00:08:48 aidanh: specifically it is due to mulibc not supporting backtrace.cc. There is some equivalent functionality it provides but it's not a drop-in replacement / I couldn't get it working. You may have more luck... 00:09:11 OICU812: seraph is not a unique pan lord 00:09:29 OICU812: you can see multiple per game, although one is guaranteed to spawn on the holy pan level 00:11:08 I think the only way to see more than one is in fact through the orb run 00:11:21 outside of the orb run, holy pan is the only way to see one 00:17:08 hm i thought they could spawn in abyss but I guess not 00:18:01 never the case apparently 00:19:32 It's a special case too so unlike most the stuff from the orb run it's not also rarely in Pan. 00:19:44 @??seraph 00:19:44 seraph (13A) | Spd: 15 | HD: 25 | HP: 178-240 | AC/EV: 20/19 | Dam: 50, 20 | 08holy, 10weapons, 10items, 10doors, fighter, see invisible, fly | Res: 06magic(160), 04fire+++, 10elec, 03poison, 04rot, 13neg+++, 08holy+++ | XP: 11197 | Sp: warning cry [11!AM, 08breath], sum.holies [06!sil], injury bond [06!sil], cleansing flame [06!sil] | Sz: Medium | Int: human. 00:19:59 yeah, only daevas and angels can iirc 00:24:19 Always thought they were kinda weak/boring. (Not enough stronger to set them apart from the rest of Holy Pan which is *word* homogeneous enough.) 00:24:25 @??daeva 00:24:26 daeva (08A) | Spd: 10 | HD: 14 | HP: 101-144 | AC/EV: 10/11 | Dam: 25, 10 | 08holy, 10weapons, 10items, 10doors, fighter, see invisible, fly | Res: 06magic(140), 03poison, 04rot, 13neg+++, 08holy+++ | XP: 1871 | Sp: smiting (7-17) [06!sil] | Sz: Medium | Int: human. 00:27:06 When I rework pan for my fork; Holy Pan will be ~60% Holy Pan spawns/40% normal Pan spawns with a few Seraphim and Lords...it's pretty high on my docket atm...I'm probably the last fork left that hasn't restructured the dungeon at all. 00:39:32 That said; they are guaranteed to be wearing an FDA or PDA that's GOOD_ITEM spec; so if placing multiple may have to nerf that to not give too good of guaranteed loot. 01:00:07 ??infestation 01:00:08 infestation[1/2]: L8 Necromancy spell in the Necronomicon. Applies the "infested" enchantment to monsters in a smite-targeted radius 2 AoE, for a duration depending on power. Infested creatures leave a temporary, but long-lived friendly death scarab when they die. Affects anything that gives XP. Enemies can be infested out of LOS. 01:00:19 ??infestation[2] 01:00:19 infestation[2/2]: http://i.imgur.com/le8OMPq.jpg 01:00:31 ??infested 01:00:31 infested ~ infestation[1/2]: L8 Necromancy spell in the Necronomicon. Applies the "infested" enchantment to monsters in a smite-targeted radius 2 AoE, for a duration depending on power. Infested creatures leave a temporary, but long-lived friendly death scarab when they die. Affects anything that gives XP. Enemies can be infested out of LOS. 01:11:24 03PleasingFungus02 07* 0.25-a0-518-g7669666: Potion of agility -> stabbing 10(5 days ago, 27 files, 86+ 60-) 13https://github.com/crawl/crawl/commit/766966629d8c 01:11:25 Branch potstab updated to be equal with master: 13https://github.com/crawl/crawl/tree/potstab 01:12:59 I've never used that spell. I've heard it's good but I never focus Necro that high until I want Necromut in late extended and am already using something like Fire Storm or just a Triple XBow for kills and don't really think about using the spellslots for it. Also; why you bringing that up here, Psymania? 01:16:29 The build has errored. (potstab - 7669666 #12704 : Nicholas Feinberg): https://travis-ci.org/crawl/crawl/builds/651343595 01:18:37 PleasingFungus Interesting; I'd have stacked the effects instead of nulling existing Agility to give the new effect (I think Bhauth's bCrawl gave Agility Swiftness and Flight or something like that); there's an error in your commit...the qw.rc change named them Potions of Camouflage where the rest of the commit calls them "Stabbing". 01:29:27 oops. false alarm 01:29:30 trying to slip by a commit after midnight when no one will notice? nice try, I'm reverting immediately! 01:32:25 rip 01:32:32 it's only 10:30 tho 01:33:58 !time 01:33:59 Time: Feb 17, 2020, 06:33:58 AM, UTC. 01:34:16 that's waking hours! 01:34:30 time for breakfast 01:37:24 Bcadren, already here and just didn't want to start another PM tab with Sequell...I can refrain from doing that anymore if it's too distracting/cluttering 01:54:39 Un...fucking...believable 01:54:44 oops, wrong channel 02:10:19 Windows builds of master branch on crawl.develz.org updated to: 0.25-a0-518-g7669666 02:14:18 Unstable branch on crawl.kelbi.org updated to: 0.25-a0-518-g766966629d (34) 02:19:50 !tell aidanh there are some variables that are pseudo-unused too. for example; mons_threat_level (const monster &mon, bool real = false); (mon-util.cc) is never called actually setting real to true...(only difference is it would use the monster's actual xp value (if randomized or vault set) instead of an approximate one based on mon info for the 02:19:50 species if it was true. 02:19:50 Bcadren: OK, I'll let aidanh know. 02:47:56 !cmd duhzwin 02:47:56 Command: !duhzwin => !lg ${*:-.