00:26:37 <amalloy> autoexplore is based on dijkstra maps, right? as described at http://www.roguebasin.com/index.php?title=The_Incredible_Power_of_Dijkstra_Maps
00:26:37 <Sequell> amalloy: You have 2 messages. Use !messages to read them.
00:53:02 <alexjurkiewicz> in 0.20 my char doesn't rest before exploring
00:53:06 <alexjurkiewicz> is this not the default??
00:54:39 <Doesnty> that is not the default
01:02:17 <alexjurkiewicz> it seems to be on CPO with a blank rc
01:04:30 <amalloy> alexjurkiewicz: Doesnty is answering you in the positive: yes, it is not the default to rest before exploring
01:12:15 <orbisvicis> how do I build crawl with debug symbols but without the extra message processing info? To trigger this coredump I need to play for some time... so "make debug install" version isnt playable
01:22:16 <alexjurkiewicz> amalloy: oh. FR to change that, seems crazy not to be
01:22:48 <amalloy> alexjurkiewicz: it's only been even an option for a couple versions. probably defaults to off to avoid surprising people with a change they didn't want
01:22:55 <amalloy> orbisvicis: make debug-lite
01:50:11 <johnny0> !tell advil the console colour calibration menu i've been working on: https://www.getlucked.com/crawl/img/ccc_wip.png
01:50:12 <Sequell> johnny0: OK, I'll let advil know.
01:54:42 <alexjurkiewicz> johnstein: feedback: should make it clear the maximum number of supported colours is 16 (looking at that screen I was like "why only 16 with xterm-256? ...oh"), and the test patterns + @ matrix seem redundant
01:54:47 <alexjurkiewicz> johnny0: ^
01:57:30 <Gretell> Windows builds of master branch on crawl.develz.org updated to: 0.21-a0-42-gb537e02
01:58:56 <johnny0> the test patterns are just to bring attention to specific squares on the grid (problematic ones like darkgrey on black)
02:08:07 <johnny0> alexjurkiewicz: it's rough trying to cram everything into 79x24 -- maybe i can get the colour info separated to the right to make it stand out more
02:11:12 <johnny0> anyway, thanks for the feedback
02:37:30 <alexjurkiewicz> 👍
02:52:00 <Gretell> Monster database of master branch on crawl.develz.org updated to: 0.21-a0-42-gb537e02
02:56:18 -!- thurin_ is now known as thurin
03:13:07 <Rotatell> Unstable branch on crawl.beRotato.org updated to: 0.21-a0-42-gb537e02 (34)
03:21:04 -!- amalloy is now known as amalloy_
03:24:07 -!- amalloy_ is now known as amalloy
04:17:02 -!- amalloy is now known as amalloy_
07:37:10 <nee````> Call me a robot, but I can not get past this captcha: https://crawl.develz.org/mantis/signup_page.php
07:38:01 <nee````> What are the cut out letters in the color boxes? I think it's 'f'.
07:38:10 <gammafunk> yeah
07:38:11 <gammafunk> those are f
07:38:17 <nee````> No matter what I enter I always get APPLICATION ERROR #1904
07:38:17 <nee````> Confirmation hash does not match. Please retry.
07:38:17 <nee````> 
07:38:17 <nee````> 
07:39:27 <gammafunk> hrm
07:42:09 <gammafunk> Napkin: It seems that the captcha for mantis registration might be broken; it doesn't seem to accept the hashes based on the images it shows
09:55:31 <Cheibriados> Invisible monster disturbances stop autoexplore/travel even when on the other side of glass. 13https://crawl.develz.org/mantis/view.php?id=11088 by elliptic
10:47:21 <Cheibriados> Autoexplore ignores sleeping monsters 13https://crawl.develz.org/mantis/view.php?id=11089 by orbisvicis
11:54:46 <Cheibriados> Removing the Robe of the Night does not refresh LOS 13https://crawl.develz.org/mantis/view.php?id=11090 by orbisvicis
12:05:08 <advil> !tell johnny0 that looks really great!  One idea for an alternative visual display would be to show a canned bit of a crawl map that has a reasonable number of items / monsters / stuff going on
12:05:08 <Sequell> advil: You have 4 messages. Use !messages to read them.
