00:00:28 -!- ZorbaBeta has joined ##crawl-dev 00:03:12 -!- monqy_ has joined ##crawl-dev 00:04:09 -!- stabwound has quit [*.net *.split] 00:04:09 -!- Guest39776 has quit [*.net *.split] 00:04:09 -!- elly has quit [*.net *.split] 00:04:09 -!- Fyren has quit [*.net *.split] 00:04:11 -!- bhaak has quit [*.net *.split] 00:04:11 -!- due has quit [*.net *.split] 00:04:11 -!- Keskitalo has quit [*.net *.split] 00:04:13 -!- Danei has quit [*.net *.split] 00:04:14 -!- kilobyte has quit [*.net *.split] 00:04:15 -!- rax has quit [*.net *.split] 00:04:15 -!- krel has quit [*.net *.split] 00:04:16 -!- rawrmage has quit [*.net *.split] 00:04:18 -!- epyon has quit [*.net *.split] 00:04:18 -!- petete has quit [*.net *.split] 00:04:18 -!- Napkin has quit [*.net *.split] 00:04:19 -!- Gretell has quit [*.net *.split] 00:04:20 -!- purge has quit [*.net *.split] 00:04:20 -!- ivan`` has quit [*.net *.split] 00:04:21 -!- Henzell has quit [*.net *.split] 00:04:22 -!- BlastHardcheese has quit [*.net *.split] 00:04:22 -!- Twinge has quit [*.net *.split] 00:04:22 -!- paxed has quit [*.net *.split] 00:04:22 -!- jlewis has quit [*.net *.split] 00:04:23 -!- Eronarn has quit [*.net *.split] 00:04:23 -!- lorimer has quit [*.net *.split] 00:04:23 -!- CIA-34 has quit [*.net *.split] 00:04:24 -!- Textmode has quit [*.net *.split] 00:04:24 -!- dtsund has quit [*.net *.split] 00:04:25 -!- FaMott has quit [*.net *.split] 00:04:25 -!- Vandal has quit [*.net *.split] 00:04:26 -!- blueDave has quit [*.net *.split] 00:04:26 -!- varmin has quit [*.net *.split] 00:04:27 -!- greensnark has quit [*.net *.split] 00:04:28 -!- monqy has quit [*.net *.split] 00:04:29 -!- ncampion has quit [*.net *.split] 00:04:30 -!- ZorbaTHut has quit [*.net *.split] 00:04:30 -!- Sequell has quit [*.net *.split] 00:04:30 -!- jle__ has quit [*.net *.split] 00:05:59 -!- paxed has joined ##crawl-dev 00:05:59 -!- elly has joined ##crawl-dev 00:05:59 -!- Guest39776 has joined ##crawl-dev 00:05:59 -!- Textmode has joined ##crawl-dev 00:05:59 -!- dtsund has joined ##crawl-dev 00:05:59 -!- jle__ has joined ##crawl-dev 00:05:59 -!- ivan`` has joined ##crawl-dev 00:05:59 -!- Fyren has joined ##crawl-dev 00:05:59 -!- ncampion has joined ##crawl-dev 00:05:59 -!- FaMott has joined ##crawl-dev 00:05:59 -!- Vandal has joined ##crawl-dev 00:05:59 -!- stabwound has joined ##crawl-dev 00:05:59 -!- rawrmage has joined ##crawl-dev 00:05:59 -!- bhaak has joined ##crawl-dev 00:05:59 -!- epyon has joined ##crawl-dev 00:05:59 -!- petete has joined ##crawl-dev 00:05:59 -!- due has joined ##crawl-dev 00:05:59 -!- Keskitalo has joined ##crawl-dev 00:05:59 -!- Napkin has joined ##crawl-dev 00:05:59 -!- Twinge has joined ##crawl-dev 00:05:59 -!- blueDave has joined ##crawl-dev 00:05:59 -!- Sequell has joined ##crawl-dev 00:05:59 -!- jlewis has joined ##crawl-dev 00:05:59 -!- Danei has joined ##crawl-dev 00:05:59 -!- varmin has joined ##crawl-dev 00:05:59 -!- rax has joined ##crawl-dev 00:05:59 -!- krel has joined ##crawl-dev 00:05:59 -!- kilobyte has joined ##crawl-dev 00:05:59 -!- Gretell has joined ##crawl-dev 00:05:59 -!- CIA-34 has joined ##crawl-dev 00:05:59 -!- greensnark has joined ##crawl-dev 00:05:59 -!- lorimer has joined ##crawl-dev 00:05:59 -!- Eronarn has joined ##crawl-dev 00:06:27 -!- monqy_ is now known as monqy 00:10:47 -!- Henzell has joined ##crawl-dev 00:19:36 Windows builds of master branch on crawl.develz.org updated to: 0.10-a0-699-g13bea13 00:32:31 -!- paxed has quit [Ping timeout: 252 seconds] 00:34:03 -!- paxed has joined ##crawl-dev 00:38:12 -!- MarvinPA_ has joined ##crawl-dev 00:40:36 -!- MarvinPA has quit [Ping timeout: 240 seconds] 01:06:29 !tell galehar I'll add a proper tutorial section to planning once I have a clearer idea myself. 01:06:30 Keskitalo: OK, I'll let galehar know. 01:08:05 -!- valrus has quit [Remote host closed the connection] 01:53:44 -!- evilmike has quit [Quit: something happened] 01:56:18 -!- st_ has quit [Ping timeout: 260 seconds] 01:57:39 -!- dtsund has quit [Quit: dtsund] 01:58:22 -!- Textmode has quit [Quit: Ex-Chat] 02:04:43 -!- Twilight has joined ##crawl-dev 02:05:09 -!- Twilight is now known as Guest61671 02:05:35 -!- Guest39776 has quit [Read error: Connection reset by peer] 02:13:18 -!- galehar has joined ##crawl-dev 02:13:56 hi 02:13:56 galehar: You have 2 messages. Use !messages to read them. 02:14:01 !messages 02:14:01 (1/2) elliptic said (6h 33m 6s ago): I don't know whether this is an AI bug or something else, but this kobold suddenly stopped moving and won't even attack me when I move next to it: http://crawl.develz.org/saves/dumps/hyperbolic-b60e431-110914-0039.tar.bz2 02:14:12 !messages 02:14:12 (1/1) Keskitalo said (1h 7m 42s ago): I'll add a proper tutorial section to planning once I have a clearer idea myself. 02:18:26 -!- Pacra has quit [Read error: Connection reset by peer] 02:22:03 -!- djinni has quit [Ping timeout: 260 seconds] 02:27:32 Unstable branch on crawl.akrasiac.org updated to: 0.10-a0-699-g13bea13 (32) 02:30:24 -!- djinni has joined ##crawl-dev 02:36:20 -!- galehar has quit [Ping timeout: 252 seconds] 02:40:41 -!- SamB_ has joined ##crawl-dev 02:43:39 -!- SamB has quit [Ping timeout: 252 seconds] 02:50:22 -!- galehar has joined ##crawl-dev 02:56:18 -!- Zaba has quit [Ping timeout: 252 seconds] 03:07:26 -!- Zaba has joined ##crawl-dev 03:08:14 -!- galehar has quit [Ping timeout: 252 seconds] 03:08:26 -!- valrus has joined ##crawl-dev 03:13:00 -!- valrus has quit [Ping timeout: 260 seconds] 03:23:55 -!- galehar has joined ##crawl-dev 03:24:06 how about branding hostile summons glyph? 03:25:01 * due for it. 03:25:12 It's really dififcult to note straight away that a monster is summoned or not. 03:27:11 diacritics/combining characters instead would be a nice touch :D 03:30:15 except that COLFLAG seems to be full (defines.h:273), so maybe it's not as simple as I thought 03:30:20 don't know much about curses 03:31:44 -!- BlastHardcheese has joined ##crawl-dev 03:33:30 curses! 03:33:46 ncurses 03:36:25 ironically that's what we were talking about. 03:36:28 uncurses? 03:37:22 that's none of MY projects 03:40:23 I got a level with >10 shafts... Is this a bug? 03:43:53 of course not, crawl just wants you dead as usual 03:44:16 not shafts, trapdoors 03:44:19 sorry 04:18:50 diacritics would be much more attractive than branding - for both allies and enemies 04:19:51 I read that as "dianetics" 04:19:58 fr: scientologist unique 04:26:04 ghallberg: I think some vaults make the builder believe that parts of the level are disconnected when they aren't. 04:26:09 ok 04:26:33 Eronarn: agree 04:32:48 -!- elliptic has joined ##crawl-dev 04:37:38 hmm. 04:41:52 Eronarn: about menv vs not_on_menv you talked about at 2am: I guess it probably would be best to get rid of the whole mindex stuff altogether 04:44:28 -!