00:03:09 Unstable branch on crawl.develz.org updated to: 0.10-a0-1032-gda6d1c2 (32) 00:18:21 Windows builds of master branch on crawl.develz.org updated to: 0.10-a0-1032-gda6d1c2 00:25:39 -!- ixtli has quit [Ping timeout: 252 seconds] 00:32:29 !tell galehar What do you think about showing all skills that are positive? the current system is pretty confusing with some starting classes, since you can start with skills you can't see... 00:32:29 elliptic: OK, I'll let galehar know. 00:34:32 -!- valrus has quit [Remote host closed the connection] 00:39:59 -!- Wensley has quit [Ping timeout: 248 seconds] 00:41:48 Huh, the new system must be a nightmare for Wn 00:42:17 why? 00:43:03 You can start with skills you have no items/spells for and have no idea about it 00:43:29 it is confusing, sure, but you can see those skills if you know to look for them with m* 00:43:55 Oh! That is still available. 00:44:01 -!- Pacra has joined ##crawl-dev 00:45:04 !tell galehar To clarify, I'm suggesting making untrainable skills that are positive visible without needing to press *... but they still wouldn't be trainable 00:45:04 elliptic: OK, I'll let galehar know. 00:50:03 -!- Zaba has quit [Read error: Operation timed out] 01:01:12 -!- st_ has joined ##crawl-dev 01:03:09 -!- Zaba has joined ##crawl-dev 01:28:00 -!- dtsund has quit [Quit: dtsund] 02:21:09 -!- syllogism has joined ##crawl-dev 02:39:10 Unstable branch on crawl.akrasiac.org updated to: 0.10-a0-1032-gda6d1c2 (32) 02:44:51 -!- R18 has quit [Remote host closed the connection] 02:54:13 -!- Textmode has quit [Ping timeout: 260 seconds] 02:58:12 -!- mikee_ has quit [Ping timeout: 240 seconds] 03:04:24 Lee's Rapid Deconstruction fails to target self (https://crawl.develz.org/mantis/view.php?id=4662) by CTB 03:04:37 -!- Galefury has joined ##crawl-dev 03:04:42 -!- Galefury is now known as alefury 03:15:16 -!- mikee_ has joined ##crawl-dev 03:30:47 -!- Twilight_1 has quit [Read error: Connection reset by peer] 03:30:53 -!- Twilight_1 has joined ##crawl-dev 03:46:16 -!- Pacra has quit [Quit: This computer has gone to sleep] 03:52:13 -!- monqy has quit [Quit: hello] 04:01:25 i just got healed by something on a zig holy level in lich form 04:01:38 in wizmode, but that shouldnt break things... 04:02:13 might have been a deadly blow, but i cant see the "die?" prompt in the message log 04:03:18 not sure that shows up 04:05:02 alefury: do a dump 04:05:15 alefury: the textlog should give yo an indication of what happened 04:05:19 k 04:05:23 it'll be somethin glike "die x/x" 04:05:27 er, died x/x 04:06:21 killed by a pearl dragon 04:06:38 there you go :) 04:06:46 you escaped past it without noticing 04:07:03 good good 04:07:08 new holy levels are nice btw 04:07:18 much more interesting than the old ones 04:07:33 not sure they are any less deadly, but at least more fun 04:07:55 the pearl dragons blocking los with their fire blasts is interesting 04:08:03 also, ophans making clouds around you is pretty nice 04:08:13 :D 04:08:35 i am originally responsible for that, though how it has ceanged recently, I am uncertain. 04:09:13 well, shedu, ophans, pearl dragons, cherubs and holy swine got added to holy zig levels recently 04:09:33 and cherubs got changed a bit, ranged weapon and flaming weapon 04:09:43 previously they were just reskinned angels 04:09:52 thats mostly it i think 04:10:05 ah 04:10:09 and that was in the initial spec 04:10:13 the shedu are a bit annoying in large quantities, but should be perfectly fine in the normal game 04:14:39 -!- Zaba has quit [Read error: Operation timed out] 04:18:43 -!- Zaba has joined ##crawl-dev 04:30:32 -!- mikee_ has quit [Disconnected by services] 04:30:46 -!- mikee_ has joined ##crawl-dev 04:35:57 -!- mikee_ has quit [Ping timeout: 255 seconds] 04:37:28 SamB: ?? and ?? might not be letters but ß is (and the Dutch are deluded into thinking ij is). It's just CIA that has a bug there. 04:38:32 alefury: right now cherubs are reskinned daevas with heal+slings/bows instead of smite :( 04:38:52 so? 04:39:07 alefury: I'm really unhappy with my change. Not reverting it only because even that is better than being exactly angels. 04:39:39 they have a different kind of ranged attack, and an extra utility spell, whats wrong with that? 04:39:49 theres lots of more similar enemies in crawl 04:39:55 -!- mikee_ has joined ##crawl-dev 04:40:13 not at such a high threat level, granted, but i dont see the problem 04:40:59 can be defended against, different loot, healing, thats a very noticeable difference 04:41:32 also no smite targeting 04:42:21 they are not that different in zigs, but in a proper levels with some cover they should be quite different from daevas 04:47:47 pearl dragon and apis are missing a corpse tile btw 04:56:51 ah I see, the tiles are generated but not linked to the monster type 05:00:03 OSX builds of master branch on crawl.develz.org updated to: 0.10-a0-1032-gda6d1c2 05:02:37 due: as your creations, should apis and pearl dragon corpses be zombifiable? It seems to me that's a fitting way for Teh Evul guys to desecrate their corpses, and we have unholy servitors. 05:12:26 oh, regarding weird holy stuff: pan lord melee attacks can be holy wrath branded, is that intentional? 05:16:24 ... 05:18:27 -!- capablanca has joined ##crawl-dev 05:20:17 -!- mikee_ has quit [Ping timeout: 240 seconds] 05:20:21 -!- capablanca is now known as mikee_ 05:38:43 -!- Textmode has joined ##crawl-dev 05:41:24 -!- evilmike has quit [Quit: something happened] 05:49:22 -!- elliptic has quit [Ping timeout: 244 seconds] 06:08:35 alefury: did you get a pan lord with holy melee? 06:08:48 alefury: there is a check for that... 06:10:39 03kilobyte * r12feed03936d 10/crawl-ref/source/ (4 files in 4 dirs): wip: holy Pan level 06:10:47 03kilobyte * r41b4775949d4 10/crawl-ref/source/ (rltiles/dc-corpse.txt tilepick.cc): Fix apis and pearl dragon corpses in tiles. 06:10:47 03kilobyte * rf9dca19b17b1 10/crawl-ref/source/ (5 files in 2 dirs): Make Elyvilon heal nearby hostiles when you kill an apis. 06:10:49 whoops 06:14:30 03kilobyte * r1533c11065f3 10/crawl-ref/source/dat/des/branches/pan.des: Revert the barely started holy Pan level, leaving the seraph tile. 06:14:31 -!- Torokasi has joined ##crawl-dev 06:51:51 -!- edlothiol has joined ##crawl-dev 06:53:09 kilobyte: i did, but it was quite a while ago 06:53:18 didnt see a commit about it, though 06:53:26 might have missed it 06:55:17 even if its possible its not a big deal imo 06:55:17 it never says its holy, it simply does more damage to demons and undead, which would be quite useful for a pan lord 06:55:17 assuming it has to claw its way to the top somehow 07:06:59 -!- Mu_ has joined ##crawl-dev 07:20:25 Unable to train spellcasting after manual use. (https://crawl.develz.org/mantis/view.php?id=4663) by Ragdoll 07:50:38 IonFrigate the Destroyer (L19 DEIE) ASSERT(y >= 1 && y <= sz.y) in 'libutil.cc' at line 906 failed on turn 80430. (Shoals:2) 07:59:43 -!- Torokasi has quit [Ping timeout: 258 seconds] 08:01:12 -!- jeanjacques has joined ##crawl-dev 08:10:34 -!- capablanca has joined ##crawl-dev 08:11:39 -!- mikee_ has quit [Disconnected by services] 08:11:40 -!- capablanca is now known as mikee_ 08:35:17 -!- mumra has joined ##crawl-dev 08:44:46 -!- MarvinPA has quit [Read error: Connection reset by peer] 08:47:52 -!- MarvinPA has joined ##crawl-dev 08:49:21 -!- Torokasi has joined ##crawl-dev 09:15:20 -!- Brannock has joined ##crawl-dev 09:26:09 ??craidargo 09:26:10 I don't have a page labeled craidargo in my learndb. 09:32:34 -!- edlothiol has quit [Ping timeout: 244 seconds] 09:37:40 -!- edlothiol has joined ##crawl-dev 09:45:33 -!- G-Flex has joined ##crawl-dev 09:46:32 I was curious: What library or libraries does DC:SS use for providing console-based output in Windows? someone's asking me about it and I realized I have no idea 09:47:40 G-Flex: just the win32 API 09:48:02 oh okay 09:48:17 I figured it probably used some sort of cross-platform deal for linux and windows 09:48:23 but I guess it's not that easy! 09:48:42 WriteConsoleOutputW() is the main workhorse 09:49:04 so basically it uses a completely different output path depending on platform 09:49:54 -!