00:31:16 -!- Rast- is now known as rast 00:34:42 it does that regularly 00:34:52 advil: btw, not sure if you saw, but you and I both now have ops in ##crawl 01:54:07 !tell dpeg gold distraction stabbing is fair when used for stabbing, but the ability to just walk away from fights halfway through is a bit strong 01:54:08 rast: OK, I'll let dpeg know. 02:38:58 !lg * won be / !trog 02:39:00 1498/7650 games for * (won be): N=1498/7650 (19.58%) 07:45:57 gammafunk: ah, good to know, I didn't actually see that. 07:46:17 on a related note, https://crawl.develz.org/tavern/viewtopic.php?f=5&t=24558 07:46:35 !lg dani 07:46:35 100. Dani the Bolt Thrower (L4 FoHu), slain by a gnoll (a +0 flail) on D:3 on 2017-11-30 11:04:10, with 84 points after 1377 turns and 0:02:11. 07:46:39 !lg dani -log 07:46:40 100. Dani, XL4 FoHu, T:1377: http://crawl.akrasiac.org/rawdata/Dani/morgue-Dani-20171130-110410.txt 07:46:54 !lg dani -2 -log 07:46:55 99/100. Dani, XL4 FoFi, T:898: http://crawl.akrasiac.org/rawdata/Dani/morgue-Dani-20171130-065727.txt 07:47:19 !lg dani -3 -log 07:47:19 98/100. Dani, XL27 HaHu, T:99396: http://crawl.akrasiac.org/rawdata/Dani/morgue-Dani-20171129-230902.txt 07:48:13 !lg dani -4 -log 07:48:14 97/100. Dani, XL6 FoCK, T:3116: http://crawl.akrasiac.org/rawdata/Dani/morgue-Dani-20171128-233950.txt 07:50:43 maybe it doesn't show in the log, I'm not sure it's through tileschat 07:50:57 !lg dani x=src -3 07:50:58 98/100. [src=cao] Dani the Slingshot (L27 HaHu of Okawaru), escaped with the Orb and 3 runes on 2017-11-29 23:09:02, with 1450649 points after 99396 turns and 5:47:01. 07:51:50 !dump dani 07:51:51 http://crawl.akrasiac.org/rawdata/Dani/Dani.txt 08:18:31 Poisonous Vapours leaks information about monster equipment 13https://crawl.develz.org/mantis/view.php?id=11302 by damerell 11:12:26 03advil02 07* 0.21-a0-510-g5070ee8: Prevent a bad linebreak in the skill menu 10(40 seconds ago, 1 file, 1+ 1-) 13https://github.com/crawl/crawl/commit/5070ee8bc0dd 11:54:10 The build has errored. (master - 5070ee8 #8974 : advil): https://travis-ci.org/crawl/crawl/builds/309586814 12:04:35 Unstable branch on crawl.akrasiac.org updated to: 0.21-a0-510-g5070ee8 (34) 12:55:04 -!- amalloy_ is now known as amalloy 13:09:02 -!- amalloy is now known as amalloy_ 13:58:15 -!- amalloy_ is now known as amalloy 14:07:20 -!- kooky is now known as wwwn 14:22:42 -!- amalloy is now known as amalloy_ 14:52:06 <|amethyst> why did the travis bot report that one as an error when the link shows success? https://travis-ci.org/crawl/crawl/builds/309586814 15:28:07 looks like it was building the PR branch, boit master, and the PR vanished out from under it 15:29:02 and.or it got confused, sinbce the UI is claioming it's building the PR branch but the IRC report says master. either way looks more like a glitch in travis-ci 15:29:17 *not master 15:29:40 ("boit"? and I thought some of the typoes I was making in a text file earlier were weird...) 15:30:58 guessing: they try to handle "pr was merged" but part of the UI doesn't get the message 15:31:41 or maybe "merged from pr" and they're out-clevering themselves 15:32:35 in any case the "backtrace" at the top of the build seems to say it's from a PR even though the rest says "master" 15:41:26 !tell MarvinPA How do you feel about artefact scarves? I find their existence to be a bit weird, since the distinction from cloaks was supposed to the egos. 15:41:27 gammafunk: OK, I'll let marvinpa know. 15:41:46 !tell MarvinPA Would you support either always making artefact scarves have a scarve ego (like how we handle amulets/rings) or just not generating artefact ones? 15:41:46 gammafunk: OK, I'll let marvinpa know. 15:49:29 I find I actually think about them the other way around: in some sense they're all artefact-oid since they all have egos and you can't enchant them 16:08:37 that reasoning could also apply to things like magical staves 16:08:45 which can be used as melee weapons but can't be enchanted 16:09:17 it's not a problem that a design like that exits, but just like how we don't have artefact magical staves (except possibly as some kind of urnand), we probably shouldn't have artefact scarves 16:09:40 but if we do, I think it'd be better if they just had a base type like how we do for jewellery 16:10:19 s/urnand/unrand/ 17:00:22 |amethyst: I think it's just because I restarted the individual jobs and they finished. The PR thing is just an interface weirdness of travis, it puts all job numbers there for some reason 17:00:59 does anyone know what crawl thinks an outline font is? alternatively, what setting "outline=true" on load_font is intended to do? 17:08:37 "not a bitmap" 17:22:24 hm, do you know that specifically about crawlcode, or do you mean in general? That distinction doesn't really seem to be what this parameter does. 17:23:05 the biggest difference is that they treat the alpha channel differently 17:23:44 actually the same font (on default settings) is loaded both ways 17:26:59 Scarfs under Ash are eligible for identification. 