- Pingas has joined ##crawl-dev 04:45:02 Eronarn: another thing: mid-game Yred is thoroughly, utterly overpowered 04:45:21 mindex dying in a fire would be ver, very nice. 04:45:49 even though you get servitors very rarely, they don't die so you end up stashing extras away. And totally ignore popcorn gifts like bone dragons. 04:46:27 Yred fails to scale late, though... a solution seems kind of obvious 04:47:09 due: too bad it's widely used. All the mon->foe stuff or death_source... 04:47:46 what's mindex used for? 04:47:49 oh. 04:47:53 nevermind :P 04:50:02 okay, then. What would be an alternative to mindex? 04:50:17 hashtable using serials? :D 04:50:36 and wasn't it added rather recently? Or am I mixing it up with something else. 04:50:46 mid, mostly. It survives being moved off level, is never reused, and unlike pointers, it is safe against going stale. 04:50:59 ah, right, mid is the new thing. 04:51:23 due: we don't have a hashtable implementation, just std::map which has logarythmic time 04:52:33 sigh 04:54:24 gitorious is giving me fatal errors with zero verbosity... 04:55:05 hey, when I was young and pretty, implementing a decent resizable hashtable from scratch was a matter of 10-20 minutes coding. Now I'm senile, but that's doable. 04:55:25 and there might be ready to use hash table templates around 04:55:58 (I tend to hate the colour they painted the bike shed, of course, and this wheel is easy to reinvent :p) 04:57:12 i avoid implementing things myself because i know i don't knkow enough to do it right. 04:59:46 heh, I remember from when I was at the Univ, during a data structures class the prof shown us an implementation of bisection, and asked: "is this one correct?". 05:00:05 OSX builds of master branch on crawl.develz.org updated to: 0.10-a0-699-g13bea13 05:00:06 Napkin: how about a nightly job to clean up all the empty directories which litter the devwiki? 05:00:19 Folks answered "yes", then the prof told us to prove there doesn't exist a set of data on which that implementation works. 05:00:30 Three different bugs in ~10 lines of code :p 05:01:23 galehar: what if doku creates them just to use them a few minutes later? 05:01:34 I'm a bit afraid about races 05:02:11 if it's going to use the namespace (a page is being created), there should be a lockfile in it 05:02:23 -!- Guest61671 has quit [Ping timeout: 252 seconds] 05:02:44 http://www.freelists.org/post/dokuwiki/Deleting-a-namespace,1 05:03:07 Eronarn: by the way, about naga constriction: today's Goblins are damn relevant 05:03:23 http://www.goblinscomic.com/09132011/ 05:05:10 std::unordered_map is a hash table, I think 05:06:47 03zaba * r81416e10b7b1 10/crawl-ref/source/dat/des/builder/layout.des: Sometimes generate doors in layout_roguey's rooms. 05:06:48 Zaba: oh, interesting. Not in the standard, though... but we can check if MSVC has it. 05:06:53 kilobyte, it's TR1 05:07:04 kilobyte, here's how people include it: http://gitorious.org/inspircd/inspircd/blobs/insp20/include/hash_map.h 05:07:15 -!- ussdefiant has quit [Ping timeout: 252 seconds] 05:07:19 kilobyte, that includes msvc support 05:08:57 -!- valrus has joined ##crawl-dev 05:09:55 -!- ussdefiant has joined ##crawl-dev 05:09:56 -!- Twilight has joined ##crawl-dev 05:10:11 -!- greatzebu has quit [Quit: greatzebu] 05:10:26 -!- Twilight is now known as Guest93088 05:15:15 -!- valrus has quit [Ping timeout: 260 seconds] 05:17:00 -!- ussdefiant has quit [Ping timeout: 258 seconds] 05:17:01 kilobyte, so basically there's (the older, non-standard) hash_map, and the TR1 unordered_map, and depending on the compiler and its version you can use either one or another. 05:17:04 -!- ussdefiant has joined ##crawl-dev 05:17:08 I don't have a gcc 4.0 installation, grabbing it as we speak 05:17:57 it could replace 95% of our uses of std::map 05:18:34 -!- Guest93088 has quit [Ping timeout: 265 seconds] 05:29:15 it appears this inspircd's include file is obsolete: it uses tricks for gcc <4.0 and for some old MSVC 05:30:47 you never get whines about "dcss doesn't compile with $compiler $version_older_than_your_mom"? 05:30:59 -!- Twilight13 has joined ##crawl-dev 05:31:39 i wanted to compile dcss on solaris 5.6! 05:32:21 wow, i didn't even think about using an OS older than your mom! 05:32:38 :> 05:32:46 greensnark enjoyed my complaining ;) 05:34:59 bhaak: since the fork, it officially supported only gcc >= 4.0. I'm not going to put the effort into porting into earlier ones :p 05:35:17 unless it would be required for some platform 05:35:28 Napkin: probably he was having as much fun as one can have watching something go BOOM 05:36:12 since the fork? how old was gcc 4.x at that time? 05:36:38 i think he's talking about the "fork to sanity"! 05:37:12 okay, i know, it's a bit different with c++ than with c compilers. but i still find it unreasonably to complain when gcc 2.95 isn't able to compile a modern C program :) 05:37:18 gcc 4.0 is dated April 20, 2005 05:37:51 Napkin: only 3 hits for "fork to sanity" :) 05:38:17 bhaak: with C, I'd stick with basic ANSI, there's little new stuff, it's a stable language 05:38:51 goddamn rmadison claimed etch has gcc-4.0 but it lied :( 05:39:23 kilobyte: no enums make bhaak sad ;-( 05:39:30 WTF, sarge doesn't have 4.0 either 05:40:21 bhaak: ok, I used to always stick with #defines, but after such exposure to C++ I would probably unconsciously go with enums... 05:40:42 kilobyte, well, if we don't have to care about gcc <4.0 and old msvc, then it's indeed easier 05:41:44 gcc-4.1 has it in not though 05:42:02 kilobyte: hmm, wikipedia claims C89 has enums, so maybe it was something else ... maybe dynamic declarations or declarations not only at the start of a block? 05:42:08 kilobyte, does work if you compile with -std=c++0x? :> 05:42:14 at least not like gcc-3.* or MSVC 8, though 05:42:17 kilobyte: we talked about yred stuff earlier 05:42:25 bhaak, pre-C99 doesn' support declarations not at the start of a block 05:42:33 servitors definitely don't help the problem - they make it a lot worse, actually - but it existed before them 05:42:48 K&R ftl! 05:43:11 i'm also a sucker for // comments, but can contain myself nowadays when doing C 05:43:32 the suggestion i made is that undead stuff on the level 'wakes up' when you get there, with deeper levels having more undead, and they don't follow you around in a mob 05:43:38 across levels 05:43:59 I can't, too much LPC coded using ed (yeah!), using /* */ comments was a bitch with a line-based editor 05:45:10 as for the menv/mid/mindex stuff, it would be nice if that were all more sensible 05:45:32 kilobyte, out of curiousity, what did you code in LPC for? 05:45:44 Eronarn: having them stick to a level is way too zombiey for me, but I'd make instead: 1. a limit on # of pets (flying skulls count as 1/3), 2. upgrades so you don't need to kill them yourself 05:45:55 Zaba: The Two Towers 05:47:20 kilobyte: ah, but many yred allies are zombie-y :) people do like having some stuff follow them, like dragons/servitors 05:47:54 but right now it's probably overstated... do people really care