00:31:54 <Kelbell> Fork (bcrawl) on crawl.kelbi.org updated to: 0.23-a0-5096-ga651bb7bf3
03:19:05 <Monkooky> I've been poking at lightning rod code- the damage is
03:19:05 <Monkooky> dice_def(dice, div_rand_round(45 + power / 4, arc + 2));
03:20:01 <Monkooky> I'm still working through trying to figure out how arc is actually computed, so I might find the answer on my own, but anyone have a notion why the denominator there is arc + 2?
03:20:43 <Monkooky> Arc is already required to be nonzero, which is also a little strange
03:23:41 <Lantell> Experimental (bcrawl) branch on underhound.eu updated to: 0.23-a0-5096-ga651bb7bf3
03:24:12 <charlie5> would it be feasible to have character movement 'smoothed' out ?
03:24:53 <charlie5> ie interpolated from current square to destination square
03:43:31 <Monkooky> would be a lot of work, and I think the static player image sliding around would look kinda out of place
03:58:22 <charlie5> ok
04:13:05 <Monkooky> I think proportional armour needs to be more effective
05:06:51 <Henzell> Unstable branch on crawl.akrasiac.org updated to: 0.31-a0-971-g75a2e9a (34)
07:15:01 <Discord|> <05i​coson> @gammafunk if I add a per-socket idle timer in the webtiles lobby, would that break beem? (Or is it now reading the json lobby?)
08:20:46 <NotCheibriados> 03advil02 07* 0.31-a0-972-gf86a31d22a: fix: don't report stopped webtiles processes as idle 10(3 hours ago, 1 file, 1+ 2-) 13https://github.com/crawl/crawl/commit/f86a31d22a53
08:20:46 <NotCheibriados> 03advil02 07* 0.31-a0-973-g6ef4313e94: refactor: improve webtiles socket/process cleanup 10(2 hours ago, 1 file, 40+ 20-) 13https://github.com/crawl/crawl/commit/6ef4313e94b4
08:20:46 <NotCheibriados> 03advil02 07* 0.31-a0-974-g2a3aa404fa: feat: a webtiles lobby idle timer 10(14 minutes ago, 3 files, 63+ 3-) 13https://github.com/crawl/crawl/commit/2a3aa404fa2d
08:21:39 <Discord|> <05i​coson> I guess that question now becomes, "does that lobby idle timer break beem"
08:34:41 <Kelbell> Unstable branch on crawl.kelbi.org updated to: 0.31-a0-974-g2a3aa404fa (34)
08:37:27 <Discord|> <09g​ammafunk> it should simply restart the lobby socket when it sees it closed, and 3hrs shouldn't cause any problems, but thanks for the heads up. I'll keep an eye out and modify its behaviour if need be
08:39:50 <Discord|> <05i​coson> ok thanks. I could special case the username too (basically add an is_bot function that is a stub for future dgl bot flag, but just checks for username "beem" for now)
08:40:26 <Discord|> <05i​coson> it'll take a while for this to filter out to all servers, since it needs a restart, but I will restart cao with these commits sooner rather than later
08:42:00 <Discord|> <05i​coson> I suppose another option would be to support a ping message or sth
08:43:02 <Discord|> <09g​ammafunk> it's had to deal with all manner of weird availability behavior due to e.g. cwz, so its lobby class has always had reasonable checks in place to restart the task when something goes wrong
08:43:51 <Discord|> <09g​ammafunk> however it's possible that eventually its model of handling lobby data won't align with how we want our webtiles servers to operate and I'd need to make it use your json output
08:44:26 <Discord|> <09g​ammafunk> but I would prefer to modify beem to work in accord to how we'd like clients to operate over having special cases for its username, if possible
08:45:20 <Discord|> <09g​ammafunk> so I should look into using said json output at some point when I have a bit of extra time to work on non-qw stuff
08:45:30 <Discord|> <05i​coson> I think eventually if we do have an admin-managed bot flag I would consider not applying the timeout to bots (similar to how it is not applied to admins)
08:47:50 <Discord|> <09g​ammafunk> I think I still haven't merged my dgl branch stuff. I've extended the bot flag functionality bit to have crawl startup arguments specific to that user class
08:48:27 <Discord|> <09g​ammafunk> the benefit being that I can have the arguments added even for bot users when they're not using the designated bot installs
08:49:07 <Discord|> <09g​ammafunk> so that e.g. qw can run without throttle if I run it on the untracked public trunk on cdi (instead of only allowing it to do so on the track bot install available only to qw/bots)
08:49:19 <Discord|> <09g​ammafunk> but that stuff is again not really relevant to anyone but me
08:50:11 <Discord|> <09g​ammafunk> so I guess I should just exclude all of it when I get around to a merge, but having official support for a bot flag would be nice, even if it does nothing fancy crawl-side
08:51:19 <Discord|> <09g​ammafunk> but that's sort of how we handle wizard users and it might be nice to have crawl startup arguments that can optionally be added for any of these user classes (and then wizard users make use of this)
11:01:06 <advil> !lm * cxc crash -log
11:01:12 <Sequell> 1130. JaaP, XL16 MiFi, T:31476 (milestone): https://crawl.xtahua.com/crawl/morgue/JaaP/crash-JaaP-20231027-160606.txt
11:01:59 <advil> I think there's some sort of tight unbounded loop there based on a reddit report
11:06:18 <advil> it does replicate on that seed with zin as the god
11:08:04 <advil> though from the code I don't immediately see why the player's god should matter
11:12:23 <Discord|> <05i​coson> ah, I see the bug
11:14:33 <Discord|> <05i​coson> just to dbl check something else, is it correct that zin hates all talismans? seems plausible, but I just want to make sure
11:16:20 <Discord|> <05i​coson> by "hates" I mean "despises"
11:21:20 <Discord|> <03s​emi_tonal> it looks like it to me - my guess from looking it at just now was that the randart name generator just happened to roll a name that specifies a god, and then in addition to that it also just happened to pick zin as the random god?
