00:02:11 -!- _dd has joined ##crawl-dev
00:03:41 <Gretell> Unstable branch on crawl.develz.org updated to: 0.12-a0-2515-g840a04f (34)
00:04:05 -!- kekekela has quit [Quit: Page closed]
00:04:52 -!- Krakhan has quit [Ping timeout: 256 seconds]
00:05:13 -!- shirish has joined ##crawl-dev
00:08:33 -!- anastasios_ is now known as Pedjt
00:20:13 <Gretell> Windows builds of master branch on crawl.develz.org updated to: 0.12-a0-2515-g840a04f
00:21:49 -!- dieseltime has quit [Remote host closed the connection]
00:22:13 -!- cocofalco has quit [Quit: Page closed]
00:22:32 -!- Shaijin has quit [Ping timeout: 245 seconds]
00:22:39 -!- shirish has quit [Remote host closed the connection]
00:30:52 -!- dcssrubot147 has quit [Read error: Connection reset by peer]
00:35:53 -!- eb has quit []
00:46:26 -!- AriaB1 has joined ##crawl-dev
00:56:00 -!- domiryuu has quit [Ping timeout: 264 seconds]
00:58:41 -!- jeanjacques has quit [Ping timeout: 248 seconds]
01:00:10 -!- LoremIpsum has quit [Ping timeout: 276 seconds]
01:00:12 -!- AriaB1 has quit [Quit: Leaving.]
01:00:50 -!- AriaB1 has joined ##crawl-dev
01:05:41 -!- Bjornvader has quit [Quit: Bjornvader]
01:12:42 -!- AriaB1 has quit [Ping timeout: 250 seconds]
01:12:55 -!- ophanim has quit [Read error: Connection reset by peer]
01:25:09 -!- Sprort has quit [Ping timeout: 276 seconds]
01:31:07 -!- Kalir has quit [Quit: CADENZA-CLASS CLOCKWORK KNIGHT--OPERATION TERMINATED]
01:39:03 -!- Poncheis has quit [Quit: ChatZilla 0.9.90 [Firefox 5.0.1/20110707182747]]
01:44:51 -!- dcssrubot402 has quit [Read error: Connection reset by peer]
01:52:31 -!- bh has quit [Quit: Page closed]
01:54:35 -!- Insomniak` has quit [Read error: Connection reset by peer]
01:58:32 -!- evilmike has quit []
02:00:42 -!- Vizer_ has quit [Quit: bye]
02:03:46 -!- minqmay has quit [Quit: Leaving]
02:06:34 -!- st_ has joined ##crawl-dev
02:23:22 -!- tophat has quit [Ping timeout: 245 seconds]
02:32:40 -!- simmarine has quit [Quit: Leaving]
02:35:59 -!- ussdefiant has quit [Ping timeout: 252 seconds]
02:45:55 -!- Caprice has quit [Quit: 'Truth' is a dangerous skill that ignores faith!]
02:52:43 -!- alefury has joined ##crawl-dev
02:58:08 -!- bonghitz has quit [Remote host closed the connection]
03:07:54 -!- DainDwarf has quit [Changing host]
03:13:19 -!- dtsund has left ##crawl-dev
03:15:28 -!- clouded_ has joined ##crawl-dev
03:15:46 -!- st_ has quit [Read error: Connection reset by peer]
03:18:21 -!- st_ has joined ##crawl-dev
03:19:59 -!- clouded_ has quit [Ping timeout: 252 seconds]
03:20:28 -!- Dixlet has quit [Read error: Connection reset by peer]
03:29:40 -!- dcssrubot623 has quit [Read error: Connection reset by peer]
03:30:52 -!- syraine has quit [Ping timeout: 245 seconds]
03:31:52 -!- Illusionbear has quit [Read error: Operation timed out]
03:39:03 -!- DracoOmega has quit [Quit: Leaving]
03:45:27 -!- superc has quit [Ping timeout: 245 seconds]
03:51:28 -!- scummos^ has quit [Remote host closed the connection]
03:52:39 -!- rossi has quit [Ping timeout: 260 seconds]
03:56:48 -!- scummos^ has quit [Ping timeout: 245 seconds]
03:59:38 -!- clouded_ has joined ##crawl-dev
03:59:47 -!- clinew has quit [Quit: leaving]
04:00:08 -!- st_ has quit [Ping timeout: 252 seconds]
04:07:29 -!- dcssrubot498 has quit [Ping timeout: 245 seconds]
04:14:45 -!- st_ has joined ##crawl-dev
04:14:57 -!- clouded_ has quit [Read error: Connection reset by peer]
04:15:37 <alefury> !tell ontoclasm https://crawl.develz.org/mantis/view.php?id=6393
04:15:40 <Henzell> alefury: OK, I'll let ontoclasm know.
04:25:15 -!- Taco_Princess has quit [Ping timeout: 252 seconds]
04:27:27 -!- Illusionbear has quit [Quit: Illusionbear]
04:34:06 -!- edlothiol has joined ##crawl-dev
04:45:30 -!- jeanjacques has joined ##crawl-dev
04:56:53 -!- Impy_ has quit [Quit: Page closed]
05:00:01 -!- Crazylemon64 has quit [Ping timeout: 245 seconds]
05:05:35 -!- monqy has quit [Quit: hello]
05:10:16 -!- MarvinPA has joined ##crawl-dev
05:14:36 -!- freefall has quit [Ping timeout: 276 seconds]
05:16:42 -!- syraine has quit [Ping timeout: 245 seconds]
05:18:37 -!- freefall has joined ##crawl-dev
05:21:37 -!- domiryuu has quit [Remote host closed the connection]
05:23:08 -!- noppa354 has quit [Quit: Page closed]
05:28:27 -!- kek has joined ##crawl-dev
05:39:49 -!- ontoclasm has quit [Quit: Leaving.]
05:40:27 -!- madreisz has quit [Ping timeout: 245 seconds]
05:45:02 -!- syraine has quit [Ping timeout: 245 seconds]
05:59:35 -!- MaxFrost has quit [Ping timeout: 260 seconds]
06:01:46 <Cheibriados> Wizard mode indicator not cleared between games in WebTiles (https://crawl.develz.org/mantis/view.php?id=6732) by Medar
06:06:44 -!- dcssrubot48 has quit [Read error: Connection reset by peer]
06:13:42 -!- syllogism has joined ##crawl-dev
06:17:26 -!- Isvaffel has quit [Read error: Connection reset by peer]
06:19:12 -!- absolutego has joined ##crawl-dev
06:26:43 -!- danbru has joined ##crawl-dev
06:30:10 -!- dagonfive has quit [Read error: Connection reset by peer]
06:40:30 -!- Nightbeer has quit [Read error: Connection reset by peer]
06:42:34 <Cheibriados> 03kilobyte 07* 0.12-a0-2516-ge6d1668: Indentation fixes. 10(4 hours ago, 2 files, 6+ 5-) 13http://git.develz.org/?p=crawl.git;a=commitdiff;h=e6d166806899
06:42:34 <Cheibriados> 03kilobyte 07* 0.12-a0-2517-g4d779ce: Free the cached abyss grid in some cases. 10(3 hours ago, 5 files, 19+ 6-) 13http://git.develz.org/?p=crawl.git;a=commitdiff;h=4d779cec5bda
06:44:42 -!- dcssrubot298 has quit [Read error: Connection reset by peer]
07:00:45 -!- dieseltime has quit [Remote host closed the connection]
07:10:50 -!- y2s82 has quit [Remote host closed the connection]
07:17:20 -!- BrightCloud has quit [Ping timeout: 260 seconds]
07:22:43 -!- freefall has quit [Remote host closed the connection]
07:23:16 -!- freefall has joined ##crawl-dev
07:27:54 -!- mikee_ has quit [Quit:  HydraIRC -> http://www.hydrairc.com <- Po-ta-to, boil em, mash em, stick em in a stew.]
07:29:35 -!- Stelpa has quit [Ping timeout: 260 seconds]
07:36:49 -!- dcssrubot383 has quit [Read error: Connection reset by peer]
07:51:47 -!- shirish has joined ##crawl-dev
07:52:54 -!- rossi has quit [Ping timeout: 245 seconds]
07:56:01 -!- lukano has quit [Quit: leaving]
07:56:45 -!- Blazinghand has quit [Ping timeout: 240 seconds]
08:03:24 -!- jeanjacques has quit [Quit:  HydraIRC -> http://www.hydrairc.com <- It'll be on slashdot one day...]
08:09:18 -!- Arkaniad has quit [Ping timeout: 256 seconds]
08:12:31 -!- Leafsnail has quit [Ping timeout: 245 seconds]
08:14:47 -!- dcssrubot298 has quit [Read error: Connection reset by peer]
08:15:26 -!- Isvaffel has quit [Ping timeout: 245 seconds]
08:15:51 -!- Arkaniad|Laptop has quit [Ping timeout: 245 seconds]
08:21:14 -!- Arkaniad|Laptop_ has quit [Ping timeout: 245 seconds]
08:22:04 -!- Stelpa has quit [Changing host]
08:22:18 -!- ajikeshi has quit [Ping timeout: 264 seconds]
08:23:22 -!- One-Eyed has quit [Ping timeout: 245 seconds]
08:33:51 -!- One-Eyed_ has quit [Quit: Page closed]
08:34:20 -!- Mottie has quit [Ping timeout: 260 seconds]
08:35:56 -!- ajikeshi has quit [Ping timeout: 256 seconds]
08:43:21 -!- ajikeshi_ has quit [Ping timeout: 252 seconds]
08:46:34 -!- ajikeshi__ has quit [Read error: Operation timed out]
08:58:04 -!- SaintWacko has quit [Quit: Page closed]
09:01:42 -!- scummos^ has quit [Read error: Connection reset by peer]
09:06:07 -!- shmupper has joined ##crawl-dev
09:06:55 -!- dcssrubot954 has quit [Read error: Connection reset by peer]
09:09:59 -!- herself has quit [Ping timeout: 246 seconds]
09:18:02 -!- eb has quit [Ping timeout: 250 seconds]
09:28:03 -!- Pthing has quit [Ping timeout: 245 seconds]
09:30:04 -!- eb has quit [Client Quit]
09:36:16 -!- inpho has quit [Quit: inpho]
09:36:41 * SamB wonders how many years until Crawl can default to building with -Og
09:37:05 * SamB doesn't even have that compiler installed yet, though
09:44:53 -!- dcssrubot658 has quit [Read error: Connection reset by peer]
09:56:23 -!- remyroy has quit [Ping timeout: 245 seconds]
10:00:25 -!- shmupper has quit [Read error: Connection reset by peer]
10:06:57 <kilobyte> -Og ?
10:11:00 <ChrisOelmueller> unlocks nasal ogres
10:11:14 -!- bhaak has quit [Ping timeout: 245 seconds]
10:14:28 -!- antlions has quit [Read error: Connection reset by peer]
10:18:48 -!- elliptic has quit [Quit: Leaving]
10:24:01 -!- Dixlet has joined ##crawl-dev
10:26:09 -!- bonghitz has quit [Remote host closed the connection]
10:26:54 -!- bmfx has quit [Read error: Connection reset by peer]
10:29:36 -!- Leafsnail has quit [Ping timeout: 245 seconds]
10:31:04 -!- bhaak has joined ##crawl-dev
10:32:17 -!- _psychoDad has quit [Quit: Page closed]
10:36:53 -!- Mandevil has quit [Quit: leaving]
10:37:01 -!- dcssrubot552 has quit [Read error: Connection reset by peer]
10:39:02 -!- shmup has joined ##crawl-dev
10:41:48 -!- inspector071 has quit [Ping timeout: 245 seconds]
10:42:18 <SamB> kilobyte: isn't that the new "optimize for debugging" flag?