} won gid=$(=allhellrunes $*) gid=$(=helluniques $*) gid!=$(!lm ${*:-.} abyss.enter won s=gid fmt:"${.}" join:"|" stub:"")|$(!lm ${*:-.} br.enter!=temple|depths|hell|geh|coc|tar|dis|zot|icecv|volcano|lab|bailey|sewer|bazaar|ossuary|wizlab|trove won s=gid fmt:"${.}" join:"|" stub:"") s=char title:"DUHZ games won for $(name_fixup $*)" stub:"No games won for $(name_fixu... 02:48:34 =replacecmd duhzwin lab| lab|gauntlet| 02:48:34 Monster database of master branch on crawl.develz.org updated to: 0.24-a0-443-g80245de 02:48:35 Redefined command: !duhzwin => !lg ${*:-.} won gid=$(=allhellrunes $*) gid=$(=helluniques $*) gid!=$(!lm ${*:-.} abyss.enter won s=gid fmt:"${.}" join:"|" stub:"")|$(!lm ${*:-.} br.enter!=temple|depths|hell|geh|coc|tar|dis|zot|icecv|volcano|lab|gauntlet|bailey|sewer|bazaar|ossuary|wizlab|gauntlet|trove won s=gid fmt:"${.}" join:"|" stub:"") s=char title:"DUHZ games won for $(name_fixup $*)" stub:"... 02:54:36 ??duhz 02:54:36 duhz[1/4]: Win a game by going through dungeon, depths, hell and zot, and nowhere else. You must get all 4 hell runes, kill all 4 hell lords, and kill the serpent of hell. You are allowed to enter temple and any portals, but not abyss (even by banishment). 03:19:53 PF, potion of bugginess is in the list of unknown potions 03:21:55 @yer 03:21:59 Stable (0.21) branch on crawl.beRotato.org updated to: 0.21.2-3-g28b50fd 03:24:39 Yermak, PF, probably means it wasn't added to the list of removed items...uhh starts at at line reading: 'const set > removed_items' in item-prop.cc 03:34:02 Fork (bcrawl) on crawl.kelbi.org updated to: 0.23-a0-2962-g32d02e0829 03:40:09 Unstable branch on crawl.beRotato.org updated to: 0.25-a0-518-g7669666 (34) 03:50:07 03gammafunk02 07* 0.25-a0-519-g16a9eb3: Mark potions of agility as removed (Yermak) 10(4 minutes ago, 1 file, 1+ 0-) 13https://github.com/crawl/crawl/commit/16a9eb376cc6 03:50:29 Yermak: fixed that, thanks 03:51:53 !tell PleasingFungus I mention in 16a9eb3 some more (minor) agility-potion-related cleanup that should be done at some point; if you don't get to that, I can do later. 03:51:54 gammafunk: OK, I'll let pleasingfungus know. 04:32:36 Experimental (bcrawl) branch on underhound.eu updated to: 0.23-a0-2962-g32d02e0829 04:54:25 Fork (gooncrawl) on crawl.kelbi.org updated to: 0.22-a0-978-g3a575b4172 04:55:53 o_o; Did mine fail O_O; I saw bCrawl 1.5 hours ago, now gooncrawl...IDK why mine would fail; but it's always a possibility 04:58:35 alexjurkiewicz: by any chance did you try https://github.com/mikroskeem/libexecinfo? 04:58:35 aidanh: You have 1 message. Use !messages to read it. 04:58:47 Unstable at 23:14, bCrawl at 00:34, gooncrawl at 1:54 ...longer delays between than normal; something is up... 05:26:36 aidanh: I don't remember, sorry. I was reading the ulibc ML which said libtraceback (i think...) was avilable on alpine, but wasn't exactly the same as GCC's. Or something like that 05:31:17 no worries, there should be some alternative 05:31:37 what distro did you decide to use instead as a base? 05:45:28 Fork (stoatsoup) on crawl.kelbi.org updated to: 0.22-s0-43-ga5c91f02cb 06:05:48 Unstable branch on crawl.kelbi.org updated to: 0.25-a0-519-g16a9eb376c (34) 06:12:07 This is on my to-do list for my fork; any intention of doing this in mainline? teach monsters to use misc Evocables. 07:24:12 Oof I got disconnected. Did my fork rebuild on Kelbi while I was disconnected? 07:27:08 Looks like no. >_> 07:52:23 aidanh: debian, since it is very close to CPO already 07:53:03 beyond size, I don't see much difference between the base OS though 07:54:07 hm, do newer debians support old kernels? 07:55:33 for reference, my wheezy vm has a 3.2.0 kernel, so your image will need to support that to run on CAO 07:56:25 i remember having difficulties even getting a shell on an ubuntu 18.04 chroot 07:57:40 *annoyed* please build already. :( There's a major common crash bug in yesterday's that needs to be gone. If it's not building I need to know why too...sorry. 07:58:34 i wasnt thinking about old kernel support so didnt try sorry 07:59:34 easiest approach might be to buid the modern toolchain from scratch in an older docker image 08:00:51 as in, build in a container? 08:01:22 we'd probably want to do something like that eventually, perhaps via github actions 08:05:03 i'm not sure how much of an issue kernel support will be in practice 08:05:19 another issue is just getting old servers to upgrade 08:07:22 that's why I'm considering chroots as opposed to full-blown docker, and alpine in particular, because it supports old kernels and has gcc 9.2 08:12:16 I made the mistake of copying some monster-main.cc code into the core game to display more monster stats on live...it worked fine on local; but apparently there are some webtiles-specific calls to monster info that cause all kinds of problems if you do that. It's fixed on local...if it'd ever server rebuild... *takes a guess as to why server 08:12:16 rebuild might have failed and pushes a fix for a bug that may or may not exist* >_< 08:13:44 Bcadren: why don't you rebuild for webtiles locally? 