if they can bring a specific W/p with them? 05:49:01 I would remove that, and make Recall work cross levels 05:49:05 i think having a self-replenishing status quo of the weak things (but still growing stronger as you go down) + some small (less than 5) number of things you care about lets you preserve keeping your bone dragon buddies without having 30 flying skulls on screen 05:50:13 it also means more time for medium power undead to shine, if you can get something quite good early, but still well before you get a dragon or a servitor 05:52:07 kilobyte: actually, what if recall only worked on your strongest servant, one at a time? 05:52:10 that might be interesting 05:52:42 as in, with repeated use? 05:54:24 kilobyte: yred gifts staying on the level you got them, plus recall-strongest-thing (whether offlevel or not), and heavily weighting gifting chaff monsters especially right as you enter a new level 05:55:44 perhaps add a small piety cost to the recall, even 05:55:53 remember there's both the enslaved soul and servitors 05:56:10 it would suck to have to choose one or the other 05:57:05 kilobyte: you could use it more than once, but you couldn't bring everything in one go (so no ->V8 with 10 servitors) 05:57:37 though i think enslave soul is a bad ability as-is :) 06:01:04 'quicksilver' breath dispels slowness from curare (https://crawl.develz.org/mantis/view.php?id=4539) by mikee 06:01:09 it should be a pokeball effect, you call/dismiss the soul as you need it 06:01:11 there's another problem: you can stash pets, and that's tedious 06:01:11 -!- ghallberg is now known as pops 06:01:11 I'd make it so you can't choose to stash them: intelligent pets would arrive onto your level after some time no matter where you left them 06:01:11 hmm 06:01:11 what if we made the 'recall one' ability passive? 06:01:11 Agreeing with kilobyte, that would also reduce tedium for those who don't scum stashes 06:01:11 I have a "totem monster" ---> totemon idea sketched out somewhere, complete with artifact totems, random base types of monsters and a load different abilities that each base type can gain. 06:01:11 so stuff just shows up next to you in a flash of unholy darkness 06:01:11 and lose recall entirely 06:01:34 it would work like: you get chaff all the time but it stays on the level; powerful undead are your escorts, and are bound to you, getting leashed back to you if you get too far from them; letting escorts die is worse than letting chaff die because you won't get them nearly as often 06:02:01 this would, of course, fix bone dragons! 06:02:18 recall is nice early on when you have to rely on zombies 06:02:53 kilobyte: we could have escorts start quite early on, and just have a suitable scale for both chaff and escorts 06:03:15 a single flying skull would be quite powerful very early game, for instance 06:03:20 @??flying skull 06:03:21 flying skull (16z) | Speed: 15 | HD: 2 | Health: 6-16 | AC/EV: 10/17 | Damage: 7 | Flags: 07undead, evil, lev | Res: 06magic(8), 05fire, 02cold, 10elec, 03poison, 12drown, 04rot+++, 13neg+++, 13torm | Vul: 08holy++ | XP: 15. 06:04:08 would be funny to see a DK tearing up D1 with one 06:09:21 How would you scale it, exactly 06:09:23 A wraith refuses to follow you past D:X? 06:09:23 RichardHawk: the idea i'm proposing, wraiths would simply not follow you; you'd get wraiths on D:whatever, and eidolons on D:whatever+10, or... whatever 06:09:23 and if you have an escort around for a long time, it would either die, level up, or decide it's too weak now and become just chaff 06:09:23 (and thus free up you to get a new escort) 06:09:24 Yredelemnul's escort agency? 06:10:13 Chaff based on depth, escorts based on strength of your character? 06:10:39 as in, XL? 06:10:51 this could actually work, unlike piety or depth 06:10:55 RichardHawk: i would base them both on the same thing, and i think actually a measure taking into account both depth and character strength would be good - so diving doesn't let you get servitors 06:10:55 XL would be an easy method, sure 06:11:10 piety is bad as you get ****** mid-lair, or with Faith even before 06:11:15 but also a high XL character gets more reasonable stuff if they're on d10 for some reason 06:11:43 kilobyte: also of note, yred doesn't currently use piety directly in gifting - it's just based on number of gifts (so total piety collected matters, because of gift timeout, but not current piety) 06:12:01 depth would encourage dipping to get a servitor then returning to safety and not doing things above your head 06:12:24 number of gifts is bad too, yeah 06:12:26 kilobyte: sure, but that's why i think a weighted value would be better... say, .8 XL, .2 depth 06:12:34 it does make late conversions worthless too 06:12:35 so you actually have an incentive to dive a bit, but not on a L1 spriggan 06:12:48 (even more than Yred is a strictly 3-rune god) 06:13:12 XL has the issue of the current XP aptitudes making precious little sense, but that's another issue 06:13:31 having escorts be both powerful and varied would help a bunch 06:13:48 I did Yred->TSO. 06:14:12 classed, intelligent undead: vamps, ghouls, maybe even a lich 06:14:16 due: <3, a DK of TSO gives you extra style points! 06:15:08 (obviously they'd need tweaked spell lists to not be super broken, but i think this lets you add new interesting stuff - 'ghoul vivisector', with pain brand claws and pain magic. 'wraith entropist', with bolt of draining) 06:15:25 kilobyte: :D 06:15:32 !hs char=HOdk 06:15:33 5. bookofjude the Orcish Porcupine (L27 HODK), worshipper of The Shining One, blasted by an Ice Fiend (bolt of cold) in the Abyss on 2009-08-17, with 506545 points after 120210 turns and 9:12:06. 06:15:37 I died, though. 06:16:19 i guess if you could have, say, 3 escorts and got an average of .5 escorts per dungeon level, that'd work well for late conversion 06:16:20 !hs char=SpDK 06:16:20 37. KiloByte the Shatterer (L27 SpDK), worshipper of The Shining One, blasted by a Fiend (blast of hellfire) in Pandemonium on 2010-01-11, with 671827 points after 199328 turns and 17:15:28. 06:16:29 regardless of piety you need to go through a certain number of levels before you are up to 'full strength' 06:18:02 you can't get piety immediately anyway 06:24:19 staff of Dispater still need HP/MP and turn.. (https://crawl.develz.org/mantis/view.php?id=4540) by cosmonaut 06:26:05 screw it, we don't compile on MSVC out of the box anyway, I'll use the MSVC 10 path and leave just a comment for MSVC 8 06:46:09 -!- syllogism has joined ##crawl-dev 06:50:36 -!- elliptic has quit [Ping timeout: 240 seconds] 07:09:19 -!- valrus has joined ##crawl-dev 07:13:55 -!- valrus has quit [Ping timeout: 260 seconds] 07:21:38 -!- elliptic has joined ##crawl-dev 07:31:43 -!- Pingas has quit [Ping timeout: 260 seconds] 07:34:04 -!- purge_ has quit [Quit: .] 07:38:26 -!- dpeg has joined ##crawl-dev 07:38:35 !seen greensnark 07:38:36 I last saw greensnark at Fri Aug 19 12:10:32 2011 UTC (3w 5d 28m 4s ago) joining the channel. 