10:46:39 -!- inpho has quit [Remote host closed the connection]
10:55:02 -!- Lassee- has quit [Client Quit]
10:58:08 -!- Lasse- has quit [Ping timeout: 246 seconds]
11:01:40 -!- ussdefiant has joined ##crawl-dev
11:03:26 -!- smeea has quit [Read error: Connection reset by peer]
11:04:07 <Henzell> Stable (0.11) branch on crawl.akrasiac.org updated to: 0.11.2-10-g01a9a8f
11:08:42 -!- rwbarton has quit [Quit: leaving]
11:14:59 -!- dcssrubot545 has quit [Read error: Connection reset by peer]
11:15:09 -!- Mottie has joined ##crawl-dev
11:16:28 -!- bonghitz has quit [Remote host closed the connection]
11:17:05 -!- Vizer has joined ##crawl-dev
11:17:25 <Henzell> Unstable branch on crawl.akrasiac.org updated to: 0.12-a0-2517-g4d779ce (34)
11:21:22 -!- st_ has quit [Read error: Connection reset by peer]
11:21:22 -!- clouded_ has joined ##crawl-dev
11:29:20 -!- Aryth has quit [Quit: KVIrc 4.0.4 Insomnia http://www.kvirc.net/]
11:31:13 -!- blabber has joined ##crawl-dev
11:32:14 <mumra> !tell bh I've been messing around with noise using perlin.cc and I'm noticing something strange: the x coord doesn't seem to make any difference (in perlin::fBM(x,y,z))
11:32:15 <Henzell> mumra: OK, I'll let bh know.
11:39:15 -!- eb has quit [Ping timeout: 240 seconds]
11:39:19 -!- rossi has quit [Ping timeout: 260 seconds]
11:39:23 -!- ussdefiant has quit [Ping timeout: 252 seconds]
11:45:36 -!- tophat has quit [Quit: Leaving.]
11:48:59 <freefall> but everything's so much simpler when you project it on an axis
11:49:32 <mumra> !tell bh another problem: for very high values of x, y or z, perlin::noise(x,y,z) always returns 0 (obviously a problem when i'm using abyss coords and offset for x,y,z!)
11:49:33 <Henzell> mumra: OK, I'll let bh know.
11:49:51 -!- alefury has quit [Quit: ChatZilla 0.9.85 [Iceape 2.0.11/20120715100958]]
11:49:56 <mumra> freefall: simpler, yes, but it creates a very horizontal-looking abyss ;)
11:50:35 * SamB goes to check the translation from Java
11:50:59 <Cheibriados> 03dolorous 07* 0.12-a0-2518-g514453d: Add formatting fixes. 10(7 minutes ago, 1 file, 6+ 5-) 13http://git.develz.org/?p=crawl.git;a=commitdiff;h=514453d2625a
11:52:28 * SamB wants to add a hook to reject any pushes with that commit message ...
11:55:12 <SamB> (Actually, some sort of hook to catch formatting issues at commit time wouldn't be a bad idea, either ...)
11:55:34 -!- DainDwarf has quit [Quit: Dain's default quit message]
11:55:39 -!- bonghitz has quit [Ping timeout: 260 seconds]
11:55:40 <Zannick> then perhaps the commit message would change to "Remove formatting errors."
11:55:46 <Zannick> :P
11:56:56 -!- shirish has quit [Read error: Operation timed out]
11:57:19 -!- ussdefiant has joined ##crawl-dev
11:57:32 -!- ajikeshi has quit [Ping timeout: 255 seconds]
11:59:04 -!- simmarine has joined ##crawl-dev
11:59:33 <SamB> I meant so the wrong formatting never got committed in the first place ;-P
12:01:11 -!- madreisz_ has quit [Quit: Page closed]
12:02:35 <ChrisOelmueller> patches welcome, git-hooks/pre-commit
12:04:57 -!- dcssrubot639 has quit [Read error: Connection reset by peer]
12:05:07 -!- ldf has quit [Quit: Page closed]
12:06:39 -!- bmfx has quit [Read error: Connection reset by peer]
12:07:32 <kilobyte> "astyle" with a set of options seems to match our formatting
12:09:06 <kilobyte> that -Og thingy seems to require gcc-4.8
12:09:56 <SamB> the trick would be to get it to check only in/near changed code, I think?
12:09:57 <kilobyte> having it be the default is simple, this kind of feature check could be done even with our current makefile
12:10:00 -!- dtsund has joined ##crawl-dev
12:10:38 <kilobyte> diff doesn't understand C++ syntax so it'd be tricky :(
12:10:45 -!- the_glow1 has quit [Ping timeout: 260 seconds]
12:13:11 <SamB> okay, so fBM isn't *in* SimplexNoise.java
12:16:47 -!- ussdefiant has quit [Ping timeout: 252 seconds]
12:17:18 -!- Vizer_ has joined ##crawl-dev
12:18:14 -!- Vizer has quit [Ping timeout: 255 seconds]
12:18:15 <SamB> ... I think fBM needs to find a different home ...
12:21:23 <Cheibriados> 03dolorous 07* 0.12-a0-2519-g5c1ffac: Comment fixes. 10(13 minutes ago, 1 file, 3+ 3-) 13http://git.develz.org/?p=crawl.git;a=commitdiff;h=5c1ffac93946
12:24:27 -!- st_ has joined ##crawl-dev
12:24:47 -!- clouded_ has quit [Read error: Connection reset by peer]
12:25:13 -!- Kalir has quit [Changing host]
12:25:27 <mumra> SamB: it's a strange bug, if i divide my x,y coords by only around 10 then everything works.
12:25:35 -!- bmfx has quit [Ping timeout: 252 seconds]
12:25:52 <mumra> SamB: then problem is this produces slightly larger geometry than for one of my purposes (but is ok for most things)
12:26:29 -!- bonghitz has quit [Ping timeout: 260 seconds]
12:26:41 <SamB> !tell SamB is this thing working, anyway?
12:26:41 <Henzell> SamB: OK, I'll let SamB know.
12:27:21 <SamB> I was going to !tell bh something but I remembered hearing that !messages was broken ...
12:27:21 <Henzell> SamB: You have 1 message. Use !messages to read it.
12:27:24 <SamB> !messages
12:27:25 <Henzell> (1/1) SamB said (43s ago): is this thing working, anyway?
12:29:11 -!- Leafsnail has quit [Ping timeout: 245 seconds]
12:29:49 <SamB> !tell bh It's not at all clear to me what fBM is or what it's doing in perlin.cc; it's not in http://webstaff.itn.liu.se/~stegu/simplexnoise/SimplexNoise.java ...
12:29:49 <Henzell> SamB: OK, I'll let bh know.
12:30:18 * SamB laughs at the comment in perlin.h
12:34:59 -!- bmfx has quit [Ping timeout: 255 seconds]
12:36:35 <Cheibriados> 03dolorous 07* 0.12-a0-2520-ga4ab263: Simplify. 10(7 minutes ago, 1 file, 4+ 8-) 13http://git.develz.org/?p=crawl.git;a=commitdiff;h=a4ab263bffb3
12:36:41 -!- shirish has joined ##crawl-dev
12:37:15 <mumra> SamB: fBM calls the noise function several times to generate perlin with octaves (the last parameter is octaves)
12:37:42 <mumra> SamB: so the problem i noticed in fBM could be caused by the other problem i saw in noise()
12:39:06 -!- alefury has joined ##crawl-dev
12:40:23 -!- ontoclasm has joined ##crawl-dev
12:42:10 -!- elliptic has joined ##crawl-dev
12:43:08 -!- lexackson has joined ##crawl-dev
12:43:13 -!- shirish has quit [Ping timeout: 248 seconds]
12:46:03 -!- Zifmia has quit [Quit: AndroidIrc Disconnecting]
12:46:43 <Cheibriados> 03dolorous 07* 0.12-a0-2521-gb34b620: Make Yred appreciate desecrating holy remains (for a small piety gain). 10(47 minutes ago, 2 files, 15+ 1-) 13http://git.develz.org/?p=crawl.git;a=commitdiff;h=b34b6208d5a8
12:48:35 <SamB> mumra: so how big are the numbers you were passing to fBM?
12:48:50 <SamB> or noise
12:49:09 <mumra> SamB: they were based on coord_def coords which are uint32_t
12:51:29 <mumra> i would hazard a guess that the upper 50% of numbers in that domain are problematic
12:52:09 <mumra> (well, probably a lot more than 50%, but that should definitely trigger the issue)
12:53:13 <mumra> and note that this coord_def was an abyssal coord which uses the full range
12:53:27 <mumra> rather than the normal GXM/GYM limit
12:53:29 -!- NeremWorld has quit [Ping timeout: 256 seconds]
12:54:11 <SamB> okay, yeah, that's well within the range where IEEE doubles can represent integers exactly ...
12:56:19 -!- shirish has joined ##crawl-dev
12:57:05 -!- the_glow has quit [Ping timeout: 248 seconds]
12:57:13 -!- eb has quit [Ping timeout: 245 seconds]
12:58:17 <lexackson> hey guys, I made an okawaru altar vault, https://crawl.develz.org/mantis/view.php?id=6679
12:58:31 <Napkin> greensnark, works like a charm, lightning fast now :)
12:58:31 <Henzell> Napkin: You have 6 messages. Use !messages to read them.
13:00:32 -!- shirish has quit [Ping timeout: 255 seconds]
13:02:26 <st_> lexackson: apparently no one but The Vault Dev who doesn't exist right now is allowed to add vaults
13:02:44 <lexackson> oh
13:03:27 <mumra> SamB: i already checked the final double values i was passing to noise and they looked fine
13:03:36 <lexackson> i guess i'll just bring it up again in a week or so
13:03:57 <lexackson> thanks st_
13:04:02 <mumra> lexackson: Grunt is a good person to pester ;)
13:04:15 <lexackson> yeah, he helped me a bit with the lua
13:04:20 <SamB> doesn't exist right now?
13:04:33 -!- eb has quit [Ping timeout: 248 seconds]
13:04:58 -!- ToastyP has quit [Quit: ChatZilla 0.9.90 [Firefox 19.0/20130215130331]]
13:12:57 -!- meowfelid has quit [Quit: ChatZilla 0.9.90 [Firefox 19.0/20130215130331]]
13:15:08 <alefury> evilmike is busy, and nobody else really feels responsible for vaults
13:15:29 -!- SaintWacko has quit [Quit: Page closed]
13:16:04 <alefury> vaults do seem to get added though, just not by any single person
13:16:16 <alefury> and it takes a bit longer
13:16:37 <alefury> imo: hangedman for vault dev!
