00:57:49 <ProzacElf> noooo
00:57:54 <ProzacElf> not from the apt table!
00:59:50 -!- amalloy_ is now known as amalloy
01:00:09 <nikheizen> sneakyness: yes, this is how i design all my chipsets
01:00:39 <nikheizen> telekinesis is integral to the QA process
01:41:55 <minmay> swiftness' description should mention that it doesn't work in water
01:43:24 <ProzacElf> i never knew that
01:58:00 <Gretell> Windows builds of master branch on crawl.develz.org updated to: 0.21-a0-81-gc5150db
02:00:34 <sneakyness> nikheizen: can u tell me which one of these processors got a better bin
02:00:38 <sneakyness> left or right
02:06:58 <nikheizen> right
02:27:58 <Cheibriados> New branch created: pull/554 (2 commits) 13https://github.com/crawl/crawl/pull/554
02:27:58 <Cheibriados> 03nikheizen02 07https://github.com/crawl/crawl/pull/554 * 0.21-a0-80-ga74b234: Four simple arrival vaults 10(23 hours ago, 2 files, 100+ 0-) 13https://github.com/crawl/crawl/commit/a74b2349e527
02:27:58 <Cheibriados> 03nikheizen02 07https://github.com/crawl/crawl/pull/554 * 0.21-a0-81-gfde10a0: A new float vault: nzn_motion_control 10(12 minutes ago, 1 file, 41+ 0-) 13https://github.com/crawl/crawl/commit/fde10a046a96
02:46:58 <nikheizen> i would much appreciate some feedback on my first few vaults if anyone wants to take a look
02:53:06 <Gretell> Monster database of master branch on crawl.develz.org updated to: 0.21-a0-81-gc5150db
03:12:26 <Rotatell> Unstable branch on crawl.beRotato.org updated to: 0.21-a0-81-gc5150db (34)
03:14:39 <Cheibriados> New branch created: pull/555 (3 commits) 13https://github.com/crawl/crawl/pull/555
03:14:39 <Cheibriados> 03kenranunderscore02 07https://github.com/crawl/crawl/pull/555 * 0.21-a0-82-g52b2803: Add a lighthouse-themed bazaar vault 10(9 hours ago, 1 file, 57+ 0-) 13https://github.com/crawl/crawl/commit/52b280398dc8
03:14:39 <Cheibriados> 03kenranunderscore02 07https://github.com/crawl/crawl/pull/555 * 0.21-a0-83-g79f7203: Fix floor colour in a bazaar vault 10(9 hours ago, 1 file, 1+ 1-) 13https://github.com/crawl/crawl/commit/79f720337914
03:14:39 <Cheibriados> 03kenranunderscore02 07https://github.com/crawl/crawl/pull/555 * 0.21-a0-84-g75fd155: Fixed a tile colour issue in a vault 10(in the future, 1 file, 1+ 1-) 13https://github.com/crawl/crawl/commit/75fd1552b883
03:15:12 <kenran> I forgot to squash the commits before pushing ^
03:15:36 <kenran> one should actually suffice, the other two were just small fixes
03:16:05 <gammafunk> no big deal, they can be squashed before merge
03:16:44 <kenran> I guess I'd have to revise it anyway if for no other reason than the monster I put in there ;)
03:16:55 <kenran> it was too tempting too use the halo
03:17:53 <nikheizen> these are some very aesthetic bazaars
03:19:53 <kenran> nikheizen: I'm going to check out yours soon as well!
03:20:13 <kenran> nikheizen: for some reason, the travis build seems to have failed for your pull request oO
03:20:23 <kenran> kind of strange, considering it's just .des files
03:20:38 <nikheizen> apparently there is a problem with the orb guardian monster
03:20:45 <gammafunk> des file changes definitely can cause build failures
03:20:50 <nikheizen> (which i didnt use?)
03:20:51 <gammafunk> we run make test for those
03:21:14 <gammafunk> which runs lua tests, and if your des file has e.g. syntax error or other kinds of errors
03:21:18 <gammafunk> crawl will abort
03:21:36 <kenran> oh cool
03:28:29 <nikheizen> that's odd, since i could generate it
03:31:27 <kenran> nikheizen: which is the problematic one, the motion_control?
03:32:34 <nikheizen> yeah
04:04:29 <nikheizen> well i definitely realized now i should have tagged it no_monster_gen, still not sure why builds are failing though.
