00:00:03 <gilian> gammafuck:  thanks!  Um... once I have the code ready for someone else to look at it... do I just submit it via git?  RTFD?
00:00:09 <Brannock> Git is fine, yes
00:00:17 <Brannock> GitHub I mean
00:00:42 <minmay> if you can detect it well enough to stick a hatch in it then you should probably delete all monsters/items/etc. in it and fill it with walls
00:00:42 <gammafunk> that's not...my...name....
00:00:42 <Brannock> hahaha
00:00:42 <gammafunk> ;_;
00:00:54 <gilian> ... :( ... that wasn't intentional...
00:01:05 <gammafunk> this is even worse than when twelwe wished me "good luck, gammafuck"
00:01:10 <gammafunk> and then I died
00:01:18 <rumflump> this helper app for git called "hub" is nice. push your changes to your own repo, add crawl's repo as a remote named "upstream", then just use "hub pull-request" and magic happens
00:01:45 <gammafunk> gilian: no worries, and yeah, like Brannock said a github PR is fine
00:02:17 <gammafunk> there are docs in docs/develop/vaults (I think) about vaults making, and a des syntax guide
00:02:47 <gilian> gammafunk:  --- yeah, I read through the vault docs.
00:04:49 <gilian> ... err some of them, at least.
00:05:10 <gammafunk> it's a very cool thing to fix, that issue
00:05:22 <gammafunk> we would love to have more intelligent vault placement in our level generation
00:05:22 <gilian> :)
00:05:32 <gammafunk> but that's another issue entirely
00:06:12 <gammafunk> well, not entirely, but it's another issue; generating fewer vetoed levels and perhaps better vault placement (in terms of placing vault poorly and making weird areas)
00:06:31 <gammafunk> if you get interested in those arcane areas of the codebase
00:06:41 <gammafunk> not a lot of people dive into them so much because it's fairly complicated
00:07:06 <gilian> Well... between Marduk and beginner on level generation... I'm towards beginner; so, I don't expect to be doing any magic.
00:08:53 <gammafunk> after Marduk comes cang level generation
00:10:07 <gilian> From cangband-?
00:10:51 <gammafunk> ??cang
00:10:51 <Sequell> cang
00:11:03 <gammafunk> @??cang
00:11:03 <Gretell> cang (04Ω) | Spd: c | HD: i | HP: 666 | AC/EV: e/π | Dam: 999 | Res: sanity | XP: ∞ | Int: god | Sz: !!!
00:11:08 <gammafunk> I don't even know what the original reference is
00:14:59 <gilian> Thanks for the help.  Have a good one.  :)
01:01:51 <minmay> fr: Marduk
01:31:42 <Eksell> Unstable branch on CRAWL.XTAHUA.COM updated to: 0.20-a0-909-g1a4ba80 (34)
02:10:25 <rumflump> ok I got a windows user who was willing to try every iteration with me
02:12:08 <rumflump> cross compiling to windows with LTO=y just causes scrolls to sometimes crash the game, it should probably not be done, since no LTO always works fine
02:20:02 <gammafunk> I don't use LTO when making the binaries
02:20:18 <Gretell> Unstable branch on crawl.develz.org updated to: 0.20-a0-909-g1a4ba80 (34)
02:20:47 <gammafunk> however I think that's because I wasn't able to get it to properly compile
02:22:53 <rumflump> https://mega.nz/#F!FhkVUbTB!OPheb7FXTPMavoGv0_ocnw the "NEW" builds are with LTO off, the other ones are with LTO on, and all four of those crash (at least for the person who tested these for me)
02:24:52 <gammafunk> well
02:24:57 * geekosaur not sure he'd want to trust LTO on a cross-compile
02:25:10 <gammafunk> if lto does cause windows crashes it may be repeatable with a VM or with wine
02:25:28 <geekosaur> maybe with clang, since there isn't much reason a cross-lld would be problematic
02:25:50 <gammafunk> wonder if mingw clang exists
02:26:58 <gammafunk> or if generally you can cross-compile to windows from clang
02:27:24 <geekosaur> in theory it should be easy. no idea if anyone has set it up
02:28:56 <geekosaur> (the compiler actually generates LLVM bytecode; the translation to "native" code should be non-problematic. or just compile to LLVM bytecode everywhere and have lld do the cross-generation. this is what LLVM is *for*)
02:31:48 <rumflump> I just try using the flags mentioned in the docs, I have no idea what any of it means :P
02:31:56 <rumflump> bytecode sounds nice though
02:33:31 <gammafunk> probably just don't use LTO
02:33:55 <gammafunk> I doubt anyone has actually looked at crawl's performance with and without it
02:34:08 <gammafunk> but it'd probably only be relevant to any significant degree to server operators
02:34:09 <rumflump> apparently! I thought the answer was don't use SOUND=y, but maybe sound was never a problem and it was LTO this whole time
02:34:12 <geekosaur> it's not. at least for llvm; it runs nowhere. it's designed so you can send a compiled program to a random machine of any architecture, and it runs lld on the bytecode to generate optimized native code
02:34:27 <geekosaur> which means the bytecode's got to contain a lot of stuff to enable lld to generate optimal native code.
