1253 * 0.25-a0-840-g6403c02: Add some missing type information 10(3 weeks ago, 2 files, 14+ 18-) 13https://github.com/crawl/crawl/commit/6403c02320da 01:40:39 03Alex Jurkiewicz02 07https://github.com/crawl/crawl/pull/1253 * 0.25-a0-841-g06ff57b: Allow overriding some HTML templates 10(2 weeks ago, 7 files, 13+ 10-) 13https://github.com/crawl/crawl/commit/06ff57b6abb2 01:40:39 03Alex Jurkiewicz02 07https://github.com/crawl/crawl/pull/1253 * 0.25-a0-842-ga6b5c7d: Always show milestone column 10(2 weeks ago, 1 file, 0+ 4-) 13https://github.com/crawl/crawl/commit/a6b5c7dcb97b 02:05:54 rebase 02:14:16 likely story 02:45:04 Monster database of master branch on crawl.develz.org updated to: 0.25-a0-651-g8c222c7a2b 03:24:04 Unstable branch on crawl.beRotato.org updated to: 0.25-a0-834-g55863b0 (34) 03:54:26 Stable (0.24) branch on crawl.kelbi.org updated to: 0.24.1-5-g77f818bfca 08:56:27 rayg00n (L18 FoAK) Crash caused by signal #11: Segmentation fault (Shoals:3) 08:57:28 rayg00n (L18 FoAK) Crash caused by signal #11: Segmentation fault (Shoals:3) 09:11:17 !crashlog 09:11:19 21890. nxtlvl, XL27 KoFi, T:50823 (milestone): https://crawl.kelbi.org/crawl/morgue/nxtlvl/crash-nxtlvl-20200419-181221.txt 10:03:12 I backported the fix for that yesterday, but I guess it would need manual rebuilds on some servers 10:03:38 oh that's a different crash 10:03:44 !crashlog rayg00n 10:03:47 No milestones for rayg00n (crash). 10:41:35 !lm rayg00n 10:41:36 2286. [2020-03-22 17:35:04] rayg00n the Spear-Bearer (L14 DrFi of Qazlal) entered the Snake Pit on turn 21404. (Lair:3) 10:41:42 !lm rayg00n FoAK 10:41:43 88. [2020-03-14 11:10:44] rayg00n the Skirmisher (L6 FoAK of Lugonu) entered a Sewer on turn 3394. (D:4) 10:41:47 !lm rayg00n FoAK x=src 10:41:48 88. [2020-03-14 11:10:44] [src=cue] rayg00n the Skirmisher (L6 FoAK of Lugonu) entered a Sewer on turn 3394. (D:4) 10:42:22 !lm rayg00n FoAK s=src 10:42:22 88 milestones for rayg00n (FoAK): 88x cue 10:42:22 Sequell was missing a bunch of crashlogs yesterday too, I think it's the ongoing Sequell problem 10:42:22 yeah 10:42:22 !lm rayg00n FoAK -log 10:42:22 rayg00n, XL6 FoAK, T:4782: https://underhound.eu/crawl/morgue/rayg00n/morgue-rayg00n-20200314-111427.txt 10:42:27 https://underhound.eu/crawl/morgue/rayg00n/crash-rayg00n-20200422-125627.txt 10:42:43 they pinged me on cxc discord, it involves using lucy corruption in shoals 10:43:04 ah, dynamic_cast in the stack trace 11:12:46 03Aidan Holm02 07[ifdefs-begone] * 0.25-a0-839-gfdb8587: fixup! Remove code for old potion types 10(6 minutes ago, 2 files, 6+ 2-) 13https://github.com/crawl/crawl/commit/fdb8587aa048 13:06:08 Stable (0.24) branch on crawl.akrasiac.org updated to: 0.24.1-5-g77f818b 15:31:50 -!- aidanh_ is now known as aidanh 17:39:25 -!- jfcaron_ is now known as jfcaron 17:55:33 03Alex Jurkiewicz02 07https://github.com/crawl/crawl/pull/1253 * 0.25-a0-843-g65e1c1d: Fix lint errors 10(53 seconds ago, 1 file, 16+ 9-) 13https://github.com/crawl/crawl/commit/65e1c1da0b10 17:55:45 https://github.com/crawl/crawl/pull/1253 should be in a mergeable state again, fixed conflicts and made some lint corrections 18:19:57 I still have to make time to test it in my dgl test setup, sorry 18:20:18 my time is fairly compressed until the end of next week, though it gets better after this week 18:22:58 -!- amalloy is now known as amalloy_ 18:34:08 -!- krator44-- is now known as krator44 19:07:14 all good, thanks for the update 22:39:02 After tasting new ghost system you start really appreciating it when you go back to the old one. I experience a strong desire to run a few scavenger characters through early dungeon. 23:04:11 monsters should probably avoid tiles affected by frozen ramparts, but they don't 23:04:22 should be similar to clouds, I would think? 