00:01:05 Unstable branch on crawl.kelbi.org updated to: 0.31-a0-910-g02257934e3 (34) 00:01:29 04Build failed for 08master @ 02257934 06https://github.com/crawl/crawl/actions/runs/6529856608 00:34:09 Fork (bcrawl) on crawl.kelbi.org updated to: 0.23-a0-5070-gfcb7ac407c 00:58:34 -!- Discord|1 is now known as Discord| 02:28:49 -!- d[-_-]b is now known as stenno 03:22:51 Experimental (bcrawl) branch on underhound.eu updated to: 0.23-a0-5070-gfcb7ac407c 05:07:15 03kate-02 07* 0.31-a0-911-g32ab0262c4: Fix the build 10(5 minutes ago, 2 files, 10+ 8-) 13https://github.com/crawl/crawl/commit/32ab0262c485 05:19:34 Unstable branch on crawl.kelbi.org updated to: 0.31-a0-911-g32ab0262c4 (34) 05:31:08 Unstable branch on crawl.akrasiac.org updated to: 0.31-a0-910-g0225793 (34) 06:30:11 <05icoson> hm that one was crashing on startup for online trunk, also 06:30:24 <05icoson> happens to early to get a log 06:31:18 <05icoson> so it's possible some rebuilds are still going to be needed 06:35:32 Unstable branch on crawl.akrasiac.org updated to: 0.31-a0-911-g32ab0262c4 (34) 06:48:31 03advil02 07* 0.31-a0-912-g472e003958: build: fix some warnings 10(20 seconds ago, 1 file, 2+ 2-) 13https://github.com/crawl/crawl/commit/472e00395843 06:54:32 <05icoson> !versions 06:54:36 <05icoson> !version 06:56:03 <05icoson> &versions 06:56:25 <04Cerebot> CAO: 0.31-a0-881-ge136f0c, CBR2: 0.31-a0-889-g1a8549509d, CDO: 0.31-a0-881-ge136f0c47a, CKO: 0.31-a0-911-g32ab0262c4, CPO: 0.31-a0-889-g1a85495, CUE: 0.31-a0-885-gbe960a18b5, CWZ: none, CXC: 0.31-a0-889-g1a8549509d, LLD: none 06:57:42 <05icoson> I guess cwz has just opted out of sequell? 06:58:06 <05icoson> I tried to ask some questions about that a few weeks ago and didn't really understand the answers 06:59:25 <05icoson> the more I do this the more I think we really have to have (official) server admins we can directly communicate with and who actually communicate with us, it would suck for a language barrier preventing a server in some region, but the fact is, we bend over backwards for cwz and it still doesn't really work 07:03:13 <05icoson> also, I guess that versions command uses milestones, so won't really answer my question (since it crashes too early for milestones) -- but based on bot logging it does look like cao was the only one on that version 07:06:14 Unstable branch on crawl.kelbi.org updated to: 0.31-a0-912-g472e003958 (34) 07:11:42 <05icoson> imo a very low minimum standard for an official server is: (i) one or more admins who monitor at least one of the #crawl-dev (via irc or discord), the discord #admin-private channel, or at least github for posts with the relevant admin group tag; (ii) public-facing milestone/logfiles for actively available versions (ideally all versions that have been active), (iii) report expected major downtime that might impact scoring/sequell/etc and 07:11:42 ask for help on this if they need it. I guess one could make the case that (i) might simply include monitoring email (which I guess maybe hong does? I don't know what his email is). But cwz is at least failing (ii)-(iii) and has been for I believe months now, and we only get garbled 3rd-party explanations, and that only when we ask 07:12:13 <05icoson> oh, (iv) put up new versions in time for tournaments 07:29:20 03PleasingFungus02 07* 0.31-a0-913-geeba0b7a58: Give Occultists a hand (dinky) 10(6 minutes ago, 1 file, 0+ 0-) 13https://github.com/crawl/crawl/commit/eeba0b7a58e0 07:29:21 03PleasingFungus02 07* 0.31-a0-914-g4c35d5ed2b: Arcanists: lbolt -> fcloud (regret-index) 10(84 seconds ago, 1 file, 1+ 1-) 13https://github.com/crawl/crawl/commit/4c35d5ed2ba0 07:36:14 Unstable branch on crawl.kelbi.org updated to: 0.31-a0-914-g4c35d5ed2b (34) 07:39:52 <06pleasingfungus> seems reasonable, but unclear to me what we want to do about it. iirc cwz has a reasonably large playerbase which would have trouble moving to another server - don’t want to punish them 07:40:16 <06pleasingfungus> !lg * recentish s=src x=cdist(name) 07:40:35 <04Cerebot> 2675630 games for * (recentish): 575063x cko [8531], 572107x cao [7686], 464971x cbr2 [7058], 313935x cue [4695], 308046x cwz [3089], 279717x cxc [2998], 96967x cpo [1735], 37774x cdi [1], 16364x lld [1007], 10686x cdo [61] 07:43:14 <06pleasingfungus> !lg * recentish s=src x=sum(dur) 07:43:18 <04Cerebot> 2675639 games for * (recentish): 575066x cko [21y+63d+18:40:28], 572108x cao [24y+200d+22:21:56], 