12:05:08 <Sequell> advil: OK, I'll let johnny0 know.
12:06:13 <advil> though I'm not sure how easy that would be
12:07:46 <Cheibriados> 03advil02 07* 0.21-a0-43-g1479571: Add some more detail to the potion simulator 10(57 seconds ago, 1 file, 14+ 2-) 13https://github.com/crawl/crawl/commit/14795715213b
12:09:29 <orbisvicis> I'm trying to get a coredump, but can't trigger a crash with the debug* builds. Any ideas ?
12:15:33 <advil> you can trigger a crash directly in wizmode, if that's relevant
12:16:19 <advil> uh though I have no idea if we generate a coredump
12:16:37 <advil> or under what circumstances
12:18:47 <orbisvicis> well it gets generated automatically for sigabrt, but I'm trying to get a backtrace for a specific issue (sigabrt from double-free when alt-tabbing from dcss)
12:21:14 <advil> hmm, can you just run it in a debugger?  (Also, what?  Is that bug in mantis?)
12:21:50 <advil> (that seems v bad)
12:23:10 <orbisvicis> well I could run it in gdb, but the without a debug build the backtrace won't be useful
12:23:28 <orbisvicis> and can't trigger the bug on debug builds - so far (happens randomly)
12:23:38 <orbisvicis> no, some issue I've been ignoring for a few years
12:23:39 <advil> oh I see what you mean
12:23:57 <advil> is this on windows?
12:24:10 <orbisvicis> linux / fedora 25
12:24:25 <orbisvicis> might be specific to gnome-shell (maybe...)
12:26:06 <advil> hmm
12:26:10 <advil> tiles?
12:26:14 <advil> I'm guessing?
12:26:36 <orbisvicis> yeah
12:27:33 <advil> we will definitely tend to miss stuff in that configuration just because no devs really play local tiles, and those of us who work on it only ever see it in debug mode. But if you can't figure it out, def put something on mantis, afaik no one is aware of this issue.
12:28:17 <advil> !tstats
12:28:38 <Sequell> Stats after 16 days (t): 2465 players, 879 runers, 427 winners, 1110 wins, 48429 games, winrate 2.29%, total player time 2y+342d+19:08:56.
12:28:55 <advil> 16 days?
12:29:20 <advil> !tstats 10
12:29:35 <Sequell> Stats after 10 days (t): 2465 players, 879 runers, 427 winners, 1110 wins, 48436 games, winrate 2.29%, total player time 2y+342d+20:05:14.
12:31:15 <advil> !tstats 10 t0.19
12:32:45 <Sequell> 90s limit exceeded: killed !tstats 10 t0.19
12:33:01 <advil> sad
12:36:01 <advil> winrates seem high though
12:57:23 <advil> travis is being mean to me
12:57:32 <advil> & I'm on a train so everything is super slow to load
13:01:10 <geekosaur> ??travis
13:01:11 <Sequell> travis[1/2]: travis runs automatic builds of Crawl and its branches, ensuring that they compile properly, and running Crawl's tests on the subsequent builds to look for errors or crashes. See: https://travis-ci.org/crawl/crawl/
13:01:16 <geekosaur> bah
13:01:19 <geekosaur> ??travis[2
13:01:19 <Sequell> travis[2/2]: <Grunt> Travis can't be bargained with. Travis can't be reasoned with. Travis doesn't feel pity, or remorse, or fear. And Travis absolutely will not stop, ever, until your project is dead.
13:01:26 <advil> heh
13:02:04 <travis-ci> The build was broken. (master - 1479571 #8424 : Kyle Rawlins): https://travis-ci.org/crawl/crawl/builds/239636813
13:02:25 <geekosaur> speak of the devil...