08:14:36 you can't just use the online servers as your compiler 08:14:36 for one thing, the latency would be unreal 08:15:30 it's been awhile since I done this; but I can try; last time I did this it didn't run...but it compiled it (using Debian via windows kernel for linux) 08:16:27 it's quite easy if you already can compile for local tiles 08:16:35 just change TILES=y to WEBTILES=y 08:18:49 I don't think that works with the standard windows compiler; but I can kick it off and try...no, it doesn't (in MSYS that just errors out before making a build target)...yea a linux default library include failure I think. 08:19:26 what is the error message? 08:20:32 tileweb.h:12:10: fatal error: sys/un.h: No such file or directory 08:21:03 Stable (0.19) branch on underhound.eu updated to: 0.19.5-3-g47ae8add2e 08:30:46 Bcadren: delete that line from both tileweb.cc and tileweb.h, and add this delcaration to tileweb.h 08:30:48 struct sockaddr_un { unsigned short int sun_family; char sun_path[108]; }; 08:31:23 or you could try using wsl to build 08:32:20 Hmph. Yea I know there was something via subsystem for linux that finished a webtiles build once; though the finished build wouldn't actually run...most my builds local (what I test with) is MSYS building for windows console; it's rare that these webtiles-specific bugs come up; but god is it bad when they do. 08:34:48 especially since I don't see the compiler error, when server rebuild fails >_> 08:37:46 you could also just set up a vm, share your source code folder with the vm, and build it for webtiles 08:39:43 but in my my experience wsl is far superior in almost every way; you get a real linux toolchain with a real distribution's packages, with near native speeds 08:43:21 hmm well with that tweak MSYS is building. Like I said I have debian in WSL (I'd set it up to try to do an official distro for BcadrenCrawl; like the mainline tourney distro...if it gets committed with the removal will that screw up the normal webservers? 08:44:42 and MSYS failed in the middle again...on a second I've never changed... 08:44:43 coord.h:64:47: error: function 'coord_def clamp_in_bounds(const coord_def&)' is initialized like a variable 08:45:18 I'm way over tired at this point. >_< I'm just going to sleep need to be up in a few hours anyways...*sigh* 08:47:22 Try kicking it off on kelbi again at some point; maybe it never attempted and maybe my bug fix for uncertain bug fixed the possible bug (at the very least if it doesn't compile; you can actually see and !tell me the error)...I really don't like leaving up a version where a SENSED_MONSTER spawning (which happens constantly as a Fo for example) 08:47:23 causes a crash. 08:51:00 please do _not_ commit the sys/un.h removal 08:52:47 !tell Bcadren please do _not_ commit the sys/un.h removal; I'd suggest you ignore changes to those files with `git update-index --assume-unchanged tileweb.*` so you don't do it accidentally 08:52:48 aidanh: OK, I'll let bcadren know. 09:08:19 03sdynet02 07https://github.com/crawl/crawl/pull/1279 * 0.25-a0-495-g3982237: bug fix - Disarm Enemy Weapons 10(4 minutes ago, 1 file, 6+ 0-) 13https://github.com/crawl/crawl/commit/3982237b0ee6 09:08:19 03sdynet02 07https://github.com/crawl/crawl/pull/1279 * 0.25-a0-496-g542edfb: Nerf recovery in the holy cloud. 10(62 seconds ago, 1 file, 1+ 1-) 13https://github.com/crawl/crawl/commit/542edfb3bc26 09:13:42 03sdynet02 07https://github.com/crawl/crawl/pull/1279 * 0.25-a0-497-gb4fbcba: Remove GDR bonus from Metamorphosis. 10(3 minutes ago, 1 file, 2+ 2-) 13https://github.com/crawl/crawl/commit/b4fbcbabf2ff 09:22:45 Stable (0.22) branch on underhound.eu updated to: 0.22.1-61-g4f546fdb5a 09:36:37 ok, I can build and run a c++17 crawl on wheezy, via an alpine chroot, although admittedly without backtrace support for now 09:37:43 if anyone who's worked with CAO / other server maintenance can think of any flaws in this grand plan, please torpedo it now before I get my hopes up :) 10:37:17 !tell Bcadren builds were slow this morning because gdb was hung again overnight 10:37:18 floraline: OK, I'll let bcadren know. 10:37:39 does anyone know what causes gdb to hang and consume endless cpu cycles? 10:57:58 floraline not sure but it definitely happens from time to time and it may depend on the nature of the crash. Is it happening a lot just for that fork? 11:09:57 !tell gammafunk leaving it in the xom effects was intentional! (but I already have another commit that cleans that out much more neatly; will push it shortly.) 11:09:57 PleasingFungus: OK, I'll let gammafunk know. 11:24:04 Stable (0.23) branch on underhound.eu updated to: 0.23.1-93-ge536e68a2c 11:53:44 drj (L13 SpAM) ASSERT(attacker) in 'attack.cc' at line 353 failed. (Gauntlet) 12:08:52 03PleasingFungus02 07* 0.25-a0-520-g08029b5: Cleanup !agi 10(33 minutes ago, 9 files, 16+ 34-) 13https://github.com/crawl/crawl/commit/08029b5080ff 12:14:33 whoops 12:15:33 03PleasingFungus02 07* 0.25-a0-521-gb5d5f98: Cleanup !agi cleanup 10(21 seconds ago, 1 file, 0+ 1-) 13https://github.com/crawl/crawl/commit/b5d5f9853597 12:28:15 The build was broken. (master - 08029b5 #12710 : Nicholas Feinberg): https://travis-ci.org/crawl/crawl/builds/651607199 12:32:51 !seen lasty 12:32:52 I last saw Lasty at Mon Feb 25 13:24:51 2019 UTC (51w 4h 8m ago) quitting, saying 'Quit: Leaving.'