13https://crawl.develz.org/mantis/view.php?id=11303 by Yermak 17:28:31 there is a certain amount of historical crupt here 17:28:36 *cruft 17:28:55 or alternatively *crap 17:31:27 yeah, I think that's part of it 17:31:30 fr: *Crap artefact property 17:31:41 the non-outline part of the code does seem to be doing the right thing for a bitmap font 17:33:47 the outline part, as best as I can tell, is manually doing some antialiasing/blurring 17:35:53 and produces fairly weird results if run on a font that is already an anti-aliased bitmap 17:36:27 heh I've spend like 2 days trying to understand how alpha blending works, and the problem is in the end just this 17:36:56 right, that's the historical cruft. sounds like it's from back when ATM was the go-to for outline fonts 17:37:51 or very early TrueType 17:38:37 ATM? Andale Tsomething mono? 17:43:58 yeah, this code is 8 years old 17:50:30 adobe type manager 17:51:15 (and anyone else here who remembers it just hissed. loudly.) 17:54:07 ah 17:58:11 I wonder if I remove this pseudo-antialiasing will someone come out of the woodwork and say they are using a font that relies on it 18:02:14 in principle I think this code could be fixed by having freetype load the font as monochrome if outline is set, but I don't see the point 18:04:45 (the problem I've been banging my head against turns out to be that crawl is applying the pseudo-antialiasing to antialiased fonts, which is messing up the alpha channel for the already anti-aliased parts, and leading to non-transparent dark colors for those pixels, which of course don't blend) 18:05:56 sounds like that code should be removed; certainly all modern systems support anti-aliased fonts, right? 18:06:19 yeah, you have to work at it to get freetype not to do it 18:07:24 right, doesn't our Tiles build actually depend on freetype? 18:07:28 yes 18:07:53 it's possible that in 1999 the beta version of freetype2 (which must be what they used) didn't antialias by default 18:08:03 sounds like that code should have been removed when we added that as dependency 18:08:06 hrm 18:08:35 Bad LRD autoaiming 13https://crawl.develz.org/mantis/view.php?id=11304 by Yermak 18:08:35 I checked and it has been freetype2 all along 18:09:20 %git 62f7040f14 18:09:20 07Enne02 * 0.4-a0-786-g62f7040: Tiles! 10(10 years ago, 2441 files, 26569+ 109-) 13https://github.com/crawl/crawl/commit/62f7040f14b3 18:09:50 advil: I'm not even sure they used a 1999 freetype, since that's 2007 18:10:11 oh duh 18:10:19 I was off by 10 18:10:24 %git 149c76283904371235ea6d5 18:10:24 07Ixtli02 {Enne} * 0.7.0-a0-1423-g149c762: Created FontWrapper, and made FT flag in makefile. 10(8 years ago, 12 files, 1009+ 882-) 13https://github.com/crawl/crawl/commit/149c76283904 18:10:26 early 2008 I guess 18:10:32 introduced the problematic code 18:10:36 wow 18:10:37 and it hasn't been touched since 18:12:36 so this is also when freetype was added to tiles? 18:12:44 I think so 18:13:14 interesting, wonder why this apparently unecessary antialiasing code was added along with 18:13:16 heh the parameter that loads it as outline is older 18:13:25 %git c070fa4aa13e00f09de5d208fdcdb 18:13:25 07jpeg02 * 0.5-a0-2694-gc070fa4: [Tiles] Keep track of which dungeon grid the player last clicked and don't display the grid description (feature, item, monster) for that grid. Fixes (part of) 2660578, i.e. monster descriptions hiding fighting messages. 10(9 years ago, 6 files, 102+ 95-) 13https://github.com/crawl/crawl/commit/c070fa4aa13e 18:13:52 well it wasn't introduced there, that was just the last change 18:14:02 so probably that freetype commit should have changed that parameter 18:14:53 %git 42f25ae23ee34f9ae3a0e31f 18:14:53 07Enne02 * 0.5-a0-665-g42f25ae: Support for inline graphics and mouse input on menus. 10(9 years ago, 17 files, 1019+ 121-) 13https://github.com/crawl/crawl/commit/42f25ae23ee3 18:15:09 anyways, I'll finish fixing this later, but at least now I know the issue 18:17:55 Unstable branch on underhound.eu updated to: 0.21-a0-510-g5070ee8bc0 (34) 18:25:30 keep in mind that linux antialiasing, to the extent it existed, was utter shit back then 18:25:55 even in freetype in 2007? 