--- Log opened Wed Dec 28 00:00:03 2011
--- Day changed Wed Dec 28 2011
00:00 <+elliptic> they should probably all just reduce MHP
00:00 <+|amethyst> what about mutation rotting of undead?
00:00 <+elliptic> that does more damage, right?
00:01 <+|amethyst> 1d3 rotting and 3 damage
00:02 <+elliptic> I guess that's intentional
00:02 <+elliptic> it isn't actually "rot" there, so being different is fine
00:03 ۰۰-- Wensley [~chatzilla@dynamic-acs-72-23-89-126.zoominternet.net] has quit [Ping timeout: 268 seconds]
00:03 <+|amethyst> hmm...
00:10 <+|amethyst> I think it's there because rotting to maxhp = 0 doesn't kill you
00:11 <+|amethyst> that can be fixed, but then you'll lose the monster attribution in zotdef
00:12 <+elliptic> hm, that makes sense
00:13 <+elliptic> it would still be good to have the more normal sources of rot made more consistent though
00:13 <+|amethyst> I think that theoretically you could be rotted to 0 maxhp without dying
00:13 <+|amethyst> currently, I mean
00:14 <+elliptic> with miasma or AF_ROT?
00:14 ۰۰-- HangedMan [63e6725f@gateway/web/freenode/ip.99.230.114.95] has quit [Ping timeout: 258 seconds]
00:14 <+|amethyst> miasma anyway
00:14 <+elliptic> actually, AF_ROT doesn't directly cause HP loss, does it?
00:14 <+elliptic> just rot status?
00:18 <+|amethyst> you have the melee damage, then a chance of getting the rot status and (if damage was > 5) getting a point of immediate rot
00:18 <+|amethyst> that point of immediate rot doesn't do damage
00:18 <+elliptic> ah
00:18 <+|amethyst> so I guess that, too, can cause you to get to 0 maxhp without dying
00:23 ۰۰-- MarvinPA [~MarvinPA@thecampbells.demon.co.uk] has quit [Ping timeout: 252 seconds]
00:26 ۰۰-- the_glow [~the_glow@host86-173-85-81.range86-173.btcentralplus.com] has joined ##crawl-dev
00:49 ۰۰-- SpaceCob [473dee4c@gateway/web/freenode/ip.71.61.238.76] has joined ##crawl-dev
00:49 < SpaceCob> question about webtiles: i have a corrupt save up there, apparently. it says "save file corrupted  - overlapping blocks" then kicks me back to the menu. how do I go about fixing that?
00:52 <+|amethyst> SpaceCob:  you can delete your save through the ssh interface
00:52 <+|amethyst> not sure about in webtiles proper
00:52 < SpaceCob> aah. i should be able to figure out how to do that
00:53 < SpaceCob> i...think.
00:53 <+|amethyst> ??cdo
00:53 <+elliptic> I don't think it is possible in webtiles, but I could be wrong
00:53 < Henzell> cdo[1/4]: Crawl server (also running development versions), located in Germany, crawl.develz.org, telnet port 345 or ssh port 22, ssh-username: crawl, ssh-key necessary: http://crawl.akrasiac.org/cao_key (openssh) or http://crawl.akrasiac.org/cao_key.ppk (putty)
00:53 <+|amethyst> elliptic:  I have a commit ready to push, but I wonder if it makes rotting less tactically relevant
00:54 <+|amethyst> I mean, it does, but I wonder if that is a bad thing
00:54 <+elliptic> |amethyst: well, I think the main tactically relevant part of rot damage right now is the 3 damage for mutating undead
00:55 <+|amethyst> getting rot while in the middle of combat and at low HP is tactically relevant
00:55 <+|amethyst> but with the change, it would not be (unless you are at low max HP too)
00:56 <+elliptic> I suppose, though we already have poison for that
00:56 ۰۰-- ussdefiant [~chatzilla@S010678cd8e7ca058.ed.shawcable.net] has quit [Ping timeout: 252 seconds]
00:56 <+|amethyst> true... and rot has long-term relevance since it's not easy to remove
00:57 <+elliptic> yeah
00:57 <+elliptic> I'd be fine with it reducing HP though
00:58 ۰۰-- ussdefiant [~chatzilla@S010678cd8e7ca058.ed.shawcable.net] has joined ##crawl-dev
00:58 <+|amethyst> hrm
00:59 < Eronarn> i'd say just make it maxHP-only, the HP loss rarely matters later on
01:03 < Gretell> Unstable branch on crawl.develz.org updated to: 0.10-a0-2967-gb42ecd2 (32)
01:04 < SpaceCob> sorry to ask but how do i delete through the ssh? I have putty running and I can play online, apparently, but i don't know how to affect the webtiles
01:04 < CIA-112> |amethyst * rc935fa8ce88b /crawl-ref/source/ (main.cc mon-behv.cc player.cc): Do not do damage when rotting, and always kill at 0 max HP.
01:05 <+|amethyst> oh, hm
01:06 <+|amethyst> SpaceCob:  hm... it looks like they use separate save directories
01:06 < SpaceCob> that would make sense
01:07 <+|amethyst> they use the same morge directory though
01:07 <+elliptic> Napkin: SpaceCob seems to have a corrupted webtiles save ("save file corrupted  - overlapping blocks") and could use assistance deleting or fixing it
01:07 <+|amethyst> s/morge/morgue/
01:08 < SpaceCob> thanks for the help
01:08 <+elliptic> (he's probably not around right now, but he'll see that at some point)
01:08 < SpaceCob> oh okay. fine by me
01:09 <+|amethyst> hm... I guess that also means there's no way to backup a webtiles save for debugging
01:19 < Gretell> Windows builds of master branch on crawl.develz.org updated to: 0.10-a0-2968-gc935fa8
01:23 ۰۰-- ainsophyao [~ainsophya@h175-177-006-200.catv02.itscom.jp] has joined ##crawl-dev
01:23 ۰۰-- blueDave [47e49759@gateway/web/freenode/ip.71.228.151.89] has joined ##crawl-dev
01:23 < blueDave> !seen amethyst
01:23 < Henzell> Sorry blueDave, I haven't seen amethyst.
01:24 < blueDave> !seen /amethyst
01:24 < Henzell> Sorry blueDave, I haven't seen amethyst.
01:24 < blueDave> !seen |amethyst
01:24 < Henzell> Sorry blueDave, I haven't seen amethyst.
01:24 < blueDave> hmm, don
01:25 < blueDave> anyhow...   main computer sick, can't work on constriction for a while  :(
01:25 <+|amethyst> blueDave:  hey
01:25 <+|amethyst> sorry :(
01:25 < blueDave> I hope it's just a heating problem
01:26 < blueDave> started getting random memory access errors, then a bluescreen
01:26 <+|amethyst> you should see the clumps of cat hair I pull out of my CPU fan every couple of months :)
01:28 < blueDave> wonder if I got the extended warranty...
01:33 ۰۰-- jooosa [joosa@heh.fi] has joined ##crawl-dev
01:35 ۰۰-- joosa [joosa@193-199-217-31.elisa-mobile.fi] has quit [Ping timeout: 252 seconds]
01:35 ۰۰-- jooosa is now known as joosa
01:36 ۰۰-- ussdefiant [~chatzilla@S010678cd8e7ca058.ed.shawcable.net] has quit [Ping timeout: 240 seconds]
01:41 ۰۰-- ussdefiant [~chatzilla@S010678cd8e7ca058.ed.shawcable.net] has joined ##crawl-dev
02:04 < CIA-112> elliptic * r8b909816834d /crawl-ref/source/player.cc: Tweak AC from forms.
02:04 < CIA-112> elliptic * r43075d9873b0 /crawl-ref/source/ (dat/descript/spells.txt spl-cast.cc spl-data.h): Make Dragon Form L7, remove draconian wizardry boost for it.
02:19 <+Napkin> moi
02:20 <+Napkin> elliptic: I can't fix the savegame ;) do you want it or should i delete it?
02:22 ۰۰-- ussdefiant [~chatzilla@S010678cd8e7ca058.ed.shawcable.net] has quit [Ping timeout: 252 seconds]
02:22 <+elliptic> Napkin: I don't know anything about webtiles, so don't give it to me ;) I have no idea what that corruption message means, so I don't know whether passing it to edlothiol makes sense... but I think SpaceCob just wanted it deleted so he could start a new game
02:26 ۰۰-- ainsophyao [~ainsophya@h175-177-006-200.catv02.itscom.jp] has quit [Remote host closed the connection]
02:28 <+Napkin> did you create a bug report, SpaceCob?
02:34 <+Napkin> SpaceCob: in case you opened a bug report, please link to this file: https://crawl.develz.org/saves/dumps/SpaceCob-8a39726-111228-0733.cs
02:34 <+Napkin> SpaceCob: save-game moved/deleted.