13:16:57 <alefury> or mumra, i dont really care
13:17:18 -!- Sabaki has quit [Ping timeout: 257 seconds]
13:17:38 <Grunt> I'm not really in a position to do much at the moment, sorry, and probably won't be for a while. :|
13:24:51 -!- Aryth has quit [Quit: KVIrc 4.0.4 Insomnia http://www.kvirc.net/]
13:25:18 -!- rossi has quit [Ping timeout: 264 seconds]
13:25:33 <SamB> hmm, that's interesting: http://staffwww.itn.liu.se/~stegu/aqsis/aqsis-newnoise/simplexnoise1234.cpp says it's in the public domain, but goes on to pull in the disclaimer from the GPL
13:26:39 -!- bonghitz has quit [Ping timeout: 245 seconds]
13:28:37 <mumra> alefury: i would recommend hangedman tbh, he does a lot a lot of vault work
13:29:16 <mumra> i think if he was given commit rights there would be mass influx of minor vault fixes
13:29:22 * SamB had been sorta thinking HangedMan should be a dev, too ...
13:35:03 -!- dcssrubot426 has quit [Read error: Connection reset by peer]
13:37:03 -!- sgjsr has quit [Quit: Page closed]
13:40:49 -!- eb has quit [Ping timeout: 248 seconds]
13:42:23 <Cheibriados> 03SamB 07* 0.12-a0-2522-g8520d3c: Add some comments to perlin.cc. 10(7 minutes ago, 1 file, 8+ 2-) 13http://git.develz.org/?p=crawl.git;a=commitdiff;h=8520d3c38808
13:47:13 -!- eb has quit [Ping timeout: 248 seconds]
13:49:45 -!- eith has joined ##crawl-dev
13:50:48 -!- tophat has quit [Quit: Leaving.]
13:51:02 -!- Sabaki_|2 is now known as Sabaki
14:00:48 -!- GreatOrbOfEyes has joined ##crawl-dev
14:02:16 <SamB> mumra: well, it looks like fastfloor does not work on stuff that isn't in the range of int (give or take a couple of units at each endpoint)
14:04:53 -!- GreatOrbOfEyes is now known as HangedMan
14:07:44 -!- shmup has quit [Ping timeout: 255 seconds]
14:10:51 * SamB opens C++0x draft to see what he can replace that with
14:12:50 <Cheibriados> 03Grunt 07* 0.12-a0-2523-g8da2df9: Prevent grunt_runaround subvaults from placing on their own. 10(6 minutes ago, 1 file, 3+ 1-) 13http://git.develz.org/?p=crawl.git;a=commitdiff;h=8da2df9ead9d
14:13:21 -!- nooodl has joined ##crawl-dev
14:18:06 <SamB> huh, C11 is in much larger print than C++11 ...
14:26:27 <Napkin> duh... frightened orc warrior, which is running away, does reaching attack while i run away from him?
14:26:53 -!- syllogism has quit []
14:27:04 -!- bonghitz has quit [Ping timeout: 245 seconds]
14:27:10 <SamB> how can you both be running away and yet be close enough for a reaching attack?
14:29:35 <Napkin> face to face, when i read he turned frightened, but hadn't moved away yet.. i move away and die. screenshot: ·o·@·
14:30:21 <Napkin> well, inverted o, since he was frightened
14:30:31 <Napkin> or similar to inverted
14:30:42 -!- the_glow has quit [Ping timeout: 264 seconds]
14:31:13 <Napkin> meleeing while running away in panic, ok.. but turning around and doing a reaching attack while running away frightened is.. a bug
14:33:32 -!- HandsOfAizul has quit [Quit: Page closed]
14:33:54 <SamB> ah
14:34:59 <alefury> actually im not sure about that :/
14:35:21 <alefury> currently enemies with ranged attacks can stop fleeing to use their ranged attacks
14:35:58 <alefury> this was deemed interesting enough to keep, while other enemies dont flee anymore (this is why dragons still flee)
14:36:14 <alefury> i guess reaching counts as ranged for this purpose
14:36:26 <alefury> whether thats a bug or not i dont know
14:36:43 <SamB> mumra: so how about if I switch these over to using "long long" instead of "int" and define fastfloor(x) = llrint(floor(x))?
14:36:44 <alefury> i think its not good in any case
14:37:58 -!- ldf has quit [Read error: Connection reset by peer]
14:38:47 -!- ChickenWing has quit [Ping timeout: 245 seconds]
14:38:47 -!- Datul has quit [Ping timeout: 245 seconds]
14:38:51 <SamB> alefury: I think it probably wasn't intended
14:39:25 <alefury> i think ranged enemies attacking you after scroll of fear is also not good
14:39:32 <alefury> really mostly that
14:39:37 -!- residualshade has quit [Ping timeout: 245 seconds]
14:39:37 -!- superc has quit [Ping timeout: 245 seconds]
14:39:37 -!- Chadul has quit [Ping timeout: 245 seconds]
14:40:02 -!- twelwe has quit [Ping timeout: 245 seconds]
14:40:02 -!- kaiserfro has quit [Ping timeout: 245 seconds]
14:40:27 -!- Elynae has quit [Ping timeout: 245 seconds]
14:44:45 <SamB> or, wait, we don't seem to actually care about more than the last 8 bits of these coordinates ...
14:46:29 -!- Stelpa has quit [Remote host closed the connection]
14:47:13 -!- HellTiger has quit [Read error: Connection reset by peer]
14:57:39 <ontoclasm> okay, now to figure out how the heck large gates work
14:57:39 <Henzell> ontoclasm: You have 1 message. Use !messages to read it.
14:57:42 <ontoclasm> !messages
14:57:43 <Henzell> (1/1) alefury said (10h 42m 6s ago): https://crawl.develz.org/mantis/view.php?id=6393
14:57:56 <ontoclasm> alefury: yeah, i saw that
15:00:11 -!- tophat has quit [Quit: Leaving.]
15:02:50 <ontoclasm> okay, i'm lost
15:02:55 <ontoclasm> this seems like it should work
15:03:38 <SamB> or, yeah, we do in the calculation of "t"
15:05:09 -!- dcssrubot556 has quit [Read error: Connection reset by peer]
15:08:30 <Cheibriados> 03dolorous 07* 0.12-a0-2524-g9848ea3: Fix spacing. 10(16 minutes ago, 1 file, 1+ 1-) 13http://git.develz.org/?p=crawl.git;a=commitdiff;h=9848ea3a0d53
15:13:35 <Cheibriados> 03dolorous 07* 0.12-a0-2525-g1d51379: Don't mark Cigotuvi's Degeneration as corpse-violating; it's never used on corpses. 10(6 minutes ago, 1 file, 1+ 2-) 13http://git.develz.org/?p=crawl.git;a=commitdiff;h=1d51379be728
15:13:37 <Cheibriados> orc alter egos (https://crawl.develz.org/mantis/view.php?id=6733) by Thrall
15:15:04 <st_> that is very cryptic
15:15:36 -!- rossi has quit [Quit: WeeChat 0.4.0]
15:16:05 <Grunt> As far as I can tell, he's claiming that AC stacks across armour changes.
15:17:42 <ontoclasm> okay, i'm trying to get sealed large gates to display properly and i can't figure out why they aren't doing so
15:18:04 <ontoclasm> the code in place for other doors seems like it would get called and work just fine
15:18:09 <ontoclasm> but it doesn't
15:18:40 <Cheibriados> 03dolorous 07* 0.12-a0-2526-g7df7149: Comment fix. 10(2 minutes ago, 1 file, 1+ 1-) 13http://git.develz.org/?p=crawl.git;a=commitdiff;h=7df71492fd67
15:27:21 -!- Namey has quit [Quit: Leaving]
15:27:42 -!- bonghitz has quit [Ping timeout: 264 seconds]
15:28:48 <Cheibriados> 03ontoclasm 07* 0.12-a0-2527-g418a51f: Tiles for sealed doors 10(31 minutes ago, 9 files, 9+ 0-) 13http://git.develz.org/?p=crawl.git;a=commitdiff;h=418a51f157de
15:28:48 <Cheibriados> 03ontoclasm 07* 0.12-a0-2528-g579a454: Fix comments 10(8 minutes ago, 1 file, 2+ 2-) 13http://git.develz.org/?p=crawl.git;a=commitdiff;h=579a45444d69
15:30:58 -!- residualshade_ has quit [Quit: Page closed]
15:32:52 <ontoclasm> still haven't fixed it
15:33:02 <ontoclasm> but it's better than question marks
15:35:46 -!- Psyknux has quit [Quit: My work is end \(^o^)/]
15:36:10 -!- Dixlet has quit [Read error: Connection reset by peer]
15:36:48 -!- Aryth has quit [Read error: Connection reset by peer]
15:40:38 -!- Serfuzz has quit [Remote host closed the connection]
15:42:38 <alefury> wow, that bug report is *really* cryptic
15:43:11 -!- Cryp71c has joined ##crawl-dev
15:43:19 <Cryp71c> Afternoon
15:43:50 <alefury> i was just thinking of you!
15:44:11 <Cheibriados> 03dolorous 07* 0.12-a0-2529-g803c793: Fix abbreviation. 10(5 minutes ago, 1 file, 1+ 1-) 13http://git.develz.org/?p=crawl.git;a=commitdiff;h=803c79300c10
15:44:41 -!- blabber has quit [Quit: leaving]
15:46:13 <Cryp71c> alefury, thanks, I'll take a look :D
15:50:55 <Napkin> the orc warrior didn't stop fleeing, alefury
15:51:11 -!- DracoOmega has joined ##crawl-dev
15:51:17 <alefury> oh, thats weird then
15:51:31 <Gretell> Leafsnail (L18 HOPr) ASSERT(monster_by_mid(m->mid) == m) in 'dbg-scan.cc' at line 455 failed. (Abyss:1)
15:51:42 <Napkin> at least he still had the background color and the mlist showed as fleeing, when my death message came
15:51:50 -!- shirish has joined ##crawl-dev
15:51:58 <Cryp71c> What's the invisible, mid-depth monster named unholy something?
15:52:04 -!- Nerm is now known as NeremWorld
15:52:07 <Cryp71c> moves pretty quick...
15:52:12 <alefury> you mean unseen horror?
15:52:13 <ontoclasm> Napkin: did he have a polearm?
15:52:14 <DracoOmega> I think monsters reaching while fleeing is a known 'issue' but people disagree on whether this is as-intended or not. I think?
15:52:14 <Henzell> DracoOmega: You have 2 messages. Use !messages to read them.
15:52:16 <lexackson> unseen horror?
15:52:17 <Cryp71c> alefury, thanks
15:52:19 <Cryp71c> yep
15:52:33 <Napkin> yes
15:52:37 <BlastHardcheese> unholy terror
15:52:51 -!- Roarke has quit [Quit: Page closed]
15:53:10 <DracoOmega> Also, I am making progress on a retreat command for allies
15:54:58 -!- Krakhan has quit [Changing host]
15:55:10 <Cryp71c> lol, antennae effect is now boosted by ash? That's cool
15:55:31 <alefury> huh?
15:56:16 <Cryp71c> alefury, the DS antennae mutation's radius is determined by max(mut_level * 2, piety / 20)
15:56:48 -!- Elynae_ has quit [Quit: Page closed]
15:56:55 <DracoOmega> Well, Ash provides that style of detection without antenna, too, you know...