13:03:04 <advil> not entirely sure it's real but I also apparently don't have ccache installed on this laptop so it's taking me forever to rebuild
13:04:32 <Henzell> Unstable branch on crawl.akrasiac.org updated to: 0.21-a0-43-g1479571 (34)
13:05:15 <advil> also that commit only changed a test, so if it's real it's not a disaster
13:09:51 <Jorgrell> Unstable branch on crawl.jorgrun.rocks updated to: 0.21-a0-43-g1479571 (34)
13:10:44 * geekosaur is confused, it still shows as building one job and all the others succeeded
13:11:19 <geekosaur> bet they turn off irc notifications again until they sort it out
13:20:28 <advil> I restarted that one job
13:20:35 -!- amalloy_ is now known as amalloy
13:20:40 <advil> since I haven't yet been able to replicate the problem locally
13:23:56 <advil> and in fact that job ran just fine this time, sigh
13:25:42 <advil> well, the non-stress-tests finished fine, at least
13:26:42 <advil> stress tests are fine too
13:29:56 -!- sneaky is now known as sneakyness
13:42:21 <Napkin> gammafunk: yeah, i broke it on purpose
13:42:35 <Napkin> gammafunk: it's rather that fonts are not available in the fonts file ;-)
13:44:23 <Yermak> !apt ha
13:44:25 <Sequell> Ha: Fighting: -1, Short: 2, Long: 0, Axes: 0, Maces: -1, Polearms: -1, Staves: -2, Slings: 4!, Bows: 1, Xbows: 1, Throw: 0, Armour: 1, Dodge: 1, Stealth: 2, Shields: 1, UC: -2*, Splcast: -3, Conj: -2, Hexes: -2, Charms: 1, Summ: -2, Nec: -2, Tloc: 1, Tmut: -4*, Fire: 0, Ice: 0, Air: 0, Earth: 0, Poison: 0, Inv: 1, Evo: 0, Exp: 1!, HP: -1, MP: 0
13:44:33 <Yermak> ops, wrong chat
14:22:18 <Lasty_> advil: in theory 0.20 should be harder than 0.19 since most changes were nerfs. Mut changes were buffs potentially..
14:22:43 <advil> yeah, thatI was surprised by the winrates that gf queried yesterday
14:22:48 <advil> *that's why I
14:22:57 <Lasty_> !tstats 10 t0.19
14:23:43 <Lasty_> I wonder if some of the QoL improvements / selective trimming is keeping winrates higher by virtue of decreased boredom. I feel like that's been my experience this t
14:23:56 <advil> could be!
14:24:01 <Sequell> Stats after 10 days (t0.19): 2611 players, 825 runers, 408 winners, 941 wins, 51636 games, winrate 1.82%, total player time 2y+296d+2:43:36.
14:24:02 <Lasty_> Playing in the t has made me really excited to get back to active development
14:24:52 <Lasty_> 150 less players so far. Huh.
14:24:52 <geekosaur> you'll get over it :)
14:25:11 <Lasty_> geekosaur: ha, probably true. Just wait til I see the player reactions to the next big project I work on :p
14:25:19 <Lasty_> That'll cure me quick
14:25:25 <advil> did you see the 0.20 is too hard thread
14:25:30 <Lasty_> lol yeah
14:25:45 <Lasty_> I was surprised to see some of the comments from certain players
14:26:01 <Lasty_> !tstats 10 t0.18
14:26:24 <advil> one interesting observation somewhere down in there is that the wand of haste removal actually means greater odds of some haste pots earlier, which could even make the game net easier
14:26:59 <Sequell> Stats after 10 days (t0.18): 2250 players, 723 runers, 372 winners, 857 wins, 44196 games, winrate 1.94%, total player time 2y+219d+4:06:26.
14:27:01 <Lasty_> Oh yeah, I was wondering about that. did we increase haste pot spawn rate?