02:34:40 <geekosaur> otoh that work has already been done for you :)
02:35:26 <gammafunk> I wrote a byte once
02:38:24 <geekosaur> (fwiw I did all the mac builds with LTO, and it actually made for visible speedups. but Apple put a lot of work into LTO, at least for Macs. (but it does seem to work fairly well on x86 linux too)
02:39:01 <rumflump> I may have misquoted you then
02:39:10 <geekosaur> hat said, I don't think I;ve done a local tiles build that way on linux, just console and webtiles
02:39:17 <rumflump> I thought "it runs nowhere" meant "LTO doesn't do much"
02:39:34 <gammafunk> I think he was talking about bytecode
02:39:37 <geekosaur> ^
02:40:03 <geekosaur> there are forms of bytecode that are runnable: Java .class files, Erlang .beam files, Python .pyo files, etc.
02:40:34 <geekosaur> LLVM bytecode is not designed to be run, it is designed to be converted to native code at the point of use, for supercomputing clusters
02:40:56 <Cheibriados> New branch created: pull/498 (1 commit) 13https://github.com/crawl/crawl/pull/498
02:40:56 <Cheibriados> 03yrmvgh02 {GitHub} 07https://github.com/crawl/crawl/pull/498 * 0.20-a0-910-g0ec22d7: don't recommend LTO 10(5 minutes ago, 1 file, 1+ 1-) 13https://github.com/crawl/crawl/commit/0ec22d7c6d90
02:48:16 <gammafunk> [skip ci]
02:48:26 <gammafunk> in those kinds of commits, if you can remember
02:48:32 <gammafunk> somewhere in the commit message
02:48:52 <rumflump> yeah, thought of that after posting, thanks for the reminder though
02:48:54 <ProzacElf> i'm on it
02:48:58 <ProzacElf> XD
02:51:19 <gammafunk> I tend to forget
02:51:56 <gammafunk> I'm not actually sure about LTO, I should look more into it maybe
02:52:12 <gammafunk> if there is some bug, it's possibly on our side, and if LTO is uncovering that, all the better
02:52:23 <gammafunk> but it maybe be due to mingw badness, or something
02:52:32 <rumflump> yeah, it's tough because, you have to make a windows build to get the crash
02:52:35 <rumflump> windows means no stack trace
02:52:56 <gammafunk> well you can get a stack trace if you run it through msys2
02:53:02 <gammafunk> I think
02:53:30 <gammafunk> main issue is whether it's recreatable
02:53:33 <geekosaur> I have a vague recollection that gcc's lto still has stability problms for windows target, whether cross compiling or no
02:53:47 <gammafunk> yeah, wouldn't surprise me
02:53:52 <geekosaur> may have been fixed at some point (but then you need to find out if the cross-compiling version has the fix!)
02:54:46 <geekosaur> s/cross-compiling version/cross-compiler/
02:57:29 <rumflump> speaking of poor excuses for a segue to a new topic, this little comment here https://github.com/crawl/crawl/blob/master/crawl-ref/source/skill-menu.cc#L534
02:58:03 <rumflump> means that if I have a skill crosstrain non-symmetrically, the menu will display it wrong but the number will actually be correct, right?