- valrus has joined ##crawl-dev 09:53:41 yeah 09:54:03 thanks 09:54:08 it's not like win32 has any POSIX functions do to that... 09:54:20 well yeah 09:54:30 but I didn't know if there were any cross-platform APIs out there 09:58:24 libtcod looks a bit promising 09:58:59 it does its own drawing rather than using win32 console, but that makes sense as win32 console sucks halfling balls 09:59:11 it is unmodified since Windows 3, too 09:59:18 that sounds aggravating 09:59:25 I much prefer having actual console output 09:59:29 win16 had no console, dude 09:59:34 libtcod can't do that on Unix, though 09:59:47 so I can change things like color definitions, font, etc. 09:59:56 and whatever else I feel like 10:00:02 NT 3.x, of course, does 10:00:04 SamB: uhm, it did... 10:00:16 kilobyte: that was an illusion 10:00:24 I play something else that has a Windows port using pdcurses and it's pretty frustrating not being able to change any of that 10:00:28 or possibly a DOS box 10:02:06 kilobyte: The console APIs were new in win32 10:02:44 https://secure.wikimedia.org/wikipedia/en/wiki/File:Win32desktop.png -- look at the right-bottom window 10:03:24 the URL says "32" but it's a regional update to 3.1 10:03:35 maybe it's win32s 10:03:39 (didn't notice the URL before writing) 10:03:46 or, you know, a DOS box 10:03:48 no, I used it for quite a bit 10:03:57 it sure looks like a DOS box 10:04:08 DOSbox is ten years or so newer... 10:04:13 not DOSbox 10:04:19 DOS box is an older thing 10:05:01 if we're talking Win 3.x then that obviously ran straight on top of DOS anyway 10:05:09 oh, Windows 1.0 had console as well 10:05:24 it's what you call the DOS sessions in Win 3.x/9x 10:05:25 so yeah, the command line interface/console was probably just DOS 10:05:37 but it's still a console 10:05:37 G-Flex: yeah, even Windows 95 did, even though the installer set up both 10:05:39 there are no Console APIs 10:05:56 Win16 programs aren't allowed to use it 10:06:11 though it *is* possible to pull of some tricks with pipes and Win32 helper programs 10:06:20 er. *off 10:06:45 If you wanted to use that thing, you had to write a DOS program 10:07:21 At some point, it became possible to run several of these in virtual machines 10:07:32 so that you could have more than one DOS box 10:07:36 hrm, Microsoft carefully drops documentation the moment they drop support for a given version, and I've dumped my Win3 books away :( 10:07:40 -!- ais523 has joined ##crawl-dev 10:07:42 I'm pretty sure Win9x didn't use virtual machines for those 10:07:52 G-Flex: very lite VMs 10:08:00 the Virtual 8086 kind 10:09:04 SamB: it's a matter like win32 vs win64... Microsoft's term for such helpers is "thunking": it translates between two APIs, which are binary incompatible but source nearly 100% compatible 10:09:07 -!- mikee_ has quit [Quit: HydraIRC -> http://www.hydrairc.com <- Nine out of ten l33t h4x0rz prefer it] 10:09:25 like, every single call was translated outright, just replacing WORD with DWORD 10:09:54 (which is a terrible misnomer, since word in i386 is 32 bits not 16... and on amd64 64 bit) 10:10:04 and because a thunk is something entirely different 10:10:27 heck messages still use the names wParam and lParam to this day 10:10:37 wParam used to be 16 bits on win16 10:20:17 crap, indeed my win16 books are gone, and my google-fu sucks too much to find this out in several minutes 10:34:55 -!- Wensley has joined ##crawl-dev 10:58:43 -!- elliptic has joined ##crawl-dev 11:21:42 Mutation replaced by racial mutation (https://crawl.develz.org/mantis/view.php?id=4664) by El Cubo 11:25:54 -!- G-Flex has left ##crawl-dev 11:31:49 -!- monqy has joined ##crawl-dev 11:44:54 kilobyte: yeah, I know 11:45:17 I could check the headers from MSVC 4.5... 11:46:16 wait, maybe it's 1.5 11:46:37 * SamB might be getting confused with borland? 11:46:55 oh, I've been thinking 11:47:14 it would be nice if there was an official mirror of crawl.git on github 11:55:10 kilobyte: yeah, I just ran "grep -nH -e Console -r ." in MSVC 1.5's include directory, and it didn't find a thing 12:06:30 -!- valrus has quit [Remote host closed the connection] 12:06:55 -!- valrus has joined ##crawl-dev 12:11:18 -!- valrus has quit [Ping timeout: 255 seconds] 12:14:31 SamB: an official mirror on github? you mean for people who are too lazy to use gitorious? :P 12:15:37 Wensley: Well, for people who think that github's UI for browsing commits might be a better match for the committers' laziness, actually 12:16:06 I'd do it myself, but I can't make a mirror and then fork it, too 12:17:53 what sort of ui do they have 12:34:12 oh, whoops 12:34:18 * SamB forgot he had a question pending 12:36:28 * SamB looks around for a good demonstration 12:38:03 -!- edlothiol has quit [Ping timeout: 244 seconds] 12:38:32 This one is kind of cute: https://github.com/mojombo/octobeer/network 12:39:05 y'know what's nice? https://github.com/torvalds/linux/network 12:39:13 takes quite a lot scrolling to truly appreciate it, though ;p 12:39:26 yeah, I kind of figured 12:39:54 mojombo/god was looking bad enough, and I've hardly heard of that project 12:40:25 ah yes, I've heard of that visualization before... but doesn't that just track pull requests? 12:40:37 and I think the only person who actually does requests those is dtsund 12:40:51 Wensley: no, that's not related to pull requests 12:40:56 it just tracks commits 12:41:05 okay, I must be thinking of a different view 12:41:20 Wensley, if it just tracked pull requests, it wouldn't work for linux :P 12:41:26 I think they added pull requests later? But I'm not sure. 12:42:21 Zaba: surely linux doesn't use github, I thought they just sent each other patches 12:42:27 and, really, the other pieces of github's UI look more polished 12:42:40 Wensley: they used to use kernel.org 12:42:45 Wensley, (git.)kernel.org is down "for maintenance" after the security breach, and linus put the official linux tree on github 12:42:53 they outgrew "send patches" quite some time ago 12:42:58 hence git's existance 12:43:12 Wensley, he does not, however, make use of pull requests 12:43:14 Wensley: Linux is using github as the official version temporarily, because kernel.org got hacked 12:43:25 right, I know about the hack, curious to see what comes of it 12:43:31 -!- mumra has quit [Quit: offski] 12:43:40 also, I figured that linus wrote git-format-patch for his own use :P 12:43:47 maybe somebody will hack github next :P 12:44:36 well, as the changes get closer to Linus, they get more and more likely to be pulled rather than sent as disembodied emails 12:45:34 Zaba: there are probably less individuals having the necessary access to github's servers to pull that off... 12:46:00 -!- Torokasi has quit [Ping timeout: 260 seconds] 12:46:01 not that anyone connected to kernel.org was helping the evildooers *on purpose* 12:46:12 erm... brute forcing someone's pasword can hardly be called "the hack" 12:46:45 I thought they were stealing SSH keys or somesuch 12:46:46 and ssh is only allowed via ssh-keys now - that came from it, Wensley 12:46:49 -!- Textmode has quit [Quit: Ex-Chat] 12:47:27 Napkin: You forgot to specify a where 12:49:03 kernel.org, of course - or was there a break-in elsewhere? 12:49:41 and gitorious was chosen, because github was terribly slow at the time of the decision - and github offered too little diskspace for crawl 12:50:10 only 50mb for free, or something like that 12:50:23 right, I can only host one branch of crawl at a time on my github account 12:50:24 at that time 12:50:33 (last I checked) 12:50:48 what access to CDO's git, Wensley? 12:51:01 *want access 12:51:07 Napkin: yeah sure thing :) 12:51:42 no wiki, no issue tracker, blabla, though - just git and gitweb ;) 12:52:48 -!- Pacra has joined ##crawl-dev 12:54:40 Wensley: you can have as many branches as you want 12:54:44 just only one repository 12:55:01 my branches aren't a tree though, they are all fragmented because I am terrible at git 12:55:35 github raised it to 300mb now 12:56:01 so like I make a branch, host it on github, it gets pulled into crawl master, and then I try to merge master into my branch locally and it gets all screwed up so I have to make a new branch 12:56:16 hehe 12:56:27 noob ;) 12:56:36 ;_; 12:56:37 Wensley: oh, yeah, that 12:57:11 that might be the fault of whoever pulled it 12:57:21 if, say, they cherry-picked for no reason 12:57:26 (I blame MarvinPA) 12:57:40 -!- MarvinPA_ has joined ##crawl-dev 12:57:45 hahaha 12:58:01 he could have forced you to make your branch merge cleanly first ;> 12:59:01 haha, I do! before I push my branch to github I always pull master and re-merge, and then only after my branch gets pulled into master does merging master into my local branch mess up 13:00:04 well, i assume the idea is that you use more branches 13:00:19 i mean.. once one of your branches is merged, there is no need to work on it anymore 13:00:31 true 13:01:09 the point is that I still have to delete my remote branch and push a new one, as far as I know, since they'll both commit the whole history 13:01:15 -!