07:38:44 !seen Napkin 07:38:45 I last saw Napkin at Wed Sep 14 10:36:38 2011 UTC (2h 2m 6s ago) saying i think hes talking about the "fork to sanity"! on ##crawl-dev. 07:40:01 wazzup? :D 07:40:32 Hallo Marc! 07:41:03 Es gab eine Frage nach der dokuwiki-Erweiterung. Weisst du, was damit los ist? Ob es geht oder gehen koennte? 07:41:27 gab es das? 07:41:33 mir hat keiner was gesagt ;) 07:41:35 -!- Pingas has joined ##crawl-dev 07:41:48 ich hab's ge!tellt :) 07:42:13 !messages 07:42:14 (1/3) dpeg said (4d 15h 27m 42s ago): Can we get the http://www.dokuwiki.org/plugin:text for the wiki? 07:42:20 ha! ?P 07:42:55 Napkin: egal, hier ist der Hintergrund: Mumra hat die Vault design-Dokumentation wikifiziert. Das hat viele Vorteile (for example, it is much more up to date now). 07:43:46 On the other hand ist kilobyte dagegen, weil txt-file lesen viel schneller geht. Best solution: use the wiki, aber txt files fuer releases produzieren. 07:43:53 Hier kommt dokuwiki ins Spiel. 07:44:39 yikes, Denglisch 07:44:55 kilobyte +1, btw 07:45:19 Napkin: you think documentation should not be online? 07:46:06 no, but i think we have a proper versioning management software already -> git ;) 07:46:18 hear me out, please 07:46:34 no, please, not again 07:46:52 we've had that discussion when the manual was moved to wiki 07:46:57 hear me out, please! 07:47:06 this is way more important: http://zeichnemit.de/ 07:47:47 My point of view is moderate: for certain tasks I want a very quick access to documentation, e.g. looking up syntax for KITEM etc. This is why (at least some) documentation should be in trunk. On the other hand is reading the elaborate texts on triggerables etc. much less painful on the wiki - compare yourself (coloured vault examples, jumping between sections). These are articles you don't really read as a quick reference. So there are good sides to both op 07:47:57 Napkin: serious talk or not? 07:47:58 Napkin: it already broke 50k 07:49:38 dpeg: following your argument, i would still keep the master copy in git and have the text automatically formated to html and shown via a webserver 07:50:11 Napkin: people spent a lot of effort to update the lua sections. This would not have happened with master in trunk. 07:50:33 I am getting in a seriously bad mood. 07:50:34 Bye 07:50:36 -!- dpeg has quit [Quit: leaving] 07:51:10 anyways, i'm ok with whatever you decide 07:51:21 oopsa - did i sound so harsh? 07:52:21 not _that_ harsh, but it sounded not as you would like to follow whatever dpeg would decide 07:52:31 not my intention.. 07:53:05 so you wanted to say "not that discussion, just decide already"? 07:53:06 -!- edlothiol has joined ##crawl-dev 07:53:15 i just remember all the reasons still and there was no need to tell me again 07:53:34 ah. no, that didn't come over like that 07:54:08 geez.. i'm at work, doing 5 things at once 07:55:04 so you shouldn't be here :) don't multitask! we humans can't do it! only computers are really good at it! 07:55:12 hehe 07:55:21 quiet, need to search for his email address now :-P 07:56:23 * bhaak opens dubstep.fm and turns up the speakers 08:00:22 Napkin: what do you think about a nightly job to clean up empty dirs in the devwiki? 08:00:30 seems like the only way to remove them 08:04:50 -!- ainsophyao has joined ##crawl-dev 08:06:09 -!- edlothiol has quit [Quit: Page closed] 08:06:42 03kilobyte * reb61e8435264 10/crawl-ref/source/spl-damage.cc: A whitespace fix. 08:06:44 03kilobyte * r6c9450d37065 10/crawl-ref/source/ (AppHdr.h store.h): Use std::unordered_map instead of std::map -- O(1) vs O(log n), in one place for now. 08:11:28 galehar: explain the problem with empty dirs 08:11:53 also, please give me an example of an empty dir 08:12:07 https://crawl.develz.org/wiki/doku.php?id=dcss:start 08:12:15 try to remove the broken links from this page 08:12:57 My understanding is that they are caused by a namespace without any page in it 08:13:07 yes 08:13:09 and it cannot be removed from the web interface 08:13:31 need to manually removed the directory from the shell 08:14:53 http://www.freelists.org/post/dokuwiki/Deleting-a-namespace,1 08:22:04 which means, that "empty" != "empty" 08:22:18 which may be difficult to match 08:22:34 -!- Twilight13 has quit [Read error: Connection reset by peer] 08:23:22 -!- Twilight13 has joined ##crawl-dev 08:30:38 -!- elliptic has quit [Ping timeout: 260 seconds] 08:34:45 -!- elliptic has joined ##crawl-dev 08:35:19 galehar? 08:35:29 did you move the content from feedback elsewhere? 08:35:35 I think it means the directory is non-empty when the user is creating a page. But the lockfile should be deleted later and the directory will be really empty. 08:35:54 the content from feedback has been merged into brainstorming a while ago already 08:35:55 we have lock files dating back to February 2010 08:36:08 you can remove it 08:36:18 no, i can't 08:36:48 why? 08:37:02 $> cat locks/b18511bb20532f0c85f96e283749f7d1.lock 08:37:03 Eronarn 08:37:11 because that's the way a lock looks 08:37:56 and this "locks" directoy is a main directory beside "pages", "attic", "meta", "cache" 08:38:28 maybe just delete all locks older than 24h 08:38:32 anyways, deleting the locks now shouldn't help anything 08:38:59 we still need to remove the empty dirs manually, right? 08:39:41 i'm quite unhappy, that the former structure of feedback is completely gone from "pages", but still completely exists in "attic" and "meta" 08:41:05 maybe we can manually clean it up (or would that break everything?) 08:41:42 on another note, I made a new wiki page about tornado's balance (proposals and data only, no discussions) 08:41:43 https://crawl.develz.org/wiki/doku.php?id=dcss:brainstorm:magic:tornado_balance 08:43:22 stupid move plugin is not moving history 08:45:26 galehar: what do "Tornado range 3" and "Tornado range 5" mean? 08:46:28 it means the monster was 3 cells or 5 cells away 08:46:39 and nobody moves during the duration of the spell 08:47:26 galehar: thanks. and what is current duration? 08:47:46 oh, sorry, on the page 08:49:21 duration 6 is what you get at power 120. So fixing it there is a boost for power < 120 and a nerf for power > 120 08:50:07 can you put the current radius formula on the page? 08:53:20 yes 08:53:41 I wonder what a Crawl board game would be like :o 08:54:18 done 08:54:18 galehar: thanks :) 08:55:10 Maybe card based and the cards would be the dungeon layout so it is always randomized 08:55:46 there are games like that already 08:55:52 yup 08:55:59 Munchkin Quest for one 08:56:27 the more I think about it though, the less well Crawl would work as anything but an electronic game 08:57:03 part of what makes this great is how huge it is, with the large variety of characters and monsters 08:57:25 The closest thing that exists to this scope is Dungeons and Dragons while using 20 supplement books 08:58:19 oh well, off to class! 