464974x cbr2 [17y+3d+14:30:49], 313936x cue [11y+306d+9:26:21], 308046x cwz [8y+211d+3:25:55], 279718x cxc [9y+150d+6:14:58], 96967x cpo [4y+34d+22:54:04], 37774x cdi [53d+10:50:56], 16364x lld [256d+6:25:09], 10686x cdo [108d+7:13:20] 07:44:01 <05icoson> there's nothing wrong with having a popular unofficial server, that's essentially what it is already if it's not posting logfiles 07:44:04 <05icoson> !lm * cwz 07:44:11 <04Cerebot> 15604329. [2023-08-07 18:38:37] ImOTab the Warrior (L18 OgWr of Uskayaw) reached level 4 of the Snake Pit on turn 34344. (Snake:4) 07:44:31 <05icoson> (I wouldn't propose delisting past ones or anything remotely like that) 07:46:20 <05icoson> so I don't actually see how anything we could do right this moment would actually punish them 07:48:25 <05icoson> I guess if hong does start publishing logfiles at some point and we refused to take them, that would be punishment; tbh though I would sort of propose that there's a point past which, this is what we should consider (not sure what that point is) 07:49:25 <05icoson> for any other server with an admin who doesn't have a language barrier that point might be well in the past 07:51:56 <06pleasingfungus> hm 07:52:20 <06pleasingfungus> if only we had sso 07:52:31 <06pleasingfungus> i wonder if there’s some way to simplify the sso problem 07:54:01 <06pleasingfungus> maybe i should write a doc with my thoughts on what the problems actually are 07:54:29 <06pleasingfungus> it’s been a long time since i thought much about it, i’m sure i’ve forgotten things 07:55:13 <05icoson> sso is going to create a whole new version of this problem, in that I haven't really thought of a way to do it that wouldn't require server admins to do a fair amount of stuff to enable it (my impression from the garbled communications I've had is that hong may have had trouble with some of the much smaller, but important, changes that server admins have had to do in the last few years, maybe around finally deprecating python 2 though I 07:55:14 don't know that for sure) 07:55:35 <06pleasingfungus> sure 07:56:02 <06pleasingfungus> but it seems like it’d be very helpful for allowing people to stand up additional/replacement servers when needed 🙂 07:56:31 <06pleasingfungus> btw, @semi_tonal @icoson ty for fixing my late night wiz commit 🙂 07:59:12 <05icoson> I wonder if there's an easy way to automate only updating to CI passing versions; ofc many CI failures are false positives but sometimes they aren't 08:00:30 <05icoson> or maybe get dgl-config to at least try to build the des cache on rebuild 08:03:09 <06pleasingfungus> good qs 08:03:39 <06pleasingfungus> i don’t think rebuilding the des cache would’ve helped here, fwiw? though it certainly would’ve in several past cases 08:03:52 <05icoson> I think this one did fail on a des cache rebuild, based on cao logs 08:04:09 <05icoson> 2023-10-16 05:42:02,541 INFO: #201500 P295106 ERR: /crawl-master/crawl-git-02257934e3/data/dat/des/portals/gauntlet.des:323: unknown monster: "wizard" 08:04:10 <05icoson> etc 08:04:28 <06pleasingfungus> hm 08:04:32 <05icoson> I didn't try it myself though 08:04:34 <06pleasingfungus> i ran it locally… odd 08:05:19 <05icoson> hm, possible there are ways it would fail on a new des cache that it wouldn't on an update? (which is non-ideal, but I wouldn't be surprised) 08:05:38 <05icoson> I believe servers are generating a fresh one for each trunk version 08:06:29 <05icoson> so "rebuild" is the wrong word 08:15:32 <05icoson> oh ha 08:15:37 <05icoson> this is simpler than it seemed 08:16:04 <05icoson> I can replicate the issue: when I build a fresh des cache on 1a8549509d1 (prior commit), then update to 129f0b0447d, I don't get an error on either running the game or calling builddb. But if I reset my des cache and then start the game or call builddb, I do get an error 08:16:34 <05icoson> but the explanation is simply that because gauntlet.des itself was not changed as part of the patch, its mtime hasn't changed, and it won't rebuild 08:17:03 <05icoson> this probably would even cause issues with local runs of make test, since I don't think that explicitly resets the des cache? 