- MarvinPA has quit [Ping timeout: 260 seconds] 13:02:02 possibly because I don't have the foresight to push the master branch of crawl to remote first, and then branch off of that (if that is what I am supposed to do) 13:02:19 i'd keep an upstream branch, which is a mirror of the original crawl master 13:02:36 branch it, work on that, let it get mergen, delete it 13:02:50 pull the original crawl master into your clean upstream branch again 13:03:04 then rebranch 13:03:26 always branch, when you work on something new 13:04:04 * SamB often finds he has to do this retroactively 13:04:05 I actually just signed up for a github account, maybe I should try forking 13:04:13 wait, I mean gitorious 13:04:28 well, forking on gitorious is, er, interesting :-( 13:04:48 I eventually went and deleted all of the branches that it gave me 13:06:00 Wensley: I end up having to rebase my local branches most of the time... 13:06:17 I don't even know what rebasing is supposed to do! 13:06:19 at least, if there's anything left that didn't get accepted 13:06:44 well, rebasing is the thing that got you into this mess in the first place, more-or-less 13:06:49 ha 13:06:56 oh, rebasing it what you do, so your changes apply cleanly, Wensley 13:07:05 I see 13:07:06 very important 13:07:24 (oops, gotta go! later) 13:07:24 Napkin: auto-merge commits are not unclean, dude 13:07:28 o/ 13:07:52 rebasing is what you have to do when the other guy done rebased 13:07:54 but they can become unclean on the next commit to upstream 13:08:08 and then you're forcing otherse to fix your conflicts 13:08:17 Usually that doesn't happen, though 13:08:23 it happens a lot 13:08:36 and and you can't expect me to rebase every damn time someone else commits 13:09:00 anyway, I think it's way overused 13:09:15 and and you can't expect others to resolve your conflicts every damn time ;) 13:09:21 I don't 13:10:11 but usually, it's not even the same file that changed 13:10:19 rebasing is really quite easy - i don't get the point 13:10:42 well, merging makes it easier to follow the history 13:10:55 rebasing edits history, I don't really like the way a VCS requires people to repeatedly change the history just to get things to merge smoothly 13:10:59 since those are your changes, you should know way better how to solve conflicts 13:11:08 ais523: the VCS totally doesn't require that 13:11:16 the thing about conflicts is, that they're always /both/ people's fault 13:11:26 and they should ideally cooperate in order to figure out the fix 13:11:29 non-sense, ais523 13:11:43 Napkin: a conflict happens because the same bit of code was changed by two different people 13:11:51 and the person who ends up sorting it out is whoever commits second, in practice 13:11:51 Napkin: anyway, most of the time it would be fairly trivial to auto-merge this stuff 13:11:59 -!- Wensley has quit [Ping timeout: 248 seconds] 13:12:30 which is to say, you could just do a "pull" and then a push and off we go 13:13:03 and it makes for a much nicer-looking, more accurate history 13:13:41 * SamB thinks Napkin did too much CVS/SVN in his youth 13:14:04 i didn't at all 13:14:13 or possibly not enough 13:14:21 i just understand the reason to accept only rebased patches :) 13:14:47 Oh, see, that's your problem right there 13:14:49 "patches" 13:15:03 * SamB deals in commits only, thanks. 13:15:14 lol 13:15:27 this is just for the sake of arguing, isn't it? 13:15:40 er, where's that xkcd 13:17:10 -!- edlothiol has joined ##crawl-dev 13:18:27 and the "too much CVS/SVN in his youth" was quite offensive 13:20:43 I think the general rule should be: did these commits get into a public repository? Merge. Did they not, or at most in a semi-public one used by no one who intends to use that as anything more than a temporary version? Rebase. 13:24:46 kilobyte: I personally think it would make sense to go with "merge" a mite more often 13:25:23 and/or get bettter mind-reading powers 13:25:55 (so you can read the intentions better) 13:26:51 * SamB thinks Napkin needs to lighten up 13:27:54 kilobyte: on the whole, of course, I agree with that rule 13:29:28 linear history is easier to read, and much easier to bisect 13:29:47 there is that 13:30:15 but I'm not sure that bisecting faux-linear history gives entirely valid results... 13:30:27 and on the other hand, rebasing breaks commit references, so if anyone saw you referring to a particular one, it's good to preserve that 13:30:31 I mean, it is easier to bisect. I don't know that it's easier to read. 13:30:46 I find it kind of disorienting when the commit dates aren't anything like in order... 13:30:51 how would it be invalid? 13:31:03 heck, I have many commits that are years old 13:31:13 work in progress, dubious ideas, etc 13:31:21 well, maybe the code wasn't broken until after being rebased 13:31:26 that sort of thing 13:31:36 having a long messy link to the past would make the commit graph nearly unreadable 13:31:51 true 13:36:34 kilobyte: just look what it did to my skill-menu help changes, though: https://gitorious.org/crawl/crawl/merge_requests/11 13:36:43 i remember when deep dwarves was an external branch... the person who wrote it merged in irregular intervals.. and i had to merge upstream master and that both into my local branch for DGL - and it was a real pain. had he rebased it might have been easier 13:37:01 Napkin: I doubt it 13:37:34 rebasing is especially painful with long-lived branches 13:37:37 Monsters wielding weapons appear as if they not after transforming (https://crawl.develz.org/mantis/view.php?id=4665) by smock 13:37:59 -!- edlothiol has quit [Ping timeout: 244 seconds] 13:39:51 (Old upsteam, Deepdwarf, new Upstream): OODDODOODODOOODDDOUUUUUU was terrible - OOOOOOOODDDDDDUUUUUU or even OOOOOOUUUUUDDDDD would have made it easier for me to fix. 13:41:41 Napkin: or are you talking about a one-time merge you did? 13:42:47 often 13:44:39 well, rebasing the same branch many times when others are already using it is unlikely to make anything *better* 13:45:30 i could have put it on top of upsteam easily then 13:45:59 i bet the person that finally merge deepdwarves cursed quite a bit, too ;) 13:46:05 could be, could be 13:46:32 but rebase *still* doesn't magically make everything better 13:47:31 the guy would probably never have just rebased when you wanted to use his latest, anyway 13:48:33 well, the fact that he never rebased, made me stop updating deep dwarf branch on DGL 13:49:06 it went fine in the beginning - actually it went great with those a remotes - but in the end i could no longer resolve the conflicts 13:49:18 *with those two remotes 13:49:47 -!- dpeg has joined ##crawl-dev 13:50:38 anyway, I'm now breaking my changes into nice bite-sized branches for your convenience 13:55:13 kilobyte: https://gitorious.org/crawl/crawl/merge_requests/10 and 12 are ready for merging, at least 13:56:49 -!- Fail_Bot has joined ##crawl-dev 14:09:29 -!- valrus has joined ##crawl-dev 14:12:55 !tell galehar Can you please look at https://gitorious.org/crawl/crawl/merge_requests/11 ? 14:12:55 SamB: OK, I'll let galehar know. 14:14:09 03kilobyte * r8462aac1127e 10/crawl-ref/source/ (mon-pick.cc mon-pick.h mon-place.cc): Get rid of redundant mons_abyss(). 14:14:09 03kilobyte * rd340038f999d 10/crawl-ref/source/dat/des/branches/dwarf.des: Purge a couple of unused functions from Dwarf lua. 14:14:09 03kilobyte * rc748314e68c0 10/crawl-ref/source/monster.cc: Make holies hang to blessed weapons. 14:14:09 03kilobyte * reb2b41b9591f 10/crawl-ref/source/godwrath.cc: Don't let Ely heal your friends when doing her revenge. 14:14:10 03kilobyte * r3d2760f8d149 10/crawl-ref/source/mon-place.cc: Remove an old hack for placing Abyss monsters. 14:14:20 03kilobyte * rc72d970d5629 10/crawl-ref/source/ (lev-pand.cc mon-pick.cc mon-pick.h mon-place.cc): Unify monster choice code for Abyss, Pan (now works) and regular levels. 14:14:20 03kilobyte * r5a2a22fca93e 10/crawl-ref/source/ (6 files in 2 dirs): Allow to (hackily) set a Pan level's monster set from vaults. 14:14:20 03kilobyte * r9af085b29e62 10/crawl-ref/source/dat/des/branches/pan.des: Set the monster set for a holy Pan level. 14:14:40 elliptic, Mu_: could one of you check if Zig levels with place:Pan still make sense? I fixed the bug that made that use place:Abyss instead, but that probably totally upsets their balance. 14:16:04 kilobyte: so can you please look at merge request 10 for me? It's got just two commits! 14:16:42 sure, just got the Pan stuff in a messy but working state 14:17:25 hmm, I wish there was a way to mark a merge request "ignore me" without deleting it... 14:17:36 so I could do that to #9 14:18:16 kilobyte: there are three different zig level sets that use place:Pan? I've looked at that lua and was pretty confused as to what it was doing... 