02:58:32 <Gretell> Windows builds of master branch on crawl.develz.org updated to: 0.20-a0-909-g1a4ba80
02:58:51 <rumflump> for a little bit, I had every ranged weap xtrain to throwing, but throwing not xtrain back to them
03:00:01 <rumflump> I searched "iterate over the result c++ stackexchange" and couldn't figure out how to make skill-menu show that properly
03:25:51 <Rotatell> Unstable branch on crawl.beRotato.org updated to: 0.20-a0-909-g1a4ba80 (34)
03:55:07 <Gretell> Monster database of master branch on crawl.develz.org updated to: 0.20-a0-909-g1a4ba80
04:23:48 -!- amalloy_ is now known as amalloy
05:22:43 -!- amalloy is now known as amalloy_
06:37:48 <alexjurkiewicz> |amethyst: tempting but i'm still not over the enum split up
06:39:57 <alexjurkiewicz> rumflump: try gdrive for downloads
06:40:20 <rumflump> wait what
06:40:43 <alexjurkiewicz> rather than mega.co.nz
06:40:50 <rumflump> oh for those builds I threw on there
06:40:59 <rumflump> what you got against nz?!
06:40:59 <alexjurkiewicz> yes
06:41:24 <alexjurkiewicz> rumflump: https://www.youtube.com/watch?v=7xUYbI64QHI
06:42:16 <rumflump> but that PSA could be lying, as youtube is a wholly-owned subsidiary of g**gl€ !!!
06:42:34 <rumflump> oh this has nothing to do with mega
10:52:02 <Cheibriados> New branch created: pull/499 (1 commit) 13https://github.com/crawl/crawl/pull/499
10:52:02 <Cheibriados> 03Gregory D. Chereshnev02 07https://github.com/crawl/crawl/pull/499 * 0.20-a0-910-g3410eb0: Invert conditional 10(20 minutes ago, 1 file, 2+ 6-) 13https://github.com/crawl/crawl/commit/3410eb08aa32
11:55:45 <SteelNeuron> !tell Lasty hey! Due to timezone problems it's probably going to be difficult to have a live chat, but if you want to read just one thing, this post is all about the issues that we never managed to agree upon: https://crawl.develz.org/tavern/viewtopic.php?p=309626#p309626
11:55:46 <Sequell> SteelNeuron: OK, I'll let lasty know.
12:06:13 -!- Mayoi is now known as erkin
12:18:05 <SteelNeuron> !tell Lasty I have noticed there is a huge misunderstanding between what people think I wanted the martial attacks to be and what I actually think they should be, and that post is my best effort to try and clarify that, so I'd appreciate if you gave it a careful read.
12:18:06 <Sequell> SteelNeuron: OK, I'll let lasty know.
12:32:05 <SteelNeuron> I would also appreciate if you read it MarvinPA since you made the changes I'm criticising there.
12:40:37 -!- yesno__ is now known as yesno
12:52:06 -!- amalloy_ is now known as amalloy
13:25:34 <Henzell> Unstable branch on crawl.akrasiac.org updated to: 0.20-a0-909-g1a4ba80 (34)
14:23:18 <gammafunk> "So i am referring to ancestor dude, dancing guy and wujin monks.
14:23:20 <gammafunk> So i am referring to ancestor dude, dancing guy and wujin monks.
14:23:41 <gammafunk> "So i am referring to ancestor dude, dancing guy and wujin monks."
14:23:47 <gammafunk> oh, oops
14:24:41 <gammafunk> anyhow I propose these new names for hep, usk, and wu
14:24:44 <gammafunk> https://www.reddit.com/r/dcss/comments/60tunv/how_about_these_new_gods/
14:30:50 <gammafunk> could explore_stop_type be converted into a bitfield?