02:44 ۰۰-- ussdefiant [~chatzilla@S010678cd8e7ca058.ed.shawcable.net] has joined ##crawl-dev
02:48 ۰۰-- SamB_ [~Sam@207-172-114-83.c3-0.upd-ubr1.trpr-upd.pa.cable.rcn.com] has joined ##crawl-dev
02:48 ۰۰-- mode/##crawl-dev [+v SamB_] by ChanServ
02:50 ۰۰-- SamB [~Sam@207-172-114-83.c3-0.upd-ubr1.trpr-upd.pa.cable.rcn.com] has quit [Ping timeout: 240 seconds]
03:17 ۰۰-- ussdefiant [~chatzilla@S010678cd8e7ca058.ed.shawcable.net] has quit [Ping timeout: 240 seconds]
03:29 ۰۰-- ussdefiant [~chatzilla@S010678cd8e7ca058.ed.shawcable.net] has joined ##crawl-dev
03:32 ۰۰-- Elynae [~mthomson@p579F666E.dip.t-dialin.net] has joined ##crawl-dev
03:38 < Elynae> what is the product latest trunk build on CDO? i currently have a block so i can't get into the game to check, and i need the product version to report it on mantis 
03:38 ۰۰-- ussdefiant [~chatzilla@S010678cd8e7ca058.ed.shawcable.net] has quit [Ping timeout: 240 seconds]
03:38 < Elynae> er, product version of the latest trunk build*
03:41 ۰۰-- Elynae [~mthomson@p579F666E.dip.t-dialin.net] has quit [Remote host closed the connection]
03:44 < st_> yeah cdo trunk is broken :(
03:59 ۰۰-- ussdefiant [~chatzilla@S010678cd8e7ca058.ed.shawcable.net] has joined ##crawl-dev
04:02 ۰۰-- dtsund [~detasund@s81-236-68-64.trico.az.wi-power.com] has quit [Ping timeout: 240 seconds]
04:02 <+kilobyte> SpaceCob: uh oh, how did we get that?
04:03 <+kilobyte> corruption of the contents of the save: fine, the game is buggy like hell, that's expected.  Corruption of the save's filesystem: that can't be!
04:06 ۰۰-- blabber [~tobi@datenschleuder.net] has joined ##crawl-dev
04:07 <+kilobyte> blabber: let me guess, my threading changes managed to break BSD builds somehow, right?
04:07 < blabber> kilobyte, to be honest i haven't tried trunk builds for a while.
04:08 <+kilobyte> (hi BTW)
04:08 < blabber> hi :)
04:08 < blabber> i may give it a try once the 28c3 has finished.
04:09 <+kilobyte> great; I don't expect BSD to have an incompatible pthreads implementation
04:14 < blabber> no. it should be quite standars compliant.
04:15 < blabber> do you guys have zhought about supporting clang builds in your mighty makefile.
04:15 < blabber> meh. that was a question ^^
04:17 <+kilobyte> would require feature tests, the zoology approach is already bursting at seams
04:17 <+kilobyte> on my TODO list, but certainly not for 0.10 :(
04:18 <+kilobyte> I mean, using that uname junk is no good, I refuse to drop the flaky cross-compiling support we have
04:38 ۰۰-- syllogism [~syllogism@89-166-16-68.bb.dnainternet.fi] has joined ##crawl-dev
04:42 ۰۰-- bmh [d8ef2c19@fsf/member/bmh] has joined ##crawl-dev
04:43 ۰۰-- bmh [d8ef2c19@fsf/member/bmh] has quit [Client Quit]
05:16 ۰۰-- wtface [~o_O@c-67-169-135-166.hsd1.ca.comcast.net] has quit [Quit: Your quote here!]
05:31 ۰۰-- Ragdoll [~Ragdoll@53518A17.cm-6-2c.dynamic.ziggo.nl] has joined ##crawl-dev
05:40 ۰۰-- ussdefiant [~chatzilla@S010678cd8e7ca058.ed.shawcable.net] has quit [Ping timeout: 255 seconds]
05:41 ۰۰-- Pingas [~Pingas@40.181.34.95.customer.cdi.no] has joined ##crawl-dev
05:42 < CIA-112> kilobyte * r346dfcfcde5d /crawl-ref/CREDITS.txt: Add dk to credits.
05:44 < Gretell> Unstable branch on crawl.develz.org updated to: 0.10-a0-2971-g346dfcf (32)
05:47 ۰۰-- wtface [~o_O@c-67-169-135-166.hsd1.ca.comcast.net] has joined ##crawl-dev
05:59 < Ragdoll> is everyone aware yet people are unable to play/start a (new) game on CDO?
06:00 ۰۰-- ussdefiant [~chatzilla@S010678cd8e7ca058.ed.shawcable.net] has joined ##crawl-dev
06:08 ۰۰-- petete [~quassel@190.190.173.140] has quit [Ping timeout: 240 seconds]
06:18 ۰۰-- evilmike [~evilmike@66.183.151.151] has quit [Quit: something happened]
06:20 ۰۰-- nooodl [~nooodl@91.177.143.102] has joined ##crawl-dev
07:03 ۰۰-- jeanjacques [~jejacques@p5481D99D.dip.t-dialin.net] has joined ##crawl-dev
07:04 < wtface> I like the new snake pit tiles
07:30 < CIA-112> Keskitalo * r76c9961f895c /crawl-ref/source/dat/des/portals/icecave.des: Fix ice cave feature descriptions.
07:31 < CIA-112> Keskitalo * r2e888efb463f /crawl-ref/source/dat/des/portals/icecave.des: Fix a few ice cave entry depths.
07:31 < CIA-112> Keskitalo * rff5b47536d9d /crawl-ref/source/dat/dlua/tutorial.lua: Don't stop displaying tutorial messages from the blue tiles after the first time the player steps on them.
07:31 < CIA-112> Keskitalo * r53cf2937f850 /crawl-ref/source/dat/des/portals/icecave.des: Make ice cave difficulty setting its own function.
07:37 ۰۰-- alefury [~galefury@B4115.karlshof.wh.tu-darmstadt.de] has joined ##crawl-dev
07:48 ۰۰-- alefury_ [~galefury@B4115.karlshof.wh.tu-darmstadt.de] has joined ##crawl-dev
07:48 ۰۰-- Netsplit *.net <-> *.split quits: alefury
07:48 ۰۰-- alefury_ is now known as alefury
08:08 ۰۰-- monqy [~swell@pool-71-102-226-192.snloca.dsl-w.verizon.net] has quit [Quit: hello]
08:14 ۰۰-- ainsophyao [~ainsophya@wl004j.ecc.u-tokyo.ac.jp] has joined ##crawl-dev
08:15 ۰۰-- edlothiol [~edlothiol@ip-95-223-51-38.unitymediagroup.de] has joined ##crawl-dev
08:15 ۰۰-- mode/##crawl-dev [+v edlothiol] by ChanServ
08:18 ۰۰-- ainsophyao [~ainsophya@wl004j.ecc.u-tokyo.ac.jp] has quit [Remote host closed the connection]
08:50 <+kilobyte> Napkin: do you set ulimit -T in the chroot?
08:55 <+Napkin> no, kilobyte 
08:55 <+Napkin> if with -T you mean hardlimit of cpu time? I don't
08:56 <+kilobyte> thread creation failed with EAGAIN, but this might have been caused by something else before
08:57 <+kilobyte> -T     The maximum number of threads
09:10 ۰۰-- wtface [~o_O@c-67-169-135-166.hsd1.ca.comcast.net] has quit [Quit: Your quote here!]
09:15 < CIA-112> kilobyte * rc529b781010b /crawl-ref/source/database.cc: Try to handle ulimit -T.
09:16 < Gretell> Unstable branch on crawl.develz.org updated to: 0.10-a0-2976-gc529b78 (32)
09:25 ۰۰-- ussdefiant [~chatzilla@S010678cd8e7ca058.ed.shawcable.net] has quit [Ping timeout: 255 seconds]
09:30 < Ragdoll> You finish putting on your +3 dwarven buckler of resistance.
09:30 < Ragdoll> nice, terence
09:43 ۰۰-- ainsophyao [~ainsophya@h175-177-006-200.catv02.itscom.jp] has joined ##crawl-dev
09:45 ۰۰-- ussdefiant [~chatzilla@S010678cd8e7ca058.ed.shawcable.net] has joined ##crawl-dev
09:47 ۰۰-- ainsophyao [~ainsophya@h175-177-006-200.catv02.itscom.jp] has quit [Remote host closed the connection]
09:55 <+|amethyst> kilobyte:  what do you mean, failed with EAGAIN?
09:56 <+|amethyst> oh
09:56 <+|amethyst> I see
09:57 <+|amethyst> ulimit -u would also do it
09:57 <+|amethyst> or /proc/sys/kernel/threads-max
09:58 <+|amethyst> In any event, that does not appear to have fixed it
09:58 ۰۰-- MarvinPA [~MarvinPA@thecampbells.demon.co.uk] has joined ##crawl-dev
09:59 ۰۰-- mode/##crawl-dev [+v MarvinPA] by ChanServ
09:59 <+|amethyst> Unable to open DB: /crawl-svn-c529b78/saves/cache.0.10-a0-2976-gc529b78/db/randart: Cannot allocate memory
09:59 <+|amethyst> Napkin:  is CDO under more memory pressure than usual?
10:02 <+Napkin> no, but seems crawl's creation of the db is again breaking the 150mb barrier, |amethyst 
10:02 <+Napkin> *100mb
10:03 <+Napkin> kilobyte, |amethyst, greensnark: do you guys want to fix that or do I need to raise the limit?