14:27:03 <advil> I didn't run the numbers to check but it has seemed subjectively like I had more haste charges in lair this t
14:27:17 <advil> I think so?  I have to admit I didn't even confirm that, just assumed it :)
14:27:23 <advil> I know !hw increased
14:27:41 <Lasty_> That alone would make a big difference
14:28:00 <Lasty_> players rarely had yellow wands early, but increased spawn rates could significantly impact the early game
14:28:05 <advil> yeah
14:28:15 <Lasty_> maybe we should scale those back down a bit
14:29:40 <Lasty_> !tstats 10 t !mi !be
14:29:47 <advil> maybe...I haven't played enough games to know but they have still felt tight in the games I have played
14:29:52 <advil> someday I'll learn how to use objstat
14:29:57 <Sequell> Stats after 10 days (t !mi !be): 2228 players, 673 runers, 323 winners, 832 wins, 40697 games, winrate 2.04%, total player time 2y+117d+4:10:42.
14:29:58 <Lasty_> haha
14:30:16 <Lasty_> !tstats 10  t0.19 !mi !be
14:31:22 <Sequell> Stats after 10 days (t0.19 !mi !be): 2323 players, 636 runers, 301 winners, 699 wins, 42766 games, winrate 1.63%, total player time 2y+67d+1:53:50.
14:31:36 <advil> huh
14:32:02 <Lasty_> !tstats 10 t (op|mu|ds|de)
14:32:13 <Sequell> Stats after 10 days (t (op|mu|ds|de)): 1248 players, 223 runers, 94 winners, 117 wins, 12411 games, winrate 0.94%, total player time 189d+20:20:39.
14:32:26 <Lasty_> !tstats 10 t0.19 (op|mu|ds|de)
14:33:25 <Sequell> Stats after 10 days (t0.19 (op|mu|ds|de)): 1220 players, 202 runers, 82 winners, 91 wins, 11901 games, winrate 0.76%, total player time 173d+1:50:44.
14:34:14 <Lasty_> Hm. The stats seem fairly cross-cutting. I was wondering if players might just be trying easier games to respond to a "harder" version, but.
14:34:20 <Lasty_> Doesn't seem like it
14:34:43 <advil> well that all does reassure me slightly that the change probably isn't due to that complicated casting fail rate cleanup I did
14:35:43 <Lasty_> haha
14:36:03 <Lasty_> Incidentally 1) I really appreciate you doing that, and 2) I've enjoyed it very much as a player who enjoys forcing high level spells
14:36:21 <advil> oh thanks!
14:36:27 <Lasty_> It's something I'd had my eyes on for a while, and it's a tough one
14:36:34 <Lasty_> so thank you for all your hard work on it
14:36:37 <Lasty_> it plays so much better now
14:37:12 <advil> good to hear -- I did a lot of testing myself but it's something that's hard to get good feedback on
14:37:42 <advil> I think |amethyst is right that to deal with code like that better we may want to figure out some real fixed point math eventually
14:39:14 <Lasty_> is it currently using floating point?
14:39:18 <Lasty_> I didn't review the code.
14:39:30 <advil> no, it's ad-hoc fixed point (with a scale parameter, like a lot of things in crawl)
14:39:55 <advil> before it was slightly more ad-hoc fixed point, with just integers at some stages, hence the steppiness
14:40:56 <advil> got to run, later, thanks for the comments on that change...in principle it's easier to tweak now if you have ideas :)
14:41:39 <Lasty_> ah, gotcha
14:41:42 <Lasty_> ttyl
14:50:09 <minmay> FR: either don't turn kirke's hogs into humans when she dies, or don't give xp/piety for killing them as hogs (as with pikel's slaves)
14:54:01 <minmay> okay, we also need to talk about all these cyan doors in elf
14:54:28 <Doesnty> doors have different colors in console?