14:44:40 <minmay> merge hep, usk, and wu into hepuskian council
14:44:56 <minmay> as you deal damage in combat, an ancestor appears and starts wall jumping to the stairs
16:39:41 <gammafunk> !tile tentacled monstrosity
16:39:41 <Sequell> https://raw.githubusercontent.com/crawl/crawl/master/crawl-ref/source/rltiles/mon/aberrations/tentacled_monstrosity.png
16:39:51 <gammafunk> command I made to link to the tile image from monster name
16:41:25 * dpeg demands equal treatment: !glyph command :)
16:41:29 <dpeg> +1 gammafunk
16:41:43 <Cheibriados> tentacled monstrosity (03X) | Spd: 10 | HD: 23 | HP: 96-149 | AC/EV: 5/5 | Dam: 22, 17, 13, 903(constrict) | 05demonic, 10doors, amphibious, evil, see invisible | Res: 06magic(160), 05fire, 02cold, 10elec, 03poison, 12drown, 04rot, 13neg+++, 13torm | Vul: 08holy | XP: 2966 | Sz: Giant | Int: human.

16:41:43 <gammafunk> %?? tentacled monstrosity
16:41:47 <gammafunk> implemented, dpeg
16:42:10 <dpeg> gammafunk: which gives a link to a web page with the glyph!!!
16:42:26 <gammafunk> I thought germans were efficient
16:42:37 <dpeg> prejudice!
16:42:41 <gammafunk> link to a url to display an ascii char in irc...
16:42:41 <dpeg> we build good tanks tho
16:42:49 <dpeg> tanks obama
16:49:37 <amalloy> !cmd !glyph .echo http://dev.host/~shmup/crawl/dcss-glyphs.html#$1
16:49:38 <Sequell> Defined command: !glyph => .echo http://dev.host/~shmup/crawl/dcss-glyphs.html#$1
16:49:44 <amalloy> !glyph lich
16:49:44 <Sequell> http://dev.host/~shmup/crawl/dcss-glyphs.html#lich
16:49:56 <shmup> weeee
16:49:59 <amalloy> dpeg: so let it be
16:50:01 <dpeg> I cannot test (on windows).
16:50:05 <dpeg> but it looks very good :)
16:50:24 <amalloy> what does windows have to do with it?
16:50:29 <dpeg> amalloy: btw, many style points for the "shmup" in there!
16:50:35 <amalloy> dpeg: shmup built it
16:51:00 <dpeg> amalloy: I am using irssi on a git bash, so I cannot click the link, but I cannot copy it linux-style either :(
16:51:05 <amalloy> ah
16:51:07 <dpeg> amalloy: shmup = shoot'em up
16:51:10 <amalloy> dpeg: i know
16:51:16 <amalloy> but shmup is the name of the person who created this page
16:51:20 <amalloy> it was not me
16:51:24 <dpeg> I hope shmup is into shmups!
16:51:29 <shmup> sorta
16:51:30 <amalloy> shmup: ^?
16:51:39 <shmup> prob used to be quite a bit more
16:52:13 <shmup> !glyph sheep
16:52:13 <Sequell> http://dev.host/~shmup/crawl/dcss-glyphs.html#sheep
16:52:16 <shmup> so rad.
16:52:20 <dpeg> shmup: I was a big fan of second generation shmups (Darius Gaiden, R-Type, Gradius and so on). I hope my kids will relive this :)
16:52:34 <Rotatell> ziggurnaut (L6 GhNe)  (D:5)
16:53:35 <shmup> nice dpeg yeah i was sorta late into shmups and when i was playing dragon spirit on `nes`, i didn't know there was a whole genre of those games
16:54:54 <dpeg> shmup: I'm sad that I missed bullet hell games, and stuff like Ikaruga. I guess these are emulated by now? So many good games...
16:55:20 <shmup> yeah i played a buncha ikaruga but one i really really liked and not sure how i stumbled over is `warning forever`. it's an odd-ball in this arena
16:55:39 <dpeg> also Japanese?
16:55:42 <shmup> many of the good bullet curtains are just PC games in general though, ala ikaruga
16:55:48 <shmup> and that
16:56:09 <dpeg> a friend once bought Kamui for me when he went to Japan
16:56:40 * dpeg never understood this 3d nonsense
17:00:41 <shmup> i use irssi via git bash every day at work (hi right now), links are no problem, dpeg
17:01:07 <shmup> double click the url, press return so quickly deselect it but upon double clicking its technically _on your clipboard_
17:01:22 <shmup> s/so quickly/to quickly
17:01:45 <dpeg> I cannot double click the link. This is a terminal of Mingw on Windows 7.