23:04:25 they do, but they have a means for evaluating that 23:04:37 and will step on the tile based on those conditions 23:04:55 oh, I haven't noticed them avoiding them, even very weak monsters who have alternative paths to me 23:05:26 yeah, I have a howler monkey that just approached me through ramparted tiles even though it could have moved diagonally and still approached 23:05:55 same with these orcs, they just don't seem to care 23:05:55 if monsters have more than half their hp, they'll step on the tiles if there's no safe path 23:06:00 there are other safe paths 23:06:13 just had an orc enter a ramparted tile while approaching me even though it didn't have to 23:06:18 they're suiciding themselves 23:08:16 it's easily reproduced, I can do it on demand 23:14:18 It's the same with Conjure Flame. Monsters start caring about it only at some amount of their hp left. 23:14:39 in my experience, monsters won't walk through clouds if they don't have to 23:14:48 they will if there's no other way to approach, and they have enough HP (or some other metric) 23:15:01 they're definitely treating frozen ramparts differently 23:15:17 they are not avoiding it under any circumstances so far 23:15:19 well did this howler monkey not have full HP? 23:15:23 or high HP rather 23:15:38 hounds and orcs are also doing it 23:15:39 You seem to have little experience. Tough monsters with full hp like hydra or giant will happily step into flame cloud even if there are other ways. 23:15:51 Yermak: Yes, I know that. I'm not saying otherwise. 23:16:05 oh, I see what you mean 23:16:09 right, it's based on monster intelligence and the proportion of HP 23:16:11 weaker creatures don't do that though 23:16:18 gammafunk: should a plain orc be walking through them needlessly? 23:16:49 if its hp is high? yes 23:16:49 even if there's another path with the same cost? 23:17:07 monsters do that for clouds too! what you're seeing is that the calculation for clouds is more exacting in terms of HP 23:17:20 for damaging walls it's simplay "is my HP over half my max HP?" 23:17:36 it's pretty suicidal on their part 23:17:46 for clouds the logic is a bit more elaborate, but it likewise does not care about any notion of alternate paths 23:18:07 it's just an "is this too much damage for my HP?" calculation 23:18:18 you could argue that the logic should be the same, I guess 23:18:30 what adjective is 50%? Is that "heavily wounded"? 23:18:56 ??almost dead[2 23:18:56 wounded[1/1]: 100% not (green); 99-81 lightly (green); 80-61 moderately (brown); 60-41 heavily (magenta); 40-21 severely (red); 20-1 almost dead (red). 23:19:04 yeah, heavily 23:19:17 thanks 23:19:31 np, I'd forgotten how clouds code does take more pains to make its hp calculations 23:20:58 right now it's just really easy to cheese a lot of monsters with it 23:23:48 and I can confirm an orc warrior at "almost dead" did avoid walking through them, so that is working 23:25:03 I will say the new spells are pretty fun for the most part, although visually judging range of hailstorm is a bit annoying 23:25:37 we could probably make the logic used for damaging walls more consistent with that used for clouds 23:25:59 unfortunately neither clouds nor walls really consider that there are alternate non-damaging paths 23:26:29 I wish there were some easy way to provide info like targeting range and MR penetration chance for non-targeted spells 23:26:29 they both just pick a logical default direction based on the difference between target (you in this case) and the monster's position 23:26:33 hailstorm, mass confusion, etc. 23:27:14 and then make a little grid of all potential next moves that could be viable and ask "is