15:56:59 <alefury> ash detection radius is piety/20, so i think it just means the larger one applies
15:58:49 -!- DainDwarf has quit [Changing host]
16:01:05 -!- Cryp71c has quit [Ping timeout: 248 seconds]
16:02:05 -!- Vanhal has quit [Quit: Page closed]
16:02:28 -!- shirish has quit [Read error: Operation timed out]
16:10:30 <Gretell> nago (L27 NaFE) ASSERT(!invalid_monster_index(mg.foe) || mg.foe == MHITYOU || mg.foe == MHITNOT) in 'mon-place.cc' at line 1680 failed. (Slime:6)
16:10:51 <DracoOmega> !lm * crash -log
16:10:52 <Sequell> 4433. nago, XL27 NaFE, T:129705 (milestone): http://crawl.develz.org/morgues/0.11/nago/crash-nago-20130304-221027.txt
16:12:00 <Gretell> Leafsnail (L19 HOPr) ASSERT(in_bounds(p)) in 'items.cc' at line 3766 failed. (Vaults:4)
16:12:12 <DracoOmega> !lm * crash -log
16:12:12 <Sequell> 4434. Leafsnail, XL19 HOPr, T:46548 (milestone): http://crawl.develz.org/morgues/trunk/Leafsnail/crash-Leafsnail-20130304-221157.txt
16:14:24 -!- shirish has joined ##crawl-dev
16:14:34 -!- ldf has quit []
16:14:39 <Cheibriados> 03dolorous 07* 0.12-a0-2530-g3f51693: Reorder conduct checks a bit. 10(5 minutes ago, 1 file, 4+ 4-) 13http://git.develz.org/?p=crawl.git;a=commitdiff;h=3f51693d96c2
16:14:46 <DracoOmega> Okay, that one IS my fault
16:14:58 <DracoOmega> But it should be a simple fix, I think
16:15:55 -!- Leafsnail has quit [Quit: When the chips are down, well, the buffalo is empty]
16:17:07 -!- Elynae has quit [Ping timeout: 245 seconds]
16:19:33 -!- shirish has quit [Ping timeout: 276 seconds]
16:19:43 <Cheibriados> Some kind of error message (https://crawl.develz.org/mantis/view.php?id=6734) by hindered
16:22:25 <DracoOmega> Actually, maybe it's more finicky than expected....
16:26:57 <DracoOmega> I mean, from the crash it obviously tried to move items to some out of bounds location, except that before it actually attempts to close a door at all, it ensures that there is a valid spot to move things to
16:28:01 <DracoOmega> This would seem to require a door being adjacent to an out of bounds location, and that out of bounds location being open terrain
16:28:12 -!- bonghitz has quit [Ping timeout: 252 seconds]
16:28:45 <mumra> SamB: (timelapse) not sure which numbers you're talking about making "long long", do you mean the coord_def?
16:29:03 <SamB> mumra: no
16:29:20 <SamB> hold on as I test this ...
16:33:45 <ontoclasm> DracoOmega: dunno if you saw this, but i added sealed door tiles
16:33:58 <ontoclasm> however, i can't figure out how to make gates show up correctly
16:34:20 <ontoclasm> they currently show up as just a bunch of separate doors
16:34:23 <DracoOmega> How do they show up incorrectly?
16:34:31 <DracoOmega> Oh, they do?
16:34:37 <ontoclasm> yeah
16:34:55 <ontoclasm> which is confusing because the code for displaying gates seems like it should work fine for them
16:34:57 <DracoOmega> That may not be the tile code's fault, incidentally
16:35:07 <DracoOmega> It may actually be that mechanically the game doesn't think they're connected
16:35:14 -!- dcssrubot118 has quit [Read error: Connection reset by peer]
16:35:15 <ontoclasm> maybe
16:35:23 <DracoOmega> (Not something you notice easily since you can't open them to test)
16:35:37 <DracoOmega> However hostile monsters CAN open them
16:35:47 <ontoclasm> i could test that
16:35:48 <DracoOmega> So if this is the case, I will fix it at some point
16:36:36 -!- Wehk has quit [Read error: Connection reset by peer]
16:36:53 -!- Wehk has joined ##crawl-dev
16:37:12 <ontoclasm> DracoOmega: no, they open together
16:37:22 <ontoclasm> _An orc opens the large door.
16:37:32 <DracoOmega> Oh, hmmm...
16:37:39 -!- Poncheis has quit [Quit: ChatZilla 0.9.90 [Firefox 12.0/20120420145725]]
16:37:41 <SamB> doesn't wizmode let you open them anyway?
16:37:51 <DracoOmega> Is there a wizmode command for that?
16:37:56 <SamB> I don't know
16:38:06 <DracoOmega> I'm not aware of one, but wizmode hides many secrets
16:38:22 -!- metasyntactic is now known as kunwon1
16:38:37 -!- Dixlet has joined ##crawl-dev
16:39:16 <DracoOmega> I do think it's something with how the TILES end of things considers them one door, though
16:39:22 -!- cocofalco has quit [Quit: Page closed]
16:39:28 <DracoOmega> Since I think I bumped into something similar earlier in development (I can't remember the context)
16:39:30 <ontoclasm> possibly
16:39:38 -!- DainDwarf has quit [Quit: Dain's default quit message]
16:39:43 <ontoclasm> how does door_vault keep them separate?
16:40:21 <ontoclasm> oh
16:40:23 <ontoclasm> it doesn't
16:40:27 <ontoclasm> interesting
16:40:45 <ontoclasm> well that's a bug too then
16:40:47 <ontoclasm> goody
16:41:40 <DracoOmega> Wait, what?
16:41:59 <DracoOmega> Are you saying that door_vault opens all at once, or that they LOOK like gates, but still open per-tile?
16:42:06 <ontoclasm> the second one
16:42:14 <DracoOmega> Oh, hmmm
16:42:25 <ontoclasm> it used to look correct iirc
16:42:33 <DracoOmega> Was that back when they were secret?
16:42:35 <DracoOmega> Or after that?
16:42:40 <ontoclasm> before that
16:42:43 <ontoclasm> when it was just doors
16:42:53 <ontoclasm> so like 0.9ish
16:43:05 <DracoOmega> They were still secret then, weren't they?
16:43:09 -!- motorbit has quit [Quit: Page closed]
16:43:09 <ontoclasm> oh
16:43:11 <ontoclasm> maybe
16:43:20 <DracoOmega> In fact, I think they're still secret in 0.11
16:43:26 <ontoclasm> yeah i guess so
16:43:33 -!- ussdefiant has joined ##crawl-dev
16:44:01 <DracoOmega> I am not sure what to do about this warden crash. I can't really figure out how the problem was encountered, just from looking at the code and the crash dump
16:44:38 <DracoOmega> A save right before it happens would be nice
16:45:22 -!- Nightbeer has quit [Read error: Connection reset by peer]
16:45:30 <ontoclasm> fyi door_vault + warden = hilarity
16:45:33 <SamB> yeah, it's too bad the servers don't store a copy of the save when it crashes ...
16:46:06 <DracoOmega> ontoclasm: Ha
16:46:14 <DracoOmega> What the warden really should do is just OPEN the whole thing
16:46:39 <ontoclasm> heh heh
16:46:47 -!- morik has quit [Disconnected by services]
16:48:57 -!- morik has quit [Disconnected by services]
16:49:08 -!- ussdefiant has quit [Ping timeout: 255 seconds]
16:49:16 -!- ussdefiant has joined ##crawl-dev
16:49:37 -!- nooodl has quit [Ping timeout: 248 seconds]
16:54:49 -!- ussdefiant_ has joined ##crawl-dev
16:56:50 -!- Zermako has quit []
16:57:16 -!- Kalir has quit [Quit: CADENZA-CLASS CLOCKWORK KNIGHT--OPERATION TERMINATED]
16:57:26 -!- ussdefiant has quit [Ping timeout: 256 seconds]
16:57:29 -!- ussdefiant_ is now known as ussdefiant
17:03:04 -!- bonghitz has quit [Remote host closed the connection]
17:13:16 -!- edlothiol has quit [Ping timeout: 256 seconds]
17:14:56 -!- absolutego has quit [Quit: leaving]
17:20:22 <Cheibriados> 03SamB 07* 0.12-a0-2531-ge213781: Widen the domain of perlin::noise to cover at least all of uint32_t. 10(3 minutes ago, 1 file, 14+ 11-) 13http://git.develz.org/?p=crawl.git;a=commitdiff;h=e21378141335
17:28:19 -!- browncustard has joined ##crawl-dev
17:30:05 <SamB> mumra: does it work better now
17:30:07 <SamB> ?
17:30:57 <HangedMan> by the way while I might fix up a million little things in other vaults and have made a ridiculous amount of vaults I am nowhere near comfortable with my sense of design and quality in vaultmaking to actually want to be a vault dev
17:32:31 -!- kekekela has quit [Client Quit]
17:32:33 <mumra> SamB: aha, that looks better (will take me a while to test because i haven't merged this branch for a while)
17:33:02 <mumra> SamB: do you know how much of a performance hit long long will have? I am calling perlin a lot per frame
17:33:19 <SamB> no, I don't
17:33:47 -!- MarvinPA has quit [Ping timeout: 245 seconds]
17:35:27 -!- s951 has quit [Client Quit]
17:40:37 <Cheibriados> New Skeletons (https://crawl.develz.org/mantis/view.php?id=6735) by Aygtets
17:44:32 <mumra> HangedMan: maybe, but which devs would even claim that right now? ;)
17:45:49 <HangedMan> must quickly write the opinionated annotated hangedman's guide to vault making philosophy
17:46:00 <HangedMan> and then get evilmike and st_ to sign that off
17:46:03 -!- MarvinPA has joined ##crawl-dev
17:46:07 <HangedMan> and then just let that be the official thing
17:46:26 -!- browncustard has quit [Quit: being stupid, sunstruck, and Dead, flew into the rocketing FIN.]
17:47:31 -!- tophat has quit [Quit: Leaving.]
17:47:56 -!- ldf has quit [Ping timeout: 245 seconds]
17:49:44 -!- _dd has quit [Ping timeout: 240 seconds]
17:52:14 -!- due is now known as due_
17:53:30 -!- jeanjacques has joined ##crawl-dev
17:57:25 <mumra> (of course i wasn't saying i wouldn't take the position if offered, but there are several people with way more vault experience than me, and i do have weeks of very low availability)
17:58:37 <DracoOmega> Monster movement code is always such a pain, since it seems to be so spread around
17:59:12 <HangedMan> s|Monster movement ||
17:59:21 <mumra> haha, HangedMan beat me to it
17:59:54 <DracoOmega> Haha. Honestly, some of the code really doesn't seem that confusing to me anymore
18:00:07 <HangedMan> you've become too corrupted
18:00:15 <DracoOmega> But I am still wondering why these monsters are continuing to move into spaces that I have explicitly told them not to
18:00:32 <DracoOmega> Reminds me (unpleasantly) of coding battlesphere
18:00:52 <DracoOmega> When it took forever to make it actually do what I was telling it to do, since it kept getting overridden by a half-dozen other points along the code path
18:01:34 <mumra> it's the same with any sufficiently large project with a long development time, you get acclimatised eventually
18:01:55 <HangedMan> corruptions
18:02:02 <HangedMan> you all need potions of cure mutation
18:03:34 <DracoOmega> mumra: Well, I'm acclimatized in the sense that I expect this and accept it. But that doesn't mean I don't always hope things take less time than they do :P
18:04:35 -!- Vizer has joined ##crawl-dev
18:04:51 -!- dtsund has quit [Ping timeout: 276 seconds]
18:05:20 -!- dcssrubot730 has quit [Read error: Connection reset by peer]
18:05:21 -!- Vizer_ has quit [Ping timeout: 248 seconds]
18:08:27 <HangedMan> by the way can somebody, anybody make local tiles actually give the build error instead of just silently instantly crashing
18:09:14 <DracoOmega> Like, for compiling des files or something?