09:01:33 galehar: manually cleaned so far 09:02:08 Napkin: thanks! 09:14:35 crap... the biggest problem with new Yred is that big uniques die in a couple of rounds, not giving you a chance to enslave them no matter how you try :p 09:34:34 -!- Fail_Bot has joined ##crawl-dev 10:08:33 -!- greatzebu has joined ##crawl-dev 10:21:59 -!- Galefury has joined ##crawl-dev 10:22:28 hi 10:23:25 galehar: there's something totally wrong with your damage numbers. I repeated my old test run (sadly, the char has no Shatter memorized so I couldn't compare that exactly), and get about the same results. 10:23:55 -!- Mu_ has joined ##crawl-dev 10:24:02 ie, Shatter has more damage than Tornado over its entire duration, on a set of monsters that consists mostly of fliers 10:24:22 I just wonder what could be wrong 10:24:53 shatter's damage has a lot of variability against fliers since it's 1d(x) 10:24:54 there is some furniture but not a lot 10:25:03 the inevitable conclusion: shatter must be way too strong -.- 10:25:07 I averaged many tries of course 10:25:16 that's why I just calculated average damage by hand. 10:25:24 Galefury: no, back then I looked mostly at Firestorm 10:25:53 actually shatter probably is too strong 10:25:58 I use actual differences in hp before and after 10:26:01 but its also super fun to play with 10:26:05 kilobyte: which numbers seems wrong to you? Tornado or Shatter? 10:26:09 -!- greatzebu has quit [Quit: greatzebu] 10:26:41 galehar: I'd have to repeat again, I looked at total damage only 10:27:07 I looked at total damage too, but against a single monster to simplify 10:27:21 (monster set: a few pan lords with a few turns worth of summons, something with a big overrepresentation of eyes) 10:29:05 heh, these don't count as fliers but as slime... still 1d damage :p 10:29:10 if monsters are spread over your whole LOS, then most of them will take very little damage from the tornado. 10:29:28 especially if you tested at low powe 10:29:31 *power 10:31:09 which is a very realistic scenario 10:31:22 you're using very high powers... I looked at a bunch of chars (from before Tornado times, though), and most seem to have 100ish power 10:31:24 in most places where using tornado seems like a good idea 10:33:11 kilobyte: 100 power is what you can expect to have when you start being able to cast it. 150 is reasonable for extended endgame with 1 enhancer and high skills 10:33:49 by making duration fixed, this should boost the spell damage output at low power 10:33:52 ... and thoroughly min-maxed INT 10:34:57 kilobyte: 130 power is normal 10:35:07 with one enhancer, 27 skills, and 30 int 10:35:09 quite normal 10:36:03 kilobyte: I've been looking again at the morphing algorithm. The fact that terrain_elements content and order is randomized every turn doesn't help. 10:36:49 by using a constant array, it seems to be better, but still too fast and it flips every few turns (I'm testing with full map morphing) 10:37:05 do you know how to slow it down? 10:37:35 galehar: monster AI currently seems to get very confused by stuff like water and lava 10:37:53 monsters will just stop moving rather than fly over it or wade in the shallow water 10:37:57 elliptic: yes, I've fixed it, will push soon 10:38:01 cool :) 10:39:48 out of first 10 morgues I've looked at, only 3 had 2 or more skills at 27, and no one was wielding an enhancer when ascending 10:40:46 the only one with int > 29 among those is a Cheibriadite, too 10:40:59 well, I don't know what collection of morgues you are looking at 10:41:37 and it isn't important to have 2 or more skills at 27... just air magic 10:41:52 (spellcasting helps a bit, but isn't that important) 10:42:06 14-runers, to reduce the amount of dedicated farmers 10:42:38 yeah, but even with that much you still got only 130, thus making 150 an unrealistic assumption 10:42:59 150 isn't realistic without two enhancers, yes, which means finding archmagi 10:43:27 still, we're debating small differences in power when galehar gets damage ration ~5-10 times smaller than I do 10:43:42 s/ration/ratio/ 10:44:30 yes 10:44:41 he already said he was just computing the damage against a single monster 10:44:54 whereas you are doing something complicated with a screen full of monsters, right? 10:44:56 the first column is optimal setup: a monster in melee range who takes full damage 10:45:12 right... I went with roughly realistic scenarios 10:45:21 really? 10:45:32 starting casting only after mingling with the first line, though 10:45:36 galehar's sounds more relevant to me, and also simpler to define 10:45:46 unless you are in zig, that is 10:45:51 or Pan 10:45:59 or, even, Snake:5 10:46:02 in pan one is mainly concerned with single targets 10:46:11 in snake:5 tornado slaughters everything 10:46:35 snake:5 is of course ridiculous 10:46:59 kilobyte: in your tests, nobody moves. Monsters which are at a range of 4 or 5 don't take much damage. In a real game, you'd be walking up to them (or them to you) 10:47:05 (you shouldn't be able to get such spells at that time... there's no way to balance both that and the late game) 10:47:48 galehar: hmm, good point. I'd have to rig hurt() itself, but that's doable 10:48:28 well, I think that's what power is for. You wouldn't cast it at the same power in snake:5 and in Zigs. But for now, the difference is too steep. 10:48:47 indeed 10:49:22 for Firestorm and Icestorm, we'll need to thoroughly nerf them too 10:49:23 snake:5 is very easy with most chars by L22 or so anyway 10:49:35 by fixing duration, we can raise damage a bit (or fix duration at 7) without making it overpowered in extended endgame 10:49:53 !lm * br.end=snake s=xl 10:50:08 13607 milestones for * (br.end=snake): 2509x 15, 2359x 16, 1749x 17, 1580x 14, 1236x 18, 900x 19, 701x 20, 593x 21, 474x 22, 453x 13, 345x 23, 223x 24, 132x 12, 120x 25, 88x 27, 76x 26, 47x 11, 12x 10, 5x 7, 5x 8 10:50:43 this is a known problem that less experienced players go into snake too early 10:51:31 well, XL22 is way on the other side though 10:51:36 making a rune required to go beyond d:24 would make it much harder to get tornado or shatter or whatever usable in time for snake:5 (and mean that snake:5 is still dangerous) 10:51:41 way on the other side of what? 10:52:05 no reason to go there earlier 10:52:43 I generally dislike the rune lock idea, but this is a strong argument for it 10:52:46 er, to go beyond d:14 10:53:39 by the way, if you want total, extreme overpoweredness, try current Yred 10:53:50 yes 10:54:10 ridiculous in midgame, also painful to play because of the interface 10:55:06 was merely ridiculous with bone dragons... with servitors, it's beyond words 10:55:11 and agree about the interface 10:55:23 the ghoul buff didn't help matters either 10:55:40 no one cares about ghouls if you have bone dragons 10:56:08 (or bone dragons with servitors) 10:56:37 we discussed a few options with Eronarn earlier, and it seems that XL is the only way to go 