04:15:58 <kenran> nikheizen: I tried running some of the make commands that travis ran, and I didn't get an error. didn't try a webtiles build yet though (I heard it's a hassle)
04:24:12 -!- amalloy is now known as amalloy_
04:31:05 <nikheizen> thanks for taking a look. i figure i most likely made a syntax error
04:36:25 <kenran> nikheizen: it seems to say the error is in arrival/small.des though, hum
04:36:48 <nikheizen> oh! well then i guess it's _arrival_settlement
04:37:11 <nikheizen> probably a wrong FTILE
04:37:22 <kenran> mind you, that's the first time I ever looked into a travis log, so I could be wrong about stuff :D
04:37:41 <kenran> those FTILEs look fine to me though, I've used both in my forest
04:40:30 <kenran> nikheizen: I'd try commenting out the lua :if and try those two SUBST lines on their own. I found that statements like this where the :if clause cannot be resolved at the time the .des is parsed don't lead to syntax errors when starting crawl
04:41:00 <kenran> I had a typo in a KMONS monster description and wouldn't get an error when starting just because it was inside an :if crawl.coinflip()
04:46:43 <nikheizen> seems to work just fine without (also thanks for telling me about coinflip
04:46:46 <nikheizen> )
04:47:09 <kenran> /home/travis/build/crawl/crawl/crawl-ref/source/dat/des/arrival/small.des:1: syntax error, unexpected MAP_LINE, expecting $end
04:47:31 <kenran> that's what seems to be the error, though I don't understand it at all. why does it say line 1?
04:50:46 <nikheizen> oh, i'm daft
04:52:34 <nikheizen> i forgot to put an endmap at the end of my last map in simple.des
04:53:43 <kenran> oh wow, right. the error message is pretty misleading then in a sense (even though now the message itself makes sense)
04:54:15 <kenran> that explains why I didn't get any errors locally: I had only copy/pasted the small.des and float.des vaults
04:55:43 <nikheizen> i guess that's because the next line loaded after line $ of simple.des is line 1 of small.des
04:55:48 <kenran> yeah
05:02:32 <nikheizen> There, should work.
05:02:40 <nikheizen> kenran: thanks for helping out.
05:03:35 <Cheibriados> 03nikheizen02 07https://github.com/crawl/crawl/pull/554 * 0.21-a0-80-g4388d81: Four simple arrival vaults 10(5 minutes ago, 2 files, 101+ 0-) 13https://github.com/crawl/crawl/commit/4388d815bffb
05:03:35 <Cheibriados> 03nikheizen02 07https://github.com/crawl/crawl/pull/554 * 0.21-a0-81-gf36eb4a: A new float vault: nzn_motion_control 10(5 minutes ago, 1 file, 41+ 0-) 13https://github.com/crawl/crawl/commit/f36eb4abee51
05:04:43 <kenran> np, you found it yourself after all :D
06:52:44 -!- svendre_ is now known as svendre
08:59:49 <advil> hmm, reddit discovered the Og M&F change
09:01:22 <nikheizen> i bet they're happy about that
09:07:02 <kenran> why did I even start reading the thread...
09:08:36 <advil> it's definitely one of those threads where anything a dev can post will just make things worse
09:14:42 <nikheizen> hmmm, how much XP is it to 26 skill at +4 compared to 18 skill at -1
09:14:53 <nikheizen> is there a calc for this
09:16:24 <|amethyst> at +0 level 18 requires 10800 skill points, and level 26 requires 27000
09:16:29 <|amethyst> !source skill_exp_needed
09:16:31 <Sequell> 1/1. https://github.com/crawl/crawl/blob/master/crawl-ref/source/skills.cc#L1523
09:17:12 <kenran> don't you need only 24 or something for GSC now? I didn't check
09:17:14 <kenran> ??gsc
09:17:15 <Sequell> giant spiked club[1/4]: (maces & flails; -7 acc / 22 dam / 1.9 base delay (1.8 in .20) / 0.7 min delay; two handed). A giant lump of wood with sharp spikes at one end. The only playable races that can wield giant spiked clubs are Ogres and Trolls.
09:17:18 <|amethyst> +4 means you need half as much, and -1 means you need 2^(1/4) as much
09:17:51 <advil> sometimes I wonder if we'd be better off if we just renamed -1 apt to 0 (etc)
09:17:53 <|amethyst> 2^(1/4) * 10800 is about 12843
09:18:17 <|amethyst> 1/2 * 27000 is 13500
09:19:16 <|amethyst> so it does take less XP for an ogre to get mindelay with a giant unspiked club than it takes for a merfolk to get mindelay with a bardiche
09:19:21 <nikheizen> so 12843 XP for a -6/20 dirt-common weapon or 13500 for a -6/18+Reaching uncommon weapon
09:19:59 <nikheizen> ogres confirmed better at clubbing things than mf are at poking stuff
09:20:04 <|amethyst> my opinion on the matter:
09:20:24 <|amethyst> most players will not know how to do that calculation, or even that they should
09:20:39 <advil> yeah, I wish we had some better way of showing stuff like this in-game
09:20:51 <|amethyst> so the change made "ogres are still good with M+F" into a spoilery thing
09:21:03 <kenran> most people think "with +4 apt, I can get any skill to 27 easily" and "at -1 it takes me ages to get any weapon to mindelay"
09:21:08 <advil> I have some ideas for showing training info for 0.21 but I'm not sure how they will work out
09:21:22 <|amethyst> a new player looks at a -1 in MF and a +0 in polearms and what are they going to think?
09:21:52 <nikheizen> that's a fair assessment, |amethyst
09:22:57 <|amethyst> I think the better change would have been to nerf GSC and GC
09:23:00 <Palyth> You could just get rid -4 through +4 and normalize them to 0 to something. Maybe should a scale factor instead of just apt rating
09:23:15 <kenran> is British English spelling preferred in the crawl source (I'm talking colour/color, randomise/randomize)?