08:19:46 <05icoson> yeah, make test uses caching des rebuild...so unless some part of the test suite invokes these exact maps (could happen by chance I think), but maybe not even then, this case will appear to pass tests locally 08:40:42 <06pleasingfungus> nice 09:27:58 <04⸸※> (the overall arcanist / occultist split is enough of an improvement over the prior situation that I'm not going to bother immediately trying to fix them up further, but I'm not going to solve the whole "we have a very limited number of ways monsters fire attacks" issue right after waking up, and there haven't exactly been any games to assess arcanists. just noting that deep elf elementalists basically never cast freezing cloud and 09:27:59 usually kill themselves when they do so, but I'm still doubtful about freezing cloud. tempted to try to get this vitrifying hex working to see if it'd be more or less dangerous than para here.) 10:18:48 <05icoson> I do think there just has been a bit less critical mass for active discussion here than there sometimes is, which then also creates a cyclical situation 10:45:14 *Random dude's experience with "new" Hexslinger* incoming: Jinxbite is okay altough the -magic resist can be obviously very dangerous and the spell becomes a liability later on but that's okay I guess for a low level spell. Hated sigil at first because of the TONS of obvious flaws I begrudgintly accepted that it's worth picking up because there 10:45:15 will be that ONE time you can use it to keep that ogre in open space pinned for a split second you need to escape. Still very high on the amnesia list. Inner Flame is still very good. Actually trying to play as AM is still pretty rough... unless you take the cheap way out and play spriggan but that feels so wrong. I guess gnoll is the next best 10:45:15 thing since it's the quick-fix solution to any gish build. Anyway, carefully needing to balance all 3 attributes, juggle the right skills at the right time, getting good armour is a pain because you can't upgrade too quick or you can't shoot fast enough or cast spells. Finding a decent launcher is painful too, +0 orcbow is your new best friend for 10:45:16 the first two runes... Guns are in an akward place, you can't hold off on upgrading the first halfway decent arbalest until you find one (IF you find one) and then it's probably not worth switching. And if you find one early you can't use it because you don't have the skill for it. The smoke is whatever but the noise is actually noticeable... Still 10:45:16 disappointed it doesn't shoot bullets, it's alchemical/magical enough without that weird design choice... 10:50:10 I unsurprisingly forgot dimensional bullseye which is just plain bad... Sounds awesome at first because you DOUBLE your damage, right? Right? In actuality you want to limit fighting multiple monsters and when you do face multiple monsters you almost 100% have something better to do than wasting a turn setting this up and the MP/shot cost is just 10:50:11 insulting... Absolute best case scenario you can use it like the old PP in a very roundabout way to hit the backline but jinxbite and inner flame usually do the trick 10:52:42 Did I mention it's a level 4 spell? Imagine all the good things you could be having insetead of nerfed PP. It's also unintuitive because you have to shoot something other than the monster with the magic bullseye you just cast on it... 11:28:21 <04⸸※> there sure are multiple reports of begrudging "hit another thing for free" being a spell, though I have to wonder if the friction of bothering for magic at all with the power of mostly going full ranged combat would leave bullseye fine at level 3 11:29:02 <04⸸※> flavour arguing back and forth aside what does firing bullets and not bolts from hand cannon actually mean mechanically 11:30:22 <04⸸※> why is there a pressure for HS to balance all 3 stats when focusing dex increases both an offense and a defense simultaneously? 11:37:08 <04⸸※> (frankly speaking I think it's more unintuitive that inner flame is something to also use 1v1 for free flame clouds and to just facetank the explosion as a strict downside than it is for a spell of Doing Trick Shots to not just fire straight on) 11:38:37 <04⸸※> ...they left before I even typed up any of that reply, didn't they 11:56:06 yes 11:56:24 drive-bys without waiting for responses are popular these days 11:59:52 <06pleasingfungus> i’m probably going to make hand cannons shoot bullets at some point 12:00:03 <06pleasingfungus> just felt silly adding a new item type for it 12:00:35 <06pleasingfungus> and don’t want them to shoot ‘sling bullets’ or slings to shoot ‘bullets’ 13:09:33 03elliptic02 07* 0.31-a0-915-g50735af31d: Make Sif gifts less frequent at lower piety. 