10:56:38 not if you have enough bone dragons/servitors, sure... but they are still really buff and increase the density of good servants 10:57:07 getting them rarely is not enough since they don't die 10:57:39 so even with 200 piety you can't get a bone dragon even by Swamp:5 10:58:24 heck, the main boss in one of Swamp endings is exactly that bone dragon 10:58:30 where by swamp:5 you probably mean something early like L16? :P 10:59:09 (I don't go there early either... some of the endings are easy, but ice ending is killer and a couple of the others are nasty too) 10:59:12 16-18 yeah... Swamp is easier than Snake, too 10:59:38 using XL sounds reasonable to me 11:00:23 depth means you dip (burning potions and wands to stay alive), get one then return to shallower levels 11:00:24 reminds me of the problems with getting demonic guardian at really low XL... quite impossible to balance giving out endgame-quality aid at low XL 11:00:43 yeah 11:02:37 really need to go home, though 11:02:43 I also liked the suggestion you made of capping the number of Y-gifted servants you have and having recall fetch them from other levels 11:02:46 we'll shout at each other later :) 11:03:09 yeah, this vs Eronarn's escort service are the two main proposals 11:03:25 and both have cross-level recalling 11:03:56 one thought: make invocation skill affect this somehow 11:04:36 so with your proposal, have the number of Y-gifted servants you can have depend partly on invoc skill 11:04:58 I know it isn't an invocation, but it still seems reasonable to me 11:05:35 -!- galehar has quit [Quit: Page closed] 11:08:52 elliptic: i think making your undead monsters more 'priestly' might be a better result of raising invoc 11:09:11 so get servitors more often, or etc. 11:09:39 'better, but also different' 11:10:56 can someone who understands vaults tell me whether reverse_minivault_9 is right? it seems like it doesn't get connected to the rest of the level, and this probably isn't good 11:11:40 -!- dtsund has joined ##crawl-dev 11:15:19 -!- st_ has joined ##crawl-dev 11:15:35 also, kilobyte: some effect of depth would be good so that e.g. if someone gets shafted they'll find slightly stronger undead (analogous to, say, xom favoring tension). should just be a small contribution in comparison to XL 11:16:49 yred isn't xom, though 11:19:13 elliptic if i remember rightly it has a staircase in it 11:20:42 Never played Skald, any pointers? 11:20:43 http://pastebin.com/qHb25nPs 11:20:56 About skills, spells, etc 11:21:16 Oops, wrong channel! 11:21:25 ^ Ignore that 11:21:31 -!- elliptic has quit [Ping timeout: 252 seconds] 11:40:10 kilobyte? beep me when you're in? 11:41:28 -!- greatzebu has joined ##crawl-dev 11:43:04 * SamB_ wonders if it's possible to switch the numeric keypad out of nethack mode temporarily 11:58:22 -!- elliptic has joined ##crawl-dev 11:59:04 Mu_: right, it is basically a staircase to a closed box with really tough monsters and doors and no loot 11:59:22 working as intended 11:59:25 :p 11:59:51 okay, so it is intended that I nearly didn't realize it was that vault? and never saw the monsters? 12:00:41 I descended into a closed room, immediately went back up and descended a different staircase, explored the level... and only later when looking at the map realized what vault it was 12:00:56 idk did it break your game or confuse you to the point of despair or what 12:01:20 no, but I don't see much point in a vault that nobody sees... 12:01:35 you could link it to the rest of the level but i think evilmike wanted the gimmick to be that you are always breaking out of the vault 12:01:46 Was the closed box diggable at least? 12:02:09 jle__: probably, I didn't check in my game because I was scared of the monsters :P 12:02:55 Mu_: what about having some of the secret doors visible when you enter? 12:04:08 maybe have the doors all linked, and make all but one of them be secret, so that you open the normal-looking door and it instantly drops all three layers of secret doors 12:04:36 Impressively evil. 12:10:29 i guess it's not really minivault9 if the doors are linked though :P 12:12:46 -!- Twilight13 has quit [Read error: Connection reset by peer] 12:13:02 -!- Twilight13 has joined ##crawl-dev 12:13:25 i'll ask evilmike i guess 12:22:42 03galehar * re79d75b06a1c 10/crawl-ref/source/mon-stuff.cc: Pathfinding: don't check the terrain on the player cell. 12:22:53 03galehar * r64567932be80 10/crawl-ref/source/art-func.h: Fix Staff of Dispater using up a turn when evoked (#4540). 12:22:53 03galehar * r18d432fd9f91 10/crawl-ref/source/abyss.cc: Simplify. 12:33:52 -!- MarvinPA_ has quit [Read error: Connection reset by peer] 12:36:23 naga with full ponderous gear has lower movement speed than others (https://crawl.develz.org/mantis/view.php?id=4541) by cosmonaut 12:36:38 -!- MarvinPA has joined ##crawl-dev 12:39:53 -!- ainsophyao has quit [Remote host closed the connection] 12:47:47 -!- Zao has joined ##crawl-dev 12:51:44 -!- Zao has left ##crawl-dev 13:03:31 -!- Galefury has quit [Quit: Stay sane inside insanity!] 13:49:49 -!- hoody_ has joined ##crawl-dev 14:19:35 -!- dpeg has joined ##crawl-dev 14:39:46 -!- syllogism has quit [] 14:52:07 Remove some ;; (https://crawl.develz.org/mantis/view.php?id=4542) by JoachimSchipper 14:57:08 Elf description fixes (https://crawl.develz.org/mantis/view.php?id=4543) by JoachimSchipper 15:00:41 FilthyApe the Conqueror (L27 MDGl) ASSERT(feat < NUM_FEATURES) in 'feature.cc' at line 24 failed on turn 159961. (Abyss) 15:02:10 Remove extra "again" in docs/develop/monster_speech.txt (https://crawl.develz.org/mantis/view.php?id=4545) by JoachimSchipper 15:02:10 "Your orc says to $foe: "I'd rather not die in your service, $player"" - makes no sense (https://crawl.develz.org/mantis/view.php?id=4544) by JoachimSchipper 15:02:50 -!- Twilight13 has quit [*.net *.split] 15:02:50 -!- stabwound has quit [*.net *.split] 15:02:50 -!- elly has quit [*.net *.split] 15:02:51 -!- Fail_Bot has quit [*.net *.split] 15:02:52 -!- BlastHardcheese has quit [*.net *.split] 15:02:52 -!- ivan`` has quit [*.net *.split] 15:02:53 -!- Twinge has quit [*.net *.split] 15:02:54 -!- jlewis has quit [*.net *.split] 15:02:54 -!- Eronarn has quit [*.net *.split] 15:02:54 -!- lorimer has quit [*.net *.split] 15:02:54 -!- CIA-34 has quit [*.net *.split] 15:02:55 -!- dtsund has quit [*.net *.split] 15:02:55 -!- Pingas has quit [*.net *.split] 15:02:55 -!- djinni has quit [*.net *.split] 15:02:56 -!- Henzell has quit [*.net *.split] 15:02:57 -!- FaMott has quit [*.net *.split] 15:02:57 -!- Vandal has quit [*.net *.split] 15:02:58 -!- blueDave has quit [*.net *.split] 15:02:58 -!- varmin has quit [*.net *.split] 15:02:59 -!- greensnark has quit [*.net *.split] 15:03:00 -!- dpeg has quit [*.net *.split] 15:03:01 -!- Zaba has quit [*.net *.split] 15:03:02 -!- paxed has quit [*.net *.split] 15:03:02 -!- ncampion has quit [*.net *.split] 15:03:03 -!- Sequell has quit [*.net *.split] 15:03:03 -!