. 12:51:26 Unstable branch on crawl.kelbi.org updated to: 0.25-a0-521-gb5d5f98535 (34) 13:02:06 New branch created: potsludge (2 commits) 13https://github.com/crawl/crawl/tree/potsludge 13:02:06 03PleasingFungus02 07[potsludge] * 0.25-a0-521-ge5ab6e9: Cleanup !agi cleanup 10(47 minutes ago, 1 file, 0+ 1-) 13https://github.com/crawl/crawl/commit/e5ab6e94d82b 13:02:06 03PleasingFungus02 07[potsludge] * 0.25-a0-522-g600ada2: Mix !might and !brill into a gross slurry 10(85 minutes ago, 38 files, 128+ 111-) 13https://github.com/crawl/crawl/commit/600ada21cc00 13:02:25 !tell ebering I made https://github.com/crawl/crawl/commit/600ada21cc00beb0772c2e7882ef0f0174e0da0d for you but midway through I realized that I didn't have any strong justification for the change. Did you? 13:02:27 PleasingFungus: OK, I'll let ebering know. 13:07:01 PleasingFungus: cool! I think removing the str/int bonuses makes sense. 13:07:46 By way of justification: !might and !bril are not equally relevant for most character archetypes 13:07:46 a character will prefer one over the other, but also might want the alternative in edge cases 13:08:17 by making them a single resource (and cutting down on its abundance) there's 1 fewer inv slot and more interesting decision making of when to use it 13:08:39 though I hadn't thought super carefully 13:08:42 aren't there a lot of characters who already won't bother carrying one or the other, though? 13:08:44 by that reasoning 13:09:01 i don't see how it invites more interesting decision making or saves inventory space 13:09:08 "alternative in edge cases" 13:09:18 it's not your #1 or your #20 go-to item 13:09:25 but it might be somewhere in the 30s and 40s 13:10:23 can you talk more about how it invites more interesting decision making? 13:11:20 Well, looking at the discord and seeing the broader issues with wizardry 13:11:40 it goes from "I have a potion of punch harder" and a potion of "cast a spell I couldn't other wise" 13:11:48 when they're separate, theyre separate resources 13:11:57 sure 13:12:09 when you combine them you have to face "is it worth punching harder now to not be able to cast this extra spell" and vise-versa 13:12:25 can't you just do both 13:12:38 like, you cast the extra spell and then punch [assuming both of those are ever relevant] 13:12:43 yes this is what occured to me when I wrote "<+ebering> though I hadn't thought super carefully" 13:13:09 c'est la vie 13:13:17 -!- bh is now known as Guest96333 13:13:46 Anyhow back to my day job 13:14:07 hf 13:34:19 Quick question: would there be any interest in a patch extending the behaviour of "regen (only affecting regen rate after your hp tops off) to regen equipment in general? Currently you can find speedrunners carrying around TLA just to heal with. While i've never seen a non-speedrunner do that, it seems like a fiddly tedious but optimal tactic in the light of piety decay. 13:35:51 there could be some decisions about when to bother in the light of the time spent on the actual equipment changing, how far to retreat before swapping, etc., but in practice the choice seems pretty dull (do it if you are going to rest a lot; otherwise, don't) 13:36:37 agreed it's very dull if it's actually happening; the question is if that dullness is common enough to be worth adding the extra complication of the regen-sync-logic to stop it 13:36:39 (going to rest a lot and are able to retreat to some relatively safe space, i suppose) 13:37:18 we already have the hp-top-up logic for "regen, as you said, so it's probably not very complex in either code or player understanding to add to other regen equipment 13:37:32 i've seen uv4 do it; i doubt it's common though 13:38:26 i think i would try to lua-automate the actual swapping equipment part for myself if it stayed in the game, which i think would make it a bit less painful 13:41:31 hm 13:41:35 seems like not something we'd want to encourage 13:41:58 i don't think i'd be opposed to a patch 13:42:10 don't feel strongly, since it is a pretty minor change either way 13:42:24 but it seems like a reasonable fix 13:42:25 thanks for bringing it up! 13:43:01 yeah at a glance it seems like a simple enough change. great, i'll write something up and see what happens 14:09:54 The build passed. (potsludge - 600ada2 #12712 : Nicholas Feinberg): https://travis-ci.org/crawl/crawl/builds/651629420 14:15:02 -!- Psymania_ is now known as Psymania 14:47:05 aidanh: yeah sorry mis-spoke. I mean if you can build a modern toolchain from source on an older server, that might the easiest way to get said modern toolchain 14:47:27 PleasingFungus: i'm sorry about the reddit thread title, I thought people wouldn't take it so seriously 14:48:16 ? 