14:57:55 <Doesnty> while we're complaining about elf why does twisted ressurection still exist
14:58:10 <minmay> they aren't supposed to
14:58:12 <minmay> but they do in elf
14:58:41 <minmay> the elf builder makes translucent rock walls and puts doors in them and those doors get coloured lightcyan
14:58:41 <Doesnty> lol
14:58:59 <minmay> which would make you expect the door to be translucent
14:58:59 <minmay> but it isn't, it's just an opaque door
14:59:15 <Doesnty> those opaque doors look weird in tiles too
14:59:25 <Doesnty> but more because it's an opaque door along a glass wall
15:54:16 <nee````> Napkin: Could you fix the captcha so I can register for the tracker? |amethyst requested me to open a pull-request for a patch I made.
16:01:20 <amalloy> nee````: you shouldn't need anything from him to open a pull request. pull requests are on github
16:09:06 <nee````> Ah, okay I understood that I needed a mantis login
16:09:43 <nee````> amalloy: Thanks
16:18:08 <amalloy> you'd need a mantis login to create an issue on mantis, but you can send a PR without that
17:04:41 <advil> well if people who find bugs and haven't reported any before can't make an account, that does seem like something to fix :)
17:07:24 <advil> though I suppose if we wait long enough probably Yermak will find every bug (unless they get tired of us not being able to keep up in fixing them all)
17:09:09 <advil> (which is to say, Yermak: I really appreciate your bug reports.)
18:03:27 <Lantell> Unstable branch on underhound.eu updated to: 0.21-a0-43-g1479571 (34)
18:58:50 <johnny0> alexjurkiewicz: i made the colour info [hopefully] more noticeable -- also, maybe this shows off the test pattern importance: https://www.getlucked.com/crawl/img/ccc_animation_no_brighten.gif
18:59:32 <johnny0> (some flashing lights, fair warning)
19:47:08 <alexjurkiewicz> johnny0: nice! one thing I'm confused by is why changing the background colour sometimes makes lightgrey-on-black not be a black background
19:47:27 <alexjurkiewicz> that gif is surprisingly long
19:47:52 <alexjurkiewicz> also please note I only use console while devving, never while playing for real, so don't take my feedback as super valuable
19:51:06 <gammafunk> johnny0: pretty cool
19:51:26 <gammafunk> when you're toggling those options, that's just doing an ingame call to (effectively) crawl.setopt()?
19:51:50 <gammafunk> as in you're just toggling the setting and having crawl reload settings
19:52:06 <gammafunk> it's sort of the beginning of in-game option settings!
19:53:22 <gammafunk> !nick uv4 ytcomultraviolent4 ultraviolent4
19:53:23 <Sequell> Mapping uv4 => ytcomultraviolent4 ultraviolent4
19:54:45 <gammafunk> !kw t0.20
19:54:45 <Sequell> Built-in: t0.20 => start>='2017-05-26 20:00:00' time<'2017-06-11 20:00:00' ((cv=0.20|0.20-a)) explbr=
19:54:53 <gammafunk> !nick wltgo
19:54:53 <Sequell> Mapping wltgo => (t0.20) demise ytcomultraviolent4 alcopop manman pekkekk gammafunk
20:17:31 -!- amalloy is now known as amalloy_
20:21:45 <johnny0> alexjurkiewicz: the foreground_colour and background_colour options does reverse mapping to the *default* default colours (lightgrey on black)
20:22:00 <johnny0> in contrast to the options.colour[] array which does one-way mapping
20:25:33 <johnny0> gammafunk: pretty much
20:25:44 -!- ProzacElf_ is now known as ProzacElf
21:07:10 -!- amalloy_ is now known as amalloy
21:43:19 <Cheibriados> Electric Eels able to move from water tile to fountain tile 13https://crawl.develz.org/mantis/view.php?id=11091 by Zigma
21:43:48 <minmay> now that's a blast from the past
21:56:42 <personjerry> hello, i can't play on CAO because it keeps saying another account is already using this save, could someone reset me?
