00:13:34 <05sastreii> Oh, in that case i don’t think it’s necessary Gods abilities don’t have backgrouds, otherwise they’ll be clustered with pixels and difficult to read (like the current makhleb marks, they’re very hard to distinguish) So, if you want i can change colors or stick to the original palette, no big deal, but imo adding other stuff will result only in confusion 00:17:39 <05sastreii> Unfortunately i cannot explain myself better than this, hope it’s clear what i mean 03:34:17 Experimental (bcrawl) branch on underhound.eu updated to: 0.23-a0-5208-geafff8c3b6 06:44:18 Great day! 06:44:46 Does Webtiles provide an option for WebSocket keep-alive on both the client and server? 06:45:04 E.g. in cases the load-balancer is configured to kill idle connections. 06:55:12 It seems there's no such implemented: https://github.com/crawl/crawl/blob/90227b92788b3662e4776d39c0c9be67120a7dc9/crawl-ref/source/webserver/static/scripts/client.js#L1715 06:55:40 When I'll have more time for it, I'll try checking out an implementation. 10:00:24 Is there a sort of "goal" to keep winrates at about 1%? 10:07:55 <07zureal> I hope not, my winrate is no where near that 10:11:38 <04dracoomega> Observing changes in winrates over time can sometimes give a benchmark about whether certain things have gotten earlier/harder and whether this gives us headroom to make particular sorts of changes that we're independently interested in making, but I don't think there's anything like a specific target we actively aim to stick at 10:13:06 <04dracoomega> (In fact, overall winrates have generally risen over the years, even while the game has been considered to have grown generally harder in many areas in that span of time, very plausibly because the playerbase itself has gotten better over time, for various reasons. So it's not really a strict measure.) 10:19:28 clowns getting more experienced as the business evolve 10:22:27 <04dracoomega> 'Expected winrate' (if such a thing exists) also varies enormously from combo to combo, and I have no idea if anyone even has a gut impression of exactly how many times harder MuBe is supposed to be than MiBe, for instance 😛 10:22:36 <04dracoomega> Beyond 'a lot' 10:26:52 !lg * t0.32 !boring / won 10:26:53 1320/76152 games for * (t0.32 !boring): N=1320/76152 (1.73%) 10:27:04 !lg * t0.32 !boring s=char / won 10:27:06 1320/76152 games for * (t0.32 !boring): 104/4364x MiFi [2.38%], 45/2513x MiBe [1.79%], 38/3018x MDFi [1.26%], 34/1455x GrEE [2.34%], 30/1481x GrFi [2.03%], 23/2056x TrFi [1.12%], 17/45x FeAl [37.78%], 16/765x MDMo [2.09%], 15/128x OnFE [11.72%], 15/401x MiGl [3.74%], 14/82x MiSh [17.07%], 14/279x MDEE [5.02%], 13/987x CoHu [1.32%], 13/1149x DECj [1.13%], 12/115x TrAl [10.43%], 12/356x MfGl [3.37%]... 10:31:25 !lg * tall !boring s=cv / won -graph 10:32:16 31757/1647532 games for * (tall !boring): https://betelgeuse.shalott.org/graphs/10cf8af2f6459eb3347cc55c4c12d19cb7790301.html 10:32:50 dracoomega has probably a few decades of experience with the tiel starting from NetHack so sure the winrate slightly higher. 10:33:00 So, no worires, Apprentice_Clown 10:34:11 May I ask regarding the art? How often new monsters/items/lore get into the title from contributions? 10:34:30 when someone contributes something 10:34:51 Is it so rare? I know some artists who may adore it. And, is it hard to get into the title in such case? 10:35:26 from what I've seen most contributions get accepted as long as they have compatible licenses 10:35:51 How many in a year or so, if I may ask? Just for a broad realization. 10:36:07 you'll probably get mroe information from actual contributors, most of whom hang out in #dcss on Discord 10:36:25 And, do you recall any contribution you personally liked? 10:36:36 Discord? The Roguelikes one? 10:36:53 Guild ID: 205277826788622337 10:38:13 yes, that one 10:38:13 I don' 10:38:22 I don't generally play tiles so I can't say much about it 10:38:48 and it looks like additions don't use consistent terminology so I can't reliably pull up commits that add new art 10:39:18 <03implojin> new tiles art is frequently contributed and used, but submissions are going to be evaluated against our existing art, which is often fairly high quality within the resolution limitations 10:39:40 <03implojin> we ask (but do not require) tiles contributions to be licensed under CC0 10:40:12 tbh I generally love new tiles but the new leather armour tiles are.... well the normal ones are okay but the artefact one is supr top heavy for some reason 13:49:34 <05sastreii> redid the marks with a strict shared palette. some of them lost personality, others gained it. swapped sprites of legion and tyrant. 13:49:34 <05sastreii> https://cdn.discordapp.com/attachments/747522859361894521/1298025408993955850/afsaf.png?ex=67180fdd&is=6716be5d&hm=0a1ae7c07151496453a663ed63f2acbdfca4ff5b5f1d662f0f0f40b25dfce7c2& 13:49:34 <05sastreii> https://cdn.discordapp.com/attachments/747522859361894521/1298025409195020430/infernal_marks.zip?ex=67180fdd&is=6716be5d&hm=28f43c30ff36a9b94e99a2dd29cdf4362727825ecea6b7823ed5fc20d85b21aa& 13:56:21 <03implojin> very infernal, much demonic 16:10:33 <04dracoomega> Hmmm... apparently monster attack flavors never trigger if the defender died to the initial hit. Which is often appropriate and also often doesn't matter, but does result in strange behavior in a handful of cases: -Broodmothers/Obsidian Bats can't summon things off hits that kill their target -AF_VAMPIRIC monsters can't drain HP off killing blows (notably: those with vampiric-branded weapons can do this) -Shadowghasts don't lose 16:10:33 their invisibility after attacking, if the shadowstab kills its target -AF_SWARM, AF_BLOODZERK, AF_BLINK similarly don't work properly 16:12:02 <04dracoomega> (I noticed this as I go to implement another forgecraft ally with an AF_ that leaves clouds around, which you can really notice not happening if it kills things, since the clouds are kind of an AoE thing also. And rather than putting that elsewhere, I kind of feel like this behavior should just be changed) 16:13:25 <04dracoomega> Use a similar split that weapon brands use, of ones that should still trigger against dead targets, and ensure that all of those play nice if the target is in fact dead 16:27:05 Thank you geekosaur! ^^ 17:51:38 03Implojin02 07* 0.33-a0-266-ga0fd5bc31a: docs: deprecate some mantis links 10(5 minutes ago, 2 files, 24+ 28-) 13https://github.com/crawl/crawl/commit/a0fd5bc31aaf 22:35:24 Unstable branch on crawl.develz.org updated to: 0.33-a0-266-ga0fd5bc31a (34) 22:57:51 Windows builds of master branch on crawl.develz.org updated to: 0.33-a0-266-ga0fd5bc31a 23:13:05 Unstable branch on cbro.berotato.org updated to: 0.33-a0-266-ga0fd5bc31a (34) 23:55:05 Monster database of master branch on crawl.develz.org updated to: 0.33-a0-266-ga0fd5bc31a