14:18:52 dpeg: the basic code support needed for a Pan level is in, even if with quirky restrictions. I guess you or one of vault makers could make a better map than me, so you may prefer for someone more creative to make an attempt. 14:19:15 dpeg: otherwise, perhaps even just putting a set of vaults on a regular level could work 14:19:40 elliptic: because saying it is confusing would be an understatement 14:20:06 elliptic: but you guys know more about Zigs than me (especially on the player's side), that's why I'm asking 14:20:19 kilobyte: the Pan lord monster set changes, will they affect the new special Pan maps we got? (Beyond copying the line from your changes.) 14:20:57 !lm * place=Zig:27 noun=~left s=name 14:20:57 432 milestones for * (place=Zig:27 noun=~left): 80x MarvinPA, 54x hyperbolic, 31x 78291, 29x flashdoom, 24x clouded, 15x sh, 15x oxeimon, 13x mikee, 11x rath, 11x inspector071, 10x zigrobin, 7x casmith789, 7x splat, 7x Piu, 5x KiloByte, 5x Surr, 4x vihar, 4x meneril, 4x sugah, 4x Sticking, 3x Tossi, 3x Maniacuz, 3x Foonesh, 3x Luca, 3x soul, 3x ggeezz, 2x Shiren, 2x Gokome, 2x felirx, 2x airman, 2... 14:21:01 dpeg: the second statement refers to a Pan-style vault in dungeon (say)? 14:21:21 dpeg: they do, I copied the old sets intact 14:22:25 elliptic: 54 full zigs on your primary name vs 5 for me say quite a bit 14:22:43 kilobyte: put another way: the predefined monsters e.g. for Mnoleg are relevant for Mnoleg monsters spawned outside of the actual map (e.g. Haunt), rigt? 14:23:02 kilobyte: I can certainly take a look at a few zigs... I was just wondering about the commit message, since I thought that pan stuff only appeared in the levels with lots of pan lords and the abyss levels (and just a couple pan lords there) 14:23:03 yeah, exactly 14:23:36 dpeg: anyone can place monsters the old way, my changes make it possible to set this from lua (ie, the vault definition) rather than being hard-coded 14:23:38 So, Dr Who anyone? 14:23:51 no spoilers no spoilers no spoilers 14:23:56 MR #10 has my changes to the skill hotkey assignments for Zot Defence 14:24:12 MarvinPA_: For DW? I'm waiting for the 720 torrentz to pop up! 14:24:15 MarvinPA_: EX-TER-MI-NATE 14:24:15 ah, heh 14:24:31 i'm waiting for it to show up on iplayer 14:24:38 :) 14:24:38 -!- MarvinPA_ is now known as MarvinPA 14:24:43 -!- gustaf_arg is now known as ghallberg 14:25:24 lol 14:25:44 * SamB wishes mantis didn't insist on displaying all those leading zeros everywhere 14:26:05 (it makes it hard to tell which tab is which, for one!) 14:26:44 ah, I see there are two Pan Zig sets, not two 14:26:55 two not two? 14:26:55 one has one lord and the rest are place:Pan 14:27:03 not three I meant 14:27:07 ghallberg: two not three, context :) 14:27:23 Context, I don't have it. Sorry :( 14:27:43 Windows installer is slow... 14:27:50 the second one has: place:Pan w:40 / pandemonium lord w:(depth-5) 14:28:30 kilobyte: ah, I see... I assumed the first one was abyss because of the bug you fixed (the monsters being abyss set) 14:29:04 so I thought that it was a matter of some abyss levels having a pan lord, whereas really it was a matter of some pan levels only having one lord 14:29:06 we may use the old way to play safe 14:29:33 it just finished compiling for me, so I'll take a look now 14:30:20 what I meant is: due to the bug the first set had a pan lord and abyssal monsters, the other set had 1/3 pan lords (on Zig:27) and the rest abyssal 14:30:35 while the original intent was to have them come from pan 14:30:38 yeah 14:30:53 did I tell you guys I fixed that timing bug with wield failures, and mitigated #4083? 14:31:10 SamB: Mantis just told us so 14:31:52 yeah, I guess it would have 14:32:22 I'm not sure how to file https://gitorious.org/crawl/crawl/merge_requests/14 though... 14:32:49 It's miscellany 14:36:50 hm, if you create a ziggurat portal in wizmode and try to leave it on zig:27, it creates a new zig:27 for you 14:37:13 I hope this has to do with wizard portals and isn't a bug with regular ones... 14:38:06 elliptic: are portals supposed to be able to generate there normally? 14:38:17 no, I mean you create a ziggurat portal in D 14:38:30 and then enter it and descend until zig:27, and then try to leave the ziggurat 14:38:54 oh, that's kind of bad regardless 14:39:23 probably not a bug in wizmode, though, so much as an indication of how much chewing gum zig is held together by? 14:39:24 I'm currently testing whether this happens with other zig portals... 14:42:08 03SamB 07refs/merge-requests/14 * r8a10f741354d 10/crawl-ref/source/actor.cc: Fix the item_def* version of actor::can_wield(). 14:44:08 03dolorous * r5cac15ceb604 10/crawl-ref/source/xom.cc: Add Beastly Appendage to Xom's spell list. 14:45:40 Hey, dolorous is around. 14:46:05 cool! 14:46:31 kilobyte: it's not really taking that long to look at MR #10, is it? 14:46:32 well, a zig created with trowel was fine 14:47:50 so I'm guessing it is just a matter of some interaction between zig code and wizard portal code and the new code handling stuff like getting abyssed from zigs 14:48:08 so, chewing gum? 14:49:32 Hmm... 14:50:05 it looks like ccache doesn't understand some of the flags our Makefile uses to generate the .d files... 14:50:13 -!- Brannock has quit [Ping timeout: 240 seconds] 14:50:14 kilobyte: changing abyss monster set to pan monster set is definitely a buff... this would probably be a good excuse to tone down the number of pan lords on the place:Pan w:40 / pandemonium lord w:(depth-5) levels 14:50:47 -!- Fail_Bot has quit [Quit: Lost terminal] 14:50:57 SamB: in the last commit of #12 failing to switch weapons costs only 0.1 time 14:51:01 could change 40 to 50, say 14:51:25 kilobyte: hmm, well, I copied the delay from the "unwield" code path 14:51:34 I don't actually understand it 14:51:35 SamB: normal swap is 5 aut, you require 3/10 of that which rounds down to 1 14:52:05 not sure what we really want, just pointing it as a bit strange 14:52:58 isn't that the same time required for w- ? 14:53:10 unwield is 3 aut i think 14:53:14 oh 14:53:49 Anyway, I didn't fully understand the calculations of the time... 14:54:13 My main concern was doing this in the right situations 14:54:33 ok, I'll change it to 3/5 then 14:54:39 And I'm also really liking the "You are now empty-handed." 14:54:43 iirc current times for things are: unwield = 3 aut, wield swap = 5 aut, wield weapon from bare hands = 5 aut, take off jewellery = 5 aut, put on jewellery = 5 aut, jewellery swap = 10 aut 14:54:51 That really helps me a lot 14:55:01 I think the jewellery model makes a lot more sense here :P 14:55:18 elliptic: quite 14:55:18 -!- dtsund has joined ##crawl-dev 14:55:48 kilobyte: I don't really get why it was all done with *= and /= 14:55:59 or where the value was originally set 14:56:28 I was mainly happy to see that it was taking time when I started with something wielded, and not when I started empty-handed 14:57:40 Oh, Re: #10, 'F' and up are growing on me 14:58:32 I think it may actually not be *too* hard to get used to the zotdef powers being there 15:00:07 03SamB * r3c8304bd49f4 10/crawl-ref/source/wiz-item.cc: Clarify prompt for &% (create item by name) wizard mode command. 15:00:07 03SamB * r1a6790fd3a2e 10/crawl-ref/ (docs/develop/levels/syntax.txt source/mapdef.cc): Refer to the full lists of possible egos from the docs on ITEM specs. 15:00:08 03SamB * raa565bec901b 10/crawl-ref/source/wiz-item.cc: Add input history to &% (create item by name) wizard mode command. 15:00:18 03SamB * r8ebce8633042 10/crawl-ref/source/item_use.cc: Do not give free unwields for certain types of wield failure; 15:00:22 SamB: this usually goes fast, yes. 15:00:49 -!- Brannock has joined ##crawl-dev 15:03:59 hmm, that's not #10 ;-) 15:04:50 Folks, I am going to give SamB commit rights, so that can push #10, #11 and all the way to ##FF in peace. Okay? 15:05:05 Also, I expect gameplay modifications in no time, it is always like this :) 15:05:46 Wait, what? I don't know how to design gameplay modifications that would be advantageous to me... 15:06:33 All devs (well, those capable of actually coding) I know of started with interface tweaks or bug fixes, but went to gameplay changes/features real soon. 15:07:18 and I was hoping someone would look at #11 -- which, by the way, you could do, dpeg -- before I tried to actually get that merged 15:07:27 SamB: if you tell me where to look 15:07:41 and, yes, I was planning to do something about the altars in zotdef 15:07:54 https://gitorious.org/crawl/crawl/merge_requests/11 15:08:24 See, it's just help text 15:08:35 but it might be a tad long-winded... 