09:23:22 <|amethyst> kenran:  australian :)
09:23:30 <kenran> oh :D
09:23:49 <|amethyst> since the -1 apt was instated so that other weapons would be competitive with those two
09:24:13 <|amethyst> kenran:  if you go with British spelling that's usually correct
09:25:10 <kenran> phew :D
09:25:19 <advil> in the actual source I think it's a bit varied
09:25:34 <kenran> as can be expected I guess
09:26:05 <kenran> still better than working with other Germans who are forced to write English source code and are kind of bad at it
09:26:20 <kenran> (not saying I'm very good, but I at least check a dictionary at times)
09:26:25 <advil> advil$ cat *.cc | grep -c color
09:26:25 <advil> 219
09:26:25 <advil> advil$ cat *.cc | grep -c colour
09:26:27 <advil> 2589
09:27:02 <kenran> is grey/gray a difference between dialects as well, or is it just ambiguous in general?
09:27:22 <kenran> because I had to use wall_brick_lightgray today, and set its colour to lightgrey ;)
09:34:06 <|amethyst> kenran:  I'd go with "grey", which is more common outside American English
09:34:09 <advil> I think "gray" is US vs "grey" in the commonwealth countries, but it seems to be less absolute than most of these spelling differences
09:34:46 <|amethyst> the tile seems to have had the same _gray name since it was added (though it was .bmp at first rather than .png)
09:35:08 <kenran> good to know, thanks
09:36:02 <kenran> another question: is there a reason the lua code I see in the .des file has 2 spaces of indentation instead of 4?
09:36:19 <|amethyst> I've wondered that myself
09:36:23 <kenran> is it a general guideline for lua, or something that is maybe needed in .des specifically (I didn't check other lua yet)
09:36:59 <|amethyst> we do seem to use two-space indent for Lua in general
09:37:23 <|amethyst> but I don't know why that is different from our C++
09:37:35 <advil> I'm perfectly happy to use 4 space indentation in lua
09:37:47 <kenran> if it's consistend so far, it would probably be a good idea to mention this special case in https://github.com/crawl/crawl/blob/master/crawl-ref/docs/develop/coding_conventions.txt
09:37:53 <kenran> consistent*
09:38:04 <advil> it's not fully consistent
09:38:37 <advil> I think it might be a lua thing to use 2 spaces? http://lua-users.org/wiki/LuaStyleGuide
09:39:50 <kenran> ah yes, that makes sense then
09:40:25 <advil> I don't like it
09:41:06 <kenran> I don't either, it just makes sense :) It's also a tad bit clunky to use when having 4 spaces set as tab width
09:46:14 <kenran> !source bazaar.des
09:46:15 <Sequell> https://github.com/crawl/crawl/blob/master/crawl-ref/source/dat/des/portals/bazaar.des
09:47:24 <kenran> the method set_bazaar_tiles sets the floor colour to some colour, then a bit further down if gets the floor colour to base the tileset on. is there a case where get_floor_colour() wouldn't be the one that was just set above?
09:49:30 <|amethyst> there is not
09:50:01 <|amethyst> it would be one extra line to write it the other way :)
09:50:48 <kenran> but wouldn't it be more easily readable (I know it's a simple case, but other people like me probably wouldn't wonder about whether such an edge case as above existed at least)
09:52:02 <|amethyst> likely
09:52:36 <|amethyst> though I'd argue it would be better to make the colour part of the object inside tileset
09:52:42 <|amethyst> then to do a random selection from that
09:53:08 <|amethyst> rather than having to make sure the colour lists in tileset and in random_bazaar_colour are the same
09:53:18 <kenran> oh that's a good point!
09:53:19 <|amethyst> s/to make/to put/
09:53:52 <kenran> I'm trying something there right now anyway, so I can try to do it (no lua experience, but it shouldn't be too hard)
09:55:28 <|amethyst> I'd say  function random_bazaar_tiles() and have that return one of those five objects (with colour="..." added); then   local tile = random_bazaar_tiles(); dgn.change_floor_colour(tile.colour);
09:56:36 <kenran> yes, I wanted to do it like this, though not as elegant. this is great
09:56:56 <kenran> I'm beginning to like lua
09:57:25 <|amethyst> you'll recover :)
09:57:31 <kenran> lol
09:57:53 <kenran> what version of lua does crawl use btw? wasn't it 5.1x?
09:59:43 <|amethyst> 5.1, yes
10:00:12 <|amethyst> or luajit, the version of which we're not as picky about
10:00:46 <|amethyst> but luajit (or at least the version in our contribs/) has some minor problems on 64-bit systems
10:01:01 <kenran> |amethyst: I can get rid of 'default' then, can't I? it wasn't picked before anyway. Or should I incorporate it into the tileset list and make it a viable option as well?