21:56:59 <gammafunk> that will happen automatically if you wait 7-10 mins
21:57:14 <personjerry> 7-10 minutes without DCSS???
21:57:16 <personjerry> what is life
21:57:31 <personjerry> 😫
21:57:32 <gammafunk> Outside, the world ends...
22:14:28 <ProzacElf> does infusion work with riposte?
22:17:55 <gammafunk> =nemelex HOSu
22:17:58 <Sequell> Defined keyword: nchoice => DEAM|DDAK|HOCj|BaEn|FoEn|MiVM|HaIE|VSEE|FeNe|GhWz|CeFE|SpTm|MuAM|TeWr|VpSk|KoAE|GrSu|OgEE|BaHu|MfCj|VpFE|DEAs|HOSu
22:24:08 <gammafunk> Lasty: Lasty_: You around? that paralysis thread has some pretty bad stuff by the OP. I'm actually not sure what's the best way to handle that one
22:25:12 <gammafunk> It's not really in the format of a GDD proposal, it's really a rant, so it could be yuifed, but at least a warning should be involved for some of those later comments
22:25:52 <Lasty> gammafunk: hey there
22:25:53 <gammafunk> MarvinPA: also pinging you in case you have an opinion on that one
22:26:15 <Lasty> I read that on my phone earlier, and didn't remember to go back and address it
22:26:32 <Lasty> yeah, I think you're right on all counts. 1) it's just a complaint. 2) It's needlessly insulting.
22:27:16 <gammafunk> What would you recommend? Unless you'd like to do the mod yourself
22:30:03 <Lasty> I think the ideal course of action would be 1) Yiuf it; 2) warn the poster about the problematic language; 3) optionally delete the problematic language.
22:30:14 <Lasty> IIRC the poster has a history of good posting, so I wouldn't come down hard here
22:37:51 <gammafunk> thx
22:41:55 <Doesnty> * Giant eyeballs are now normal speed, but must stare thrice to paralyze.
22:42:16 <Doesnty> i love the inconsistent english here
22:59:44 <alexjurkiewicz> it's interesting there's currently a thread on tavern saying the new version is too *hard*
22:59:54 <alexjurkiewicz> how long has it been since a version was considered a nerf not a buff
23:00:36 <Gramm> but i only get like 20 teleports per game now
23:02:49 <gammafunk> alexjurkiewicz: it happens nearly every version, according to many players
23:02:52 <MarvinPA> alexjurkiewicz: that's not surprising at all
23:02:55 <gammafunk> game getting to hard, too elitist
23:03:56 <gammafunk> too removed from the realities of *fun* video games that are *good* and *balanced*
23:04:11 <alexjurkiewicz> !tstats
23:04:14 <gammafunk> in conclusion, bring back Mountain Dwarves
23:04:19 <alexjurkiewicz> i ruined the tstats sadly
23:04:20 <gammafunk> winrate is up this version
23:04:24 <gammafunk> you did?
23:04:25 <alexjurkiewicz> my one game has been MiMo challenge
23:04:27 <Sequell> Stats after 16 days (t): 2539 players, 897 runers, 434 winners, 1153 wins, 50730 games, winrate 2.27%, total player time 3y+25d+4:57:37.
23:04:42 <gammafunk> !tstats 10
23:04:45 <gammafunk> !tstats 10 t0.19
23:04:46 <alexjurkiewicz> that sees like a high winrate!
23:04:47 <gammafunk> !tstats 10 t0.18
23:04:58 <Sequell> Stats after 10 days (t): 2505 players, 887 runers, 431 winners, 1124 wins, 49243 games, winrate 2.28%, total player time 2y+358d+23:35:26.