15:09:47 i definitely approve of overwhelming us with merge requests as a method of getting commit rights :P 15:10:18 03dolorous * r332cf850cc5e 10/crawl-ref/source/mon-place.cc: Properly add cherubs to the HOLY_BEING_WARRIOR class. 15:10:20 MarvinPA: yes, I believe the documentation explicitly suggested it ;-) 15:10:23 :D 15:10:35 SamB: seems okay, my only comment would be that short sections in a help screen should generally come before the long section 15:10:59 uhm... that commit was bad 15:11:15 (dolorous' one) 15:11:23 kilobyte: just reply on c-r-d 15:12:09 yeah 15:12:09 * dpeg has to recall how to add developers. 15:12:35 First, put a sword onto their shoulder? 15:12:49 dpeg: I think you go to the repository's front page 15:13:01 then click on the ~something that shows up near the due 15:13:02 Is it still the SF abomination page? 15:13:12 SamB: I know, I know, as long as it is still SF :) 15:13:18 SF? 15:13:28 I meant on gitorious 15:13:30 SamB: you can add yourself if you already know the drill :P 15:13:35 ... right 15:14:11 I'll need some sort of a password, probably ;-P 15:14:51 * dpeg is fighting with the hydra that is SF right now. 15:15:39 hey, I didn't sign up to administrate any mailing lists! 15:15:41 -!- valrus has quit [Remote host closed the connection] 15:16:46 SamB: do you have an SF account? If so, please tell me the nick. If not, get one! 15:18:07 I have to admit I don't understand why the recourse to SF is necessary - I thought our trunk files are elsewhere now. Anyway, the grandfather did it with SF, the father did it with SF, so no reason to change it right now. 15:19:12 -!- Wensley has joined ##crawl-dev 15:19:32 SamB: lemme review your remaining patches, though 15:20:00 dpeg: sbronson 15:20:08 03elliptic * rae913d2de1c8 10/crawl-ref/source/dat/dlua/ziggurat.lua: Decrease the number of Pandemonium lords on ziggurat levels slightly. 15:21:00 dpeg: the repositories are all on gitorious, yes, as far as I can see 15:21:06 SamB: I assigned "fan mail" to you, ok? 15:21:24 the what? 15:22:02 We get billions of letters starting with "I want a child from you guys!", I cannot take it anymore. 15:22:11 SamB: also, you should be able to push now. 15:23:04 I'll keep that in mind for when I've got something I want to push 15:23:30 kilobyte: you may as well mark MR #9 as closed now 15:24:17 since I wasn't planning to do anything more with it, having broken out most of what was in there into other MRs 15:24:46 or, wait, I guess I could? 15:25:44 oh I guess, Mantis rights are also in order 15:26:07 so now I get to be a, er, "developer" instead of an "updater"? 15:27:34 yes! 15:28:18 SamB: you have just now sacrificed your little heart to the black cult of the Evil Crawl. We take it gladly, cheers! 15:28:44 dpeg: can you alter him on Mantis? 15:29:19 I might have the rights to do so as well but not sure 15:29:47 kilobyte: done 15:30:18 SamB has not only the right but also the responsibility to push his commits and to close his requests! :) 15:30:35 -!- Twilight_1 has quit [Read error: Connection reset by peer] 15:30:55 -!- Twilight_1 has joined ##crawl-dev 15:33:15 hurrah :) 15:34:23 ghrmdammit, can't find things to complain about worse than "an enum without a comma in the last item would produce too long conflict-inducing diffs" :p 15:34:30 Yes. SamB, most of the above was joking! (Not the Evil Crawl part, though. I'll send you form which you should sign with your blood and send back asap.) Welcome! 15:35:10 -!- valrus has joined ##crawl-dev 15:38:21 -!- SamB has quit [Ping timeout: 252 seconds] 15:38:48 He's a bit afraid now, I think. 15:39:08 -!- pointless_ has joined ##crawl-dev 15:42:08 03dolorous * r8e6742e84c1a 10/crawl-ref/source/mon-place.cc: Revert "Properly add cherubs to the HOLY_BEING_WARRIOR class." 15:42:27 -!- SamB has joined ##crawl-dev 15:43:58 SamB: welcome back : 15:44:42 kilobyte: Yeah, well, it was like that when I started. Plus, I couldn't remember if C++ allowed trailing commas or not ;-D 15:49:32 -!- valrus has quit [Remote host closed the connection] 15:52:19 hmm, is an amulet of rage of any use to a berserker? 15:53:08 IIRC it lowers the chance of falling unconscious afterwards 15:53:29 last I checked that was the case 15:53:38 okay 15:53:53 There is no problem with changing that and making it "useless" with Trog. 15:54:07 dpeg: but there's also no problem with doing nothing 15:54:08 03SamB * r113f1aaa7334 10/crawl-ref/source/abl-show.cc: Try assigning zotdef abilities to hotkeys 'F' and up (#4587). 15:54:08 03SamB * rb18f3ae0153c 10/crawl-ref/source/ (abl-show.cc abl-show.h): Give Zot Defence it's own section in the ability menu. 15:54:09 03SamB * rb5abf20537f7 10/crawl-ref/settings/advanced_optioneering.txt: Reformat advanced_optioneering into a valid configuration file. 15:54:09 03SamB * r706235789402 10/crawl-ref/source/mon-place.cc: Add a couple of probability annotations to mon-place.cc. 15:54:09 03SamB * reb0c26f40f1a 10/crawl-ref/source/describe.cc: Redraw quiver after inscribe command. 15:54:09 03SamB * ree6346c50f1e 10/crawl-ref/INSTALL.txt: Suggest ccache. 15:54:10 03SamB * r26d698f34d7f 10/crawl-ref/source/Makefile: Detect "debug" and "profile" targets precisely. 15:54:15 03SamB * r9d61502713aa 10/crawl-ref/source/actor.cc: Fix the item_def* version of actor::can_wield(). 15:55:09 SamB: yes 15:55:33 cool 15:55:52 not anymore (d39ed373) 15:56:40 the amulet might be used to train Evoc, but that's so minor it's not worth confusion 15:58:39 how does invocations work with the new system? have to have a god, or have to have a god that uses invo? 15:59:38 skill menu help stuff is in galehar's land, and doxygen in due's 16:00:09 wensley: I think it is supposed to be "have an invocation that uses invo skill" 16:00:25 ah, so you might have to gain some piety first 16:00:48 SamB: feel free to push that stuff if you feel comfortable with it 16:01:07 SamB: you did everything right in the latest crop of patches... 16:02:44 dpeg: so, you said something about not putting long bits before short bits? 16:03:03 Wensley: Didn't you read the help text? ;-P 16:03:25 wait, so why exactly I'm closing bugs fixed by your patches? This is responsibility I can shove onto you! 16:03:26 ##crawl-dev is my help text :P 16:03:28 SamB: yes. Imagine you're a first time reader of the help screen. You don't want to read a long paragraph only to discover a short, important paragraph comes later. Just switch them. 16:03:56 dpeg: I well, I just wanted to know which short paragraph(s) you thought needed moving 16:05:36 SamB: there is this long list about how to enable skill training. It should come at the end. 16:05:38 -!- PlasticMars has joined ##crawl-dev 16:05:55 dpeg: Yeah, I realized that was the long bit 16:08:29 !tell rax Hi, when I try to reload my save game in CAO it says that another game is already in progress using the save. The advanced options screen seems to be not working so I can't delete the save. Could you help? 16:08:30 PlasticMars: OK, I'll let rax know. 16:09:22 PlasticMars: this means the game is still running, at least from the operating system's point of view (not just DGL's) 16:09:45 were red devils renamed? because I'm trying to spawn one in wizmode, and instead of giving me "monster not found" it's just spawning random monsters, which seems odd 16:09:47 PlasticMars: when did you last played it? 16:09:58 Forgot to mention, it's been stuck like this for a couple of days now. 16:10:45 ...oh wait, I'm misled XP 16:12:06 PlasticMars: so it's likely your game is taking 100% CPU and slowing down everyone... Nap.kin watches CDO quite closely, and there were a couple bugs of this kind recently 16:12:44 kilobyte: is there some reason you didn't want to attract his attention? 16:14:12 -!- edlothiol has joined ##crawl-dev 16:15:25 SamB: not wasting several minutes or more of his time 16:15:28 PlasticMars: did you try doing nothing? for ten seconds? 16:15:31 edlothiol: You're French? 16:15:53 kilobyte: ah 16:15:59 dpeg: no, I'm just in France 16:16:08 I'm german 16:16:56 edlothiol: I recalled that, was just wondering. 16:17:35 I'm currently in france as an exchange student 16:17:42 edlothiol: is it good? 16:18:13 SamB: Do you mean nothing at the "There is already a game in progress" screen? 16:18:32 PlasticMars: yeah 16:18:38 dpeg: yeah :) my french was very rusty, but it's getting better 16:18:39 you did say CAO, right? 16:18:50 SamB: Trying that now. CAO, yeah. 16:19:20 edlothiol: you can practise with galehar :) 16:20:11 SamB: Nothing happened. 