10:03 <+|amethyst> aha
10:03 <+|amethyst> hm.
10:04 <+|amethyst> does the db get regenerated for each player?
10:04 <+Napkin> it's create on first start
10:04 <+Napkin> so, only by the first player
10:04 <+Napkin> of a new version
10:04 <+|amethyst> you could have the scripts do the builddb 
10:05 <+|amethyst> without the ulimit in place
10:05 <+Napkin> i could (is there an option for that?) but is that really the desired way to go?
10:05 <+|amethyst> crawl --builddb ... not sure about whether that's the best way to do it
10:05 <+Napkin> i would assume it isn't
10:06 <+Napkin> unless this is not just another bug?
10:06 <+|amethyst> hm
10:07 <+|amethyst> which ulimit are you using?  I want to test here
10:07 ۰۰-- HangedMan [63e6725f@gateway/web/freenode/ip.99.230.114.95] has joined ##crawl-dev
10:07 <+Napkin> 100mb
10:07 <+Napkin> that's -v
10:07 <+|amethyst> oh
10:08 <+Napkin> oh, i set it to 150mb now, and still starting crawl fails:
10:08 <+Napkin> terminate called after throwing an instance of 'std::bad_alloc'
10:08 <+Napkin> terminate called recursively
10:08 <+Napkin>   what():  std::bad_alloc
10:08 <+Napkin> terminate called recursively
10:08 <+Napkin> terminate called recursively
10:08 <+Napkin> ----
10:08 <+|amethyst> -v might be the wrong thing to use
10:09 <+kilobyte> Napkin: busy at work, going home now...  this makes sense, the 9 concurrent threads obviously can take more memory
10:09 <+Napkin> actually, look at by stack trace
10:10 <+|amethyst> Napkin:  is CDO 64-bit?
10:10 <+Napkin> http://crawl.develz.org/morgues/trunk/Napkin/crash--20111228-150743.txt
10:10 <+Napkin> is that really related to what we're talking about here?
10:10 <+Napkin> yes
10:10 <+kilobyte> even on my phone (256MB ram) the threaded version is faster but apparently it hits the limit here
10:11 <+kilobyte> can #ifndef DGAMELAUNCH it
10:11 <+kilobyte> (got to go though, sorry.  brb)
10:12 <+|amethyst> Napkin:  do you have any other memory-related ulimits ( -m, -d, -s )?
10:12 <+|amethyst> I can't duplicate with ulimit -v $((100*1024)) but I'm on a 32-bit machine
10:13 <+Napkin> could you answer my question, |amethyst?
10:13 <+|amethyst> Napkin:  related to your crash dump?
10:13 <+Napkin> yeah, just making sure you read my doubt here
10:14 <+Napkin> i just removed the cache folder and the ulimit settings (only -v and -c in use, both at 150mb at the moment)
10:15 <+|amethyst> it's kind of hard to tell with so many of the stack frames being unannotated
10:16 <+|amethyst> but that crash dump is at least consistent with new failing for lack of memory
10:17 <+Napkin> that's the stack trace of a fresh start, without memory ulimit and without cache folder (i hope): http://crawl.develz.org/morgues/trunk/Napkin/crash--20111228-151534.txt
10:17 <+|amethyst> hm...
10:17 <+|amethyst> that doesn't make sense
10:18 <+Napkin> there are 2 versions installed today, which were manually triggered
10:18 <+Napkin> why?
10:18 <+Napkin> was the bug there since this morning?
10:20 <+|amethyst> not sure why the firt one this morning was triggered
10:20 <+|amethyst> threading went in at the normal update
10:21 <+|amethyst> that second crash dump appears to be failing on   dbfile += ".db"  which still sounds like a memory issue
10:23 <+|amethyst> wha... now my rebuild failed for lack of memory
10:28 <+|amethyst> oh, duh, I tried rebuilding from the same shell where I set the ulimit for testing :)
10:29 ۰۰-- ainsophyao [~ainsophya@h175-177-006-200.catv02.itscom.jp] has joined ##crawl-dev
10:31 <+|amethyst> okay, if I set a 60 MB ulimit here (32-bit userspace) I get a similar error
10:31 <+Napkin> ok, this is a guaranteed first start (without cache, without memory limit): http://crawl.develz.org/morgues/trunk/Napkin/crash--20111228-153029.txt
10:34 <+|amethyst> Napkin:  hmm
10:34 <+|amethyst> Napkin:  I definitely can't reproduce a crash without ulimits
10:35 <+|amethyst> wait... I have a single core
10:36 <+|amethyst> your error is still std::bad_alloc though, which is odd
10:37 <+Napkin> last working version on CDO is ae15d2e
10:38 <+|amethyst> which was shortly before 0e9f143 - Threading support
10:39 < CIA-112> kilobyte * r89365825582d /crawl-ref/source/database.cc: Don't bother with parallel db rebuild on DGAMELAUNCH.
10:43 <+Napkin> also, dgamelaunch is compiled with --with-rlimit-core=157286400 --with-rlimit-as=104857600
10:43 <+Napkin> which never seemed to work (to be inherited by the crawl processes), hence the manual ulimit settings in the starting scripts
10:43 <+|amethyst> hrm
10:43 <+Napkin> maybe it does work now?
10:44 <+|amethyst> heh
10:44 <+|amethyst> is someone doing a rebuild with that new commit?
10:45 <+Napkin> kilobyte: and they run
10:45 <+Napkin> with insanely low ulimits
10:45 <+Napkin> ---
10:45 <+Napkin> what do you want? the 50 instances of crawl to use more than 150mb at most?
10:49 <+kilobyte> didn't start the build
10:50 <+kilobyte> Napkin: it'd have to be OpenMoko to run into swap, so the limit is artificially low for regular builds
10:51 <+kilobyte> however, on DGL something that low makes sense
10:51 <+|amethyst> just how much per-thread data is there, anyway?
10:52 ۰۰-- ussdefiant [~chatzilla@S010678cd8e7ca058.ed.shawcable.net] has quit [Ping timeout: 244 seconds]
10:52 <+kilobyte> optimizing those several seconds away is not worth the effort if it causes problems
10:52 < Gretell> Unstable branch on crawl.develz.org updated to: 0.10-a0-2977-g8936582 (32)
10:53 <+kilobyte> |amethyst: a sqlite instance plus almost nothing else
10:53 <+|amethyst> err, now I can't log in at all
10:53 <+|amethyst> Permission denied (publickey).
10:53 < chrisoelmueller> same
10:54 <+|amethyst> I guess Napkin is changing something in dgl?
10:54 <+kilobyte> [~]$ ssh cdo
10:54 <+kilobyte> Permission denied (publickey).
10:55 <+Napkin> there's no other way to ensure it's my crawl trying to create the db
10:55 <+Napkin> opened again
10:55 <+|amethyst> oh :)
10:55 <+|amethyst> yay
10:57 ۰۰-- ixtli [~textual@ool-4357c7fc.dyn.optonline.net] has joined ##crawl-dev
10:57  * Napkin is still grumpy about "insanely".. Pah! :-P *
10:58 <+kilobyte> Napkin: "insanely" as in "insanely for regular machines in the wild"
10:59 <+|amethyst> hm... how to get gdb to run something with ulimits
10:59 <+|amethyst> I guess I can break in main and call setrlimit by hand
11:06 ۰۰-- Pacra [~Pacra@c-68-44-35-89.hsd1.nj.comcast.net] has quit [Quit: This computer has gone to sleep]
11:09 <+|amethyst> hm... I'm crashing when loading maps, not databases, but I guess that's a 32- vs 64-bit issue (i have to lower the ulimit to 60 MiB to get a crash at all)
11:13 <+Napkin> kilobyte: @regular: ServerLimit 300 ;) less and people were complaining that mantis is too slow! which in turn needed max_connections 500 for mysql. crawl & co is harsh for a single regular server ;)
11:15 ۰۰-- ussdefiant [~chatzilla@S010678cd8e7ca058.ed.shawcable.net] has joined ##crawl-dev
11:17 <+|amethyst> okay, while doing the db rebuild my address space appears to get up to about 90 MiB; while running the game, it's closer to 64 MiB
11:24 <+|amethyst> the difference being about five blocks of 8 MiB each
11:25 <+|amethyst> looking like "f7385000-f7b85000 rw-p 00000000 00:00 0"   which isn't very helpful to me
11:29 ۰۰-- Ragdoll [~Ragdoll@53518A17.cm-6-2c.dynamic.ziggo.nl] has quit [Quit: Up, up and away!]
11:30 <+|amethyst> bit in any event, it seems like each thread is taking 8 MiB of address space, which would explain the problem
11:31 <+|amethyst> not sure how to tell whether those are actually resident
11:31 < Gretell> 78291 (L23 DsSu) ASSERT(attk_flavour == AF_REACH || weapon && weapon_reach(*weapon)) in 'melee_attack.cc' at line 4033 failed. (Zot:1)
11:31 <+|amethyst> wha?