10(3 minutes ago, 1 file, 1+ 1-) 13https://github.com/crawl/crawl/commit/50735af31dfb 13:22:12 Unstable branch on crawl.kelbi.org updated to: 0.31-a0-915-g50735af31d (34) 13:48:37 <09hellmonk> they should shoot hands 13:48:39 <09hellmonk> clearly 13:58:01 <06pleasingfungus> wow 13:58:08 <06pleasingfungus> now that's galaxy brain 14:00:48 03PleasingFungus02 07* 0.31-a0-916-g8387aacbc1: Give wizards their tongues back 10(71 seconds ago, 1 file, 5+ 1-) 13https://github.com/crawl/crawl/commit/8387aacbc100 14:06:12 Unstable branch on crawl.kelbi.org updated to: 0.31-a0-916-g8387aacbc1 (34) 14:09:11 03advil02 07* 0.31-a0-917-ga885b83477: feat: clo to force a full des cache reset 10(3 minutes ago, 7 files, 19+ 2-) 13https://github.com/crawl/crawl/commit/a885b83477d9 14:23:11 <03semi_tonal> %git fb8d33413f86 14:23:13 <04Cerebot> hellmonk {hellmonk3} * 0.28-a0-1274-gfb8d33413f: vampire castle bailey (2 years, 1 month ago, 1 file, 70+ 0-) https://github.com/crawl/crawl/commit/fb8d33413f86 14:23:38 <03semi_tonal> this vault probably shouldn't give zombies weapons, since they're not supposed to be able to use items (and it then doesn't properly display them) 14:25:28 <05icoson> I'm not seeing any zombies with weapons in the des? 14:27:49 <03semi_tonal> oh i guess they got added at some point later 14:27:49 <05icoson> oh it's not in that commit 14:29:27 <05icoson> %git 507d92f8637e054f2 14:29:28 <04Cerebot> hellmonk {hellmonk3} * 0.28-a0-1284-g507d92f863: various bailey fixes (1 year, 10 months ago, 1 file, 42+ 41-) https://github.com/crawl/crawl/commit/507d92f8637e 14:29:38 <05icoson> looks like it was thematic, the orcs became undead orcs 14:41:23 <09gammafunk> I have a vault that gives zombies weapons 14:42:02 <09gammafunk> Maybe that part got removed, but it's gammafunk_cloud_chamber 14:48:56 <03semi_tonal> hmm, i feel like that probably shouldn't use them either but they're not as simple to replace there 14:50:55 Unstable branch on crawl.kelbi.org updated to: 0.31-a0-917-ga885b83477 (34) 14:56:10 <06pleasingfungus> hm, zombies can't use items? don't they use starting items? 14:56:16 <06pleasingfungus> am i out of date again 15:01:42 <04⸸※> (they drop their equipment as they're animated but vaults can place them with equipment, yes) 15:06:32 <09gammafunk> Actually I have a plan for that vault use a runed door for the inner door and I have more poison monsters available (I think) 15:07:37 <09gammafunk> main reasons I used zombos are because I needed rpois monsters that couldn't open that inner door 15:07:57 <09gammafunk> But making it runed solves this 15:08:48 03kate-02 07* 0.31-a0-918-gebf5b86a2d: Remove weapons from some zombies in a bailey 10(33 minutes ago, 1 file, 1+ 1-) 13https://github.com/crawl/crawl/commit/ebf5b86a2d69 15:21:07 Unstable branch on crawl.kelbi.org updated to: 0.31-a0-918-gebf5b86a2d (34) 15:25:22 Unstable branch on underhound.eu updated to: 0.31-a0-917-ga885b83477 (34) 15:30:00 <06pleasingfungus> hm, vaguely wonder if that's a bug 15:30:06 <06pleasingfungus> i know that sort of thing kept breaking 15:37:23 <03semi_tonal> i'm pretty sure it was changed intentionally, although i don't remember when (but quite a long time ago i think) 16:21:09 07|amethyst02 * 0.15-a0-2177-g8515cbe870: Allow giving derived undead equipment in mspecs (#8807) 10(9 years ago, 1 file, 5+ 1-) 13https://github.com/crawl/crawl/commit/8515cbe8708e 16:21:09 %git 8515cbe8708e 16:21:15 hmm 16:56:31 03PleasingFungus02 07* 0.31-a0-919-g3060d423dc: Unban Ozocubu's Armour + Statue Form (Lici) 10(87 seconds ago, 2 files, 0+ 10-) 13https://github.com/crawl/crawl/commit/3060d423dca6 17:12:08 Unstable branch on crawl.kelbi.org updated to: 0.31-a0-919-g3060d423dc (34) 17:12:14 03PleasingFungus02 07* 0.31-a0-920-g28cbb04a8c: Remember a dream 10(71 seconds ago, 3 files, 9+ 1-) 13https://github.com/crawl/crawl/commit/28cbb04a8c86 17:21:24 heh, 3 deep in swamp monsters, a stone statue fills beasties with inner flame ... iron shot one of them and *boom* *Boom* *BOOM* 17:21:41 no more beasties 17:22:09 i tired a few of the variants last night ... pretty dreadful ... will stick with vanilla :) 17:33:23 Unstable branch on crawl.kelbi.org updated to: 0.31-a0-920-g28cbb04a8c (34) 23:42:37 -!- elliptic_ is now known as elliptic 23:52:12 Unstable branch on cbro.berotato.org updated to: 0.31-a0-920-g28cbb04a8c (34)