16:20:18 PlasticMars: oh dear 16:20:43 dpeg: hmm, am I just blind, or am I really not listed on https://gitorious.org/+dcss-developers ? 16:21:20 PlasticMars: is the wording that, or "Another game is already in progress using this save!" ? 16:21:32 dpeg: yeah, although writing french is not that much of a problem, as opposed to talking :D 16:21:35 I didn't change that. You're also not (yet) a dev in the CREDITs file. My concern was to make everyhing work, not to stroke egos =) 16:21:49 edlothiol: I agree. 16:21:51 dpeg: well, I'm pretty sure I need that to actually commit 16:21:57 er, and push 16:22:15 SamB: could you try? I never had business with gitorious.org 16:22:33 kilobyte: "Another game is already in progress using this save! Hit any key to exit..." 16:23:08 oh, good, kilobyte added me 16:23:28 SamB: it turns out I have the rights 16:23:35 that it does 16:24:17 PlasticMars: right, so this means there actually exists another game, perhaps wedged up, it's not just a bookkeeping error 16:24:46 99% chances it's the recent skill training bug 16:25:25 rax needs to unleash her kill -9 powers onto the old process 16:25:35 kilobyte: which bug caused *that*? 16:25:42 I seem to have missed that one 16:26:21 not sure exactly which, galehar and Napkin know 16:27:51 PlasticMars the Digger (L1 TrEE) (D:1) 16:28:11 Kamina the Eclecticist (L11 DEFE) (D:10) 16:29:00 3 stuck processes killed 16:29:06 all with -SEGV 16:29:18 Napkin: you rock 16:30:06 nah, i'm relaxing! would i be rocking i would hear the highlighting ;) 16:30:13 *i wouldn't hear 16:30:42 PlasticMars: try again now 16:31:09 It works! Thanks very much. 16:31:25 -!- pointless_ has quit [Quit: ChatZilla 0.9.87 [Firefox 3.6.7/20100713130626]] 16:34:56 -!- Xiberia has quit [Remote host closed the connection] 16:39:16 -!- Xiberia has joined ##crawl-dev 16:40:13 dpeg: so, exactly how far down should I move this big paragraph in the help text? 16:43:07 -!- st_ has quit [Ping timeout: 255 seconds] 16:51:39 * SamB wonders if there could be any advantage in using GCC's precompiled headers 16:53:00 -!- PlasticMars has quit [Ping timeout: 252 seconds] 16:53:27 SamB: just push it, I'll edit :) 16:53:48 dpeg: I already decided to break that off into a new section at the very end 16:53:52 There is this roguelike podcast business which somehow bypasses Crawl completely. I am not sure what to make of it. 16:53:57 SamB: fine! 16:54:08 and discovered that the heading at the very top is off-center compared to the text 16:58:42 Yred's Enslave Soul creates permanently invisible spectrals. (https://crawl.develz.org/mantis/view.php?id=4666) by ebarrett 17:22:08 03SamB * rc20bc5cf9e32 10/crawl-ref/source/dat/database/help.txt: Expand on the skill-menu help for the new system; update other pieces. 17:55:36 !seen galehar 17:55:37 I last saw galehar at Fri Sep 30 22:58:07 2011 UTC (23h 57m 30s ago) quitting with message Remote host closed the connection. 18:06:39 SamB: help looks good, will perhaps tweak a very little later, when I am on a linux machine. 18:07:03 dpeg: what, no SSH? 18:07:22 ;-P 18:08:08 03SamB * r30a7400d0cca 10/crawl-ref/source/ (6 files): Fix Doxygen warnings. 18:08:09 03SamB * r0c4fa0191e13 10/crawl-ref/source/ (actor.cc food.cc): Doxygen stuff. 18:08:09 03SamB * r003a1029de02 10/crawl-ref/source/Makefile: Don't include system headers in the output of "inline depgen". 18:08:42 SamB: my ssh on windows is so unstable, don't like to do any Crawl stuff from here 18:09:00 "inline depgen" contains a glaring typo! =) 18:09:01 dpeg: ah 18:09:07 does it? 18:09:18 inline dpeg 18:09:19 dpegen :) 18:09:49 so, YOU want to write all those little .d files now? 18:10:03 you don't want to let the compiler continue to be responsible for this task? 18:10:07 _You_ don't want me to do that, trust me. 18:10:20 well, then I don't think I spelled that wrong, did I? 18:10:39 * dpeg has been passing the bucks for more than sixty years. 18:18:30 -!- evilmike has joined ##crawl-dev 18:21:37 oh, I see someone already implemented PCH 18:23:58 -!- MarvinPA_ has joined ##crawl-dev 18:25:59 dpeg: what's all this podcast nonsense? also, just finished the quotes patch that I've been working on for two weeks, would you like to take a look at it? 18:27:25 -!- pointless_ has joined ##crawl-dev 18:27:44 -!- MarvinPA has quit [Ping timeout: 276 seconds] 18:30:09 03galehar * rc423a7aa4fb9 10/crawl-ref/source/ (evoke.cc skills.cc skills.h): Don't disable spellcasting when we stop reading a manual (#4655). 18:30:17 03galehar * rb5a91744b0e1 10/crawl-ref/source/itemname.cc: Revert "No scroll is useless if you don't know spellcasting." 18:30:17 03galehar * r4875a3edf438 10/crawl-ref/source/itemname.cc: Revert "Bad scrolls are useless to troglodytes since they don't want spellcasting (#4233)." 18:31:20 Wensley: sure, would you send it my way? 18:31:37 certainly, check your mailbox in five minutes 18:31:40 Wensley: http://roguelikeradio.blogspot.com/ 18:32:08 Just listened to session 1. 18:35:29 i'm interested to see where that goes once they run out of new roguelikes 18:36:02 evilmike: I think they can go for quite a while. 18:36:19 a few months, at least 18:36:34 yes 18:36:48 Perhaps they'll approach Pat Rankin for the final session :) 18:37:36 he's the only "active" nethack dev right? 18:37:47 the only one communication, yes 18:37:57 *communicating 18:39:20 dpeg: sent 18:39:41 evilmike: are you familiar with recent games? I am absolutely not. I listen when people talk about it, but I never play anything. 18:39:47 now I can finally do that sif thing! ...tomorrow :P 18:40:06 Wensley: hehe, just reading your warning. We'll take care of that by removing MD! 18:40:17 I'm familliar with some. I've played dungeons of dredmor, which the podcast is going to do next 18:40:17 I prefer hill orcs anyway :P 18:40:23 Wensley: didn't dare to ask you anymore. 18:40:32 in my opinion it's not a very good game, although it might be good for someone *completely* new to roguelikes 18:40:49 Wensley: what rights do I have? I don't care about the quotes too much, but can I object to stuff? 18:41:05 dpeg: object to anything you like and new material will be found 18:41:08 dungeons of dredmor is also a good example of how not to present information to the player 18:41:19 sooooo many stats in that game, resists are numerical, etc... 18:41:24 dpeg: what are these "recent games" of which you speak? 18:41:40 dpeg: and if you don't care about the quotes at all, well, then the patch is good to go in! :) 18:41:57 evilmike: Yes. I don't mind all the hullaballoo about the new games on the block, although I am interested (narcism speaking) in whether/how they're influenced by Crawl. In the worst case, exactly in the same way how Nethack influenced us (i.e. only as a warning). 18:42:02 Wensley: I think it would be a bad idea for him to object to anything he likes 18:42:11 he should probably only object to things he dislikes 18:42:21 ;-P 18:42:32 there was one roguelike in development (unreleased I think...) which had screenshots that makes it look like bizarro-crawl 18:42:37 I don't like things that I don't like except things I like. 18:42:40 same interface, probably uses some of the crawl source code 18:42:49 evilmike: was that kharne? 18:42:56 kharne sounds right 18:43:08 SamB: I play Crawl and Go. My scale of "recent" is not that of most people. But I guess the cool games are Desktop Dungeons, 100 Rogues, Cardinal Quest... shorter, lighter, much easier accessible. In other words: they feature no gods and thus are poor excuses for games :) 18:43:11 yup, it's kharne 18:43:26 evilmike: I think he just stole the interface, from what I remember kharne is offline-only and from reading the dev log the author talked mostly about angband 18:43:27 dpeg: heh 18:43:42 (100 rogues is very bad) 18:43:59 well, maybe not very bad. but very disappointing 18:44:32 evilmike: do you know how often I am confronted with players who ask for precisely this numerical information? I don't know about the other devs, but I am really stubborn about this. I believe that hiding numbers is often good, and unhiding them should be seriously thought about (sometimes necessary, of course). 18:45:23 dpeg: I think it's fine to show number as long as the player has some useful context in which to evaluate them... the problem is in supplying that context. 