this logical move in this set? if not, choose a logical backup, otheruse use that logical move" 23:27:45 so if the higher-hp monster could take a trivial alternate path to avoid a cloud/wall, it won't choose to do so 23:28:28 it's just that the HP conditions of clouds are a bit more stringent, so it will more often be forced to choose an a backup location (that will avoid the cloud) 23:29:02 yeah, and ramparts puts a bunch of damaging tiles right around you, where monsters will usually be walking 23:29:24 it's normal cold damage right? Asking because there's no message about the monster resisting 23:29:28 so I think unifying the logic for the HP calculation could help ramparts a bit, and for the other bigger issue, that would take a bit more research 23:29:45 it is just BEAM_COLD, yeah 23:29:51 would maybe open up more strategic uses for it, as it would keep monsters away from you sometimes 23:30:47 well, I think the important reason for the change is consistency, to help players be able to reason about what monsters would do 23:31:12 logical expectation is that they'd evaluation damaging walls similarly to how they would damaging clouds 23:31:28 yeah, that's pretty much what I expected 23:31:41 intelligent monsters with like 3 max HP walking onto damaging tiles was unexpected 23:31:43 so yeah, I'll look into making a fix for that; thanks for the report 23:31:50 thanks 23:38:35 gammafunk: it's known that Sequell is missing some games/milestones, right? 23:38:43 elliptic: yeah, I've asked greensnark about it, but no word back yet 23:38:52 CAO scoring should be working ok though 23:42:18 I see, nice 23:42:27 i guess an alternate implementation could be to make monsters ignore damaging walls and clouds. That would be even easier for the player to reason about! 23:42:42 it's weird because sequell seems to have some recent games/milestones from the same logfile 23:44:02 I guess not that much more recent 23:44:30 elliptic: yeah, it's intermittently missing stuff 23:44:48 greensnark had mentioned disk issues on the server and the need to move Sequell to another server, some time ago 23:44:57 not sure if this is related, or if it's a network problem 23:45:06 on a different subject, did something change about trog piety "recently" (by which I mean in the past year)? 23:45:33 not to my knowledge 23:45:51 https://crawl.kelbi.org/crawl/morgue/elliptic/morgue-elliptic-20200420-031506.txt 23:46:08 this character found it actually impossible to reach 6* piety 23:46:42 even with faith on for quite a while and not really using abilities, etc 23:47:24 and I kept on falling to 4* piety, and then getting a gift immediately upon going back up to 5* 23:48:16 I did notice my last trog character did seem to have trouble keeping piety above 5* 23:48:35 it definitely felt like something major had changed with how gifts interacted with piety 23:48:45 and I wasn't sure if it was an intentional trog nerf or not :P 23:48:50 but yeah I can't think of any outright change, nor does a quick 'git log -i --grep piety' pull up anything 23:49:00 (on the other hand, oka piety feels normal to me) 23:49:28 the only recent substantive change was trog not gifting ranged weapons nor ammo 23:49:28 that and restricting the set of brands used for gifts 23:49:33 but neither of those should affect piety gain 23:49:59 wonder if there was some trog piety gain adjust in that first commit, let me see 23:51:39 was there some way for gifts to be vetoed which was getting hit more frequently before those changes, perhaps? 23:57:07 it looks like that commit made gifts guaranteed as soon as you got to 5* piety 23:57:17 specifically for trog 23:57:26 the logic just totally changed 23:58:21 in _give_trog_oka_gift in religion.cc