11:22:05 <Discord|> <03s​emi_tonal> oh hmm or maybe that wouldn't do it
11:23:51 <Discord|> <06p​leasingfungus> gotta be my bug, sorry
11:24:01 <Discord|> <06p​leasingfungus> advil: ya
11:24:17 <Discord|> <06p​leasingfungus> zin excommunicates you for using any talisman
11:27:27 <Discord|> <05i​coson> yeah the bug was from https://github.com/crawl/crawl/commit/4004991382f120dfaece3d1c88eefbfeb2bcbd76
11:27:47 <NotCheibriados> 03advil02 07* 0.31-a0-975-g5d48084ff3: fix: levelgen freeze with zin + randart talisman 10(8 minutes ago, 4 files, 11+ 8-) 13https://github.com/crawl/crawl/commit/5d48084ff38f
11:27:59 <Discord|> <05i​coson> has anyone noticed weirdnesses with prompts in console lately? It doesn't seem to be properly clearing the last line of a prompt message?
11:28:21 <Discord|> <05i​coson> can't remember this code being changed in a long time
11:34:43 <Discord|> <06p​leasingfungus> !lg . !tiles
11:34:44 <Discord|> <04C​erebot> No games for pleasingfungus (!tiles).
11:36:45 <Discord|> <05i​coson> also, the serpent talisman of Vehumet's Approval {Int-4 Dex+5} seems like a bit of a sad item  (that's the one that was crashing in that seed)
11:40:20 <Kelbell> Unstable branch on crawl.kelbi.org updated to: 0.31-a0-975-g5d48084ff3 (34)
11:58:05 <Discord|> <06p​leasingfungus> wow, it’s basically a free dex ring
11:58:10 <Discord|> <06p​leasingfungus> seems good to me
12:01:54 <Monkooky> oh yeah I was gonna argue about proportional armour before I fell alseep
12:04:15 <Monkooky> proportional armour is maybe ok as written for player targetted effects- but for monster facing effects it more or less makes armour ignorable
12:09:19 <Monkooky> Proportional armour is, per documentation, intended to be used for effects with many small hits, for which armour would be too effective- while reasonable, I would argue that armour should still be strong versus such effects.
12:14:06 <Discord|> <09h​ellmonk> is proportional ac used for anything other than polar vortex and summon forest damages
12:14:17 <Monkooky> yes, lightning rod damage
12:14:30 <Monkooky> which it really shouldn't be
12:14:50 <Discord|> <09h​ellmonk> lightning rod should either check half or none ac
12:14:52 <Discord|> <09h​ellmonk> agree with that
12:15:20 <Discord|> <09h​ellmonk> believe lately the push has been to make all lightning stuff check half
12:16:56 <Monkooky> yeah I was trying to figure out lightning rod dam so I could figure out if swapping to half ac was a buff or nerf, and by how much
12:17:20 <Monkooky> (if it's hitting for more than 20, swapping to half ac is a buff)
14:03:26 <NotCheibriados> 03PleasingFungus02 07* 0.31-a0-976-g643d43726f: Unban felid claw muts (Lici) 10(2 minutes ago, 1 file, 2+ 6-) 13https://github.com/crawl/crawl/commit/643d43726f5f
14:12:51 <NotCheibriados> 03PleasingFungus02 07* 0.31-a0-977-gc230f8ef24: Ban a grate warden play (Midn8) 10(59 seconds ago, 1 file, 6+ 3-) 13https://github.com/crawl/crawl/commit/c230f8ef24ec
14:21:00 <Kelbell> Unstable branch on crawl.kelbi.org updated to: 0.31-a0-977-gc230f8ef24 (34)
15:26:02 <Lantell> Unstable branch on underhound.eu updated to: 0.31-a0-977-gc230f8ef24 (34)
21:45:58 <Monkooky> if (!apply_chunked_AC(1, foe->evasion(false, mon)))
21:46:12 <Monkooky> what on earth is going on in the summon forest code
21:49:58 <Discord|> <09g​ammafunk> yeah I recall reading this in the past
21:50:26 <Discord|> <09g​ammafunk> it's an "evasion" check as to whether the damage applies, as I recall
21:50:46 <Discord|> <09g​ammafunk> it's not done in the usual way evasion is checked, but is using the EV stat
21:51:28 <Discord|> <09g​ammafunk> the core issue is that the trees are not actual monsters so you can't use the usual setup
21:52:15 <Discord|> <09g​ammafunk> although I guess proportional AC might fundamentally require a different type of check
21:54:26 <Monkooky> this is... not very similar to an actual evasion check
21:54:59 <Monkooky> like, EV 30 mons will have a ~32% dodge chance versus forest
21:57:25 <Discord|> <09g​ammafunk> well, I'm not sure if applying EV normally will be better for the spell's balance
21:57:31 <Monkooky> it does work as a bodge for EV checks, but I feel like there's better bodges
22:14:26 <Monkooky> with the current numbers summon forest is going to have better accuracy than a 27/27 UC dude with +4 slay against any targets with more than 9 ev
22:17:08 <Monkooky> this is a significant enough 'accuracy' that it should either be changed or made more learnable
23:21:08 <Cbrotell> Unstable branch on cbro.berotato.org updated to: 0.31-a0-977-gc230f8ef24 (34)
23:34:49 <Gretell> Unstable branch on crawl.develz.org updated to: 0.31-a0-977-gc230f8ef24 (34)
23:56:07 <Gretell> Windows builds of master branch on crawl.develz.org updated to: 0.31-a0-977-gc230f8ef24