00:20:42 !tell MarvinPA I wrote up a rework of Fedhas here https://github.com/crawl/crawl/wiki/Fedhas-Rework 00:20:43 gammafunk: OK, I'll let marvinpa know. 00:21:51 !tell MarvinPA Some of the proposed abilities probably need some work, and I'm certainly open to only using a subset of this proposal or making major alterations, but I want to get some agreement as to what parts of these proposals would be good to go forward with 00:21:51 gammafunk: OK, I'll let marvinpa know. 00:23:17 Any other dev is of course welcome to comment if they care a lot about fedhas 00:34:31 gammafunk: fwiw seems overall reasonable to me. i prefer LRD-targeted growth to corruption-style. i assume you'd notice this when implementing, but note that for barachi LOS goes up to 8, where your current formula for aura size ends at 7 00:35:07 of course mpa is a much better critic of design/balance than i am 00:35:48 oh, I wonder how that's handled for tso 00:35:58 maybe it's just a higher multiplier so it doesn't matter 00:36:06 s/tso/tso+dith/ 00:36:29 could just give frogs an extra tile at that final piety level 00:37:51 yeah, or just implement it as min(max_los(), piety * 2 + 1) 00:38:18 cleaner implementation, but probably worse than just giving them an extra tile at 4* 01:38:23 Unstable branch on crawl.develz.org updated to: 0.24-a0-530-g18ba70b (34) 01:43:10 For the screenshot part of the json morgue i was thinking i could give the actual tiles/glyphs references instead of a string 01:43:24 but again i'm worried about the eventual size of the json 01:43:45 have you guys any thoughts on it? 01:44:10 the idea would be that the morgue viewer would be able to recreate the exact tile map 02:10:15 Windows builds of master branch on crawl.develz.org updated to: 0.24-a0-530-g18ba70b 02:13:23 bgiannan: there's backwards compatibility issues there 02:14:13 why would there be? those would only be in the json morgue, the txt morgue would be left untouched 02:14:13 and both would be dumped 02:16:58 e.g. if you embed packed_cells / tile flags 02:17:21 oh you mean in future versions 02:17:26 currently, those are only used during rendering, so they can be changed whenever is convenient 02:17:31 yep 02:18:57 maybe i could add an info about which tileset to refer to 02:19:13 the viewer would have to be able to reference tileset of 0.23, 0.24 etc. 02:19:33 one possibility is to embed the level data, which would also allow you to adapt your screenshot to the viewer's screen size, but that also has back-compat issues 02:20:55 well, if you want pixel-exact screenshots, the most robust way is to embed the tilesheet images within the morgue 02:21:23 yeah but that would take too much space 02:21:29 you could include only the parts that are actually visible on screen 02:21:53 now that you say this 02:22:02 there's already a .map file that is generated 02:22:16 so we could very well add a .png 02:22:32 which could be tiny given its size and the relative low count of colors 02:23:24 guess i should also think about it for my "snapshot" PR 02:23:37 as in just take a screenshot and store it directly, without any extra steps? 02:23:44 yes 02:24:32 that would definitely be simple, but you probably would see larger file sizes 02:25:38 your png encoder wouldn't be able to factor out any of the repetition, and I'd imagine almost all screenshots have a huge amount of repetition 02:26:19 you'd also not be able to adapt your screenshot to different viewport sizes, but that may not be very important 02:26:40 well i think for a morgue file the viewport should always be the same 02:29:51 right, but we might want to change the UI or viewport size in the future 02:33:56 Btw if one of you guys wants to dive into the PR. It's not mergeable in its current state I think. I duplicate a lot of code but i'm not sure i should factorize too much because the purpose of the json functions is often not the same as it's txt counter part even though they seem to do the same thing right now. 02:34:11 its* 02:34:54 Also I don't really know c++ so maybe i wrote some enraging stuff 02:37:18 I'll have a look when I have time, which won't be for at least two weeks 02:37:38 no worries, enraging c++ will fit right in 02:39:25 I'm thinking also of writing a txt to json converter. Not everything is "parsable" from the txt (that's why i'm doing the json in the first place). But we could have some basic json converted files with less information. Maybe the hiscore, stats, overview and notes section for example.. 02:40:20 these way we could still view older morgue files from the morgue viewer 03:00:26 Monster database of master branch on crawl.develz.org updated to: 0.24-a0-443-g80245de 03:11:56 -!- amalloy is now known as amalloy_ 03:12:11 Unstable branch on crawl.beRotato.org updated to: 0.24-a0-530-g18ba70b5 (34) 03:15:34 -!- amalloy_ is now known as amalloy 03:30:34 -!- amalloy is now known as amalloy_ 06:03:22 03giann02 07https://github.com/crawl/crawl/pull/1122 * 0.24-a0-539-g51e0576: JSON Morgue: resistances in stats section 10(3 minutes ago, 2 files, 192+ 0-) 13https://github.com/crawl/crawl/commit/51e057643a05 06:58:55 isn't it inconsistent that when a scroll of silence effects are over the game says "Your hearing returns"? Shouldn't it be something like "Your voice returns"? 