03:33:50 Experimental (bcrawl) branch on underhound.eu updated to: 0.23-a0-5261-gd9800d219b 04:20:38 New branch created: pull/5373 (1 commit) 13https://github.com/crawl/crawl/pull/5373 04:20:38 03Brian Power02 07https://github.com/crawl/crawl/pull/5373 * 0.35-a0-845-g1711930599: i18n: Send HUD species+god line from executable to webserver 10(15 minutes ago, 3 files, 22+ 4-) 13https://github.com/crawl/crawl/commit/17119305994c 05:37:38 New branch created: pull/5374 (322 commits) 13https://github.com/crawl/crawl/pull/5374 05:39:53 03Amelia02 07https://github.com/crawl/crawl/pull/5374 * 0.35-a0-547-g5d94e9f4b9: Jessica Modified 10(8 weeks ago, 3 files, 17+ 3-) 13https://github.com/crawl/crawl/commit/5d94e9f4b9fc 05:39:53 03Amelia02 07https://github.com/crawl/crawl/pull/5374 * 0.35-a0-548-gab0625cd59: Update mon-gear.cc 10(8 weeks ago, 1 file, 1+ 1-) 13https://github.com/crawl/crawl/commit/ab0625cd59c5 05:39:53 03Amelia02 07https://github.com/crawl/crawl/pull/5374 * 0.35-a0-549-g39714843b5: Update mon-gear.cc 10(8 weeks ago, 1 file, 1+ 1-) 13https://github.com/crawl/crawl/commit/39714843b539 05:39:53 03RoGGa02 {GitHub} 07https://github.com/crawl/crawl/pull/5374 * 0.35-a0-559-gfcb7cde2c6: Merge pull request #4 from AmeliaGryffon/Jessica-Modified 10(7 weeks ago, 0 files, 0+ 0-) 13https://github.com/crawl/crawl/commit/fcb7cde2c6ec 05:39:53 03RoGGa02 {GitHub} 07https://github.com/crawl/crawl/pull/5374 * 0.35-a0-564-gdbd24358d6: Merge branch 'crawl:master' into master 10(7 weeks ago, 0 files, 0+ 0-) 13https://github.com/crawl/crawl/commit/dbd24358d655 05:39:53 03RoGGa02 {GitHub} 07https://github.com/crawl/crawl/pull/5374 * 0.35-a0-577-g8dd1c3c4c5: Merge branch 'crawl:master' into master 10(7 weeks ago, 0 files, 0+ 0-) 13https://github.com/crawl/crawl/commit/8dd1c3c4c5d7 05:39:53 03Lucien-main02 {GitHub} 07https://github.com/crawl/crawl/pull/5374 * 0.35-a0-585-gd54d7d30a2: Merge branch 'crawl:master' into master 10(7 weeks ago, 0 files, 0+ 0-) 13https://github.com/crawl/crawl/commit/d54d7d30a2ea 05:39:53 03Lucien02 07https://github.com/crawl/crawl/pull/5374 * 0.35-a0-586-gf9bc037fe1: [Silent Spectre] [BcadrenCrawl Port] Initial YAML setup. 10(7 weeks ago, 1 file, 71+ 0-) 13https://github.com/crawl/crawl/commit/f9bc037fe1f5 05:39:53 03Lucien02 07https://github.com/crawl/crawl/pull/5374 * 0.35-a0-587-ge7c45348c5: [Silent Spectre] [BcadrenCrawl Port] Undead State: Ghost added. This also allows Poltergeists to transform. 10(7 weeks ago, 8 files, 23+ 5-) 13https://github.com/crawl/crawl/commit/e7c45348c5b9 05:39:53 03RoGGa02 {GitHub} 07https://github.com/crawl/crawl/pull/5374 * 0.35-a0-586-ga3022414e8: Update .gitignore with additional file patterns 10(7 weeks ago, 1 file, 56+ 0-) 13https://github.com/crawl/crawl/commit/a3022414e800 05:39:53 ... and 312 more commits 05:41:35 -!- The topic of #crawl-dev is: Crawl Development | https://github.com/crawl/crawl | Logs: http://s-z.org/crawl-dev/, temporarily http://crawl.akrasiac.org/logs/cheibriados/ | People with +v have commit access, devs on bridged discord as well | General Crawl-related chat to #crawl | Long stuff to a pastebin service, please 05:55:00 03RoGGa02 {GitHub} 07https://github.com/crawl/crawl/pull/5374 * 0.35-a0-1167-g257696973e: Merge branch 'master' into master 10(53 seconds ago, 0 files, 0+ 0-) 13https://github.com/crawl/crawl/commit/257696973ea5 06:07:50 03CrawlOdds02 07* 0.35-a0-845-g3729dcff44: Cull redundant stairs better 10(6 hours ago, 1 file, 3+ 8-) 13https://github.com/crawl/crawl/commit/3729dcff44ee 06:07:50 03CrawlOdds02 07* 0.35-a0-846-g17d105d106: Do not cull random vault stairs 10(6 hours ago, 1 file, 12+ 11-) 13https://github.com/crawl/crawl/commit/17d105d1061b 06:07:50 03CrawlOdds02 07* 0.35-a0-847-gf146058b2a: Fix an overflow vault that never placed (rypofalem) 10(6 hours ago, 1 file, 1+ 2-) 13https://github.com/crawl/crawl/commit/f146058b2a49 06:35:38 03CrawlOdds02 07* 0.35-a0-848-gd8b905dbbe: Fix a ziggurat entry vault (rypofalem) 10(21 minutes ago, 1 file, 1+ 1-) 13https://github.com/crawl/crawl/commit/d8b905dbbe11 06:53:00 <11O​dds> We get lots of vetoes in spider in general, I think with webs disconnecting levels 06:53:42 <11O​dds> Looks like grunt_spider_rune_circles is quite vetoed, but not to a super abnormal level and your 0/19 roll was quite unlucky 06:56:06 <09g​ammafunk> webs count as non-passable terrain for connectivity? that seems wrong... 