23:05:09 <gammafunk> yeah it's up compared to previous, which is always the case due to the experience of the player-base always growing, but it's up a bit higher this time
23:05:32 <gammafunk> not meleebug levels, at least, like 0.5% I think
23:06:27 <alexjurkiewicz> !tstats 10 t0.19
23:06:28 <Sequell> 90s limit exceeded: killed !tstats 10 t0.18
23:07:20 <MarvinPA> also re: that tavern paralysis poster, i'd noticed them being pretty low-quality elsewhere, would definitely give them a proper warning for the reported stuff (i just split that out now)
23:07:33 <Sequell> Stats after 10 days (t0.19): 2611 players, 825 runers, 408 winners, 941 wins, 51636 games, winrate 1.82%, total player time 2y+296d+2:43:36.
23:07:34 <gammafunk> thanks, if you want to do that MPA, please feel free
23:07:41 <MarvinPA> sure, can do
23:08:37 <Sequell> Stats after 10 days (t0.19): 2611 players, 825 runers, 408 winners, 941 wins, 51636 games, winrate 1.82%, total player time 2y+296d+2:43:36.
23:13:37 -!- amalloy is now known as amalloy_
23:26:29 <elliptic> gammafunk: it isn't meleebug levels but it is still worrying, I was hoping that the winrate would come down more after a couple days
23:27:04 <elliptic> looking at the changelog I don't see anything that seems like it would plausibly make things a lot easier
23:27:23 <elliptic> and a few things that seem like they should make things harder by a non-negligible amount (good wand removal, rMsl spell removal)
23:27:48 <advil> it seems somewhat high across the board, I haven't seen any obvious factors yet
23:28:26 <elliptic> so that large a change in winrate either means I'm missing something in changelog, we have a meleebug-esque bug (though not as bad), or the playerbase/player habits have changed drastically
23:29:44 <nikheizen> it's pretty obviously because of scarves
23:31:43 <alexjurkiewicz> well, it's a smaller bug than meleebug
23:32:15 <elliptic> yes
23:32:17 <alexjurkiewicz> it could just be random noise
23:32:46 <alexjurkiewicz> 1153 wins vs 941 wins, 434 winners vs 431 winners
23:33:30 <advil> maybe it's the noise meter :D
23:33:36 <alexjurkiewicz> winners are the 98.5%!
23:33:41 <alexjurkiewicz> i mean...
23:39:00 <gammafunk> elliptic: yeah, it certainly seemed worrying, at least. I wonder if it's something as simple as more heal-wounds/haste earlier on
23:39:37 <gammafunk> oh, possibly the mut pot changes making good muts more common, I guess, but that seems like a bit of a stretch
23:40:02 <elliptic> gammafunk: yeah I thought of those things, it's still a large shift though
23:40:18 <elliptic> !lg * tall !boring !@bot s=cv / won o=%
23:40:40 <gammafunk> I suppose it could always be some bug where melee is artificially more effective due to a bug (or AC/EV more effective)
23:41:09 <elliptic> yeah, or just some large class of monsters doing less damage
23:41:13 <Sequell> 12186/747632 games for * (tall !boring !@bot): 56/1128x 0.16-a [4.96%], 1741/57675x 0.16 [3.02%], 22/749x 0.19-a [2.94%], 1104/48550x 0.20 [2.27%], 52/2290x 0.20-a [2.27%], 1358/68718x 0.18 [1.98%], 1339/69901x 0.17 [1.92%], 1401/78642x 0.19 [1.78%], 987/55490x 0.15 [1.78%], 61/4122x 0.17-a [1.48%], 603/42224x 0.13 [1.43%], 637/45377x 0.12 [1.40%], 741/55197x 0.14 [1.34%], 169/13842x 0.4 [1.22%], ...
23:41:26 <elliptic> ??meleebug2
23:41:26 <Sequell> meleebug2 ~ meleebug[1/2]: The git commit e0bdd66 roughly doubled player melee damage. This was not fixed until around halfway through the 0.16 tournament. http://s-z.org/neil/images/logicbug.jpg
23:41:31 <elliptic> ??meleebug[2]
23:41:31 <Sequell> meleebug[2/2]: The git commit 2733cf4 made all monsters wielding weapons take 5 aut longer to attack for about 2 weeks (in 0.18-a).