14:48:26 i haven't been on the dcss reddit in years. should i? 14:48:34 no 14:48:41 problem solved, then 14:48:55 floraline well either mine never rebuilt or the build failed and there's a bad webtiles-unique bug right now. 14:49:43 ^^ 14:58:56 PleasingFungus: "Is this a real commit? It reads like an early April Fools' joke. 14:59:12 i don't think there's any difference between real and not real things 14:59:21 they're exactly the same. 14:59:29 am i real? seems unlikely, frankly 14:59:56 I was the same as something once 15:00:21 also idk if you saw my conversation with ebering earlier, but I'm currently leaning towards not merging that change and instead making some other tweaks (removing the str from !might, making !brill do something that's not spell failure) 15:02:21 the problem these days is that you write too few changes and too many good ones. it's giving people bad expectations that they can look at every commit and expect it to have a good reason to exist 15:02:24 just preposterous i think 15:04:10 https://github.com/crawl/crawl/commit/e219424e2279bb451731ea910ed07e43e114a213 if there were more commits like this one around, crawl would be a happier place 15:07:36 They're good commits, Brent 15:09:46 %git 37d83eae1e5c 15:09:46 07gammafunk02 * 0.22-a0-446-g37d83ea: Make Lee's Rapid Deconstruction never break walls 10(1 year, 9 months ago, 7 files, 12+ 54-) 13https://github.com/crawl/crawl/commit/37d83eae1e5c 15:09:55 The gold standard of good commits you must all aspire to 15:10:47 PleasingFungus: I confess I haven't thought out all the ramifications of a merger, but removing int bonus from brilliance sounds good 15:11:04 When you say making brill do something that's not spell failure, I assume you'd keep its spellpower enhancer? 15:11:47 my intuition is that it should just be an archmagi (maybe for a shorter duration or a limited number of spells, if that's too strong) 15:12:12 ??potion of brilliance 15:12:12 brilliance[1/2]: Temporarily gives the maximum {wizardry} bonus, a power boost equivalent to three spell school levels, and Int+5. 15:12:19 the current power boost is quite small 15:12:21 ah, so it's a custom bonus, interesting 15:12:27 I had always assumed it was archmagi! 15:12:31 it would make sense if it was 15:12:39 one day I'll learn how to play dcss 15:12:54 more optimistic than i am 15:14:32 i think a small spell failure chance reduction would be fine too, if we wanted that. a large reduction encourages you to memorize spells that can only be cast when you have !brill on, which I think is maybe not something we want... it seems like an effect that works better with equipment, where it's more part of an overall playstyle than a brief 15:14:32 boost 15:14:47 I don't think archmagi for a single use is too strong; thankfully even if it was we could make the potion a bit less common 15:15:11 archmagi + hunger reduction seems like a nice feelsgood combo 15:15:17 right, lowering miscast chance is 'fun' in that "now I can do something I couldn't before" sense, but it has some unfortunate implications 15:15:37 i'm torn on how i feel about that... someone was arguing it was fun to have a limited number of shots of, say, dragon's call 15:15:45 <- someone 15:15:48 ty 15:15:54 well, we do have a god that does that now! 15:16:07 someone in discord mentioned that :) all these newfangled innovations 15:16:15 the most beatifull goddess in the game.... 15:16:17 <- also someone 15:16:25 gammafunk i don't want any furry content in this channel 15:16:26 keep it to discord 15:17:00 FWIW, if it's a good mechanic on a god it seems like it might be a good one on a potion too; but I can think of relatively few spells where having a rationed number of casts is going to make a big difference 15:17:27 without the +5 str, would !might be significantly weaker? 15:17:30 And I grant on new Sif you're not obliged to stuff it into your memorised spells to use it 15:17:35 the potion would need to work like the god, if that where to be true 15:17:42 i have zero intuition for how much str actually matters 15:17:43 specifically the god lets you cast spells from your library 15:17:50 not from those memorized 15:17:56 ooh, that's a great use of the new library thing 15:17:59 who needs book amnesia! 15:18:15 isn't one of the docs either on running a webserver or building webserver style to test? I can't find it. 15:18:40 i don't know how you'd make it work (i think you can't) but it'd be insanely thematic if sif piety was based on library size 15:19:46 yeah, sadly it's basically just exploration piety with a different theme; I hate exploration piety, since it's a worse version of piety from kills! 15:20:08 Bcadren: I dunno where it _is_, but make with "WEBTILES=y" then "python webserver/server.py" then connect to localhost:8080 15:20:16 I am not against interesting piety systems that really seem to introduce cool decisions; Ru might be the best example 15:20:29 wu jian is also different 15:20:36 you mean usk? 15:20:38 uh 15:20:41 yes 15:20:49 and yeah, another Lasty innovation 15:20:53 sorry, i've never played with either, so they meld together in my mind... 15:21:00 why does he make all the cool alternate piety schemes! 