18:26:04 I think it was pretty well supported at that time 18:26:13 a few of us used various freetype patches that improved things, but it took a long time for them to upstream (2013/2014ish), I think due to patent issues 18:26:21 it was present 18:26:28 it was not at all good 18:26:33 I definitely was using freetype on linux in the 2000s 18:28:05 I do recall it looking maybe less good than commercial OS' antialiased fonts, but it was still an improvement 18:28:25 I can't remember the name of the old freetype patches, in the early days you had to apply them separately and not distribute the result 18:29:23 but also some of the things that they didn't add was because they were legal in the US but not some other countries (and others were legal in some other countries but not here, such as the patent on the TTF hint opcodes) 18:30:23 and the legal tangles meant that base freetype was really bad and some distributions took the tack of "legal where we're building the distro, we'll take our chances" etc. 18:31:30 I just can't believe I'm no longer running Xwindows 19:20:28 -!- jeefus is now known as jefus 19:26:34 Issue with wands and autopickup 13https://crawl.develz.org/mantis/view.php?id=11305 by Yermak 19:26:38 gammafunk, ^ 19:55:09 -!- Fixer_ is now known as Fixer 19:58:49 -!- gammafunk is now known as Undertaker 19:58:59 -!- Undertaker is now known as gammafunk 21:02:09 -!- amalloy_ is now known as amalloy 21:07:52 -!- Webmant9 is now known as Webmant 21:20:40 03amalloy02 07* 0.21-a0-511-g5c865af: Improve level 2 MUT_HOP description (longer than what?) 10(21 seconds ago, 1 file, 1+ 1-) 13https://github.com/crawl/crawl/commit/5c865af6885d 21:22:27 speaking of weird messages, does anyone mind if i change "looks as sick as possible!" to "looks extremely sick!" or something? as sick as possible is bizarre to me 21:22:47 let alone how you can *look* as sick as possible 22:04:50 -!- amalloy is now known as amalloy_ 22:09:13 its weird to see the sa thread so positive on hellcrawl since i cut even more shit that the real devteam 22:10:25 03advil02 07* 0.21-a0-512-g0ee8fc4: Don't load any fonts as "outline" fonts 10(4 hours ago, 1 file, 2+ 2-) 13https://github.com/crawl/crawl/commit/0ee8fc4e44d2 22:10:25 03advil02 07* 0.21-a0-513-gdcb3f5e: Remove old outline font code 10(7 minutes ago, 5 files, 26+ 67-) 13https://github.com/crawl/crawl/commit/dcb3f5eb2531 22:10:37 that sa thread is something 22:12:22 does not make me particularly interesting in that thread, or sa as a place in general 22:12:30 hehe 22:12:31 "devs are ruining crawl forever by removing things btw did u guys hear about this fork that cut 5 species 3 backgrounds 6 runes 40 floors and the concept of upstairs" 22:12:34 *in participating in 22:12:49 yeah it's pretty terrible, but it is a thread with an interesting history 22:13:10 still good to get feedback from another source though 22:13:17 yeah 22:13:35 in its heyday it was fairly important, since a few devs came from that community 22:13:38 though one of the striking things about it is how a lot of them seem to think they're at the center of the crawl world 22:13:47 I don't need to take feedback w/ a grain of salt because I am already salty 22:13:48 I barely knew that community existed until recently 22:14:09 yeah that's a funny thing about online communities, I've noticed that each of the dcss discussion places do this to some extent 22:14:14 true 22:14:19 I think it's stronger there because of just what SA is as a website 22:14:36 and people have been playing the game a while and posting about it, maybe that's the biggest thing 22:14:46 tbf I think they're probably better than the reddit 22:15:20 There are a few good players who post on SA that I know of, but none of the "top players" 22:15:55 yeah I guess like uv4 and a couple other people post on the reddit 22:16:00 by good I just mean notable for having lots of wins with a good winrate, tournament participation at a high level 22:16:15 yeah reddit I can't really say what is the "average level" there 22:16:26 the reddit is somewhat new while tavern/SA/irc have been around forever 22:16:41 I like that reddit isn't too toxic though 22:16:41 feels like its mostly people going "lol check out my D:2 gds how will I die?" or "is this species/background viable" 22:16:53 yes, a lot of basic beginner questions 22:17:02 imo you should make DCSS 0.21 less toxic 22:17:06 by removing poison magic 22:18:56 if he advances to lasty3 that will be an entirely new upgrade! 22:19:45 https://www.youtube.com/watch?v=8TGalu36BHA 22:20:27 yes 22:20:29 just like that 22:20:40 not to be confused with over 9000 of course 23:12:32 -!- Krakhan|2 is now known as Krakhan 23:18:49 -!- mikee__ is now known as mikee_ 23:30:52 -!- mikee___ is now known as mikee_ 23:41:28 -!- mikee___ is now known as mikee_ 23:59:37 Unstable branch on CRAWL.XTAHUA.COM updated to: 0.21-a0-513-gdcb3f5e (34)