18:09:34 <HangedMan> yes
18:11:59 <mumra> HangedMan: run tiles from command prompt, it outputs stuff then
18:12:19 <HangedMan> default things should not be dumb
18:12:51 <mumra> SamB did mention something recently about this, like popping messages up in a message box
18:13:37 <HangedMan> samB best dev
18:13:49 <SamB> well, I just said it would make sense to tell the user where crash dumps went
18:15:13 <ontoclasm> oh dear https://crawl.develz.org/mantis/view.php?id=6735
18:15:32 <SamB> it hadn't occurred to me that people start programs from the GUI; I was just thinking of the Windows command line
18:15:46 <mumra> SamB: the command line already tells you where crash dumps went
18:15:52 <HangedMan> ontoclasm do you not like that skeleton drake
18:15:59 <SamB> mumra: Windows fails at this
18:15:59 <HangedMan> which is a thinner white drake with red eyes
18:16:03 <HangedMan> and weird wings
18:16:04 <SamB> for graphical programs
18:16:08 <mumra> SamB: not on my windows apparently
18:16:30 <mumra> SamB: anything that goes to std output gets printed in the command line i started crawl from
18:16:36 <mumra> whether that's msysgit or normal command line
18:16:45 <DracoOmega> SamB: Besides, I can assure you that 99.9% of people running windows tiles crawl are not running it from the commandline
18:16:58 <ontoclasm> skeletons made by someone who doesn't understand bone structure :(
18:17:02 <HangedMan> and can't you just wait to have to redraw all of these sekeletons instead of accepting weird blurry monstrosities
18:17:11 <HangedMan> like wow that troll
18:17:15 <DracoOmega> Well, one or two of them are kind of okay, maybe
18:17:19 <DracoOmega> The troll is scary, though
18:17:21 <SamB> mumra: that happens with Windows' own console windows?
18:17:36 <mumra> SamB: i quite often test things by using printf(..) and watching the command line window because it's easier to see than dprf
18:17:49 <mumra> SamB: that's what i mean by "standard command line"
18:18:03 <mumra> i think? i mean the normal "Command Prompt" app
18:18:19 <ontoclasm> i have to put them in though because how could we get by without different turtle / spiked turtle skeletons
18:18:41 <DracoOmega> I can think of a few ways we could...
18:18:53 <HangedMan> people might confuse alligator skeletons with alligator snapping turtle skeletons!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
18:18:59 <SamB> I thought tiles was built as a "graphical" .exe, which is not permitted to inherit any console handles from it's parent process
18:19:04 <HangedMan> otherwise I mean
18:19:41 <DracoOmega> There are some mistakes I believe we are allowed to let our users make
18:19:45 <mumra> SamB: hmm, maybe there could be a difference between release builds and "make TILES=y" ?
18:21:24 <mumra> but i'm not really sure that there's a distinction like that between command line vs graphical apps
18:22:30 <DracoOmega> Why must there be so many different places where monster movement is executed?
18:22:45 <DracoOmega> Apparently this problematic ally isn't even USING the same function a bunch of other things seem to be. For... some reason?
18:23:02 * SamB checks if the PECOFF spec specifies the meaning of that header flag in any detail
18:23:28 <DracoOmega> I had this same problem before. Maybe I should make a list of all the different places in code that this can happen
18:23:34 -!- magicpoints has quit [Ping timeout: 256 seconds]
18:25:19 -!- scummos has quit [Ping timeout: 260 seconds]
18:29:36 -!- Zifmia has quit [Ping timeout: 245 seconds]
18:32:10 <SamB> mumra: ah, it's the Optional.Subsystem field in the PE header
18:32:30 <SamB> IMAGE_SUBSYSTEM_WINDOWS_GUI vs. IMAGE_SUBSYSTEM_WINDOWS_CUI
18:33:36 <HangedMan> mumra: so uh hypervaults can leave monsters in walls if they're bands and they end up in the border of a given vault
18:33:51 <HangedMan> does the next vaults incarnation fix this
18:34:14 <MarvinPA> newhypervaults?
18:34:24 <mumra> HangedMan: hmm, not sure what would cause that, does this not happen with normal vault placement?
18:35:17 <mumra> MarvinPA: the next version drops the word 'vaults' ;)
18:35:41 <HangedMan> mumra: well it used to but that was fixed
18:36:08 <DracoOmega> The monsters are actually IN the walls?
18:36:11 <mumra> perhaps the fix is not in the code path that i use for placing vaults
18:36:13 <HangedMan> yeah
18:36:22 <HangedMan> saw it with an ugly thing band
18:36:41 <mumra> because i have to use a slightly circitous route to get them to place exactly where i want
18:36:43 <DracoOmega> Would that imply that the walls were drawn AFTER monster placement or something? Because surely the game wouldn't place a monster into a wall that already existed, would it?
18:36:49 -!- Datul has quit [Quit: Page closed]
18:37:08 <mumra> the only time walls are drawn in a place where monsters could theoretically *already* have placed
18:37:12 <mumra> is in "open" areas
18:37:31 <HangedMan> well as part of dracoomega's requests I've made most of the vaults try to place a band
18:37:34 <mumra> i.e. where rooms are placing in big floor areas, rather carving into the walsl
18:38:13 <mumra> really it sounds like a bug in monster placement rather than anything to do with my lua
18:38:24 <HangedMan> probably
18:38:35 <mumra> if a vault places a band it should only place it within the vault's bounds
18:38:41 <HangedMan> yeah
18:38:43 <mumra> it sounds like it's placing stuff outside of those bounds
18:39:06 <|amethyst> yes, that's what it would do
18:39:19 <|amethyst> it just cares about the monsters being in LOS of the band leader
18:39:55 <mumra> aha ... i see what's going on then
18:40:03 <|amethyst> with normal placement this is handled by placing walls first then monsters
18:40:12 <|amethyst> but that doesn't help if walls of adjoining vaults aren't there yet
18:40:17 -!- lexackson has quit [Quit: Leaving]
18:40:18 <|amethyst> s/walls/features/
18:40:48 <mumra> |amethyst: yes, the current V code is placing the vault before the walls
18:40:55 <mumra> i don't know why it was this way around
18:41:05 -!- Arkaniad|Laptop_ has quit [Quit: I've killed the senator.]
18:41:19 <mumra> the good news is that in v3 i already switched this around, mostly for code sanity than to fix anything i was aware of
18:41:20 <|amethyst> maybe something to do with connectivity?
18:41:25 <|amethyst> ah
18:41:45 <mumra> no, nothing to do with connectivity, it was just the order i wrote things in, didn't think it would make any difference either way
18:41:58 <DracoOmega> mumra: So does that mean this bug is already fixed in some new version you're working on?
18:42:08 <mumra> DracoOmega: yes, that's right ;)
18:42:13 <DracoOmega> Well, isn't that convenient? :)
18:42:23 <mumra> (well, assuming the los check is working, and there's nothing else i'm not aware of!)
18:42:53 -!- Zifmia_ is now known as Zifmia
18:42:55 <mumra> it's actually really easy to fix in the current version, just move vault placement to after the big loop where walls are drawn
18:43:04 <DracoOmega> Well, band placement code seems robust elsewhere
18:43:28 <DracoOmega> These modifications to the room vaults is so that Vaults won't have so many band monsters spawning without their bands, since several of them work poorly on their own
18:43:49 <|amethyst> DracoOmega:  it does?  pure luck I assure you :)
18:44:07 <DracoOmega> |amethyst: Haha. Well, I haven't seen or heard of any problems, so.... :0
18:44:10 <DracoOmega> :)*
18:44:40 <mumra> DracoOmega: yes, it's a good idea, i discussed it before i think
18:45:09 <mumra> really it would be nice if '0' would automatically place bands given enough space, since i can't think of a case where this wouldn't be wanted
18:45:30 <DracoOmega> Well, some vaults use highly clustered 0s and it might be kind of bad there
18:45:42 <mumra> "given enough space"
18:45:56 <DracoOmega> Well, I think it's hard to algorithmically define that, since we're not just talking about 'physical room to place the monster'
18:45:57 <mumra> (although how that would be defined is arbitrary, yes)
18:46:17 <DracoOmega> Since you often want a little breathing room around them (but sometimes not, if they're in enclosed rooms, say)
18:46:39 <HangedMan> 0 should just stay 0 because always placing a band throws a lot of old vaults out of whack and most of the time the point is a mix of random floor spawns instead of just a regular floor band
18:46:42 <mumra> you could always use "0 individual" to veto bands in specific places
18:46:55 <mumra> or "0 lone" to save keypresses
18:47:43 <HangedMan> also while 0 is a bit variable in difficulty 0 band would go crazy with variability and make conjurations even better for clusters than it needs to be and stuff
18:48:13 <HangedMan> anyway general disapproval I have to go
18:48:19 -!- HangedMan has quit [Quit: leaving]
18:49:35 -!- TacoPrincess has quit [Ping timeout: 260 seconds]
18:55:38 -!- nicolae- has joined ##crawl-dev
18:56:27 <SamB> so, ... can we make him vault dev without his consent?
18:56:41 <nicolae-> yes
18:56:58 <|amethyst> as long as we don't tell him
18:57:11 <DracoOmega> Haha
18:57:17 <|amethyst> we just run all new vaults by HangedMan and if he says okay somebody commits
18:58:12 <nicolae-> hahaha, nice
18:58:48 <nicolae-> how long would it take for him to notice, i wonder
18:59:21 <mumra> he might read the logs
18:59:25 <mumra> can they be doctored?
18:59:31 <DracoOmega> Haha
18:59:41 <nicolae-> this conspiracy goes all the way to the top
19:06:52 -!- Nakatomy has quit [Remote host closed the connection]
19:07:46 -!- tophat has quit [Read error: Connection reset by peer]
19:17:17 -!- jdpage has quit [Ping timeout: 255 seconds]
19:17:56 -!- alefury has quit []
19:18:03 -!- pelotron has quit [Ping timeout: 260 seconds]
19:19:21 -!- Crazylemon64 has quit [Quit: Crazylemon64]
19:19:28 -!- Vizer_ has joined ##crawl-dev
19:21:19 -!- Vizer has quit [Ping timeout: 260 seconds]
19:24:35 -!- eith has quit [Ping timeout: 246 seconds]
19:26:55 <ontoclasm> chainsawing complete
19:27:45 <DracoOmega> Chainsawing?
19:28:27 <ontoclasm> to get (some of) those tile submissions into shape
19:29:36 <DracoOmega> Ah
19:29:45 <DracoOmega> Well, a couple seems not too far off from acceptable
19:29:52 <DracoOmega> (And several more are... um...)