10:01:23 <|amethyst> nothing that keeps it from working, but the custom allocator that does memory throttling doesn't work
10:01:31 <|amethyst> kenran:  I would get rid of it
10:01:50 <kenran> I had to google luaJIT
10:02:00 <|amethyst> kenran:  that was a safety fallback in case someone added a new colour without updating tilesets
10:02:10 <kenran> yeah, it is unnecessary now
10:03:57 <|amethyst> kenran:  also, the refactoring you are doing there should be a separate commit from your vault edits
10:04:14 <|amethyst> kenran:  and from any other actual changes
10:04:39 <kenran> I'm planning another pull request
10:04:44 <|amethyst> cool
10:05:19 <kenran> would it be okay to have one commit with this refactoring, and one that changes things up a tiny bit to restore some of the older vaults to what they were supposed to look like?
10:05:32 <kenran> with the theme of the PR being the latter
10:05:47 <kenran> or should I separate this into 2 pr's?
10:06:09 <|amethyst> I think two commits in one PR are fine here
10:06:40 <|amethyst> but if they are independent (don't touch the same lines and can be applied in either order), there's not a problem with making two PRs
10:07:02 <kenran> then they're most likely not independent
10:07:33 <kenran> rather the refactoring is a useful step before doing my actual changes, or I'd clutter the code more
10:07:56 <|amethyst> yeah, then definitely the same PR
10:07:58 <kenran> I really appreciate all the advice in here btw
10:08:45 <|amethyst> we appreciate the vaults and the cleanup :)
10:10:00 <kenran> he :D have you seen the lighthouse vault? I wonder what you guys think about the ophan being in there. I'm not really relying on it, and I think all other monsters have been removed from bazaars over time
10:11:12 <|amethyst> this is weird
10:11:22 <|amethyst> when I click a link in github,
10:12:09 <|amethyst> I get a blue progress bar across the top of the screen that goes pretty slowly (takes about 5-10 seconds) and never completely fills, and the page doesn't update
10:12:20 <|amethyst> when I "open in new tab" the link opens right away
10:12:29 <kenran> that sounds very strange
10:12:30 <|amethyst> I suspect I need to upgrade my browser a lot
10:12:39 <kenran> what browser are you using?
10:13:11 <|amethyst> chromium
10:13:21 <|amethyst> with a number much much smaller than 59 :)
10:13:29 <|amethyst> 39 specifically
10:13:39 <|amethyst> haven't done updates in some time here
10:13:56 <kenran> the longer you wait, the more hazardous it becomes :D
10:14:23 <|amethyst> particularly since I run Debian unstable
10:14:33 <kenran> ohhh
10:14:49 <kenran> I've killed my debian unstable a couple of weeks ago with something like that
10:15:04 <kenran> now I'm using arch. not sure yet if it won't come to the same end
10:15:28 <|amethyst> running unstable was a good idea several years ago when I wanted to learn how Debian worked from the inside out
10:15:55 <kenran> wait, I'm talking crap. I wasn't using unstable, I was using testing, which should be a whole lot more stable
10:16:00 <|amethyst> but now that I don't have time for that and/or have lost the interest...
10:18:04 <|amethyst> I think this install dates to 2003 or so
10:18:53 <|amethyst> of course I've done plenty of updates since then, but unstable does tend to accumulate cruft after that long :)
10:20:00 <kenran> and you're still on the same system?
10:22:25 <|amethyst> I've upgraded hardware several times, but not reinstalled the OS
10:23:26 <|amethyst> in fact, it's a completely different computer from 2003, nothing in common, but I just duplicated the file system to the new disk
10:26:43 <|amethyst> oh, looks like it was 2002, not 2003
10:26:59 <|amethyst> I've been using GNU+Linux on my home computer since 1996 or so, but I've only installed three times (slackware in 96, red hat in 97, and Debian in 2002
10:27:17 <|amethyst> I've done plenty of installs on work machines, VMs, etc, of course
10:27:20 <kenran> then debian surely seems to be the right os for you :D
10:29:54 <|amethyst> maybe not unstable though :)
10:30:04 <|amethyst> anyway, I must be going, happy hacking!
10:30:22 <kenran> thanks, I'm off to the playground as well
13:08:56 -!- amalloy_ is now known as amalloy
13:10:15 <amalloy> kenran: re the lighthouse bazaar, it looks very pretty and evokes the feeling of a lighthouse, but leaves me thinking like...why are there shops in a lighthouse?
13:12:27 <johnstein> so no reports of cao issues? scoring looks ok?
13:16:17 <Cheibriados> New branch created: pull/556 (1 commit) 13https://github.com/crawl/crawl/pull/556
13:16:17 <Cheibriados> 03cut1less02 07https://github.com/crawl/crawl/pull/556 * 0.21-a0-82-g3deb030: Let shallow water be like floors for water species 10(22 minutes ago, 1 file, 5+ 1-) 13https://github.com/crawl/crawl/commit/3deb030efb34
13:27:26 <gammafunk> johnstein: check the overview; scoring is frozen I think
13:28:30 <gammafunk> last win is early on the 22nd
14:02:25 <johnstein> gammafunk, I have basic instructions from |amethyst and I'm wondering if I need to tread cautiously here.
14:02:37 <johnstein> step 2) tail -f /home/rax/scoring.log, wait for the "Finished batch" line, and when you get it kill the 'python scoresd' process
14:03:01 <johnstein> 2017-06-21 18:11:29,618 [INFO] Done processing 154 lines.