17:01:53 <shmup> i misread i think then
17:02:14 <dpeg> it's my wife's machine, should really go linux here, too
17:02:54 <shmup> this is a .net shop so i do what im told but the linux subsystem on win10 has been kinda fun to play with
17:03:21 <shmup> ok burger time
17:03:25 <dpeg> dig in!
17:03:33 <minmay> clean asia was good
17:06:03 <G-Flex> people know about clean asia?
17:06:32 <G-Flex> I thought it was just me
17:14:47 <CanOfWorms> okay, stomp can injure allies and anger them and there's no prompt
17:14:50 <CanOfWorms> good to know...
17:53:19 <Cheibriados> 03amalloy02 {GitHub} 07* 0.20-a0-910-gb611367: Remove duplicate word in a comment 10(9 seconds ago, 1 file, 1+ 1-) 13https://github.com/crawl/crawl/commit/b611367d8c0e
17:54:14 <|amethyst> I wonder how how many word duplications like that we have
17:55:09 <gammafunk> we should do a regex pass over the entire source tree
17:55:15 <gammafunk> remove any instance of duplicated words
17:55:51 <gammafunk> also, amalloy, get good
17:56:02 <|amethyst> If we had had a program like that a few weeks ago...
17:56:04 <gammafunk> removes a word and can't even get some millimarvins xD
17:56:11 <ProzacElf> burnt!
17:56:42 <|amethyst> I think that that would be a great idea
17:57:45 <amalloy> |amethyst: http://www.goodreads.com/quotes/58832-good-item-seven-the-had-had-and-that-that-problem
17:59:35 <|amethyst> amalloy: That's what Prince Said said!
17:59:44 <ProzacElf> oh snap
17:59:45 <ProzacElf> lol
18:09:23 <Jorgrell> Unstable branch on crawl.jorgrun.rocks updated to: 0.20-a0-910-gb611367 (34)
18:20:11 <gammafunk> https://www.reddit.com/r/dcss/comments/60x93n/dcss_tiles_without_mouse_commands/
18:20:41 <gammafunk> apparently our tutorial is pushing users to the mouse!
18:20:57 <gammafunk> I haven't looked at the tutorial messages and how they list the command keys
18:33:15 <amalloy> !source tutorial1.txt
18:33:16 <Sequell> Can't find tutorial1.txt.
18:33:21 <amalloy> !source dat/descript/tutorial1.txt
18:33:22 <Sequell> Can't find dat/descript/tutorial1.txt.
18:34:03 <amalloy> https://github.com/crawl/crawl/blob/master/crawl-ref/source/dat/descript/tutorial.txt
18:37:59 <amalloy> <nomouse> and <console> tags in that file look like the problem
18:38:41 <amalloy> most stuff has unconditional text, plus text conditional on localtiles, so tiles players get everything in those cases. a few snippets have text conditional on *not* localtiles, which is what bothers the guy on reddit
18:39:02 <amalloy> gammafunk:
18:39:10 <gammafunk> cool
18:39:23 <gammafunk> working on something else right now, but maybe I can get to it later
19:02:29 <Lantell> Unstable branch on underhound.eu updated to: 0.20-a0-910-gb611367 (34)
19:12:57 -!- amalloy is now known as amalloy_
20:41:52 <alexjurkiewicz> !tell hellmonk can cpo get one-way hellcrawl? I want to play it
20:41:53 <Sequell> alexjurkiewicz: OK, I'll let hellmonk know.
20:41:56 <alexjurkiewicz> !seen hellmonk
20:41:56 <Sequell> I last saw hellmonk at Wed Mar 22 05:45:37 2017 UTC (18h 56m 19s ago) quitting, saying 'Quit: Page closed'.
21:00:40 <gammafunk> one-way hellcrawl
21:01:03 <gammafunk> "Going...down!"
21:02:43 <dpeg> I am surprised, the no_stairs monster tag doesn't work for me. (Anymore? I think it used to work at some time.)
21:03:48 <dpeg> My no_stairs big kobold happily uses both portal exit and staircases.
21:06:09 <gammafunk> it may need something custom, not sure
21:06:23 <gammafunk> there's also the issue of player expectation
21:06:41 <gammafunk> perhaps the monster should simply dissapear when the exit is reached?