19:30:19 <ontoclasm> yeah, i ended up doing 4; the others seem a) bad and b unnecessary
19:30:52 <ontoclasm> we don't need different tiles for skeleton ogres and trolls
19:31:44 <Cheibriados> 03ontoclasm 07* 0.12-a0-2532-g7d555e8: Skeleton tiles (Aygtets, 6735) 10(5 minutes ago, 6 files, 14+ 3-) 13http://git.develz.org/?p=crawl.git;a=commitdiff;h=7d555e852b6d
19:33:33 <DracoOmega> Man, it's amazing how much nicer that winged quadraped one looks compared to the starting version >.>
19:34:11 <ontoclasm> hah
19:34:28 <DracoOmega> I think the skeletal turtle still looks kinda iffy, though, but am not sure what to suggest
19:34:30 -!- Kellhus has quit [Quit: Page closed]
19:34:56 <ontoclasm> yeah, i'm not really sure what to do with that
19:35:00 <ontoclasm> i mean
19:35:26 -!- dcssrubot772 has quit [Read error: Connection reset by peer]
19:35:34 <ontoclasm> turtles are like %90 shell, how the hell would a skeletal one look
19:36:09 <nicolae-> well, the shell is part of the skeleton iirc
19:36:24 <DracoOmega> Yes, basically
19:36:45 <ontoclasm> well yeah
19:37:08 <ontoclasm> but i mean... it'd be alike a skeletal crab or insect
19:37:20 <ontoclasm> it's already got an external skeleton
19:37:23 <DracoOmega> http://upload.wikimedia.org/wikipedia/commons/9/9a/Galapagos_Turtle_skeleton.jpg
19:37:33 <ontoclasm> like*
19:37:35 <nicolae-> but it's got an internal one too, it's just an outgrowth of the vertebrae i think
19:37:53 <ontoclasm> yeah
19:38:01 <ontoclasm> well
19:38:06 <ontoclasm> i suppose i could try
19:38:21 <DracoOmega> Well, it's probably a fairly low priority, all things considered :P
19:38:38 <ontoclasm> pf, everything i do is low-priority
19:39:00 <nicolae-> maybe smooth out the colors in the skeleton tile shell area a bit? like, lighten the cracks between the shell plates, so the shell part is smooth
19:39:26 <DracoOmega> ontoclasm: I beg to differ!
19:44:20 <ontoclasm> https://dl.dropbox.com/u/1000017/crawl/skeleton_turtle.png
19:44:22 <ontoclasm> ?
19:44:46 <DracoOmega> I think the shell looks way better
19:44:52 <DracoOmega> (Also do this with the living one :P)
19:44:58 <ontoclasm> hah
19:45:08 <nicolae-> yeah that looks cool
19:45:20 <DracoOmega> The head still looks a little awkward, I think. Maybe it's the angle? I'm not quite sure
19:45:25 -!- noppa354 has quit [Quit: Page closed]
19:45:32 <ontoclasm> yeah, they have sort of... cartoony heads
19:45:40 <nicolae-> i think the head is also staring right at the player in that one
19:45:48 <DracoOmega> "Soon."
19:46:10 <ontoclasm> really i should just go back and start with the living snapping turtle
19:48:36 <DracoOmega> Perhaps. But certainly that shell design is in the right way, in my opinion :)
19:49:06 <ontoclasm> thanks :D
19:54:48 <greensnark> Looks like an ice turtle
19:56:23 <DracoOmega> Well, still some mysterious bugs, but at least I've fixed that bit where they refuse to move around hostiles that are directly in their path
19:57:45 <nicolae-> who refuses to move around hostiles?
19:58:27 <DracoOmega> I'm working on a 'retreat' command for allies
19:58:46 <nicolae-> oh, cool
19:58:52 <DracoOmega> Because currently it is very hard to get them to do this, as they stop to attack anything that hits them
19:59:25 <nicolae-> you're like some kind of code machine
19:59:47 <DracoOmega> Haha
19:59:54 <DracoOmega> Well, thanks, I think :)
20:00:33 <nicolae-> well it's like every week you've coded another bunch of features
20:01:29 <BlastHardcheese> fr code felids that are not crap
20:02:16 <DracoOmega> Well, I think it would be kind of hard to make people who dislike those like them, without changing something fundamental about what the race is
20:03:13 <elliptic> hm, I bet DTEEs are dangerous for felids
20:03:42 <DracoOmega> Yes, they might be. I'm not sure their LRD does a whole lot of damage when using plain rock, but they also have no AC
20:03:51 -!- Zermako has quit []
20:05:06 <elliptic> my only real complaint about felids is how awful the non-caster non-berserker starts are
20:05:18 <elliptic> because you have no HP and no ranged attacks
20:05:34 <elliptic> I don't really have an idea for how to fix this though
20:05:50 -!- kunwon1 has quit [Read error: Operation timed out]
20:06:05 <BlastHardcheese> that and most of the stuff in the game being irrelevant to them
20:06:22 <BlastHardcheese> which makes every felid feel like the same character
20:07:17 <elliptic> that's an exaggeration; felid of trog feels a lot different from hybrid felid feels a lot different from pure caster felid
20:07:31 <SamB> what does hybrid do?
20:07:45 <elliptic> die a lot trying to melee things with buffs on
20:07:45 <DracoOmega> elliptic: I wonder if that stabbing skill merge might make felid melee starts feel less terrible?
20:08:13 <elliptic> DracoOmega: nah, you will not get enough stealth stabs to matter early game
20:08:24 <DracoOmega> Yeah, I sort of suspected so
20:09:03 <DracoOmega> Give felids spriggan hp? :P
20:09:21 <SamB> felids should start with decent stealth ...
20:09:21 <elliptic> giving them more HP would of course help, but would also make them feel less unique
20:09:37 <nicolae-> doesn't "not able to do most things" make the unique already though
20:09:40 <nicolae-> them
20:09:56 <DracoOmega> Well, I think they'd still feel quite unique with the various other bonuses and restrictions they have
20:10:16 <DracoOmega> And I'm not convinced that an hp buff on that order would make them too strong, either
20:10:24 <DracoOmega> But I do see your point
20:10:34 <SamB> cat's are generally known to walk pretty quietly
20:10:43 <DracoOmega> They do have good stealth
20:11:08 <DracoOmega> It's just that even good stealth races can't really rely on lots of stealth openers early on
20:11:08 <SamB> and not get enough stabs
20:11:13 <elliptic> SamB: they have great stealth, the thing is just that stealth stabs are not at all reliable until you get endgame-quality stealth
20:11:31 <ontoclasm> https://dl.dropbox.com/u/1000017/crawl/snapping_turtle.png ?
20:11:42 <SamB> hmm
20:11:48 <nicolae-> cool turtle
20:11:50 <DracoOmega> I, uh... I think he's too flat
20:11:56 <elliptic> stabbing 20% of creatures isn't particularly significant; stabbing 90% is
20:11:58 <nicolae-> though there is that, yes
20:12:10 <DracoOmega> And perhaps not differentiated enough, color-wise (looks a bit like a lizard to me?)
20:12:13 <ontoclasm> well, snapping turtles are flat eye are ell
20:12:26 <SamB> you misspelled that
20:12:30 <DracoOmega> (I still like the shell design from the skeleton more :P)
20:12:30 <ontoclasm> but yeah, i guess i could plump him up a bit
20:12:54 -!- kunwon1 has joined ##crawl-dev
20:13:58 <ontoclasm> DracoOmega: oh? i did them the same way
20:14:07 <elliptic> I'd be interested in what kilobyte thinks about felid early game
20:14:07 <ontoclasm> maybe it didn't come out as well this time
20:14:11 <DracoOmega> Really? They look fairly different to me
20:14:17 <elliptic> since he is the felid design expert
20:14:28 <DracoOmega> It may have to do with size and proportions
20:14:35 <ontoclasm> maybe
20:16:16 <DracoOmega> Hmmm... perhaps I should take a break from this for now. Seems there is lots more fiddly bits left
20:16:34 <DracoOmega> I should probably leave some of them until I can look at them with a clearer head :P
20:17:13 -!- Pthing has quit [Read error: Connection reset by peer]
20:17:49 <DracoOmega> So... today I added a half-dozen things to my to-do list, while completing half of one from it :)
20:18:19 <ontoclasm> learn add #crawl_dev
20:18:54 -!- Cryp71c has joined ##crawl-dev
20:19:10 <DracoOmega> At least some of those other things will be quicker/easier
20:19:13 <DracoOmega> (So he says)
20:22:39 -!- magistern has quit [Quit: This computer has gone to sleep]
20:29:18 <ontoclasm> ugh, i give up
20:29:24 <ontoclasm> turtles are mpossible to draw
20:29:28 -!- lobf has quit [Quit: lobf]
20:31:53 <ontoclasm> https://dl.dropbox.com/u/1000017/crawl/snapping_turtle2.png
20:31:59 <ontoclasm> :/ not happy about that
20:32:03 <nicolae-> what if you drew a turtle... WITH a turtle?
20:32:14 <ontoclasm> using TURTLE, the program
20:32:17 <nicolae-> yes exactly
20:32:24 <nicolae-> RT 90 FD 100
20:33:42 <Cryp71c> Anyone particularly familiar with the map stuff?
20:33:55 <Cryp71c> map_knowledge.h in particular?
20:34:35 -!- Residualshade has quit [Client Quit]
20:40:02 -!- dtsund has joined ##crawl-dev
20:54:44 -!- dcssrubot9 has quit [Ping timeout: 240 seconds]
20:55:25 -!- santiago has quit [Read error: Connection reset by peer]
21:06:36 -!- ussdefiant_ has joined ##crawl-dev
21:07:56 -!- ussdefiant has quit [Ping timeout: 245 seconds]
21:08:02 -!- Wehk has quit [Read error: Connection reset by peer]
21:08:02 -!- ussdefiant_ is now known as ussdefiant
21:08:42 -!- mikee_ has joined ##crawl-dev
21:12:39 -!- dtsund has left ##crawl-dev
21:16:51 -!- dtsund has joined ##crawl-dev
21:19:42 -!- dtsund has left ##crawl-dev
21:20:44 -!- st_ has quit [Ping timeout: 252 seconds]
21:25:01 <mumra> DracoOmega: i had a similar few days. came up with masses of ideas for hyper layouts but go no further with implementation ...
21:25:16 <mumra> ... but started two other projects that have absolutely nothing to do with hyper ...
21:26:16 <mumra> (well, technically the procedural layout stuff i'm doing can be combined with hyper layouts but it's basically completely separate)
21:32:31 -!- MaxFrost has quit [Ping timeout: 245 seconds]
21:32:46 -!- monqy has joined ##crawl-dev
21:35:25 -!- Illusionbear has quit [Quit: Illusionbear]
21:37:01 -!- bonghitz has quit [Remote host closed the connection]
21:39:05 -!- Taco_Princess has quit [Changing host]
21:56:03 -!- kek has quit [Remote host closed the connection]
21:56:30 <DracoOmega> ontoclasm: Oh, by the way (I was away for a bit) I do like that second turtle a fair bit more. The only real issue I think is that it's kinda small
21:57:00 <DracoOmega> (Really sorry for always quibbling here)
21:57:06 -!- Sticking has quit [Ping timeout: 252 seconds]
21:59:27 <SamB> needz moar pik-celz?
21:59:29 -!- pelotr0n has quit [Ping timeout: 248 seconds]
22:02:27 -!- Sequell has quit [Read error: Operation timed out]
22:03:38 -!- varmin has quit [Ping timeout: 272 seconds]
22:04:06 -!- shirish has joined ##crawl-dev
22:07:31 -!- LoremIpsum has quit [Ping timeout: 245 seconds]
22:08:28 -!- dcssrubot426 has quit [Ping timeout: 252 seconds]
22:08:52 -!- nicolae- has left ##crawl-dev
22:09:59 -!- iasov has quit [Read error: Operation timed out]
22:10:21 -!- Taco_Princess has joined ##crawl-dev
22:11:09 -!- Mhcerri has joined ##crawl-dev
22:12:19 -!- varmin has joined ##crawl-dev
22:12:57 <mumra> SamB: by the way that perlin fix definitely seems to be working
22:13:51 <Mhcerri> Hi, I'd like to set up a local server, but I couldn't find docs about that. Does anyone know where I can find more info about that?