11:31 <+|amethyst> !lm @78291 crash -log
11:31 < Sequell> 16. 78291, XL23 DsSu, T:134311 (milestone): http://crawl.develz.org/morgues/trunk/78291/crash-78291-20111228-163106.txt
11:32 <+|amethyst> oh, old version
11:33 <+kilobyte> yeah, some people tend to refuse upgrades
11:36 ۰۰-- ussdefiant [~chatzilla@S010678cd8e7ca058.ed.shawcable.net] has quit [Ping timeout: 240 seconds]
11:46 ۰۰-- Ragdoll [~Ragdoll@53518A17.cm-6-2c.dynamic.ziggo.nl] has joined ##crawl-dev
11:48 ۰۰-- ussdefiant [~chatzilla@S010678cd8e7ca058.ed.shawcable.net] has joined ##crawl-dev
11:52 ۰۰-- st_ [~s@cpc10-stre5-2-0-cust69.1-1.cable.virginmedia.com] has quit [Ping timeout: 252 seconds]
12:02 <+|amethyst> is there any reason for uniq_price_ribbit2 and uniq_roxanne not to be transparent?
12:02 <+|amethyst> s/price/prince/
12:03 <+|amethyst> likewise uniq_dummy*
12:03 <+Zaba> |amethyst, uniq_dummy never gets placed
12:04 <+Zaba> |amethyst, as for the other two - no reason, somebody just accidentally left it out
12:05 <+Zaba> |amethyst, "never gets placed" as in the builder just places no vault when it picks a vault tagged 'dummy', not to say that it is useless.
12:05 <+|amethyst> aha
12:05 <+|amethyst> hm... also can_overwrite... I guess that could be a problem for roxanne since her vault is 3x3, but what about prince ribbit's 2x1 vault?
12:06 <+Zaba> |amethyst, I'd say no to both just to be on the safe side
12:07 <+|amethyst> ignacio is also missing can_overwrite despite being 1x1... but I guess that keeps him out of the rune vaults
12:12 ۰۰-- ussdefiant [~chatzilla@S010678cd8e7ca058.ed.shawcable.net] has quit [Ping timeout: 276 seconds]
12:12 ۰۰-- ussdefiant [~chatzilla@S010678cd8e7ca058.ed.shawcable.net] has joined ##crawl-dev
12:18 ۰۰-- dtsund [~detasund@s81-236-68-64.trico.az.wi-power.com] has joined ##crawl-dev
12:31 ۰۰-- alefury [~galefury@B4115.karlshof.wh.tu-darmstadt.de] has quit [Ping timeout: 252 seconds]
12:40 <+|amethyst> hm.. I think escaping is accidentally more easy than blueDave intended
12:41 <+|amethyst> it uses transform_size() for the player... which, if you are not transformed, returns SIZE_CHARACTER, which is larger than even SIZE_HUGE
12:42 <+|amethyst> I'll fix that, but that will change balance
12:43 ۰۰-- alefury [~galefury@B4115.karlshof.wh.tu-darmstadt.de] has joined ##crawl-dev
12:43 <+|amethyst> oh, wait
12:44 ۰۰-- upsy [michael@cpc1-aztw17-0-0-cust621.aztw.cable.virginmedia.com] has joined ##crawl-dev
12:44 <+|amethyst> hm
12:52 ۰۰-- nooodl [~nooodl@91.177.143.102] has quit [Ping timeout: 240 seconds]
12:55 < Eronarn> |amethyst: speaking of size, how are na/ce handle
12:55 < Eronarn> d
12:55 <+|amethyst> yeah, it definitely needs rebalancing
12:56 < Eronarn> i feel like centaurs shouldn't be hard to grapple just because they have a horse half
12:56 <+|amethyst> Eronarn:  they were considered medium for constriction, I've got a commit that makes them large
12:56 <+|amethyst> I'd think constricting a centaur should be no easier than constricting a horse
12:57 < Eronarn> |amethyst: you can just constrict the human part
12:57 < Eronarn> that will be enough to kill them
12:58 <+|amethyst> but would it keep them from running away?
12:58 < Eronarn> hmm, that's a good point; if you define constriction as needing to affect their form of motion as well as deal damage, then i guess ce would be large
12:59 < Eronarn> (also, we should add horses to crawl)
12:59 <+|amethyst> also, who knows where centaurs keep their vital organs :)
12:59 <+|amethyst> I really want to use the word "horso" to describe the lower part of a centaur
13:00 < HangedMan> unicorns
13:00 < Eronarn> unicorn would be a decent ely holy
13:00 <+|amethyst> I like
13:01 < Eronarn> however i'd want to see different halo colors for the gods before adding more non-tso holies
13:01 < Eronarn> which reminds me
13:02 < HangedMan> what happened to the silver stars getting neat halo colours
13:02 < Eronarn> we should consider using ele colors for stuff like silence/halo; it'd help the issues with floor colors
13:04 <+|amethyst> sanctuary is already elemental, isn't it?
13:04 < Eronarn> dunno
13:05 < Eronarn> hmm, one thing we could do is have the halo border be aggressively elemental colored
13:05 < Eronarn> that might get annoying though
13:05 <+|amethyst> hm
13:05 <+|amethyst> this is only an issue in console, right?
13:05 < Eronarn> yeah
13:05 ۰۰-- jeanjacques [~jejacques@p5481D99D.dip.t-dialin.net] has quit []
13:05 <+|amethyst> why not change the background colour for halo etc
13:06 < Eronarn> it looks kind of ugly, unfortunately
13:06 < Eronarn> it could be a worthwhile option, though
13:07 < Eronarn> if we had combining characters, halos would become so much easier to do :(
13:07 < Eronarn> a () around each haloed square, etc.
13:11 < CIA-112> |amethyst * r11e5df14021e /crawl-ref/source/ (actor.cc melee_attack.cc monster.cc player.cc): Use full-body size for constriction purposes.
13:11 < CIA-112> |amethyst * rb879844085a0 /crawl-ref/source/ (7 files in 2 dirs): Merge more duplicated constriction code.
13:11 < CIA-112> |amethyst * r5db6206e8ddd /crawl-ref/source/dat/des/builder/uniques.des: Make two unique vaults transparent.
13:12 <+|amethyst> as I mentioned, that last one makes it much harder to escape constriction
13:13 <+|amethyst> but I think the formulae need revamped anyway
13:14 <+|amethyst> err, that first one (CIA reported the commits in reverse order)
13:15 <+|amethyst> yeah, around 10 attempts for a spriggan to escape a ball python
13:15 ۰۰-- ainsophyao [~ainsophya@h175-177-006-200.catv02.itscom.jp] has quit [Remote host closed the connection]
13:17 <+|amethyst> also, things of SIZE_TINY (including player bats and spiders) are completely incapable of escaping constriction
13:17 <+|amethyst> this was the case before my commit
13:23 ۰۰-- ainsophyao [~ainsophya@h175-177-006-200.catv02.itscom.jp] has joined ##crawl-dev
13:29 <+Napkin> cia seems to always do that, |amethyst 
13:29 ۰۰-- nooodl [~nooodl@91.177.143.102] has joined ##crawl-dev
13:29 <+Napkin> anyone around who can try to build on macosx?
13:29 <+Napkin> i think it's broken for a few days now
13:30 <+|amethyst> Napkin:  actually, it seems to be more or less random order, if you look at Keskitalo's commits earlier
13:31 <+Napkin> true
13:34 ۰۰-- Pacra [~Pacra@c-68-44-35-89.hsd1.nj.comcast.net] has joined ##crawl-dev
13:55 < SpaceCob> Thanks for deleting the corrupted file for me. I saw your questions that you asked last night...sorry I was asleep. I never put in a bug report and yeah, I just wanted it deleted
13:55 < SpaceCob> I just checked, I can make a new game. So yay! I really appreciate it. Now I can go back to splatting innocent Tengus.
13:56 <+Napkin> hehe, enjoy :)
14:01 ۰۰-- SpaceCob [473dee4c@gateway/web/freenode/ip.71.61.238.76] has left ##crawl-dev []
14:05 ۰۰-- ais523 [~ais523@unaffiliated/ais523] has joined ##crawl-dev
14:13 ۰۰-- ainsophyao [~ainsophya@h175-177-006-200.catv02.itscom.jp] has quit [Remote host closed the connection]
14:20 ۰۰-- jeanjacques [~jejacques@p5481D99D.dip.t-dialin.net] has joined ##crawl-dev
14:41 ۰۰-- ainsophyao [~ainsophya@h175-177-006-200.catv02.itscom.jp] has joined ##crawl-dev
14:48 ۰۰-- BlastHardcheese [chris@pdpc/supporter/active/blasthardcheese] has quit [Quit: Talk is cheap because the supply exceeds the demand.]