- jle__ has quit [*.net *.split] 15:04:03 -!- dpeg has joined ##crawl-dev 15:04:03 -!- Twilight13 has joined ##crawl-dev 15:04:03 -!- dtsund has joined ##crawl-dev 15:04:03 -!- Fail_Bot has joined ##crawl-dev 15:04:03 -!- Pingas has joined ##crawl-dev 15:04:03 -!- BlastHardcheese has joined ##crawl-dev 15:04:03 -!- Zaba has joined ##crawl-dev 15:04:03 -!- djinni has joined ##crawl-dev 15:04:03 -!- paxed has joined ##crawl-dev 15:04:03 -!- Henzell has joined ##crawl-dev 15:04:03 -!- elly has joined ##crawl-dev 15:04:03 -!- jle__ has joined ##crawl-dev 15:04:03 -!- ivan`` has joined ##crawl-dev 15:04:03 -!- ncampion has joined ##crawl-dev 15:04:03 -!- FaMott has joined ##crawl-dev 15:04:03 -!- Vandal has joined ##crawl-dev 15:04:03 -!- stabwound has joined ##crawl-dev 15:04:03 -!- Twinge has joined ##crawl-dev 15:04:03 -!- blueDave has joined ##crawl-dev 15:04:03 -!- Sequell has joined ##crawl-dev 15:04:03 -!- jlewis has joined ##crawl-dev 15:04:03 -!- varmin has joined ##crawl-dev 15:04:03 -!- Eronarn has joined ##crawl-dev 15:04:03 -!- lorimer has joined ##crawl-dev 15:04:03 -!- greensnark has joined ##crawl-dev 15:04:03 -!- CIA-34 has joined ##crawl-dev 15:05:06 -!- pops is now known as ghallberg 15:05:07 -!- galehar has joined ##crawl-dev 15:07:12 Cannot find Orc branch (https://crawl.develz.org/mantis/view.php?id=4546) by Tonfa 15:10:55 -!- valrus has joined ##crawl-dev 15:15:18 -!- bhaak has quit [*.net *.split] 15:15:19 -!- due has quit [*.net *.split] 15:15:19 -!- Keskitalo has quit [*.net *.split] 15:15:20 -!- Danei has quit [*.net *.split] 15:15:20 -!- kilobyte has quit [*.net *.split] 15:15:21 -!- rax has quit [*.net *.split] 15:15:21 -!- krel has quit [*.net *.split] 15:15:28 -!- due has joined ##crawl-dev 15:15:30 -!- Danei has joined ##crawl-dev 15:15:30 -!- bhaak has joined ##crawl-dev 15:15:35 -!- due has quit [Changing host] 15:15:35 -!- due has joined ##crawl-dev 15:15:45 -!- valrus has quit [Ping timeout: 260 seconds] 15:18:47 -!- rawrmage has quit [*.net *.split] 15:18:49 -!- Fyren has quit [*.net *.split] 15:18:50 -!- epyon has quit [*.net *.split] 15:18:50 -!- petete has quit [*.net *.split] 15:18:51 -!- Napkin has quit [*.net *.split] 15:18:52 -!- Gretell has quit [*.net *.split] 15:18:57 -!- Fyren has joined ##crawl-dev 15:18:57 -!- Napkin has joined ##crawl-dev 15:18:57 -!- rax has joined ##crawl-dev 15:18:58 -!- epyon|masamune has joined ##crawl-dev 15:18:58 -!- epyon|masamune is now known as epyon 15:19:17 -!- rawrmage has joined ##crawl-dev 15:19:17 -!- petete has joined ##crawl-dev 15:19:18 -!- petete has quit [Remote host closed the connection] 15:19:22 -!- Gretell has joined ##crawl-dev 15:23:54 -!- krel has joined ##crawl-dev 15:36:44 -!- kilobyte has joined ##crawl-dev 16:11:03 How difficult of a project would a GUI for vault building be? 16:11:48 I did something relatively similar in wxPython for an RPG I designed, I imagine a lot more parsing of crawl data than I had to do. 16:11:50 Fail_Bot: I am not sure it is worth it. 16:11:59 The syntax is fluid, for example. 16:11:59 And how useful would it be? 16:12:33 I cannot say... I draw all my maps in a plain editor, even without colouring :) 16:12:51 Have to ask Mu_ and mumra, those made huge and complicated vaults. 16:13:24 It would certainly be a big project 16:14:26 yes. 16:14:31 I would feel its appeal would be more entry level, and getting more people to access vault design, than catering to the experts, who can already hand write it. 16:14:31 Hi galehar! 16:14:54 hi dpeg! 16:15:48 I just feel like crawl is great, but from the development side the accesability is a little steep IMO. Perhaps its that way for a reason though 16:15:57 galehar: you're right that ?MM shouldn't hide branch entries... I guess my suggestions of ?MM giving away features (staircases, altars) but nothing more doesn't fly well with you? (My motivation is that loot closets would still be hidden and left to T&D users.) 16:15:59 Fail_Bot: well, it might bring more people to vault design, sure. 16:16:36 Fail_Bot: I think what would help most is a shell where players can edit a vault and immediately play it. I have no idea how feasible that is, however. 16:16:55 The step from editing to testing seems to cause more concerns. 16:17:03 Fail_Bot: yes, that's very true. The code has poor documentation and it's hard to get started. The reason is simply that coders hate writing docs of course :P 16:17:22 galehar: vault making documentation is much, much better now 16:17:29 absolutely 16:17:39 Having it all online with the links and coloured maps should make it much easier. 16:18:01 Especially since many players will have completely missed docs/develop/levels/syntax.txt :) 16:18:39 Fail_Bot: Maybe you could make a tutorial? 16:18:47 a GUI would be useful, but has dpeg said, it's not really worth it if you factor the amount of work vs how useful it would be. But if you have fun coding that kinda tool, it might be worth it for you! 16:20:00 I'm just trying to think of something useful to streamline the vault process. Essentially a Vault IDE. I like what dpeg said about testing because I thought that too, that vault testing was a little obtuse. 16:20:54 dpeg: I don't like ?MM giving just features right. T&D is very useful against traps (and might be even more with the new traps on the way). ?MM is a limited resource, making it more useful doesn't walk on T&D feet. 16:20:57 jle__:I plan on writing one to be honest, but part of the project is working through the vault design myself, and sorting it all out 16:21:01 It might be sufficient if there was a command line option crawl -V Failbot_altar_map which starts Crawl with a random char and the map on D:1 16:21:04 and searching for secret doors isn't very fun 16:21:13 galehar: okay, go ahead. 16:21:38 There is also the issue that proper treatment would mean that the player only ever gets _one_ chance at door/trap detection. 16:22:11 right, that would be interesting. And we could detect mimics too! 16:22:16 Fail_Bot: and an IDE could Crawl with that option (and other options the user specifies, like species, etc.) 16:22:24 galehar: yes 16:22:39 It would be nice if ?MM provided a last-ditch resort for finding that darn elusive secret door :) 16:23:35 All of that was out of the question as long as Mapping was a spell. Now, of course... 16:24:20 true. But now that it's a limited resource (and not very common), it's different 16:24:31 yes 16:24:50 Awesome.... My suggestion got commited.:D 16:26:25 elusive secret doors 16:26:27 hm... 16:26:51 -!- Galefury has joined ##crawl-dev 16:27:44 Noooo, don't code secret doors that run away from you. 16:28:32 maybe if the player has fully explored all of the level range and all the up/down stone staircases in that range they could be given a hint to the location of the missing branch entrance 16:28:48 Having only try to detect a trap/door/mimic needs restrictions, however. For example, you need to be able to search for the Temple/sewer/etc. secret doors etc. In other words, use one try on all hidden features except for 'permanently findable' doors that can be used in vaults and are used in random map building. What do you think? 