14:03:01 <johnstein> 2017-06-21 18:11:29,619 [INFO] Rendering best-players-total-score
14:03:01 <johnstein> 2017-06-21 18:12:52,290 [INFO] Rendering overview
14:03:01 <johnstein> 2017-06-21 18:13:00,772 [INFO] Rendering all-players
14:03:01 <johnstein> so it died in the middle of rendering
14:03:15 <johnstein> step 3 is to restart the python script
14:04:23 <johnstein> anyone familiar with this particular process? if not, I'll work with |amethyst before I try anything. sounds like you don't want to re-launch the script unless the previous batch job has been completed. so maybe I need to go in and clean something up first?
14:18:32 <Cheibriados> New branch created: pull/557 (1 commit) 13https://github.com/crawl/crawl/pull/557
14:18:32 <Cheibriados> 03Umer Shaikh02 07https://github.com/crawl/crawl/pull/557 * 0.21-a0-82-g0a4ebe0: Remove manual's reference to Ogres' especially good maces aptitude 10(9 minutes ago, 1 file, 4+ 4-) 13https://github.com/crawl/crawl/commit/0a4ebe0c6d0a
14:29:12 <johnstein> got a question on the new 2 headed ogres
14:29:13 <johnstein> !lg * ((cv=0.20||cv=0.19||cv=0.18||cv=0.17)) ((killer=two-headed_ogre||killer=hill_giant)) place=D:4 s=killer,cv -graph
14:29:14 <Sequell> 253 games for * (((cv=0.20 || cv=0.19 || cv=0.18 || cv=0.17)) ((killer=two-headed_ogre || killer=hill_giant)) place=D:4): https://shalott.org/graphs/dd445175700474432782de585a42845e457566b1.html
14:29:30 <johnstein> what's the reason they are popping up so much earlier?
14:29:36 <johnstein> !lg * ((cv=0.20||cv=0.19||cv=0.18||cv=0.17)) ((killer=two-headed_ogre||killer=hill_giant)) place=D:5 s=killer,cv -graph
14:29:37 <Sequell> 465 games for * (((cv=0.20 || cv=0.19 || cv=0.18 || cv=0.17)) ((killer=two-headed_ogre || killer=hill_giant)) place=D:5): https://shalott.org/graphs/cc9385c3c32337d9add2cf5064656670a7739367.html
14:29:39 <johnstein> !lg * ((cv=0.20||cv=0.19||cv=0.18||cv=0.17)) ((killer=two-headed_ogre||killer=hill_giant)) place=D:6 s=killer,cv -graph
14:29:40 <Sequell> 632 games for * (((cv=0.20 || cv=0.19 || cv=0.18 || cv=0.17)) ((killer=two-headed_ogre || killer=hill_giant)) place=D:6): https://shalott.org/graphs/8f26d8d3c573ac0a8f5b12d864cc599c690c023a.html
14:30:08 <johnstein> just an intentional part of the overall rebalance after hill giants were removed?
14:35:10 <amalloy> johnstein: yes, 2HO were moved in to replace hill giants
14:35:49 <johnstein> amalloy, yea, I knew that. but looks like they are killing a ton more players on D:3/4/5 than hill giants ever did. was just wondering if that was an intentional part of the change
14:38:06 <amalloy> oh, i hadn't looked at the graphs. i'd say on D:5 it's not noticeable, and D:4 doesn't seem like that huge a change either, but D:3 does seem like a big jump
14:38:12 <johnstein> by D:6 it looks pretty similar
14:38:33 <amalloy> er, subtract 1 from all those numbers
14:38:38 <johnstein> (and I'm not saying it's bad or good, just was wondering if that was an intentional change)
14:38:46 <amalloy> or add 1, or something
14:38:57 <johnstein> yea. add 1.
14:39:19 <minmay> <johnstein> what's the reason they are popping up so much earlier?
14:39:19 <minmay> because their depth was reduced
14:39:22 <johnstein> to 8 I thinl
14:39:24 <johnstein> think, right?
14:39:29 <johnstein> so is it just OOD?
14:39:30 <amalloy> !git 5be650752770eb0226b8f70edfbf9eaf8a4da687
14:39:30 <Sequell> %git 5be650752770eb0226b8f70edfbf9eaf8a4da687
14:39:30 <Cheibriados> 07Brannock02 * 0.20-a0-283-g5be6507: Generate 2HO more frequently 10(6 months ago, 1 file, 1+ 1-) 13https://github.com/crawl/crawl/commit/5be650752770
14:39:33 <minmay> yes
14:39:50 <amalloy> change from SEMI to FLAT is interesting
14:40:09 <johnstein> so it was 2 things. 1, make them take over for hill giants, which is clear from the D6 chart. and 2) make them more lethal than hill_giants ever were for some floors above?