18:45:30 Wensley: I think 100 Rogues is really good! 18:45:50 bleh!! bought it, played it, not impressed :) 18:46:09 that's a lot of rogues 18:46:21 how did you find time to finish them all? 18:46:26 commercial roguelikes tend not to be as good as open source ones. I can think of only one or two exceptions 18:46:28 Design-wise, they avoided a fair number of classical traps and had some genuinely good ideas on their own. 18:46:35 evilmike: which ones? 18:47:02 shiren the wanderer, for the snes 18:47:05 Wensley: uh, 161 kb, that will take a bit 18:47:11 dpeg: yes :P 18:47:20 the other one i was thinking of is diablo, but that's more of a roguelike-like 18:47:56 evilmike: Erik (first non-coding DCSS dev) told me a lot about that one, including a long list of interesting monsters. That opened my eyes as to how cliche and boring our monsters are. (That was before new slimes, harpies, etc.) 18:48:23 Only case I can think of where unhiding numbers in 18:48:23 evilmike: btw, is Diablo 2 out? jpeg and I are preparing the next round of the survey and it'd be good to know. 18:48:27 whups 18:48:32 haha diablo 2 came out more than 10 years ago 18:48:39 evilmike: sorry, diablo 3 18:48:48 SamB: that "non-inline dependency genetation" is nearly completely broken 18:48:49 I don't think that one is out yet 18:48:52 evilmike: I don't play these games at all, forgive me. 18:48:52 I think diablo 3 is coming out next year 18:48:54 Only case I can think of off the top of my head where unhiding numbers in Crawl might be good is spell/ability success rates 18:49:07 kilobyte: hmm? 18:49:08 SamB: s/nearly // 18:49:12 only diablo 1 was really roguelike-ish though, d2 moved away from it and more into its own "action rpg" territory 18:49:14 dtsund: interesting, godo to know 18:49:18 it doesn't do anything except letting you compile 18:49:24 yes, percentages are numbers with context, thus I think it's fine to unhide them. dtsund has the right idea 18:49:28 kilobyte: yeah? 18:49:40 (You're welcome to steal my code for this, of course) 18:49:54 really? 18:50:05 it's a stub needed to get clang builds... but then any change forces a recompilation of everything 18:50:14 okay ... 18:50:15 There is the psychological argument that a scale of 0-100 is way too fine-grained. That is why we moved from old aptitudes to new ones, for example. 18:50:16 dpeg: with regard to 100 rogues, I felt that the character classes were pretty shallow, the physical interface tedious, the graphical interface hideous, and the hunger mechanic was incredibly misguided. I did think some of the enemies were interesting, though 18:50:19 kilobyte: how does that work? 18:50:21 myself, I'd like to be able to know how much MR I have in crawl, or at least have it not hidden behind the @ command 18:50:30 it's such an important resist 18:50:31 fixing them would be good, of course, but claiming that they do something isn't really an argument 18:50:33 it sure seems to run often for me 18:50:37 dpeg: did you hear, we recently moved skills from a scale of 27 to a scale of 270 :P 18:50:41 evilmike: That's not a number with context, though. 18:50:46 Wensley: I was against it :) 18:51:00 kilobyte: why do I see all these "DEPGEN" lines if it doesn't serve any purpose? 18:51:09 dtsund: I suppose so. But neither is EV really... try to find a player who really knows how EV works 18:51:10 I have no more idea, unspoiled, what "MR of 70" means than I do "quite resistant" 18:51:20 I think if MR could be normalized to be comparable to EV/AC/SH it could be friendlier 18:51:30 people know that 30 EV is really good, but they don't really know more than that 18:51:35 dpeg: meh, just use unskilled/basic/skilled/expert 18:51:40 dtsund: in fact, "quite resistant" will be more helpful 18:51:42 SamB: the makefile is buggier than hell :( The dependency is re-generated but not used. Don't ask me why... 18:51:48 Does Crawl let the player even know GDR exists? 18:51:52 It is quite important. 18:52:04 kilobyte: probably they get included too early 18:52:09 RichardHawk: it used to be incredibly unimportant, iiuc. And yes, it is in the manual. 18:52:11 SamB: however, the problem with not including system headers is that you'll get a miscompilation if you upgrade a header 18:52:30 Wensley: re 100rogues: I see, but my assessment is quite different. :) 18:52:31 dpeg: I agree, regarding the MR adjective. 18:52:33 kilobyte: hmm 18:52:35 which happens these days on Debian unstable like once a week due to the multiarch transition 18:52:40 Also I appear to have triggered an incredible bug. 18:52:44 dpeg: I'd love to hear it. you have my email address :P 18:52:44 Yred gifts armour now. 18:52:45 http://i.imgur.com/NRvDw.png 18:52:46 so what DOES normal depgen do? 18:53:15 I suppose part of my complaint about MR is that I don't find the adjectives helpful. "Quite resistant" might as well be "not resistant". Everything below "extremely" is rather poor 18:53:17 kilobyte: how does that go bad unless the new headers are ABI-breaking? 18:53:23 * SamB is confused 18:53:27 RichardHawk: Does xv on that Crown reveal an enemy name or the Crown of Dyrovepreva? 18:53:43 evilmike, same problem I personally have with spell success names 18:53:58 It could be the code assigning the wrong tile to something 18:53:59 "oh, it says my spell success rate is "good". I'll start casting it a lot!" 18:54:12 dtsund: The enemy name. When I move the cursor around the tiles change to various armour 18:54:57 I am fine with more clever use of adjectives. There must be a way that does usage proper justice. 18:57:18 * SamB wonders how much time make is wasting on its built-in implicit rules... 18:57:51 a start could be to just shift the MR adjectives "up" by one, so the old "quite" becomes the new "somewhat", etc 18:58:01 maybe remove one of the upper adjectives, and add in a new low one, "barely" 18:58:54 evilmike: in case I forget (quite probable), please remind me. I'd like to bounce this proposal around here. 18:59:01 but this might be a bad idea. The adjectives aren't exactly *wrong*... at "very" you'll resist most magic, it's just that since paralysis is in there, "very" is pretty bad 18:59:31 -!- Torokasi has joined ##crawl-dev 18:59:39 (problem with paralysis) 18:59:53 maybe just do numbers, but say out of what? 19:00:02 there's no upper limit, really 19:00:07 hmm 19:00:09 the AC/EV/SH range, for a start 19:00:21 then it could be displayed next to them, perhaps 19:00:41 not sure if that would be better than what we have, though 19:01:02 is MR really that applicable, in general? 19:01:10 if not, perhaps it could be 19:01:17 the numbers go a lot higher than AC/EV/SH 19:01:28 yes, would need some reworking 19:01:29 maybe display log MR ;-P 19:02:11 but at least ac/ev/sh share a little common context, i.e. "30 is probably enough" 19:02:42 not that that's all that intuitive, but it's better than just showing the pure MR number as it is right now 19:03:46 I think it'd be interesting to show numbers like that in some context relative to your level 19:03:54 (still might not be as good as showing the adjective, as we currently do) 19:04:09 just to give people a sense of "oh man all my defensive stats are low for my level, I need to train things up" 19:04:26 rather, a solution could be to show numbers like that in some context relative to the enemies in sight 19:04:44 like how enemies now have their names displayed in certain colors depending on their dangerousness 19:04:45 Forget about numbers... right now I think the best might actually be to just reword the adjectives. 19:04:47 I don't like that idea so much - I tend to do my equipment optimizing when I'm not in battle 19:04:58 instead of using "quite, very, extremely..." use "poor, good, very good"... 19:05:03 I don't want to have to find a pack of ogre mages in order to determine if my MR is good enough to deal with ogre mages 19:05:11 ("nope, it wasn't, and now you're in the abyss") 19:05:18 SamB: hmm... you might be right. There are convoluted cases with headers that hide versioned symbols behind your back, but the chance for that to break is really small, at least if no downgrade or revert among system headers happens. 19:05:25 ZorbaBeta: that's how you'd determine it normally, in any case :P 19:05:42 wensley, well, right now, I look at my MR numbers, then go check the wiki for information on whether that's good enough :V 19:06:04 better, if you know you're fighting ogre mages, just put on all your MR gear and wish for the best 19:06:12 kilobyte: anyway, a "make clean" will take care of you, since ccache is pickier than that 19:06:25 yeah 19:06:41 hmm, I wonder about the "downgrade" part 19:07:00 it'd result in a broken build 19:07:12 do you have an idea how costly full dependency information is? 19:07:17 that would probably break your executable, too 19:07:38 I mean, if that versioned stuff comes into things somehow 19:07:59 I thought the symbol versioning didn't come in until link time, though... 