15:15 ۰۰-- ainsophyao [~ainsophya@h175-177-006-200.catv02.itscom.jp] has quit [Remote host closed the connection]
15:21 ۰۰-- ainsophyao [~ainsophya@h175-177-006-200.catv02.itscom.jp] has joined ##crawl-dev
15:27 ۰۰-- petete [~quassel@190.190.173.140] has joined ##crawl-dev
15:27 <+|amethyst> hm... I'm working on clearing non-adjacent constrictions whenever an actor is moved
15:28 <+|amethyst> but that breaks the "along for the ride" stuff
15:28 <+|amethyst> I guess a new flag for moveto() would fix it :(
15:29 <+|amethyst> s/fix it/allow it to be fixed/
15:35 < Eronarn> |amethyst: yeah, that's what i noted earlier - we really should have a flag for that
15:35 < Eronarn> at least as an option
15:35 <+|amethyst> hm... even with a flag on moveto() it would be difficult
15:36 <+|amethyst> since you want to clear most of the constrictions
15:36 < Eronarn> |amethyst: what should happen is that any time someone moves, constrictions get broken, unless the certain flag is present
15:37 < Eronarn> then it can check if it's broken, different ways depending on which flag
15:45 < CIA-112> edlothiol * r2cdada0216b5 /crawl-ref/source/ (rltiles/dc-feat.txt tilepick.cc tilepick.h tileview.cc): Fix #5133: secret door positions revealed next to walls with redefined tiles.
16:04 <+SamB_> Eronarn: set on what? 
16:04 <+SamB_> should the flag be set on the actor that moved?
16:04 ۰۰-- SamB_ is now known as SamB
16:07 <+|amethyst> I'm setting it on the moveto() for the actor that is originally teleported
16:08 <+|amethyst> move actor to new position (without breaking constrictions), pull along constrictors/constrictees (breaking non-adjacent constrictions; they're adjacent to the original actor); then break the actor's non-adjacent constrictions
16:13 < petete> should sublimate blood have a special case for casting it over a blood fountain?
16:15 < Eronarn> SamB: i mean a flag describing which type of movement
16:15 < Eronarn> like: teleported, stepped, was shoved
16:16 <+|amethyst> I'm not doing that, though it may be a good idea
16:16 <+|amethyst> I'm just adding a bool to moveto(), move_to_pos(), and move_player_to_grid()
16:17 <+|amethyst> which, if set to false (default is true), will avoid checking nonadjacent constrictions
16:18 <+|amethyst> then the cases that can drag things along will pass false (and then check constrictions *after* pulling other things along)
16:19 < Eronarn> |amethyst: better to make a flag than a bool. for example, it's relevant to clinging too
16:20 < Eronarn> if the functions are going to be changed at all you may as well just change them to what they should be
16:22 <+|amethyst> that's a much bigger change though
16:22 <+|amethyst> since basically all the callers will have to be updated
16:25 <+|amethyst> I think your suggestion is a good one, but it would take me quite some time to make and fully test the changes
16:25 < Eronarn> you can make it optional, same as a bool
16:25 < Eronarn> MOVED_MYSTERY or whatever
16:37 <+SamB> Might not work too well...
16:40 < CIA-112> edlothiol * r464bc2c33aa7 /crawl-ref/source/item_use.cc: Fix #5135: Octopodes get unnecessary prompt with one ring.
16:41 <+|amethyst> !lg Perryman s=-rune
16:41 < Sequell> Unknown selector rune
16:41 <+|amethyst> doh
16:47 ۰۰-- evilmike [~evilmike@66.183.151.151] has joined ##crawl-dev
16:47 ۰۰-- mode/##crawl-dev [+v evilmike] by ChanServ
16:49  * SamB wishes putty could be configured to ask pageant to load a specific key file ...
16:52 ۰۰-- ainsophyao [~ainsophya@h175-177-006-200.catv02.itscom.jp] has quit [Remote host closed the connection]
17:13 < phyphor> whu?
17:19 ۰۰-- jeanjacques [~jejacques@p5481D99D.dip.t-dialin.net] has quit []
17:28 < CIA-112> |amethyst * r9cea76154799 /crawl-ref/source/ (9 files): Do not lose constriction when teleporting.
17:28 < CIA-112> |amethyst * r4fedd6d63ed1 /crawl-ref/source/ (main.cc mon-stuff.cc spl-transloc.cc teleport.cc): Avoid doubled "along for the ride" messages.
17:28 <+|amethyst> wha?
17:28 < CIA-112> |amethyst * ra8af7f392eff /crawl-ref/source/ (7 files): Clear non-adjacent constictions when moved.
17:29 <+|amethyst> ah, there we go
17:29 <+|amethyst> Eronarn:  I didn't make the change you suggested yet.  I'd gladly accept a patch of course :)
17:29 < Eronarn> eh, i have been too busy to do much crawl lately
17:29 < Eronarn> where 'lately' is 'over six months, at least'
17:30 <+|amethyst> I'll be in such a state in a few days
17:35 <@due> I have been like that for a while
17:37 ۰۰-- upsy [michael@cpc1-aztw17-0-0-cust621.aztw.cable.virginmedia.com] has quit [Quit: Leaving]
17:47 <+|amethyst> oh, missed a case
17:48 <+|amethyst> monster swapping apparently doesn't use moveto() ?
17:52 ۰۰-- wtface [~o_O@c-67-169-135-166.hsd1.ca.comcast.net] has joined ##crawl-dev
18:04 ۰۰-- alefury [~galefury@B4115.karlshof.wh.tu-darmstadt.de] has quit [Ping timeout: 252 seconds]
18:15 < HangedMan> the bug where items in the abyss are shifted to other tiles when in any water rather then just deep water means that N78291 just had evilmike_abyss_rune_tentacle_tunnel put the rune outside the vault
18:18 < CIA-112> |amethyst * re270302f761b /crawl-ref/source/actor.cc: Fix a new constriction crash (oops).
18:18 < CIA-112> |amethyst * r7657a9ca8bec /crawl-ref/source/ (6 files): Fix constriction crash on monster swapping etc.
18:18 ۰۰-- petete [~quassel@190.190.173.140] has quit [Quit: No Ping reply in 180 seconds.]
18:19 ۰۰-- petete [~quassel@190.190.173.140] has joined ##crawl-dev
18:29 <+|amethyst> presumably statues, liche, blade, and appendage forms should allow constriction
18:30 < HangedMan> yeah
18:34 < Eronarn> |amethyst: it might be worth searching for anything that adjusts monster/player position that doesn't use moveto
18:34 <+|amethyst> Eronarn:  that's what I did
18:34 < Eronarn> ah, cool
18:35 < Eronarn> good on you :)
18:35 <+|amethyst> I left wizmode &u and &d as is
18:35 < Eronarn> (it's kind of scary how much of crawl's code i remember after not touching it for so long...)
18:36 < HangedMan> crawl is infectious
18:36 < Eronarn> oh that reminds me
18:36 < Eronarn> the new data that is now collected in games
18:36 < Eronarn> is tension in there?
18:37 < Eronarn> because if so i'd like to try and get lava orcs in a testable state while i'm on winter break
18:38 ۰۰-- monqy [~swell@pool-71-102-226-192.snloca.dsl-w.verizon.net] has joined ##crawl-dev
18:40 < HangedMan> naga and octopode ghosts don't get constriction, what
18:40 < Eronarn> yeah, i was thinking of that way before it got coded
18:40 < Eronarn> you can't constrict ghosts, but it might make sense for them to be able to constrict you
18:40 < Eronarn> this would also let us have e.g. a chain-rattling ghost
18:40 < HangedMan> hehehe
18:41 <+|amethyst> form changes (polymorph, player form spells) don't stop constriction
18:41 < Eronarn> |amethyst: insubstantiality
18:41 <+|amethyst> my comment was unrelated to the ghost one
18:41 <+|amethyst> other than being about constriction :)
18:42 < Eronarn> oh, oops
18:42 < Eronarn> now that i read it again i see what you meant. yeah i ran across that earlier
18:42 < Eronarn> constriction is quite buggy
18:43 < Eronarn> i do hope it gets into this version, even if only in a limited form so octo PCs can have it
18:43 <+|amethyst> I don't think there's much choice
18:43 <+|amethyst> reverting it would be kind of difficult now
18:44 <+|amethyst> not impossible I guess
18:44 < Eronarn> |amethyst: it could just be disabled and not reverted :)
18:44 <+|amethyst> true
18:45 < Eronarn> oh yeah, i had an idea earlier that might fix the issues with dragonform: remove it
18:45 <+|amethyst> what issues?
18:45 < HangedMan> "become huge form"
18:45 < Eronarn> instead, have this: http://magiccards.info/roe/en/184.html
18:45 < Eronarn> yes
18:46 < CIA-112> |amethyst * rc110c26c6902 /crawl-ref/source/melee_attack.cc: Allow constriction in some forms (Fixes #5136).
18:46 < Eronarn> see also: http://magiccards.info/zen/en/162.html
18:47 ۰۰-- Ragdoll [~Ragdoll@53518A17.cm-6-2c.dynamic.ziggo.nl] has quit [Quit: Up, up and away!]
18:47 < Eronarn> |amethyst: the issues of what school to make it, what type of dragon, resists/defense, equipment
18:48 < Eronarn> also note that having Gigantiform would mean octos could constrict more kinds of stuff :D
18:49 <+kilobyte> two spells: Enlarge, Reduce
18:49 <+kilobyte> +hunger +maxhp -ev; -maxhp +ev
18:49 < Eronarn> embiggen :D
18:50 < Eronarn> actually though, in a break from my normal stance, i wouldn't advocate symmetry
18:50 < Eronarn> hex/tmut: Shrink Ray
18:50 <+|amethyst> What wording to use for the A screen display of constriction?