14:40:26 <minmay> I don't think they are more lethal than hill giants unless you are a naga or barachian
14:40:43 <minmay> D:3 hill giant was rarer than D:3 2hogre is now, but a 2hogre cannot one-shot you with a throwing net
14:41:27 <johnstein> sounds like players aren't learning that lesson
14:41:44 <johnstein> though I suppose the winrate increase says otherwise :)
14:41:46 <amalloy> well, they never do. 2HO is more dangerous if you hold down tab
14:42:47 <johnstein> I suppose D3-D5 2HO means skip that floor? is that what the good players do?
14:43:03 <kogasa> i wouldn't
14:43:03 <amalloy> well, you don't have to skip the whole floor, just try to route around them
14:43:07 <minmay> it means don't melee it
14:43:07 <kogasa> unless the floor was tiny
14:43:13 <minmay> it's speed 10 and cannot possibly have a ranged attack
14:43:25 <kogasa> it's basically d:2 terence except more instantly lethal
14:43:42 <kogasa> and terence can probably pick up wands i guess
14:43:59 <johnstein> sounds like I need to dial down my conservatism. trying speedruns is helping, but I almost tabbed into this one
14:44:28 <Cheibriados> New branch created: pull/558 (1 commit) 13https://github.com/crawl/crawl/pull/558
14:44:28 <Cheibriados> 03cut1less02 07https://github.com/crawl/crawl/pull/558 * 0.21-a0-82-g57cdc10: Remove Mottled Draconians from table of aptitudes 10(12 minutes ago, 1 file, 1+ 0-) 13https://github.com/crawl/crawl/commit/57cdc1070694
14:44:40 <kogasa> i only skip floors for things that will threaten to kill me if they spot me
14:44:53 <kogasa> sigmund, grinder, rupert, mara, etc.
14:45:25 <amalloy> it's been so long that i've forgotten: how do i look at my 0.20 &dump instead of trunk?
14:45:45 <amalloy> i seem to remember running into two 2HOs quite early in this game
14:45:49 <johnstein> makes sense. I always get nervous having a Big Bad dude on a floor with me since I'm worried they'll sneak up and ambush me when dealing with another group
14:46:42 <|amethyst> &dump amalloy * 0.20
14:46:43 <Sequell> unknown/amalloy/amalloy.txt
14:46:46 <|amethyst> hm
14:47:06 <|amethyst> &dump amalloy cao 0.20
14:47:07 <Sequell> http://crawl.akrasiac.org/rawdata/amalloy/amalloy.txt
14:47:13 <|amethyst> hm, no
14:47:31 <|amethyst> oh, doh
14:47:35 <amalloy> !cmd &dump
14:47:35 <Sequell> Command: &dump => !lm ${1:-.} fmt:"\$(=morgue_link ${2:-${src}} \$(if \$(match -a ${3:-${cv}}) trunk ${3:-${cv}}))/${name}/${name}.txt" stub:"No milestones for $(name_fixup ${1:-${user}})."
14:47:42 <johnstein> !lg * killer=two-headed_ogre
14:47:43 <Sequell> 19709. Vizrt the Spear-Bearer (L11 NaVM of Okawaru), slain by a two-headed ogre (a +0 giant club) on D:9 (minmay_the_grid_octo_tiled_octocenter; minmay_the_grid_square_5x5_lines) on 2017-06-23 18:42:34, with 8813 points after 11854 turns and 0:34:27.
14:47:56 <amalloy> !lm . x=cv
14:47:57 <Sequell> 7867. [2017-06-19 03:02:42] [cv=0.20] amalloy the Spear-Bearer (L17 DrCj of Hepliaklqana) left the Vaults on turn 44476. (Vaults:1)
14:47:58 <|amethyst> morgue directories are shared among versions
14:48:09 <|amethyst> so you have just one dump per server
14:48:13 <amalloy> oh, i see
14:48:15 <johnstein> !lg * killer=two-headed_ogre
14:48:17 <Sequell> 19709. Vizrt the Spear-Bearer (L11 NaVM of Okawaru), slain by a two-headed ogre (a +0 giant club) on D:9 (minmay_the_grid_octo_tiled_octocenter; minmay_the_grid_square_5x5_lines) on 2017-06-23 18:42:34, with 8813 points after 11854 turns and 0:34:27.
14:48:20 <amalloy> so i'd just log in and #
14:48:51 <|amethyst> yeah
14:49:14 <johnstein> |amethyst, not sure if you saw my comments on cao scoring above. I hadn't gotten around to a !tell
14:49:29 <amalloy> i guess i was wrong anyway. i didn't see any 2HOs until like D:9 or D:10
14:49:52 <|amethyst> johnstein:  I think it's fine to just start it back up
14:49:54 <johnstein> also, I wasn't able to get away from my D4 2HO. ran into it after walking on a tele trap and ran into another enemy and got ambushed, like I feared
14:50:37 <|amethyst> johnstein:  sounds like it wasn't in the middle of a database update, just regenerating pages
14:50:54 <johnstein> |amethyst, ok I'll do that now. step 1 says to run as root. that's for all steps? or just to edit?