23:41:55 <advil> I guess we should put together some fsim tests?  I suppose you all probably discussed things like this after meleebug
23:42:21 <gammafunk> you mean like systematically test player vs monsters in fsim?
23:42:37 <advil> yeah
23:42:50 <gammafunk> there are a lot of challenges in doing that, probably, at least in terms of detecting a meaningful change
23:43:45 <advil> true, but it seems doable
23:43:51 <gammafunk> !lg !bot !@goodplayers !boring t / won
23:44:05 <gammafunk> ??tournament[2
23:44:05 <Sequell> tournament[2/7]: The 0.19 tournament ran from 20:00 UTC Friday 4 Nov 2016 to 20:00 UTC Sunday 20 Nov 2016. Tournament home page: http://dobrazupa.org/tournament/0.19/
23:44:09 <elliptic> !lg * tall !boring !@bot s=${regexp_replace(cv, ".a", "")} / won o=%
23:44:11 <advil> decide on some baseline monsters, and check if the avg damage under various configurations is within certain thresholds
23:44:40 -!- rumflump is now known as rumflump\OotC
23:44:51 <Sequell> 305/39805 games for bot (!@goodplayers !boring t): N=305/39805 (0.77%)
23:45:33 <advil> need to write a lua interface to fsim
23:45:34 <gammafunk> !lg !bot !@goodplayers !boring t0.19 start<'2016-11-15 00:00:00' / won
23:45:50 <Sequell> 12186/747642 games for * (tall !boring !@bot): 1797/58803x 0.16 [3.06%], 1156/50854x 0.20 [2.27%], 1358/68718x 0.18 [1.98%], 1400/74023x 0.17 [1.89%], 1423/79391x 0.19 [1.79%], 987/55490x 0.15 [1.78%], 603/42224x 0.13 [1.43%], 637/45377x 0.12 [1.40%], 741/55197x 0.14 [1.34%], 169/13842x 0.4 [1.22%], 473/40176x 0.11 [1.18%], 370/33683x 0.10 [1.10%], 286/28399x 0.5 [1.01%], 232/27043x 0.9 [0.86%], 2...
23:45:58 <gammafunk> I'm wondering a bit if it could also be a function of who's playing in this t
23:46:10 <gammafunk> there are certainly a lot of top players butting in serious hours this time
23:46:15 <gammafunk> *putting in
23:46:28 <elliptic> gammafunk: ideally for comparing with !@goodplayers we would use the goodplayers nick as of the time of last tourney for that one
23:46:34 <gammafunk> yes
23:46:46 <gammafunk> which can probably be done since the list of t players isn't too big
23:46:49 <Sequell> 213/39029 games for bot (!@goodplayers !boring t0.19 start<'2016-11-15 00:00:00'): N=213/39029 (0.55%)
23:47:00 <johnstein> at least it's not meleebug level
23:47:17 <johnstein> I need a better way to pull the data
23:47:19 <gammafunk> yeah it's clear that the disparity is only going down, so we're not talking meleebug levels
23:47:54 <johnstein> I just PM sequell late at night with a TON of !tstat calls for every 0.1 days
23:48:14 -!- amalloy_ is now known as amalloy
23:49:20 <elliptic> the last time we had a winrate change this large between versions was 0.14 -> 0.15
23:49:34 <elliptic> I forget what caused that one but I think we had some explanation
23:50:18 <elliptic> oh right, we removed inventory weight and item destruction
23:50:20 <advil> weight/item dest removal
23:50:22 <advil> yeah
23:51:09 <gammafunk> so it was due to having more consumables in inventory on-average
23:51:21 <advil> well, or just not hating the game quite as much
23:51:27 <elliptic> oh, also shields were buffed
23:52:05 <elliptic> (shield enchantment was effectively doubled)
23:59:31 <Eksell> Unstable branch on CRAWL.XTAHUA.COM updated to: 0.21-a0-43-g1479571 (34)