22:14:41 <Mhcerri> I'm interested in both ASCII ans tiles version, with both sharing the same save files if possible
22:15:09 <DracoOmega> I think |amethyst is the best person to talk to about that, but I don't know that he's around at present
22:15:22 <SamB> Mhcerri: this is certainly a good place to ask, but you might need to be patient
22:15:44 <DracoOmega> He's the one who set up and maintains CSZO
22:15:51 <Mhcerri> For example, crawl.develz.org has a nice initial menu to choose the game version and to edit configs when you login through shh/telnet... I couldn't find the code of it (not sure if it is in the git repo)
22:16:12 <|amethyst> Mhcerri:  That stuff is in the dgamelaunch-config repository
22:16:19 <mumra> Mhcerri: the webtiles stuff is in the crawl source in crawl-ref/source/webserver
22:16:20 -!- Gretell has quit [Ping timeout: 248 seconds]
22:16:42 <|amethyst> Mhcerri:  https://github.com/greensnark/dgamelaunch-config
22:17:11 <|amethyst> Mhcerri:  CSZO and CAO have more additions to that config that I haven't published yet
22:17:23 <|amethyst> hm
22:17:35 <Mhcerri> mumra: right, but does that include account management and things like that?
22:17:47 <|amethyst> oh wait
22:18:05 <|amethyst> https://github.com/neilmoore/dgamelaunch-config
22:18:06 -!- broquaint has quit [Ping timeout: 264 seconds]
22:18:09 <SamB> there's also dgamelaunch itself, isn't there ?
22:18:11 <Mhcerri> amethyst: hmm great. Let me take a look at it
22:18:11 <|amethyst> I did push my stuff
22:18:33 <mumra> Mhcerri: it includes everything you see in the webtiles interface including registration
22:18:45 <|amethyst> Yeah, you'll need dgamelaunch.  We have a patched version at http://git.develz.org/?p=dgamelaunch.git;a=summary
22:19:01 <|amethyst> I also have another patch for dgamelaunch at  http://dobrazupa.org/0001-Properly-salt-passwords.patch
22:19:09 <|amethyst> which is necessary for the CAO/CSZO configuration
22:19:22 <|amethyst> There's still a bunch of other stuff you'll need to do
22:19:44 <|amethyst> Create a user or two, set up a chroot, create the appropriate directories, etc
22:20:06 <|amethyst> I haven't documented that stuff yet but can give advice
22:20:28 -!- eb has quit []
22:20:52 -!- shirish has quit [Remote host closed the connection]
22:21:25 <Mhcerri> mumra amethyst: and all these scripts and tools are designed to run on a Linux box, right?
22:22:02 <SamB> it's possible they might be usable on other *nixen
22:22:20 <|amethyst> Mhcerri:  yeah... some of it you might be able to get working on OS X or another Unix, but IIRC webtiles had some Linux dependencies
22:22:32 <|amethyst> edlothiol is the person to talk to about webtiles
22:22:45 <SamB> especially if their names start with Debian/ ...
22:22:49 <|amethyst> in particular, he has a repo for a patched version of python-tornado
22:22:50 <SamB> er
22:22:53 <SamB> Debian GNU/
22:23:16 <|amethyst> Mhcerri:  I put a few of my notes in http://dobrazupa.org/setting-up-dgamelaunch-and-webtiles    but it might refer to repositories that don't (yet) exist
22:24:34 <Mhcerri> samb amethyst: my intention is to run in a Linux machine... I will first try an arm (but I don't have much hope on that), and if that doesn't work I will try a x86 machine that I have
22:24:36 <|amethyst> In particular, the dgamelaunch.git repo mentioned in that file doesn't exist (use the one from git.develz.org plus my patch)
22:25:18 <|amethyst> Mhcerri:  It should be possible, but maybe webtiles will need some tweaks
22:25:24 -!- Sequell has joined ##crawl-dev
22:28:05 <Mhcerri> I noticed that CDO's ASCII and webtiles versions do not share the same saves between them... Do you know if that is possible?
22:29:09 <SamB> there wouldn't be any reason for the save formats to be fundamentally incompatible
22:29:37 <elliptic> Mhcerri: CAO and CSZO share saves between console and webtiles
22:29:54 <elliptic> CDO is just older, and that wasn't possible when it was originally set up
22:30:25 <Mhcerri> elliptic: I didn't know that
22:30:32 <SamB> were the save formats actually incompatible back then?
22:30:56 <SamB> (due to divergent hacking, I would guess?)
22:31:03 <elliptic> not sure
22:32:33 <SamB> I suppose just having builds out of sync might be a problem, though, for trunk ...
22:32:46 <|amethyst> that, for one thing
22:32:58 <Mhcerri> Great!! I think that is pretty much what I needed to take the first step. I will give a try and start on it. Thank you very much for your help, guys
22:33:25 <|amethyst> and, while there was infrastructure in place to support multiple versions of trunk in webtiles, edlothiol had to do a bit more hammering on it for CSZO
22:35:22 -!- geedmat has quit [Quit: Page closed]
22:37:03 <|amethyst> Mhcerri:  And if you want to add anything to that "setting-up-dgamelaunch-and-webtiles" document, please do so and send me the updates
22:37:40 <|amethyst> Mhcerri:  thanks for the interest
22:37:43 <Mhcerri> amethyst: sure. No problem. Thanks again
22:38:31 -!- monqy has quit [Quit: hello]
22:38:48 -!- bh has joined ##crawl-dev
22:38:52 -!- Mhcerri has left ##crawl-dev
22:40:01 -!- ussdefiant has quit [Ping timeout: 245 seconds]
22:40:46 <bh> moin
22:40:47 <Henzell> bh: You have 3 messages. Use !messages to read them.
22:40:51 <bh> zounds.
22:40:53 <bh> !messages
22:40:54 <Henzell> (1/3) mumra said (11h 8m 40s ago): I've been messing around with noise using perlin.cc and I'm noticing something strange: the x coord doesn't seem to make any difference (in perlin::fBM(x,y,z))
22:41:02 <bh> !messages
22:41:03 <Henzell> (1/2) mumra said (10h 51m 30s ago): another problem: for very high values of x, y or z, perlin::noise(x,y,z) always returns 0 (obviously a problem when i'm using abyss coords and offset for x,y,z!)
22:41:09 <bh> !messages
22:41:10 <Henzell> (1/1) SamB said (10h 11m 21s ago): It's not at all clear to me what fBM is or what it's doing in perlin.cc; it's not in http://webstaff.itn.liu.se/~stegu/simplexnoise/SimplexNoise.java ...
22:41:44 <mumra> hi
22:42:09 <mumra> SamB pushed a fix for it, basically fastfloor was truncating the values
22:42:13 -!- ussdefiant has joined ##crawl-dev
22:42:22 <bh> oh, great
22:42:27 <bh> I can add a comment to fBM
22:43:31 <SamB> yeah, I replaced a few occurances of "int" with "long long" ... but I guess you can look at the commit for the details
22:43:47 <mumra> i basically see what fBM is doing (although the octaves don't seem to be producing exactly the sort of variation i was hoping for)
22:44:48 <mumra> bh: how do you think long long will affect performance? basically i'm working on a ProceduralLayout that uses perlin to generate geography
22:44:55 <bh> what kind of variation do you want?
22:44:59 <mumra> so it's calling perlin quite a lot of times per ProceduralSample
22:45:13 <bh> SamB: why long long instead of uintXX_t?
22:45:32 <bh> mumra: how rapidly does the terrain change? How well can you bound the morphing?
22:45:47 <mumra> bh: crinkly mountainous stuff; even with several octaves, perlin is producing smooth rounded areas
22:46:19 <mumra> but i am still experimenting, i can combine different perlin scales manually to achieve what i'm after, or even get worley involved
22:46:19 <bh> mumra: oh! that sounds like a scale issue
22:46:35 <SamB> bh: I think I was going to use something from <cmath> that specifically returned "long long" but changed my mind when I saw how it optimized
22:46:40 <mumra> well, at low scales there is high variation across the whole map, at high scales there are big smooth areas
22:46:43 <SamB> feel free to switch it over
22:46:53 <mumra> so i need to combine a high scale and a low scale i think
22:47:24 -!- Arkaniad has quit [Remote host closed the connection]
22:48:04 <bh> how many octaves are you calling fBM with?
22:48:12 <mumra> i tried up to 10
22:48:13 <bh> take a look at this: http://www.iquilezles.org/www/articles/warp/warp.htm it might give you some ideas
22:48:24 <SamB> mumra: note that perlin.cc actually implements simplex noise, which isn't called Perlin because that was already taken ...
22:48:43 <mumra> yeah i only noticed that after i saw your fix
22:48:50 <bh> same idea, better dimensionality
22:49:06 -!- broquaint has joined ##crawl-dev
22:49:06 <mumra> yeah, simplex works in a very similar way
22:49:39 <mumra> mmm, that's a nice distortion image
22:50:19 <SamB> mumra: I added a comment about that in a slightly earlier commit, yeah
22:51:35 -!- dcssrubot413 has quit [Read error: Connection reset by peer]
22:52:09 -!- GuraKKa has quit [Quit: Leaving.]
22:52:10 <bh> mumra: here's some non-crawl procedural generation porn: http://i.imgur.com/uGKjAfO.png
22:52:25 <mumra> bh: this is the kind of thing i'm doing - http://pbrd.co/VxCTNl - http://pbrd.co/VxD3Em
22:52:57 <mumra> there are a bunch of different perlin layers defining "height", "hot", "wet", etc. and then the terrain is picked based on combinations of those values
22:53:03 <bh> that's super pretty looking
22:53:17 <bh> lots of doors though :)
22:53:25 -!- blackcustard has quit [Read error: Connection reset by peer]
22:53:54 <bh> are you taking the unmodified x/y coordinates and feeding them to fBM?
22:54:04 -!- ToastyP has quit [Quit: ChatZilla 0.9.90 [Firefox 19.0/20130215130331]]
22:54:18 <mumra> where the doors are are supposed to be citified areas, but it's pretty difficult to get doors to place sensibly
22:54:36 <mumra> bh: i'm multiplying the x/y coordinates by a given factor and offsetting them
22:54:40 <bh> you could do it in post
22:54:44 <mumra> the multiplier and offset are different for each layer
22:54:59 <bh> you probably want to divide them, unless your factor is less than one :)
22:55:39 <mumra> the factors are floats and usually < 1 (except for the "jitter" map which is supposed to be extremely random)
22:55:52 <mumra> probably best if you see the code: http://pastebin.com/1imNjN1i
22:56:13 <mumra> it's a huge mess right now but i've been experimenting with a lot of things ;)
22:56:37 -!- dtsund has joined ##crawl-dev
22:57:49 -!- Kalir has quit [Changing host]
22:57:52 <bh> it's finnicky to get this stuff right. You might want to try rendering your layout in processing first, or using kilo's dungeon layout tester thing
22:57:57 <bh> it'll give you quicker feedback
22:57:58 <mumra> bh: i also wanted some advice on how to set the changepoint, basically i have no idea what the value means or what sort of ranges are sensible
22:58:19 <bh> mumra: for Perlin noise, it's really hard. That's why I use worley noise so extensively
22:58:32 <mumra> bh: well it's pretty quick to compile, start the game and hit &Ctrl-R&{y
22:59:16 <mumra> hmm, good points
22:59:48 <mumra> (although i had troubles compiling kilobyte's layout tester...)