14:50:57 <|amethyst> johnstein:  and all pages get regenerated after the fetch when you start up
14:51:05 <|amethyst> johnstein:  all steps
14:51:08 <johnstein> ty
14:51:11 <|amethyst> probably it shouldn't be running as root
14:51:27 <|amethyst> but I didn't feel like changing permissions etc to fix that
14:55:23 <|amethyst> hm, there was once an entry under |amethyst but I think it's gone now
14:55:52 <|amethyst> something like "I don't always make commits on production machines, but when I do, I do it as root"
14:56:14 <johnstein> "I don't always run kill -9 -1, but when I do, I do it as root"
14:58:58 <ProzacElf> lol
15:02:59 <|amethyst> !learn add johnstein "I don't always run kill -9 -1, but when I do, I do it as root"
15:03:00 <Sequell> johnstein[10/10]: "I don't always run kill -9 -1, but when I do, I do it as root"
15:03:13 <|amethyst> immortalised
15:03:49 <johnstein> I was very relieved that things seemed to go back to normal after r.ax rebooted
15:04:10 <|amethyst> I'm not surprised though
15:04:28 <johnstein> I was worried about some database thing getting corrupted
15:04:31 <|amethyst> what you did is cleaner than a power loss, since at least the kernel is still running and can flush buffers
15:04:39 <johnstein> since database stuff is still kinda magical to me still
15:04:45 <johnstein> oh. hmm
15:04:52 <|amethyst> it wouldn't have been corrupted, but it could have lost data
15:04:54 <geekosaur> logical consistency's still a potential problem though
15:04:59 <|amethyst> which should be fixed
15:05:19 <|amethyst> the problem is that we commit the file offsets first, then start processing the lines
15:05:30 <johnstein> how did the crawl games in progress not get messed up?
15:05:45 <johnstein> or does the -9 -1 still tell it to save the current status?
15:05:46 <|amethyst> some of them probably did get reset to the beginning of the level
15:05:49 <johnstein> ah ok
15:06:01 <|amethyst> no, with -9 there is no opportunity to do anything
15:06:16 <geekosaur> right, anything after the last save (which will be done on any level change inscluding those done by the game itself) was lost
15:06:18 <|amethyst> https://www.youtube.com/watch?v=IuGjtlsKo4s
15:06:21 <johnstein> ok. so there may be one player out there who made a miraculous save just prior to the outage who now had to go do the same thing again
15:06:29 <johnstein> though I suppose they'd have perfect knowledge
15:06:30 <geekosaur> right
15:06:32 <|amethyst> (warning: mild misogyny)
15:06:50 <geekosaur> also they might *need* perfect knowledge
15:07:09 <geekosaur> you could change the outcome by doing stuff int he wrong order so the rng rolls more/less
15:23:54 <kenran> amalloy: I even thought "why would there be shops on an island", but then I saw other bazaar vaults do it and thought "ok, that's fine". and as I said, I'm not too adamant on keeping the actual light source. I actually think the halo is nice but interferes a bit with the wall colouring.
15:25:34 <kenran> so yeah, it would be possible to switch the ophan (and probably the yellow permrock as well) to something else (another shop?) and make it be just an island bazaar
15:26:00 <Cheibriados> Trunk Graphics Issue: 13https://crawl.develz.org/mantis/view.php?id=11121 by hollowman
15:26:00 <kenran> I'd love to get some more opinions on the whole idea though if possible :)
16:49:17 <n1> There's an alignment issue in the table under Items in the in-game manual
16:49:52 <n1> ah I think I figured it out, those characters are escaped in crawl_manual.rst
17:10:01 -!- jefus- is now known as jefus
17:14:53 <Cheibriados> New branch created: pull/559 (1 commit) 13https://github.com/crawl/crawl/pull/559
17:14:53 <Cheibriados> 03Nikhil Addleman02 {GitHub} 07https://github.com/crawl/crawl/pull/559 * 0.21-a0-82-gcc51955: Fix alignment of Items table in crawl_manual.rst 10(23 minutes ago, 1 file, 2+ 2-) 13https://github.com/crawl/crawl/commit/cc51955d1d64
17:19:14 <n1> it's worth going over the manual more thoroughly but I'm not sure if some of the issues I see are features of this particular SSH client, or actual problems
17:20:07 <kenran> I just picked up two plate armours that were coloured blue. Both of them were neither cursed, nor enchanted, nor had they an ego. does this sound buggy to you?