18:51 <+|amethyst> I was thinking "You can constrict your foes." but that doesn't differentiate between nagas (one foe) and octopodes (many foes)
18:51 < Eronarn> 'You can use your constriction_part() to constrict enemies.' is fine
18:51 < Eronarn> maybe 'many enemies at once' for octos
18:53 < Eronarn> (growth spell... should it be self-only? it might be cool to have a 'form spell' usable on others)
18:54 < Eronarn> actually, hey, now that we have beastly appendage...
18:54 < Eronarn> it'd be cool to have a way to cast that on your allies
18:56 < chrisoelmueller> Mass Append
18:58 < Eronarn> maybe Monstrous Horde, as a reference to DS
19:06 ۰۰-- blabber [~tobi@datenschleuder.net] has quit [Quit: leaving]
19:06 < CIA-112> kilobyte * re0ee701efd3d /crawl-ref/source/util/columnise-credits.pl: Re-enable columnise-credits on non-broken platforms.
19:06 ۰۰-- HangedMan [63e6725f@gateway/web/freenode/ip.99.230.114.95] has quit [Ping timeout: 258 seconds]
19:08 <+|amethyst> kilobyte:  thanks
19:08 <+|amethyst> kilobyte:  I hadn't even noticed that I accidentally committed that
19:10 < CIA-112> |amethyst * r085d75f20223 /crawl-ref/source/ (mutation.cc output.cc): Add constriction to the A and % screens.
19:11 ۰۰-- HangedMan [63e6725f@gateway/web/freenode/ip.99.230.114.95] has joined ##crawl-dev
19:13 ۰۰-- syllogism [~syllogism@89-166-16-68.bb.dnainternet.fi] has quit []
19:13 ۰۰-- alefury [~galefury@B4115.karlshof.wh.tu-darmstadt.de] has joined ##crawl-dev
19:17 <+kilobyte> edlothiol: should mcache.clear_all() be called during loading saves?
19:18 <+kilobyte> it's currently done only #if TAG_MAJOR_VERSION == 32
19:20 ۰۰-- alefury [~galefury@B4115.karlshof.wh.tu-darmstadt.de] has quit [Ping timeout: 252 seconds]
19:25 <+kilobyte> shouldn't MUT_TENTACLES grant constriction?
19:25 <+edlothiol> kilobyte: yeah, I think it should actually... when I wrote that I thought it would only be necessary after loading an old mcache, but of course when loading a new level there will still be the old mcache, which should be cleared
19:28 < HangedMan> can demonspawn be granted tentacles
19:28 <+|amethyst> kilobyte:  hm... what about beastly tentacles?
19:29 <+kilobyte> edlothiol: hmm, just pushed it, but won't this be an issue for restart_after_game too?
19:29 <+kilobyte> ie, it probably should be cleared whenever you enter a level, no matter if it was loaded or created
19:30 <+kilobyte> |amethyst: this is what I mean
19:30 <+|amethyst> kilobyte:  I mean from beastly appendage
19:31 <+kilobyte> yeah
19:31 <+|amethyst> ah
19:32 <+kilobyte> I think there are no other sources of tentacles -- you either start with them or get them from Beastly
19:32 < CIA-112> kilobyte * rcc153de22e4a /crawl-ref/source/tags.cc: Don't re-tile levels that have tile data already (in console).
19:32 < CIA-112> kilobyte * r6f94a5458bfb /crawl-ref/source/tags.cc: Drop mcache on save load even for TAG_MAJOR_VERSION > 32.
19:33 <+MarvinPA> they're already way better than the other mutations you can get from appendage though, aren't they?
19:33 ۰۰-- Wensley [~chatzilla@dynamic-acs-72-23-89-126.zoominternet.net] has joined ##crawl-dev
19:33 <+MarvinPA> oh actually i guess not anymore since they don't boost base uc damage
19:33 <+kilobyte> no, horns were the best by far (and got nerfed).  Then constriction came.
19:34 <+MarvinPA> tentacles were definitely the best for a good while, since they gave you an aux and also increased your base uc damage
19:35 <+kilobyte> IIRC that was changed (might be wrong though)
19:35 <+MarvinPA> yeah, it has been now
19:35 <+MarvinPA> i dunno, possibly i've just totally lost track of when things happened :P
19:35 <+MarvinPA> anyway yeah
19:35 <+MarvinPA> constriction on mut_tentacles sounds good!
19:36 <+|amethyst> kilobyte:  err, I just got a compilation error
19:36 <+MarvinPA> could make them an alternative to claws as a demonspawn body facet, too
19:36 <+|amethyst> tags.cc:3921:9: error: ‘mcache’ was not declared in this scope
19:36 <+|amethyst> tags.cc:3928:5: error: ‘mcache’ was not declared in this scope
19:37 ۰۰-- alefury [~galefury@B4115.karlshof.wh.tu-darmstadt.de] has joined ##crawl-dev
19:38 <+|amethyst> kilobyte:  (compiling without tiles)
19:39 <+edlothiol> kilobyte: hmm... actually, it may be better to let the mcache's garbage collection run its course... tile_fg will be reset anyway, which should reduce all refcounts to 0
19:40 <+kilobyte> right, new levels are likely to have similar monsters
19:42 < CIA-112> kilobyte * r765af0c3a705 /crawl-ref/source/ (stairs.cc tags.cc): Clear mcache on all levels (including restart_after_game).
19:42 <+kilobyte> so there's no need to clear the mcache at all, right?
19:44 <+edlothiol> right, the only problematic situation is when loading old games with saved mcache -- currently the mcache is read and then cleared, a better way would probably be to not touch the actual mcache and just discard the saved information immediately
19:47 <+kilobyte> in those games TILE_WALL_MAX is always wrong anyway, yeah, and that's the only thing stored after mcache
19:48 ۰۰-- Pingas [~Pingas@40.181.34.95.customer.cdi.no] has quit [Read error: Connection reset by peer]
19:58 ۰۰-- ainsophyao [~ainsophya@h175-177-006-200.catv02.itscom.jp] has joined ##crawl-dev
20:04 <+|amethyst> is there any way to get tentacles 1 and tentacles 2?
20:05 < Eronarn> kilobyte / |amethyst: it might be worth talking about the tentacles mut, it wasn't well planned when i added it
20:06 < Eronarn> and i don't think they have been changed too much
20:07 < Eronarn> also, for DS... tentacles need to wield weapons, so how're they constricting?
20:07 < Eronarn> my thought last time i thought about this was a split into MUT_ARM_TENTACLES, MUT_LEG_TENTACLES, MUT_EXTRA_TENTACLES
20:08 <+|amethyst> hm
20:08 <+|amethyst> oh, I know what I was confused about a bit ago... tentacles vs pseudopods
20:08 < Eronarn> since universal reaching seems to have not caused problems, we could maybe go ahead and give it to octos
20:09 < HangedMan> ds gets larger tentacles then octopode so they could hold down an opponent with their off-tentacle?
20:09 <+kilobyte> |amethyst: only tentacles 3, from both sources
20:09 <+kilobyte> HangedMan: no tentacley demonspawn
20:09 < HangedMan> boo
20:10 <+|amethyst> kilobyte:  not pseudopods then?
20:10 < HangedMan> does every monsterous demonspawn have to have clawas
20:10 < Eronarn> which could mean: arm tent = no gloves, reaching, punch becomes tentacle slap, constrict if unarmed; leg tent = no boots, kick becomes tentacle slap, constrict; extra tentacles = tentacle slap aux, constrict
20:10 < HangedMan> claws
20:10 <+|amethyst> I guess they're too amorphous to constrict
20:10 < Eronarn> does that sound like a reasonable breakdown?
20:10 <+kilobyte> Jiyva has pseudopods which are distinct
20:11 < Eronarn> yeah pseudos should stay separate. though tentacle mut should become reasonable enough to be a random mut that jiyva could give out
20:11 <+|amethyst> leg tentacles would also need special handling for merfolk
20:11 <+kilobyte> I see Beastly tentacles cause problems here... let's not waste any time but replace them with claws
20:11 < Eronarn> |amethyst: tentacles are good for swimming :) they can have them in both forms
20:12 <+|amethyst> that still needs special-casing
20:12 < HangedMan> and bring back the issue of the base damage appendage being the most desirable one?
20:12 < Eronarn> "Your tail is long and rubbery" vs. "Your legs are long and rubbery" should be all the special casing needed
20:12 <+|amethyst> and you'd need to replace the tail slap instead of the kick
20:14 < Eronarn> wrt beastly: spliting tentacle mut into three would handily solve that
20:14 < Eronarn> it'd just give the extra tentacle mut
20:14 < HangedMan> said extra tentacle mut doesn't block any slots, right?
20:15 < HangedMan> unless it gives deformed as a side effect it could be "optimal" to learn it for the aux on anybody not using other forms
20:15 < Eronarn> HangedMan: well, my justification for octos not having cloaks is that tentacles get tangled. you could make extra tentacles 3 block cloak (they come out of your back)
20:15 < HangedMan> mm, interesting
20:17 < Eronarn> like dr. octopus ;)
20:18 < CIA-112> kilobyte * rd74c999e13a0 /crawl-ref/source/ (stairs.cc tags.cc tilemcache.cc tilemcache.h): Don't read the mcache in any case; drop unmarshalling code.