15:21:02 haha 15:21:10 ask him to fix ash imo 15:50:14 Making a fresh clone see if WSL can compile for webtiles; it's failing to read the one MSYS is touched, probably MSYS making it windows style endings... 16:03:54 PleasingFungus: I just got a telepathygram from Lasty. Your Ash Piety Level = Your bondage level. Ash periodically (in the style of Ru) offers to curse something; you can accept or reject the curse but if you reject Ash waits longer; Ash only offers curses if you're not fully bound (no building up savings!). Ash abilities (Scry, Knowledge Transfer) scale in some fashion with bondage level (duration, 16:04:00 effectiveness &c.). Uncursing can be done at any time. 16:04:38 This idea probably needs further workshopping, but its the first one I've had since thinking about removing curses otherwise that seems like it has legs 16:04:43 s/probably// 16:09:47 the radius of scry could scale with bondage level quite easily. almost 1:1 equip slots to los radius 16:10:46 03Alex Jurkiewicz02 07https://github.com/crawl/crawl/pull/1253 * 0.25-a0-428-ge9d5112: webtiles: load_games improvements (aidanh) 10(6 minutes ago, 1 file, 104+ 20-) 13https://github.com/crawl/crawl/commit/e9d51125ffa8 16:11:22 wow, telepathygram 16:11:30 sounds like the sort of thing ash *would* grant... 16:12:16 hm, so ash is the one that chooses what to curse? 16:12:38 that part of the telepathygram was unclear 16:13:10 it would be odd to 'reject' a free curse otherwise... i guess maybe if you didn't want to curse anything at all 16:13:38 i think either way is plausibly interesting 16:14:35 i don't think this needs to be locked to other curse changes, which is nice... in the current regime, you could just give penance for using ?rcurse 16:15:03 (and then if we wanted to remove non-ash curses later, that'd work fine) 16:15:33 well, I'd like to remove the id minigame, which makes non-ash curses need to go 16:15:40 i had heard that rumor 16:16:20 i just like that this isn't order-dependent (can be done before or after other curse changes) 16:16:22 i guess before or at the same time as... 16:32:41 !tell pleasingfungus Thanks, got it built local; still won't actually run, but it was enough to see and fix the compilation errors. 16:32:41 Bcadren: OK, I'll let pleasingfungus know. 16:36:15 floraline or anyone else with build access; it will build now (fixed two compilation errors, one Unix specific which Windows got around; one webtiles specific). :) please kick it off... 16:54:35 aidanh warning: type qualifiers ignored on function return type [-Wignored-qualifiers] 17:11:58 Bcadren: is that from your code? if it's from vanilla, what line? 17:12:17 Bcadren: if you're trying to solve it, googling the error points to a good stack overflow post 17:16:44 alexjurkiewicz they are very common in OLD vanilla; there's a commit I skipped on October 10th of last year that switched from using ints to refer to item slots to passing around item_def pointers...the old version of swapping items with ints has that error everywhere. (I skipped the commit because; BcadrenCrawl has dual wielding which already 17:16:45 completely changed the wield_weapon function and made it about three times as long; so it'd be a very very hard merge. 17:21:26 deciding if I want to fix that compile error or switch to item_def pointers like mainline did. 17:24:20 My feeling is any time you gratuitously diverge from vanilla you're just making a rod for your own back in terms of applying future commits 17:27:17 Fair point; but . . . yea probably switching to item_def pointers would be better; even if the practical difference is small (the initial commit that does it is straight up called "wield weapons directly from the floor" or something like that...which is the only practical difference...but still dual wielding means I have to do that whole thing 17:27:18 manually the existing commit is close to useless for the amount of conflicts it has with the (much longer) wield weapon. funciton. 17:28:05 Bcadren: I'll remind you again that this is -dev and primarily for the discussion of the development of the main dcss branch 17:28:27 It's fine to check in and ask questions about the codebase but you're veering pretty far off course with this 17:56:26 glaas, PleasingFungus, I might be biased, because as a fellow-speedrunner I'm strongly opposed against all regen items gaining the regen amulet property, but look at this: there is already behavioural difference between jewellery and armour (and all the rest regen properties could only be met on armour), do we really need to ignore this fact and force this change? 17:58:13 (Yes, I carry tla on normal characters, too, in case I need to get to timed portal.) 