17:23:17 <kenran> oh, they don't say 'shiny' or anything, they're just coloured blue
17:24:01 <amalloy> !lm kenran
17:24:08 <Sequell> 13089. [2017-06-23 21:18:34] Kenran the Grappler (L11 BaFi of Gozag) left the Lair of Beasts on turn 10634. (Lair:1)
17:24:09 <kenran> nevermind, I found the culprit
17:24:42 <kenran> it's another line that has been broken up on cdo :D
17:32:42 <gammafunk> rc setting to color all armour blue, making you think it has ego/enchant
17:32:50 <gammafunk> the optimism RC file
17:33:18 <kenran> since I copy more than 50% of my settings from you, it could also have been a trolling attempt
17:33:58 <kenran> though I'd have seen that in tiles as well :)
17:44:38 <amalloy> looking at the PR to remove mottled drac from aptitude table, i have concluded that https://github.com/cut1less/crawl/blob/57cdc1070694a929160ff6c79aa18f2c4054ee2c/crawl-ref/source/util/gen-apt.pl#L65 is a bit silly. it seems to be making draconian colours right-aligned while the rest of the table is left-aligned
17:44:43 <amalloy> i don't mind that i guess, but the weird thing is that it is going to a lot of effort to make them all print as long as "Draconian Black", an arbitrary choice since Black happens to sort first
17:46:16 <amalloy> oh no, it makes them all print as long as "Draconian Base", a string which is never actually printed. it would be a lot simpler to just print all draconian colours as sprintf("%20s", $flavour)
17:47:00 <amalloy> also i think this would break if we ever introduced an Alabaster Draconian, since that sorts before Base
17:48:51 <amalloy> (oh, and to be clear this is not a weirdness introduced by the PR: it was always there)
17:51:23 <Cheibriados> 03cut1less02 {amalloy} 07* 0.21-a0-82-gdb84782: Remove Mottled Draconians from table of aptitudes (#558) 10(8 seconds ago, 1 file, 1+ 0-) 13https://github.com/crawl/crawl/commit/db8478240255
18:02:26 <Lantell> Unstable branch on underhound.eu updated to: 0.21-a0-82-gdb84782 (34)
18:04:12 <gammafunk> kenran: you're bad at copying RC files
18:09:32 <kenran> unfortunately yes :D
18:09:46 <Jorgrell> Unstable branch on crawl.jorgrun.rocks updated to: 0.21-a0-82-gdb84782 (34)
18:12:47 <gammafunk> my RC is actually broken right now, I've been in the process of overhauling portions of it
18:12:59 <gammafunk> makes testing crawl a bit weird since I have to use a default rc
18:14:11 <kenran> what do you mean by broken?
18:23:37 <gammafunk> my local one, it's got incomplete changes
18:23:39 <gammafunk> hence broken
18:24:02 <gammafunk> you know, ever work on a source code file to make changes, but not finish them? it won't compile or work!
18:24:31 <Cheibriados> 03kenranunderscore02 {amalloy} 07* 0.21-a0-83-g1e75fd5: Tree form now fully suppresses the blink and teleportitis mutations 10(2 days ago, 1 file, 6+ 0-) 13https://github.com/crawl/crawl/commit/1e75fd560e9d
18:24:31 <Cheibriados> 03kenranunderscore02 {amalloy} 07* 0.21-a0-84-g85d6d27: Print a better message when an attempted blink is not possible 10(2 days ago, 1 file, 2+ 1-) 13https://github.com/crawl/crawl/commit/85d6d27d4c04
18:29:52 <gammafunk> rip my blinking bush proposal
18:51:33 <cut1less> amalloy: i think "base draconian" does occur, just not in the expected place -- gen-apt.pl generates the species names from aptitudes.h rather than species-data.h.
18:51:39 <cut1less> see https://github.com/crawl/crawl/blob/master/crawl-ref/source/util/gen-apt.pl#L229
18:52:09 <amalloy> cut1less: it doesn't appear in the output aptitudes table, is what i'm saying
18:52:26 <amalloy> but then all other draconian flavours are formatted to be the same length as "draconian base" in that table
18:58:03 <cut1less> ahhhh, i see what you're saying. yeah, choosing an arbitrary length would be better, although i think 20 would be too much (lemme check quickly)
19:01:12 <cut1less> yeah, the 20 in sprintf("%20s", $flavour) is slightly too much; 14 matches what it is now, which looks reasonable
19:03:57 -!- GiantOwl is now known as Kalir
19:09:36 <Jorgrell> Unstable branch on crawl.jorgrun.rocks updated to: 0.21-a0-84-g85d6d27 (34)
19:10:05 -!- amalloy is now known as amalloy_
19:57:58 <Cheibriados> 03cut1less02 07https://github.com/crawl/crawl/pull/556 * 0.21-a0-83-gaa40100: Fix and clean up spacing/parentheses 10(10 minutes ago, 1 file, 3+ 3-) 13https://github.com/crawl/crawl/commit/aa4010060339
22:01:33 -!- amalloy_ is now known as amalloy
22:05:50 <johnstein> thoughts on why lairratio is timing out?
22:05:55 <johnstein> !lairratio .
22:08:29 <Sequell> 90s limit exceeded: killed !lairratio .
22:09:08 <johnstein> I logged into shalott and nothing seemed off.
23:22:20 -!- GiantOwl is now known as Kalir
23:24:56 <amalloy> !cmd !lairratio
23:24:56 <Sequell> Command: !lairratio => .echo $(let (x (!lm ${1:-.} $* br.enter=lair ktyp!= x=cdist(gid) fmt:"${x}" stub:"0") y (!lg ${1:-.} $* fmt:"${n}" stub:"0") player (name_fixup $1)) "${player}$(if "${*}" " (${*})" "") has reached Lair in $x of $y attempts: $(int (+ 0.5 (/ (* 100.0 $x) $y)))%")
23:59:33 <Eksell> Unstable branch on CRAWL.XTAHUA.COM updated to: 0.21-a0-84-g85d6d27 (34)