23:00:11 <bh> worley is easy. If you're in a cell, and the distance to the nearest cell is d, you know that you can never change cells before depth has increased by d
23:00:37 -!- Poncheis has quit [Quit: ChatZilla 0.9.90 [Firefox 12.0/20120420145725]]
23:01:30 <mumra> about worley ... why are there two values in distance[] ?
23:01:50 <bh> dist[0] = distance to the nearest point, d[1] = distance to the next nearest point
23:02:07 <mumra> ohh
23:02:29 <mumra> and what do i use as my actual pseudorandom value?
23:03:41 <bh> hm?
23:03:53 <mumra> i mean, i want a pseudorandom noise function to return a pseudorandom value
23:04:00 <mumra> do i use distance for that?
23:04:06 <bh> id
23:04:34 <mumra> right, the comment for id confused me and made me think it had some other purpose
23:05:12 <mumra> also, why is there more than one id?
23:05:29 <bh> like distance: id[0] = nearest cell, id[1] = next cell
23:05:57 <mumra> so id[0] will always return a value based on the nearest cell? it doesn't change smoothly over the domain?
23:06:10 <bh> yep, it will change discontinuously
23:07:25 -!- dcssrubot455 has quit [Read error: Connection reset by peer]
23:07:34 <mumra> so when i see a worley image like this: http://www.google.co.uk/imgres?imgurl=http://upload.wikimedia.org/wikipedia/commons/thumb/0/00/Worley.jpg/200px-Worley.jpg&imgrefurl=http://en.wikipedia.org/wiki/Worley_noise&h=200&w=200&sz=6&tbnid=WZoPlCuSlQnplM:&tbnh=90&tbnw=90&zoom=1&usg=__XcELL_CvtvnjZ23FY_UDphcjMl8=&docid=ahPwkaTONLoL9M&sa=X&ei=2ns1UYu6KIiEOIYs&ved=0CEUQ9QEwAw&dur=2338
23:07:56 <mumra> which appears to vary continously, is that not the same thing? or am i just not understanding something?
23:08:09 <bh> ah. In that case they're using a color function based on distance
23:08:42 <bh> http://graphics.ucsd.edu/courses/rendering/2006/agoldberg/CSE168Files/Worley.jpg
23:09:22 <bh> In this picture, Worley is doing something like: "If dist[1] - dist[0] < 0.1 then white else pickColor(id[0])"
23:10:14 -!- shirish has joined ##crawl-dev
23:10:55 <jeanjacques> Napkin: my cdo trunk game needs to be recovered (or terminated)
23:12:01 <elliptic> cdo seems to be having issues in general at the moment
23:12:24 <mumra> right, yes; this is why i'm using perlin - it gives me a value i can use straight away ;)
23:12:33 <elliptic> the session freezes if I press w to watch games in progress
23:12:49 <elliptic> and footv is stuck waiting for a ttyrec from cdo
23:12:52 <mumra> but with this info i can use worley as well, should get good results
23:13:41 <mumra> s/perlin/simplex but whatever
23:13:51 <|amethyst> elliptic:  I think napkin was tweaking a few scripts so that Sequell wouldn't make apache generate as many directory listings
23:13:58 <|amethyst> elliptic:  could be related to that
23:13:58 <bh> mumra: I think perlin is better for things that are invariant based on depth. Otherwise computing the changepoint is just hard, and then you regenerate gobs of terrain on every turn
23:14:16 <|amethyst> elliptic:  or writing a new script or something
23:14:23 <elliptic> yeah
23:14:45 <bh> mumra: If you're interested in this stuff more generally, find a copy of Texturing & Modeling: A Procedural Approach
23:14:55 <elliptic> this behavior with hanging on w has happened previously, though not recently... but I forget the explanation for what had happened
23:14:56 <mumra> bh: i have it varying very little by depth, i just don't understand what the changepoint means.
23:15:28 <bh> The return value is an abyss depth. It says "This square will not change before we reach this depth"
23:15:29 <|amethyst> mumra:  in the abyss, you'd like to know "what is the next point in time at which this grid might change"?
23:15:51 <|amethyst> mumra:  otherwise you have to recompute every grid cell every turn
23:15:55 <bh> 5600 cells.
23:16:01 <mumra> bh: it's something i've been interested in for years, but when i've done stuff previously i've only messed around with perlin, so i just understand how to get results from it quickly
23:16:50 <mumra> |amethyst: i got that part. but i return the value of 10 and cells are changing a lot more frequently than every 10 turns.
23:16:50 <bh> mumra: amitp has a bunch of handy stuff too -- www-cs-students.stanford.edu/~amitp/gameprog.html
23:17:20 <bh> mumra: it's an absolute depth, not a relative depth. You're responsible for determining your lower bounds
23:17:44 -!- ruwin has quit []
23:18:00 <mumra> so does offset + 10 mean "in 10 turns time" ?
23:19:20 <bh> yes. I should document all this :) I only half expected anyone to write a generator
23:19:20 <|amethyst> I'd guess it means in 10 abyss-"depth"-units
23:19:26 <|amethyst> oh, it is turns?
23:19:32 <bh> |amethyst: depth units
23:19:53 -!- MarvinPA_ has joined ##crawl-dev
23:20:08 <|amethyst> mumra:  see increase_depth for how time turns into depth... it's not linear at all
23:20:51 <|amethyst> mumra:  I thought I added some comments when I wrote that 0.2 + 2.8 * sin(theta/2)^10 formula
23:20:58 <|amethyst> but apparently they've gone missing
23:21:25 -!- MarvinPA__ has joined ##crawl-dev
23:21:40 <mumra> yeah i guessed they would vary, but if i have an approximation i can at least experiment and find a value that works
23:21:46 <bh> |amethyst: I probably did that while I was ripping up the tracks
23:21:56 -!- Naruni has quit [Remote host closed the connection]
23:22:17 <SamB> bh: perhaps you'd better craft some new ones
23:22:17 <bh> mumra: You'll want to find it analytically, I'm afraid. If you overestimate the value you'll get jumps on loading a game
23:22:35 -!- MarvinPA has quit [Ping timeout: 260 seconds]
23:22:35 -!- MarvinPA__ is now known as MarvinPA
23:23:57 <|amethyst> mumra:  when it was a double, one tick was 0.002 to 0.007 depth units
23:24:10 -!- Naruni has joined ##crawl-dev
23:24:27 <|amethyst> mumra:  (well, that's an average for a given abyss speed: top speed was three times that)
23:24:34 -!- MarvinPA_ has quit [Ping timeout: 245 seconds]
23:24:48 <|amethyst> mumra:  but it was rescaled when integerized
23:25:22 <|amethyst> oh, rescaled by 200 it looks like
23:25:31 <mumra> hmm, could all be massively problematic with this generator
23:25:50 <mumra> for now i might remove the variance by depth altogether
23:26:05 <bh> mumra: yeah. none of my layouts vary by depth unless they're using worley noise
23:27:54 -!- dtsund has quit [Ping timeout: 276 seconds]
23:29:32 <|amethyst> DracoOmega:  they only know about in-sight stuff, though
23:29:34 <|amethyst> doh
23:30:32 <bh> |amethyst: I wish there was a way to remove abyss maprot. alas.
23:31:34 <MarvinPA> for testing?
23:32:11 <MarvinPA> you can with &{ or &}, i forget which is which
23:32:15 <MarvinPA> unless you mean removing it for real
23:32:33 <mumra> bh: it's a shame because really awesome things happen as it's shifting ... but i guess it'll still be good with everything else shifting anyway
23:32:49 <mumra> MarvinPA: that reveals the map but it starts disappearing as soon as you move or anything
23:32:59 <MarvinPA> oh i thought that removed it, huh
23:33:22 <mumra> what would be really useful for abyss testing is like "perma-mapping" so everything outside of LOS is revealed every turn
23:33:28 <bh> MarvinPA: for real
23:33:35 <|amethyst> It removes unmappability, but there's still rot
23:33:44 <|amethyst> mumra:  x-ray mode
23:34:44 <mumra> more like satellite imagery but yeah ;)
23:34:47 <|amethyst> just hack it into you.can_see :)
23:35:05 <|amethyst> that's how arena works, right?
23:37:24 <mumra> see_cell rather than can_see presumably
23:38:05 <ontoclasm> https://dl.dropbox.com/u/1000017/crawl/snapping_turtle2.png https://dl.dropbox.com/u/1000017/crawl/snapping_turtle_alligator.png
23:38:33 <bh> |amethyst: quick, what the critical probability for 2d percolation?
23:38:36 <bh> what's
23:39:02 <|amethyst> I dunno, I don't drink coffee
23:39:13 <bh> ontoclasm: the head of turtle2 reminds me a bit of a rodent
23:40:21 <ontoclasm> hm
23:40:25 <DracoOmega> Well, one definite plus is that the two look less similar to each other
23:40:34 <mumra> |amethyst: that hack resulted in "ERROR: range check error (-19 / 17)"
23:40:53 <mumra> |amethyst: i am guessing arena gets away with it because it's not abyss
23:40:53 <|amethyst> oh right
23:41:17 <|amethyst> arena gets away with it because the player doesn't move
23:41:17 -!- MarvinPA has quit [Ping timeout: 245 seconds]
23:41:22 <mumra> right
23:41:23 <|amethyst> and other stuff
23:41:32 <|amethyst> 17 is the diameter of LOS
23:41:40 <|amethyst> there's some cache or another, I forget
23:41:55 <mumra> also, it didn't actually work anyway, it still only showed the LOS circle before the crash
23:42:14 <|amethyst> figured as much
23:42:26 <mumra> i can just call magic_mapping every turn anyway
23:42:32 <|amethyst> there are checks for arena all over the place
23:42:47 <|amethyst> I guess that works
23:45:34 <mrwooste1> a
23:46:44 -!- mrwooste1 has quit [Quit: leaving]
23:48:06 -!- Cryp71c has quit [Ping timeout: 264 seconds]
23:49:43 -!- hurdos has quit [Ping timeout: 245 seconds]
23:52:21 -!- monqy has joined ##crawl-dev
23:55:37 <mumra> bh: what would i set changepoint to if the layout is constant? i'm seeing offset + 4096 in some layouts, is that the recommended option?
23:55:52 -!- superc has quit [Ping timeout: 245 seconds]
23:56:13 <bh> mumra: That's one way to do it. I picked a number so large that it's unlikely the square will be reevaluated before an abyss shift.
23:56:34 <bh> |amethyst: do you think it's worth adding a sample cache for the case where a player runs back and forth provoking abyss recenterings?
23:57:25 <DracoOmega> Would this require the playing moving back and forth across the same, fairly wide, swath of terrain?
23:58:04 <DracoOmega> In which case it's a really lousy thing for them to ever be doing, so it doesn't seem worth optimizing for it unless it's simple
23:58:29 <SamB> it seems likely that an attemt to optimize that would only end in tears
23:58:43 -!- dtsund has joined ##crawl-dev
23:58:55 <bh> we have a huge improvement over "Evalute 5600 features every turn!"
23:59:25 <bh> mumra: If you can't put any bound on something, I think it's still safe to use (changepoint = offset + 1) as long as it's sufficiently rare
23:59:57 <bh> Your bound doesn't even need to be tight. If you're sufficiently conservative things should be fine