18:00:13 well I think the point of the full-health activation is to disincentivise tedious behaviour, no? I don't believe the difference in behaviour is about making the property distinctive on different equipment 18:00:53 if we wished to do that, probably having the regen amount be higher on some equipment would be best 18:01:35 I think the only reason why it's not on armor is because the turn cost for the swap was thought to be enough of a disincentive to carrying around "regen junk" 18:01:36 no, but we'd make this difference less meaningful with this change. 18:02:00 but that turn cost is probably not sufficient 18:02:06 no-no, it was high enough to prevent sapping in a fight. 18:02:15 swapping 18:02:27 sure, but not enough for other situations, like the portal one you describe 18:02:51 Swapping anything to regen faster is tedious 18:02:53 carrying tla to help rush for portals doesn't feel like the right thing to encourage for body armour 18:03:24 I guess you could argue it's a tad more interesting on the amulet, since we do want non-strategic jewellery swaps to some degree 18:04:15 There are plenty things in crawl that doesn't feel right and should be discouraged. And I personally like this one little strategy aspect. 18:04:35 For example trying to match Jiyva 6 star with you being at Slime:5 to minimize the loot eaten. 18:05:52 venus (L12 MuFE) ERROR in 'tags.cc' at line 4559: Invalid item: glowing pink potion (D:9) 18:05:59 yeah, that's another thing we'd probably like to reform :) 18:06:02 ruh roh 18:07:07 ...That's an easy one though; there are no basic jellies by default on Slime:$; just suppress them spawning randomly there from Jiyva's power and paint some flavour around it...champion gets first dibs on the loot, whatever...done. 18:08:39 venus (L12 MuFE) ERROR in 'tags.cc' at line 4559: Invalid item: glowing pink potion (D:9) 18:08:44 !crashlog 18:08:55 Yermak: It's sometimes hard to distinguish between stuff that feels emergent like how you feel about regen armor or jiyva loot preservation and stuff that feels just like tedious effort for smallish gain to most others; usually there's some kind of compromise solution out there 18:09:10 21702. venus, XL12 MuFE, T:28874 (milestone): https://crawl.kelbi.org/crawl/morgue/venus/crash-venus-20200217-230510.txt 18:09:38 venus (L12 MuFE) ERROR in 'tags.cc' at line 4559: Invalid item: glowing pink potion (D:9) 18:09:38 probably an agility potion that had already generated before a game transfer? 18:09:38 !blame PleasingFungus 18:09:38 I pronounce PleasingFungus... Guilty! 18:09:48 I agree. You swap to tla a couple of times per game. I don't think it's tedious. 18:10:10 venus (L12 MuFE) ERROR in 'tags.cc' at line 4559: Invalid item: glowing pink potion (D:9) 18:10:29 <|amethyst> the speedrunning case, I don't care about much, because speedrunning is bound to involve doing non-obvious things, and that's (I guess) part of what makes speedrunning fun 18:11:14 <|amethyst> for the timed portal case, could make portals not tick down when they are in view, though that does eliminate the "fun" of seeing the portal dissapear when you almost make it 18:11:39 venus (L12 MuFE) ERROR in 'tags.cc' at line 4559: Invalid item: glowing pink potion (D:9) 18:11:55 gammafunk: probably. if i had to guess, a potion that was on the floor instead of in inventory when the save transfer happened? 18:12:00 I suggested once, that portals don't disappear when timer is out until you leave the floor. I think it's beautiful change. 18:12:09 amalloy: yeah, that's what I said (or was trying to say) 18:12:24 although I'm not sure how PF's removal makes that a problem 18:12:32 Yermak: That would discourage the comedic deaths caused by trying to find the portal without fully regenerating HP/MP 18:12:56 But the real reason behind timer is for players not to get back when they're stronger. 18:13:22 would there even be a point to the timer if it worked that way? 18:13:29 just have the portal poof upon level change 18:13:45 that's another option 18:13:48 <|amethyst> gammafunk: that makes shafts even less fun 18:14:16 well, if you get shafted and are running away from the level, you're probably not going to go back for the portal 18:14:27 even under the current system 18:14:30 <|amethyst> and ducking back upstairs to rest before you have had a chance to explore, or to take another stair when the first one sucked 18:15:01 disable shaft traps while there is an unvisited portal :O secret tech 18:15:01 venus (L12 MuFE) ERROR in 'tags.cc' at line 4559: Invalid item: glowing pink potion (D:9) 18:15:01 yeah, but that's a fairly small instances when you'd actually escape a shafting but want to go back for the portal 18:15:21 I think I would always want to, if I wanted to before being shafted 18:15:26 it also is a thing where stronger characters are more likely to be able to do this, which is sort of going against what having a temporary portal is trying to achieve 18:15:37 I once got shafted 2 floors away from portal onto a floor with another one. Completed deep one, and met the third one on a floor in between. 18:16:01 How do double-portal levels work in this proposal? 18:16:34 All works exactly like before. Only you have one more chance once the time is out if you don't leave the floor. 