20:18 <+|amethyst> hm... I've got MUT_TENTACLES constricting now
20:18 < Eronarn> i don't think that casting a L1 spell should let you constrict
20:18 < Eronarn> though a higher level tmut that did would be reasonable
20:19 < Eronarn> and constriction for DS could also be reasonable (maybe monstrous only)
20:20 <+kilobyte> Eronarn: trivially fixable by giving claws instead
20:20 < Eronarn> kilobyte: giving clawsi s boring and means it's only useful with UC
20:21 <+kilobyte> or we could make a new mutation for the hands slot
20:21 <+|amethyst> wouldn't you only get claws if you were unarmed to being with?
20:21 <+kilobyte> or use tentacles 1 or 2
20:21 <+|amethyst> s/being/begin/
20:21 ۰۰-- ais523 [~ais523@unaffiliated/ais523] has quit [Remote host closed the connection]
20:22 <+kilobyte> Beastly cares about worn armour, not weapons
20:22 < Eronarn> has it been discussed having the spell give you a random thing?
20:22 < Eronarn> ??beastly appendage
20:22 < Henzell> beastly appendage[1/1]: Turns one of your body parts into the "monstrous" equivalent (horns, tentacles, etc). Is not powerful enough to meld armour. Level 1 pure Transmutations. Book of Changes.
20:22 ۰۰-- ainsophyao [~ainsophya@h175-177-006-200.catv02.itscom.jp] has quit [Remote host closed the connection]
20:22 <+kilobyte> Eronarn is right, claws do nothing if you use a weapon
20:22 < Eronarn> well, they do help offhand punch damage, but mostly nothing still :P
20:22 ۰۰-- ussdefiant [~chatzilla@S010678cd8e7ca058.ed.shawcable.net] has quit [Ping timeout: 252 seconds]
20:23 < Eronarn> what all effects can beastly give currently?
20:23 < alefury> kilobyte: giving claws would let people butcher with a blunt weapon, right?
20:23 < alefury> cursed blunt weapon, that is
20:23 < HangedMan> I'm not entirely sure how powerful auxes are, but wouldn't waiting for appendage to give claws be desirable for the base damage boost
20:23 < HangedMan> horns, tentacles, and talons
20:24 <+kilobyte> (and a custom new mutation for octopodes)
20:24 ۰۰-- ussdefiant [~chatzilla@S010678cd8e7ca058.ed.shawcable.net] has joined ##crawl-dev
20:25 < HangedMan> oh right, and tentacle spike
20:25 < HangedMan> can they mutate tentacle spike regularly?
20:25 < Eronarn> would anyone have issues with the extra tentacles mut i mentioned, which is blocked by cloak slot?
20:25 < Eronarn> enough to batter, not enough to constrict - and then higher levels, via monstrous DS, could constrict (while losing cloak)
20:26 <+|amethyst> losing cloak without losing body armour is kind of weird
20:26 <+kilobyte> Eronarn: sounds like a good use for existing MUT_TENTACLE on levels 1/2
20:26 < HangedMan> draconians get that
20:26 <+|amethyst> draconians lose body armour without losing cloak
20:26 <+|amethyst> this would be the opposite
20:26 < Eronarn> then you've got one for boots, cloak, helmet... one for gloves would round it out
20:27 < HangedMan> oh whoops
20:27 < Eronarn> this could reasonably be claws, but that does pose some problems the others wouldn't
20:28 < Eronarn> maybe a new mut? you have fanged hands on your palms :)
20:28 < Eronarn> doesn't affect punching, but gives you a hand-bite aux
20:29 <+kilobyte> tendrils?
20:29 < Eronarn> ooh
20:29 < Eronarn> what about scythe like blades sprouting off of your wrists
20:29 < HangedMan> plant demons
20:29 < Eronarn> blocks gloves but won't affect punches
20:30 < Eronarn> since you have to backhand to use them
20:31 <+|amethyst> hm
20:32 <+|amethyst> what body part do nagas constrict with?
20:32 <+|amethyst> I mean, how to describe it in the A screen
20:32 < Eronarn> tail
20:32 <+|amethyst> that's what I have now, but you.has_tail is false
20:32 < Eronarn> or 'snake-like lower-half'
20:32 <+|amethyst> and they don't get a tail slap
20:32 < Eronarn> that's for mf tail only
20:32 <+|amethyst> also drac
20:32 < Eronarn> nagas can get a tail attack with the stinger mutation
20:33 <+kilobyte> here's how it's done:
20:33 <+kilobyte> ??constriction
20:33 < Henzell> constriction[1/4]: http://www.goblinscomic.com/09132011/
20:33 < Eronarn> they just don't get a base one
20:34 < Eronarn> also, here's a point: we could just use this opportunity to fix claws :P
20:34 < HangedMan> nooooooooo
20:34 <+kilobyte> claws are used by 6984677386 races
20:34 < Eronarn> 'sharp fingernails' mut (rake aux, no gloves, unchanged punch) and 'real' claws like trolls and felids and ghouls have
20:34 <+kilobyte> they're also an aux that actually makes sense to use UC only
20:34 < Eronarn> which flat out replace punching
20:35 < Eronarn> kilobyte: the idea is that having sharp fingernails should let you use them as an attack but you need pretty big claws to be able to just stab with your hand
20:35 < Eronarn> and the two don't have to be the same mutation
20:36 < Wensley> it might make sense to make distinctions between different types of tails
20:36 < CIA-112> |amethyst * r1763a442fe56 /crawl-ref/source/mutation.cc: Nagas constrict with their entire lower body, not just the tail.
20:36 < Eronarn> in a fight you might rake with your fingers, rather than punch, but only do this if you're going to be able to hit their eyes
20:36 < Eronarn> if you have sharp fingernails, you can lacerate, so you will occasionally want to scratch generally instead of punchg
20:36 < Eronarn> if your hands are BUTCHER KNIVES, you want to just punch with extended fingers all the time
20:37 < HangedMan> scratching a giant eyeball does not sound very difficult
20:37 ۰۰-- nooodl [~nooodl@91.177.143.102] has quit [Ping timeout: 240 seconds]
20:37 < Eronarn> fr: 1/100 chance of 'poke out' being the attack verb for killing an eye with unarmed
20:39 < Eronarn> but yeah, here's a specific proposal: sharp fingernails mut, 3 levels, blocks gloves at 3, no effect if wearing gloves, chance to replace offhand punches with a rake aux based on UC + mut level
20:39 < Eronarn> and no effect on punches generally
20:40 < HangedMan> and any changes to the current claws mutation itself?
20:40 < Eronarn> current claws species have something else entirely, which does affect 'punches' (because they stab/slice on all attacks)
20:41 < Eronarn> though you could let them get a rake aux too (e.g., thjis could be what causes bleeding, rather than any claws hit)
20:46 ۰۰-- joosa [joosa@heh.fi] has quit [Ping timeout: 255 seconds]
20:57 ۰۰-- ainsophyao [~ainsophya@h175-177-006-200.catv02.itscom.jp] has joined ##crawl-dev
21:03 ۰۰-- ainsophyao [~ainsophya@h175-177-006-200.catv02.itscom.jp] has quit [Remote host closed the connection]
21:25 ۰۰-- ussdefiant [~chatzilla@S010678cd8e7ca058.ed.shawcable.net] has quit [Ping timeout: 268 seconds]
21:27 <+SamB> kilobyte: soo, I'm using putty, with TERM=putty-256color and "Window > Colours > Bolded text is a different colour" unchecked.
21:27 <+SamB> this does not exactly work, for some reason :-(
21:27 ۰۰-- Vandal [~Vigilant@cpe-65-185-156-241.woh.res.rr.com] has quit []
21:29 <+|amethyst> SamB:  crawl doesn't do 256-colour at all AFAIK
21:29 <+SamB> |amethyst: yes, but it's only using the first 8!
21:29 <+SamB> and trying to use bold to get the other 8
21:30 <+SamB> (with no effect at all for black)
21:30 <+|amethyst> that's how you have to do it on most non-256-colour terminals
21:30 <+SamB> perhaps so...
21:31 <+SamB> but it still sucks
21:31 <+SamB> and it's not even necessary on a proper 16-color terminal
21:31 <+|amethyst> I guess curs_fg_attr could check t_co
21:32 <+|amethyst> or colors or whatever it's called
21:34 ۰۰-- ussdefiant [~chatzilla@S010678cd8e7ca058.ed.shawcable.net] has joined ##crawl-dev
21:36 <+|amethyst> hm... right, curses has the weird thing where you need to set up colour pairs
21:37 <+SamB> |amethyst: the what now?
21:37 <+SamB> that's for a wierd kind of terminal that nobody actually uses, I thought
21:38 <+|amethyst> such terminals are the reason the API is that way, but that is in fact the curses API
21:39  * SamB spends more time reading the terminfo documentation, as a rule
21:40 < HangedMan> ??catoblepas[4]
21:40 < Henzell> catoblepas[4/5]:   Your earth elemental is engulfed in calcifying dust. Your earth elemental turns to stone!