19:08:02 evilmike: I am always ears! 19:08:47 the word "versioned" doesn't seem to be in my GCC manual... 19:09:02 right now I'm trying to think if anything else in crawl uses the "quite...uncannily" scale instead of the "poor...excellent" one. I know stealth does 19:09:47 -!- Twinge has quit [Ping timeout: 258 seconds] 19:10:53 of course, I will freely admit my system is nearly 100% testing (though not all from anywhere near the same date...) 19:13:21 -!- pointless_ has quit [Quit: ChatZilla 0.9.87 [Firefox 3.6.7/20100713130626]] 19:13:33 -!- syllogism has quit [] 19:19:45 -!- Wensley has quit [Ping timeout: 252 seconds] 19:20:43 SamB: the trick is: library version A declare a function named foo() which is actually foo14() under the covers. Version B has foo() be actually foo15() which is not compatible, but foo14() is still available (just not in headers) so ABI is not broken. 19:21:20 kilobyte: you're sure this happens because of header changes? 19:21:28 glibc is especially notorious for doing this... and this way they can do a massive library with no user-visible ABI breaks for a long time 19:21:53 lemme check how exactly they do this 19:24:11 -!- ais523 has quit [Remote host closed the connection] 19:24:45 -!- Twinge has joined ##crawl-dev 19:29:35 here's an in-depth explanation: http://sourceware.org/glibc/wiki/Development/Versioning_A_Structure 19:31:22 -!- Wensley has joined ##crawl-dev 19:32:26 kilobyte: is there any particular reason why PCH is neither the default nor mentioned in the blurb at the top of Makefile? 19:34:53 no idea, sadly 19:36:10 -!- valrus has joined ##crawl-dev 19:36:55 I admit I never used precompiled headers. Do they work with mad C++ templating? 19:39:00 THE TEMPLATING, IT's INSANE 19:39:10 at a very undeducated glance at the documentation, we'd have to sort headers to not include actual code -- would this even work with C++ classes? 19:39:52 kilobyte: there's only one precompiled header allowed, anyway 19:40:00 and AppHdr is our designated one 19:40:04 and precompiling system headers would save 0.000000000000001% of compilation (only slightly exaggerating), it's the templates which slow everything down 19:40:41 yeah, it doesn't include any of the big offenders like player.h or env.h 19:40:48 hmm 19:43:30 where did you read that code is forbidden? 19:44:12 "info gcc", section 3.20 "Using Precompiled Headers" 19:44:49 BTW, sorry for doing anything that slowly tonight. Can't move too much. Explanation: http://angband.pl/tmp/knees.jpg 19:45:12 hmm, I must be blind 19:45:59 I might be misreading something, that's likely 19:46:11 kilobyte: :) 19:46:15 also, nights 19:46:17 -!- dpeg has quit [Quit: zzz] 19:46:37 yes, but I find it's always best to start with the assumption that I'm blind when I don't see something that someone else does 19:47:11 that way, I don't look so stupid when it turns out to be right there ;-) 19:47:43 if you have ANY clue about precompiled headers, ignore me :p 19:48:06 well, I didn't see any such statement 19:48:44 and it seems like it would be kind of a showstopper ... 20:13:08 by the way, theres a small abuse with the new exp system that lets people dodge the restrictions in some cases 20:13:25 if you master a skill and everything else is turned off the exp is stored 20:13:34 turn something on, kill something, and its dumped there 20:14:08 03dolorous * r0594cdf657f7 10/crawl-ref/source/transform.cc: Remove now-unused beak-related parts of Beastly Appendage. 20:14:14 maybe make people chose a skill to enable if they somehow manage not to have anything enabled 20:14:34 hm, i guess ill just put this on mantis where it belongs 20:15:26 alefury: last I checked that didn't work... did something change? I thought it turned on the first trainable skill you have 20:15:37 no it doesnt 20:15:44 at least not in the version i played 20:15:54 which is 3 days old or so, it has the restrictions 20:16:09 what does this have to do with restrictions? 20:16:28 i have no idea, im just telling you what version i saw the bug in 20:16:30 -!- Ryak has joined ##crawl-dev 20:16:32 oh, okay 20:16:49 also you asked if something changed, that would be one thing 20:17:11 also the restrictions are what makes this a problem 20:17:20 previously you could store some exp, but it would be pointless 20:17:26 now theres a way to take advantage 20:20:47 alefury: hm, I'm unable to reproduce this 20:21:02 it just happened to me when i killed asmodeus 20:21:22 spellcasting got mastered at some point in geh, i noticed after killing asmo 20:21:33 i turned on conj and ice, killed something, and both gained a level 20:21:33 and fighting didn't turn on? 20:21:36 nope 20:21:52 strange 20:22:10 the char is from before the restrictions got introduced, and fighting was turned off 20:22:18 oh, a transferred char? 20:22:20 there are some strange happenings with that update 20:22:25 so that might be the cause of it 20:22:26 lots of stuff is messed up with transferred chars 20:22:31 so this doesn't surprise me 20:22:35 should i report it anyway? 20:23:06 sure, good to have a record of it... but mention that it was transferred from before the skill restrictions 20:23:11 okay 20:24:41 -!- valrus has quit [Remote host closed the connection] 20:26:47 good night 20:27:03 -!- alefury has quit [Quit: Stay sane inside insanity!] 20:27:45 hmm, we probably should include in character notes when a version changes 20:28:07 this will really spam notes of debug chars, but meh 20:29:40 that's a good idea, yeah 20:29:49 mastering a skill allows storing exp (https://crawl.develz.org/mantis/view.php?id=4667) by Galefury 20:29:53 shouldn't be too spammy even 20:30:36 i think i've had that issue after mastering a skill too, yeah 20:31:07 no idea if it was on a character that was transferred through versions, it's pretty likely though i guess 20:32:42 what do you guys think of the updated numbers on that patch i submitted? 20:34:15 -!- MarvinPA_ is now known as MarvinPA 20:52:16 03kilobyte * r41b259ab8140 10/crawl-ref/source/rltiles/tool/ (tile.cc tile.h tile_list_processor.cc tile_list_processor.h): A new rltiles command %texture that can use an image as a mask for compositing. 20:52:20 03kilobyte * r24f30a6719b2 10/crawl-ref/source/tags.cc: Add a comment about the save browser's needs. 20:52:20 03kilobyte * rc4a508dadd4e 10/crawl-ref/source/main.cc: Spam character notes whenever the game version changes. 20:55:48 -!- Brannock has quit [Ping timeout: 245 seconds] 21:02:06 -!- Twilight_1 has quit [Read error: Connection reset by peer] 21:06:12 -!- Wensley has quit [Ping timeout: 240 seconds] 21:13:18 -!- edlothiol has quit [Ping timeout: 245 seconds] 21:34:46 -!- jeanjacques has quit [] 21:37:04 -!- Mu_ has quit [Quit: Defecator, may everything turn out okay so that you can leave this place.] 21:44:18 -!- pointless_ has joined ##crawl-dev 22:03:56 -!- Torokasi has quit [Ping timeout: 260 seconds] 22:05:56 -!- MarvinPA_ has joined ##crawl-dev 22:09:23 -!- MarvinPA has quit [Ping timeout: 276 seconds] 22:18:30 -!- pointless_ has quit [Quit: ChatZilla 0.9.87 [Firefox 3.6.7/20100713130626]] 22:35:47 -!- elliptic has quit [Quit: Leaving] 22:39:12 -!- mikee_ has joined ##crawl-dev 22:40:10 03SamB * r84f8b01d9ba4 10/crawl-ref/source/abl-show.cc: Allow a?? to show as much as it can about abilities with no descriptions. 22:44:21 -!- mikee_ has quit [Read error: Connection reset by peer] 22:45:32 -!- capablanca has joined ##crawl-dev 22:45:44 -!- capablanca is now known as mikee_ 22:46:41 -!- ixtli has joined ##crawl-dev 22:52:13 -!- Wensley has joined ##crawl-dev 22:55:39 -!- ixtli_ has joined ##crawl-dev 22:55:45 -!- ixtli has quit [Ping timeout: 255 seconds] 22:55:45 -!- ixtli_ is now known as ixtli 22:57:41 -!- UbAh has joined ##crawl-dev 22:59:04 -!- mikee_ has quit [Read error: Connection reset by peer] 22:59:30 -!- elliptic has joined ##crawl-dev 22:59:46 -!- mikee_ has joined ##crawl-dev 23:02:09 03SamB * red835a370aec 10/crawl-ref/source/ (abl-show.cc religion.cc): Doxygen: hyperlink together several closely-related arrays. 23:05:01 -!- mikee_ has quit [Read error: Connection reset by peer] 23:05:56 -!- elliptic has quit [Remote host closed the connection] 23:06:02 -!- capablanca has joined ##crawl-dev 23:06:23 -!- capablanca is now known as mikee_ 23:15:40 -!- elliptic has joined ##crawl-dev 23:18:00 -!- RichardHawk has quit [Quit: Recalculating TATARI at "the crossing crossing"] 23:36:34 -!- elliptic has quit [Quit: ChatZilla 0.9.87 [Firefox 3.6.17/20110420140830]] 23:39:55 -!- elliptic has joined ##crawl-dev 23:46:23 -!- ixtli_ has joined ##crawl-dev 23:46:41 -!- ixtli has quit [Ping timeout: 252 seconds] 23:46:41 -!- ixtli_ is now known as ixtli