08:22:02 03giann02 07https://github.com/crawl/crawl/pull/1122 * 0.24-a0-540-gcfb45d8: JSON Morgue: equipment in stats section 10(6 minutes ago, 2 files, 138+ 39-) 13https://github.com/crawl/crawl/commit/cfb45d86e037 10:08:43 03giann02 07https://github.com/crawl/crawl/pull/1122 * 0.24-a0-541-gd47b0ed: JSON Morgue: status, mutations, orb and runes in stats section 10(6 minutes ago, 3 files, 178+ 106-) 13https://github.com/crawl/crawl/commit/d47b0ed44ce2 10:31:49 -!- neunon_ is now known as neunon 11:22:44 hello, I'm getting "Error initialising console input mode." when trying to launch crawl from a version compiled straight from the repo (w7 x64, msys2) 11:23:35 Stable (0.23) branch on underhound.eu updated to: 0.23.1-82-g5129a83184 11:48:48 cbus: have you tried running the latest trunk build and see if you get the same error? (https://crawl.develz.org/trunk/) 11:49:03 will give it a try 11:50:02 03giann02 07https://github.com/crawl/crawl/pull/1122 * 0.24-a0-541-g63cab6e: JSON Morgue: status, mutations, orb and runes in stats section 10(2 hours ago, 2 files, 177+ 105-) 13https://github.com/crawl/crawl/commit/63cab6e0d90d 11:52:15 same thing if I run it in MinTTY running msys2 but works if I just launch the exe from explorer 11:52:40 git version behaves the same way 11:55:59 ah what if you do "start crawl.exe"? 11:56:36 or maybe just `start crawl` 11:56:38 just launch the exe from explorer or such? then it works just like the trunk 11:56:49 when built with msys2 it runs in a command.exe window, and does not use a unix tty 11:56:59 so will not work when run directly in a tty 11:57:01 launches in a new cmd-ish window and wokrs 11:57:20 yeah, that is the expected behavior 11:57:46 if you did want it to work in a tty that *may* be possible building in the linux for windows subsystem, I haven't tried it myself though 12:15:30 I wonder what happens if you try to run it in windows terminal, looks like there's now a preview version out 12:16:26 wow, on github under the MIT license even 12:17:14 mostly playing online but wanted a local version to play around with lua, but ended up doing it in a proper arch vm 12:22:38 -!- amalloy_ is now known as amalloy 12:24:49 bgiannan: why would it say your voice returns? the effect of ?silence is to nullify all sound in the affected area. is there some message printed somewhere that says it affects your voice? 12:27:15 Passwall despawns an enemy 13https://crawl.develz.org/mantis/view.php?id=12041 by Pekkekke 12:27:33 well it doesn't affect your ears either 12:27:44 they work fine, there's just no sound around anymore 12:36:51 huh. has passwall always been able to displace enemies? that's news to me, but i don't see it mentioned in any commit messages in the last few years 12:44:05 yeah, that's pretty old 12:46:20 %git 0eccbe92 12:46:20 07haranp02 * 0.5-a0-2708-g0eccbe9: FR [2163236]: Passwall should wake monsters it displaces. Fix Passwall instadeaths not working for deep dwarves. 10(10 years ago, 2 files, 10+ 5-) 13https://github.com/crawl/crawl/commit/0eccbe92f13c 12:46:27 earliest direct reference to it that I see 12:47:12 Can anyone review my PR https://github.com/crawl/crawl/pull/1109? 12:49:47 %git 1d0f57c 12:49:47 07greensnark02 * 1d0f57cbceb7: Merged stone_soup r15:451 into trunk. 10(13 years ago, 180 files, 39016+ 25829-) 13https://github.com/crawl/crawl/commit/1d0f57cbceb7 12:50:35 amalloy: looks like that passwall behavior appears in that commit, I'm not sure what exactly that commit title means though 12:52:16 god knows. i like git archaeology, but decoding a git-svn commit that's recently become a teenager is not my jam 12:53:03 those might be the days where people were actually using diffs or something 12:56:28 i remember being excited about svn as a replacement for cvs 12:59:07 heh 12:59:33 I had all my files in cvs for a while as an undergrad 13:02:49 %git 041af8f7281887 13:02:49 07greensnark02 * 041af8f72818: Massive merge of 4.1 view code (mostly untested, Stuff will BREAK). Player move effects go to player.cc (move_player_to_grid). Multidrop remembers items that were not dropped because something interrupted the drop. Passwall is kinder to monsters at the receiving end. Sif Muna piety decay halved. scrolls_burn is now replaced by expose_player_to_element(BEAM_FOO, str). your_sign/your_colour moved into struct player as symbol/colour. 10(13 years ago, 44 files, 2356+ 2957-) 13https://github.com/crawl/crawl/commit/041af8f72818 13:03:44 looks like the relevant commits from that merge are preserved on the `stone_soup` branch 13:03:53 that's where shifting came in, before that it blinked the monster 13:06:48 I think the blinking behavior comes from the original implementation, pre-stone-soup 13:11:04 circa 4.0 b20 21:17:13 -!- zoid is now known as `zoid 21:24:40 advil: i just had a thought about game seeds. it'd be cool if they included the game version in them, so that the game could tell you "hey, this seed you got from someone else isn't going to produce the same game for you, because you're running a different version" 21:28:08 otoh of course players will want to type in all kinds of silly things as seeds, and it'd be nice for those to be valid. maybe a canonicalized form that includes the version, and an informal version that's more lax? 21:30:50 -!- amalloy is now known as amalloy_ 22:52:44 -!- `zoid is now known as zoid 23:35:16 Would be nice if using '%' when selecting a race, would actually show the aptitudes of the selected choice instead of opening a long help file