21:41 < Wensley> I blame MarvinPA for being the one who told me not to bother with implementing a proper rPetrify and just add catlobes to the hardcoded list of monsters that can't be petrified
21:41 < Wensley> whether or nor MarvinPA is actually the one who told me to do that, I blame him nonetheless :P
21:42 < HangedMan> stone golems are also there
21:42 < HangedMan> it wouldn't be hard to add earth elementals to that list anyway!
21:42 < Wensley> there should probably be a lot of things there
21:43 < HangedMan> well the comments there say they assume skeletons can be fossilized, and that materials like clay can harden
21:43 < Wensley> I suppose it's not really the end of the world to be lazy and have a hardcoded list rather than making it a true resist
21:43 < HangedMan> what else would even be relevant
21:43 <+|amethyst> A spirit drawn from the elemental plane of earth, which exists in this world by inhabiting a lump of earth and rocks.
21:44 <+|amethyst> So apparently it's not all stone
21:44 < HangedMan> I only asked about this because I want to push some earth elementals and catlobes together in a vault
21:44 < Wensley> I think it is eminently reasonable to give earth elemental rpetrify
21:45 < Wensley> nomes will have it, one day
21:46 <+SamB> too bad the curses model is more complicated than both the Tektronix- and HP-like models :-(
21:47 < HangedMan> I'd also ask to improve the ai so that catoblepae would breathe their clouds through rPetrify monsters but that'd be a headache presumably
21:47 ۰۰-- ussdefiant [~chatzilla@S010678cd8e7ca058.ed.shawcable.net] has quit [Ping timeout: 240 seconds]
21:48 < Wensley> don't they already breathe through anything
21:49 < HangedMan> they can, but in a corridor I couldn't make them do it
21:52 ۰۰-- ussdefiant [~chatzilla@S010678cd8e7ca058.ed.shawcable.net] has joined ##crawl-dev
21:52 <+|amethyst> what about non-rPetrify monsters?  would they care?
21:54 ۰۰-- alefury [~galefury@B4115.karlshof.wh.tu-darmstadt.de] has quit [Quit: ChatZilla 0.9.88 [Firefox 8.0/20111104165243]]
21:55 < Eronarn> having a list of which monsters are stone would be good for LRD of course
21:59 <+SamB> or, like, a flag
22:00 < HangedMan> a catoblepas breathed through a rat to get me, killing it
22:01 < HangedMan> but it won't do that for a stone golem even though the golem wouldn't be hurt
22:01 ۰۰-- edlothiol [~edlothiol@ip-95-223-51-38.unitymediagroup.de] has quit [Ping timeout: 248 seconds]
22:02 ۰۰-- edlothiol [~edlothiol@ip-95-223-51-38.unitymediagroup.de] has joined ##crawl-dev
22:02 ۰۰-- mode/##crawl-dev [+v edlothiol] by ChanServ
22:03 < HangedMan> but it wouldn't breathe through a dragon okay I have no idea how this works
22:04 <+|amethyst> hm, monster AI does not generally look at resists
22:04 <+|amethyst> or, at least, bolt::nasty_to does not
22:04 ۰۰-- ussdefiant [~chatzilla@S010678cd8e7ca058.ed.shawcable.net] has quit [Ping timeout: 252 seconds]
22:05 <+|amethyst> except for holy flame and pain/agony
22:06 < Wensley> petrify doesn't even have a proper resist, I believe the only check for immunity is when you are already in the cloud
22:09 <+|amethyst> hm
22:09 <+|amethyst> weird
22:10 <+|amethyst> not sure how it's determining what it can breathe through
22:11 <+|amethyst> it's breathing through a naga hd:10  but not a clay golem hd:1
22:11 < HangedMan> yes to rat, gecko, human, no to giant or dragon
22:12 <+|amethyst> oh
22:13 <+|amethyst> it uses mons_power, which depends only on the base type's HD, not the monster's actual HD
22:20 ۰۰-- ussdefiant [~chatzilla@S010678cd8e7ca058.ed.shawcable.net] has joined ##crawl-dev
22:23 <+|amethyst> hm... petrifying cloud does damage... where is that defined?
22:24 <+|amethyst> oh, I see
22:30 ۰۰-- Vandal [~Vigilant@cpe-65-185-156-241.woh.res.rr.com] has joined ##crawl-dev
22:35 <+|amethyst> it seems like the simplest thing to do would be to reduce the beam's foe_ratio
22:36 <+|amethyst> but that will also make friendly catlobes more likely to breathe through you
22:37 < HangedMan> friendly catlobe show up in what circumstances
22:38 < HangedMan> shadow creatures, xom polymorph, polymorphing your own allies
22:39 <+|amethyst> enslavement
22:40 < HangedMan> aside from being trapped in a corridor one would be able to move out of the cloud and then enjoy a petrifying enemy, so I don't think it'd be that bad
22:41 < Eronarn> !learn add octopode Yes, they eat fruit. http://www.etsy.com/listing/73961114/tentacles-and-fru[D[D[Dit-pear
22:41 < Henzell> octopode[8/8]: Yes, they eat fruit. http://www.etsy.com/listing/73961114/tentacles-and-fru[D[D[Dit-pear
22:41 ۰۰-- santiago [~quassel@190.190.173.140] has joined ##crawl-dev
22:41 <+|amethyst> !learn edit octopode[8] s/\[.*D//
22:42 < Henzell> octopode[8/8]: Yes, they eat fruit. http://www.etsy.com/listing/73961114/tentacles-and-fruit-pear
22:42 ۰۰-- petete [~quassel@190.190.173.140] has quit [Ping timeout: 240 seconds]
22:42 < Eronarn> whoops
22:44 < Eronarn> http://www.etsy.com/listing/78527282/1885-armed-calamary-octupus-victorian
22:45 < Eronarn> Glorious Victorian-era black and white book plate from an 1885 American zoology book that was “Fully illustrated with scientific accuracy.”
22:46 <+|amethyst> HangedMan:  okay, reducing the foe ratio from the default 80 to 30 causes the catlobe to breathe through two but not three stone golems to hit an XL 15 player
22:46 < HangedMan> awesome
22:47 < HangedMan> wonderful material to work with
22:47 ۰۰-- santiago [~quassel@190.190.173.140] has quit [Ping timeout: 240 seconds]
22:53 < CIA-112> |amethyst * r8116a39d4572 /crawl-ref/source/mon-cast.cc: Make catlobes more likely to breathe through allies.
22:53 < HangedMan> danke!
22:54 <+|amethyst> no problem
22:54 <+|amethyst> I'm making it your responsibility to test the lair ending and make sure its balance hasn't changed too much
22:54 <+|amethyst> I'm not a good enough player to judge
22:54 <+|amethyst> :)
22:54 < HangedMan> oh right there are other vaults with catlobes
22:54 < HangedMan> will do
22:56 <+|amethyst> I think it probably shouldn't matter much at all, since you'll usually deal with the other monsters before heading down the corridor towards the catlobe
22:56 < HangedMan> yeah
22:59 <+|amethyst> if anything, it eliminates a potential scum where you let a death yak or two chase you to keep the catlobe from firing
22:59 <+|amethyst> some may call that a "tactic" instead :)
23:04 ۰۰-- HangedMan [63e6725f@gateway/web/freenode/ip.99.230.114.95] has quit [Ping timeout: 258 seconds]
23:13 ۰۰-- HangedMan [63e6725f@gateway/web/freenode/ip.99.230.114.95] has joined ##crawl-dev
23:13 ۰۰-- stabwound [~stabwound@unaffiliated/stabwound] has quit [Quit: ZNC - http://znc.sourceforge.net]
23:14 ۰۰-- stabwound [~stabwound@unaffiliated/stabwound] has joined ##crawl-dev
23:32 < Wensley> I approve of this buff/nerf
23:33 <+elliptic> |amethyst: if there is a way of using the catlobes to petrify the death yaks then that might be worthwhile
23:34 <+elliptic> also there are three catlobes in the lair ending
23:34 <+elliptic> however the change sounds good to me :)
23:34 <+|amethyst> it should be possible now, assuming you aren't extremely low-level
23:35 <+elliptic> generally the death yaks are the danger in that vault, not the catlobes
23:35 < Wensley> !lg * map=~catoblepas s=killer
23:35 < Sequell> 20 games for * (map=~catoblepas): 9x a catoblepas, 9x a death yak, a griffon, a vapour
23:35 < Wensley> a dead heat
23:35 <+elliptic> hm, that surprises me
23:36 <+elliptic> I guess there have been a lot of people dying to catoblepae because they don't realize that they need to play slowly and step out of the breath
23:37 < Wensley> "play slowly and step out of the breath" sounds like a UK public service slogan
23:37 < Wensley> in any case yes, catlobes are retroactively designed to defeat autofight
23:40 ۰۰-- MarvinPA_ [~MarvinPA@85.211.123.203] has joined ##crawl-dev
23:40 ۰۰-- mode/##crawl-dev [+v MarvinPA_] by ChanServ
23:43 ۰۰-- MarvinPA [~MarvinPA@thecampbells.demon.co.uk] has quit [Ping timeout: 240 seconds]
--- Log closed Thu Dec 29 00:00:23 2011