06:57:34 <11O​dds> Hmmmm it does rather. I should check the history of that one... 07:00:55 <11O​dds> Ah. This looks like an unintended consequence of 8d147f2, which refactored traps. Before that all traps except dispersal and permament teleport were passable for connectivity 07:14:44 <11O​dds> I wonder what the right rule is there. For example I'm not sure we should treat Zot traps as passable for connectivity purposes, as the old rule did. 07:48:45 <08n​icolae> which traps disappear when they go off? 07:49:07 <08n​icolae> perma teleport shouldn't be connectivible (that's a word) since it never goes away, i think same for zot traps 07:49:18 <08n​icolae> but like a one-time teleport could be connectivible 07:49:32 <08n​icolae> because you can get rid of it forever if you're willing to Takes Your Chances 07:51:59 <11O​dds> Not sure I see that that's the right heuristic? Rather than e.g. "it's potentially extremely deadly to walk through" 07:53:04 <11O​dds> I'd rather walk through a net trap (they don't go away, right?) than a temporary tele trap 07:57:34 <08n​icolae> hmmmm, fair 07:58:32 <08n​icolae> the vault design lobe of my brain is thinking "well if you were scared about potentially extremely deadly circumstances ya shouldna come in the damn dungeon" but i acknowledge that this is not the best mentality for game design as a whole 08:01:43 the vaultal lobe 08:04:13 <11O​dds> I suppose a good reason for permanence being the test is that permanent traps continue to annoyingly block autotravel even after you walk through them 08:08:48 <08n​icolae> scientists say they've never seen anything like it 08:18:15 New branch created: pull/5375 (3 commits) 13https://github.com/crawl/crawl/pull/5375 08:18:15 Branch pull/5375 updated to be equal with unrand-slots-melding: 13https://github.com/crawl/crawl/pull/5375 08:18:15 New branch created: pull/5376 (1 commit) 13https://github.com/crawl/crawl/pull/5376 08:18:15 Branch pull/5376 updated to be equal with pain-bond-fix: 13https://github.com/crawl/crawl/pull/5376 09:43:38 <08o​____0> Yeah like in the chart where I did 1000 iterations of spider4, it played 14 times instead of the expected ~91 (based on vault weights). There seems to be a baseline ~50% failure on spider4 levels but grunt_spider_rune_island and johnstein_spider_rune_arachne_lair failed much more often 09:54:06 <08o​____0> There being a level rule that webtraps never completely block certain (all?) paths is interesting. I am not sure if it needs to be the case for spider. 09:54:36 <08o​____0> Web traps are permanent and do not block autotravel! 09:56:06 <11O​dds> Yeah I think webs should not block connectivity. I'm personally inclined to think that all other traps should, but don't feel strongly about the temporary ones (teleport, alarm) 09:57:02 <08o​____0> (I am realizing maybe I took that too literally and you didn't mean having some isPermanent() style test in the code to determine passability) 😅 10:19:22 <08o​____0> Oh and it's rune_island not rune_circles, just so we are talking about the right vault 😅 In the 1000 iterations of spider4 it only placed in 1.4% of them 10:22:56 <11O​dds> Oh oops I started testing the wrong vault halfway through 10:23:46 <11O​dds> Yeah I'm seeing it vetoed around 90% of the time which is very high indeed 10:24:53 <08o​____0> I am glad it's webs that are the issue so it was probably mostly spider vaults effected 😌 11:45:04 <11O​dds> I think grunt_spider_rune_island is basically a case of "massive vaults with solid walls cause disconnections", which is true in general but even truer in spider because small pockets of webs don't get filled in. So it will somewhat but not completely go away if we make webs connectivity-passable. 12:06:22 <09g​ammafunk> One of my pipe dreams for DCSS is rethinking its approach to layout so that it has some interesting scheme for placing vaults in layouts that's more purposeful in terms of reasoning about the relative vault content (e.g. difficulty, size, notions of 'type' or theme). The current scheme of choosing a layout, generating it, and then mostly randomly choosing vaults and randomly plopping them down so they fit leaves a lot to be desired 12:07:24 <09g​ammafunk> But these types of ideas probably take a lot of work at every aspect of level gen, from deciding the type(s) of algorithms to choose/place vaults to how vaults are specified and tagged to how layouts are constructed in order to facilitate these choices 12:08:39 <09g​ammafunk> I'm not sure if crawl's current vault placement algorithm can be tweaked specifically to reduce certain problematic cases for primary vaults (i.e. rune vaults) causing disconnects due to layout and other vaults 15:39:58 <04d​racoomega> @Odds Just reading through the form melding PR now. It's some pretty involved, complicated work you've done, so this isn't comprehensive, but I do have a few questions/comments. What is the reasoning for using _slot_is_banned for coglin rings, but not for other always-banned things (like felid armour)? Is this just because jewelry is the only one that can come from unrands as opposed to forms? The comment about whatever hack 15:39:59 you're doing regarding serpent form and Skull of Zonguldrok (presumably) is a little vague to me. (Like, I assume it has to do with you losing the hat from the skull, but gaining another hat from being a serpent, but I think the wording may be slightly wrong here? (Unless I'm just really reading it wrong.) _dag_ordered_slots() seems like a pretty heavyweight calculation to have to do potentially multiple times when previewing a form, considering (if I 15:39:59 understand right) the properly ordered list doesn't depend on player state at all, but rather what unrands exist in Crawl as a whole, and thus cannot actually change at runtime. Seems like it could be constructed just once and simply referred to afterward? 15:43:38 Unstable branch on underhound.eu updated to: 0.35-a0-848-gd8b905dbbe (34) 16:36:12 <08w​ormsofcan> now i'm imagining an unrand that gives you a bonus slot but it's random each game 17:33:50 <09h​ellmonk> does anyone have the notes for the lower level manifold assault variant 17:53:09 <04d​racoomega> It wasn't really a lower level manifold assault variant, per se. But when I was working on that plan to adjust spell difficulty, skill point gain, and skill cost curves a bunch (which I have more mixed thoughts on now than I did when I first started working on that), level 7 spells in general would have become meaningfully harder to cast. So the idea was to move Manifold Assault itself down to level 6 (which would have been 17:53:10 slightly easier than level 7 is now, but harder than level 6 is now) and nerf it slightly, while adding a new 'super manifold assault' (kinda) at level 8 (which would be harder to access than current level 8 is). The main thoughts on changes were to make it no longer able to target things that were adjacent to you (which I still kind of like, to be honest, as it also gives potentially interesting counterplay against Incarcerators) and for the cost to 17:53:10 go up with each sequential cast (so that it was a bit less one-note in how a strong character tended to use it - you'd be more encouraged to do small bursts with doing other things inbetween, rather than 'empty all your MP to blow up everything, rinse, repeat' 17:53:52 <09h​ellmonk> oh, ok 20:23:22 <02M​onkooky> Hm, maybe some degree of charge by doing melee attacks? 20:23:55 <02M​onkooky> That'd need to 22:30:47 Unstable branch on crawl.develz.org updated to: 0.35-a0-848-gd8b905dbbe (34) 22:41:44 Windows builds of master branch on crawl.develz.org updated to: 0.35-a0-848-gd8b905dbbe 23:12:34 Unstable branch on cbro.berotato.org updated to: 0.35-a0-848-gd8b905dbbe (34) 23:25:23 <11O​dds> Thanks for taking a look, and understood that this isn't comprehensive * Yeah, _slot_is_banned is for cases where a species bans a slot even when that is granted by an unrand. Coglins' jewellery are the only species/slot pair with that property at the moment; most unrand slot adders did that before checking mutations (including e.g. the fisticloak added a slot before felids' MUT_NO_ARMOUR kicked in, though of course felids can't actually 23:25:24 wear it). It's not so clear to me whether felids would be banned from getting the slot from a hypothetical armour-granting piece of jewellery. * The serpent form hack is actually pre-existing (it already doesn't count the skull in the counted_melded_unrands case), but yeah I can clarify the comment a bit here. * Agreed on _dag_ordered_slots. Actually maybe I'll just make it an explicit list with a code comment in the appropriate pointing a developer to 23:25:24 update it, it's probably overly fancy for constructing something static that changes on a timescale of years 23:45:31 Monster database of master branch on crawl.develz.org updated to: 0.35-a0-848-gd8b905dbbe