18:16:52 venus (L13 MuFE) ERROR in 'tags.cc' at line 4559: Invalid item: glowing pink potion (D:9) 18:17:10 If you always want to leave a level that you've been shafted to, rest, and go back to portal, then you'd also always want to just go for the portal in the first place :p 18:17:13 (if you don't leave the floor after time is out, you can leave when timer still ticks) 18:17:13 i guess portals close if you use stairs only. Using portals or being shafted could leave the "timed" portal open 18:17:32 (and you're probably also aways making a bad decision) 18:17:46 so if you found a shaft on a level with a portal, you would be able to save the portal for later secret tech 18:17:54 portals could only consider leaving the current permanent level, not leaving for another portal 18:18:20 alexjurkiewicz: Then I should run around the whole level in an effort to get explo trap shafted :-) 18:19:06 yes 18:19:40 that doesn't sound too bad to me tbh. The one ugly wart with this idea is |amethyst's point about sometimes wanting to immediately leave the floor and use different stairs 18:20:23 maybe the portal is announced once you step off the stairs on a level 18:20:23 <|amethyst> it's not necessarily that bad 18:20:23 well you wouldn't actually want to go get shafted to preserve a portal; portals are simply not worth that 18:20:23 <|amethyst> you have a tradeoff to make 18:20:25 it would allow players to take that option as a bad decision, which of course they are welcome to do 18:20:43 so there's no real problem with that from a standpoint of what you should do 18:21:36 <|amethyst> bug request: make taking the timed portal itself delete the portal before determining its destination 18:21:51 <|amethyst> and send the player to minus world 18:22:18 Place: D:-32767 18:22:25 we'd need to add piranha pipes for that??? 18:22:29 Unstable branch on underhound.eu updated to: 0.25-a0-521-gb5d5f98535 (34) 18:22:48 also mushrooms so that large players can duck and trigger a wall glitch 18:22:57 that's probably doable in smb1 while small, I guess 18:23:38 <|amethyst> gammafunk: add Pirah� pipes, which some people argue do not allow recursions 18:23:48 <|amethyst> s/ions/ion/ 18:25:11 this is like when |amethyst told me about sporulate and I had to google it 18:26:25 <|amethyst> I am a veritable font of useless knowledge 18:26:29 <|amethyst> some of it's even true! 18:27:18 !crashlog venus 18:27:20 1. Deedee, XL15 SETm, T:42741 (milestone): http://crawl.akrasiac.org/rawdata/Deedee/crash-Deedee-20110520-064539.txt 18:27:23 er 18:27:26 hm 18:27:26 PleasingFungus: You have 1 message. Use !messages to read it. 18:27:47 https://crawl.kelbi.org/crawl/morgue/venus/crash-venus-20200217-230510.txt 18:27:50 odd 18:27:52 !crashlog 18:27:53 21708. venus, XL13 MuFE, T:29290 (milestone): https://crawl.kelbi.org/crawl/morgue/venus/crash-venus-20200217-231618.txt 18:28:01 you'd think those would give the same thing! 18:28:05 <|amethyst> !kw venus 18:28:06 No keyword 'venus' 18:28:11 !nick venus 18:28:12 Mapping venus => deedee 18:28:13 <|amethyst> or, that's now kw 18:28:14 doh 18:28:49 !lm deedee 18:28:50 447. [2011-06-18 07:12:50] Deedee the Farming Archmage (L27 SpAr of Nemelex Xobeh) escaped from the Abyss! (Abyss) 18:28:58 sounds like a dummy !agi potion somehow generated in some way i didn't anticipate 18:29:03 maybe that nick mapping should go 18:29:12 they could return at ANY MOMENT 18:29:33 i don't suppose we can get a save dump 18:30:01 PleasingFungus not sure how you did the save compat but don't forget that levels may generate before being visited now 18:30:04 sure 18:30:07 &dump * name=venus 18:30:27 ok maybe that's not a good way of doing that 18:30:40 !seen venus 18:30:40 Sorry gammafunk, I haven't seen venus. 18:31:10 ah, maybe a discord mapping 18:31:10 though I guess you probably would've handled floor potions somehow 18:31:10 really need to have cerebot support that directly 18:31:22 unknown/willhelm/willhelm.txt 18:31:38 i re-used the old !agi enum for !stab and added a new POT_AGILITY as a temporary placeholder for xom & ds powered by pain - it wasn't supposed to ever generate 18:31:59 then i removed the new pot_agility entirely, and this happened 18:32:02 i guess i can add a minor tag 18:32:09 ah I see 18:32:22 we're running out of those! 18:32:28 and when we do 18:32:32 oh boy 18:32:33 MAjoR VERSIOn 35! 18:32:38 im scared 18:32:43 the faster we run out the faster major version gets bumped :-) 18:32:44 I was just saying the other day, maybe this security bug thing is sign it's a good time to break save compat 18:33:01 (not actually sure doing so helps us wrt that bug) 18:33:23 PF returns for a while? time to bump the major version 18:33:37 doesn't seem very relevant to the security bug 18:33:41 i respect your enthusiasm tho 18:33:58 well, we do have to force transfer trunk games to fix the bug 18:34:22 so in the no-save-compat world, those games would just be lost instead? 18:34:23 actually, I guess breaking save compat sort of gets in the way of that more than it would help 18:35:04