21:06:48 <gammafunk> with a little message or something
21:07:01 <gammafunk> I guess you'd have to define "the end of the portal" as an event
21:07:08 <dpeg> yes, can be done, right now I just wonder why it doesn't work :)
21:08:18 <dpeg> syntax.txt has: Sigmund att:friendly no_stairs
21:13:17 <dpeg> another fun fact: friendly orc priests don't smite, those bastards
21:15:33 <dpeg> place D:4 zombie produced an imperial myrmidon zombie -- should that be?
21:16:16 <dpeg> pretty sure that not: it gave 6 XLs
21:18:30 <hellmonk> pretty sure I got an early D wolf spider zombie once
21:18:45 <hellmonk> zombie generation is weird imo
21:19:28 <dpeg> hellmonk: it's the first time for me to be so grossly off... I can manually define a zombie list, no problem
21:19:45 <hellmonk> oh, are you reworking that part of monster gen?
21:20:47 <minmay> what exactly did you use to make the zombie
21:22:30 <dpeg> hellmonk: no, I am using it for my pet peeve portal vault
21:22:55 <dpeg> minmay: place D:4 zombie
21:22:57 <hellmonk> ah
21:23:28 <dpeg> the most mind-boggling thing is that no_stairs reliably won't work, and I believe it did work before
21:23:31 <minmay> dpeg: "place D:4 zombie" isn't a valid mons spec...
21:24:06 <minmay> or at least shouldn't be according to debug mode
21:24:17 <minmay> "place:D:4 zombie" is
21:24:47 <dpeg> minmay: yes, that's why I have. Sorry for typo,
21:28:37 <minmay> !source mapdef.cc:4033
21:28:38 <Sequell> https://github.com/crawl/crawl/blob/master/crawl-ref/source/mapdef.cc#L4033
21:28:53 <minmay> so this isn't really related but can someone explain this line
21:31:44 <dpeg> anyway, more tomorrow, I need to make my map definitions tighter for further layouts anyway
21:31:55 <dpeg> double "anyway" :(
21:37:00 <johnstein> alexjurkiewicz: why are you in the acknowledgments for Peter Tyson's O'Reilly DF book?
21:37:52 <johnstein> (it says testers, so I'm assuming you tested the things in the book)
21:38:14 <johnstein> probably should have !telled him. hmm
21:39:53 <gammafunk> lol, an oreilly DF book
21:40:09 <gammafunk> just going to become out of date in 6 mo
21:54:37 <ProzacElf_> oreilly?  what, just like 200 pages of bitching about the liberal slant of DF?
21:54:38 <ProzacElf_> =p
22:08:06 <gammafunk> DF is apparently adding procedurally generated magic, but where procedurally generated means something even more elaborate than you might guess
22:29:51 <johnstein> it's a pretty good book and the ebook gets updated about once a year
22:30:18 <johnstein> the basics are pretty consistent. I'm currently dipping my toe back into it
22:31:11 -!- Shard1697_ is now known as Shard1697
22:32:14 <Shard1697> shmup chat happened a long while ago but I just wanted to say that Crimzon Clover is one of the few games that I consider 'perfect'
22:32:43 <Shard1697> as in, I do not think it could have accomplished its goals any better than it did
22:37:53 <minmay> <Shard1697> as in, I do not think it could have accomplished its goals any better than it did
22:38:12 <minmay> That's a very generous definition of 'perfect'
22:38:29 <minmay> Like, Monster Girl Quest and Don't Shit Your Pants probably meet that definition
22:48:37 <alexjurkiewicz> minmay: be that as it may, I think "perfect within the confines of its genre" is generally understood to mean something different
22:48:37 <Sequell> alexjurkiewicz: You have 1 message. Use !messages to read it.
22:49:04 <alexjurkiewicz> hellmonk: i'll wait til your go then. I will be a little unavailable over the weekend btw
22:52:42 <alexjurkiewicz> johnstein: because I am very good at dwarf fortress
22:54:54 <gammafunk> have you won dwarf fortress tho
22:55:07 <gammafunk> what's your score
23:18:32 <minmay> j
23:37:31 <Dracunos> :wq
23:43:53 <alexjurkiewicz> gammafunk: i never finished HFS