16:29:05 Zannick: I thought about this, but it gets easily out of hand. 16:29:08 like ash gives it as a temptation gift 16:29:13 :) 16:29:27 Zannick: sometimes, you cannot find a bubble with white stairs. 16:30:10 dpeg: right, just restrict it to branch entrances and maybe altars behind secret doors 16:30:16 "having only _one_ try" up there, sorry 16:30:42 i saw a report fly by earlier about a missing orc branch 16:31:25 Zannick: it'd be simpler if we told players about the level on Ctrl-O (plus a message) if they have entered all levels in that branch entry's range plus a depth of four or so. 16:31:38 dpeg: yeah 16:31:55 If the player then enters the level of the (not yet found) branch, give a message. But I am not sure it's worth it :) 16:32:37 but i think that would be weird to fall down a shaft far enough that retreating all the way back up eventually yields you the exact floor without exploring any of them 16:32:47 * dpeg wonders if he should wikify the random god discussion now. 16:33:15 Zannick: drawbacks, drawbacks... 16:33:30 dpeg, galehar: mapping does reveal what is on the other side of a secret door if there are stairs in that component of the map 16:33:53 elliptic: cool! What do you think of my modified-detection proposal? 16:33:56 so it already works for detecting branch entrances 16:34:11 probably too harsh for some trap types, too :| 16:34:11 vault making tool would be cool i guess 16:34:57 hey, someone is editing that page <3 16:34:58 dpeg: not sure how it would play 16:34:59 elliptic: hey, great 16:35:42 elliptic: It does not seem to be a concern right now, but I don't want players to do the "step-wait-wait-wait" dance of Nethack. 16:36:10 Disc of storms is worth 5400 gold?! 16:36:27 dpeg: so instead the step-wait-step-wait dance? 16:36:28 nrook and I didn't get to misc prices yet. 16:36:29 some shops are more expensive than others 16:36:32 Ah 16:36:37 dpeg: Makes sense 16:36:54 galehar: if you only have one chance for detecting most hidden things, there is no reason just to run around? 16:37:41 yeah, but it's not that simple. Do you detect all things in LOS or just adjacent? 16:37:55 dpeg: I feel like traps are something that are really hard to make work right in a roguelike 16:38:06 LOS if it is one shot...Otherwise you run around. 16:38:07 it kinda does remove the choice of walking carefully or rushing, even if it's not the most interesting one 16:38:18 yes, I know, it uses LOS but that might change how you want to explore 16:38:20 it is a mess 16:38:40 Perhaps radius by skill? 16:38:46 at least, with the current system we can tell player that they should have mashed 5 when they noticed traps in the labyrinth room ;) 16:38:52 Fail_Bot: that is not the issue ... detection works already like that 16:38:57 -!- MarvinPA_ has joined ##crawl-dev 16:39:17 elliptic: the new traps will at least be much better than the current ones, I am sure. 16:39:30 by the way, about traps, there is an interface annoyance that has been bugging me for a while... if you've walked on a square, you normally know that there is no trap there 16:39:37 but you can't tell in-game which squares you've walked on 16:39:39 If I'd start a roguelike from scratch, I'd probably skip traps. 16:40:11 elliptic: and same for monsters you've seen move 16:40:26 yes 16:41:11 in some situations (running away with low HP through already explored areas), knowing which squares are guaranteed not to have traps can mean the difference between life and death 16:41:17 elliptic: this won't be a concern if we stick with single use traps. 16:41:23 uh, why? 16:41:31 sorry, blunder 16:41:51 I have died to axe traps when trying to find a cosy place for resting (without immediate danger). Not a fun death. 16:42:16 yeah 16:42:23 -!- MarvinPA has quit [Ping timeout: 260 seconds] 16:42:48 it is rare of course but really not fun when it happens because it was usually avoidable if you remembered which squares were safe 16:44:18 elliptic: no idea how to indicate that in console 16:44:23 I've considered the option of providing this information in-game (on the | key, say), but it would probably be annoying to use 16:44:37 I was just about to suggest using a key to display it 16:44:44 yes, colouring floor squares seems the only acceptable solution 16:45:34 elliptic: the new traps would still help, as they're not about plain damage. Knowing where to step would still be better, but there'd be fewer deaths. 16:46:08 hopefully, yes 16:47:26 -!- Galefury has quit [Read error: Connection reset by peer] 16:52:43 03galehar * r0484b6ef69b4 10/crawl-ref/source/ (abyss.cc abyss.h main.cc misc.cc): Move the call to abyss_morph into run_environment_effects 16:52:53 03galehar * r32602a727d58 10/crawl-ref/source/ (effects.cc stairs.cc tag-version.h tags.cc): Variable Abyss speed. 16:52:53 03galehar * r0fcafa34460d 10/crawl-ref/source/ (abyss.cc player.cc): Make the abyss morphing preserves worley's noise layout. 16:52:54 03galehar * re1af22bd561b 10/crawl-ref/source/ (8 files): Clean up all the old abyss code. 16:53:56 -!- Galefury has joined ##crawl-dev 16:53:56 galehar: didn't you want to pass the worley noise to someone else? :) 16:54:38 < Fail_Bot> How difficult of a project would a GUI for vault building be? 16:54:49 if you want to start with something simpler, i had started a syntax highlighter for vaults for the wiki 16:55:02 Eronarn: ah, you made that. Many thanks! 16:55:16 -!- SamB_ is now known as SamB 16:55:32 well, it's not even close to done, unless someone took over it :P 16:55:47 how come "Make Ponderous" isn't treated as a god ability? 16:56:10 SamB: missing description, or what context? 16:56:49 well, it gets the letter "f" and doesn't show up on ^ at all 16:58:28 dpeg: yeah, at some point I thought it was too much maths for me. But I was already too far, and it kept bugging me. In the end, it wasn't that hard :) 17:02:23 -!- Galefury has quit [Read error: Connection reset by peer] 17:07:43 SamB: please file a bug report 17:07:52 galehar: great to hear. 17:07:59 galehar: you've seen my question to you? 17:08:15 it is intentional that it doesn't use a 17:08:24 because a should be something that you actually use a lot 17:08:43 plenty of other gods have abilities that aren't on a-e, too 17:08:47 -!- ussdefiant has quit [Ping timeout: 260 seconds] 17:08:49 -!- Galefury has joined ##crawl-dev 17:08:59 however it should be mentioned on ^ 17:09:00 dpeg: no, what question? 17:13:09 elliptic: fine 17:31:34 -!- hoody_ has quit [Remote host closed the connection] 17:32:02 -!- hoody_ has joined ##crawl-dev 17:35:00 -!- MarvinPA has joined ##crawl-dev 17:36:55 -!- hoody_ has quit [Ping timeout: 260 seconds] 17:37:57 -!- MarvinPA_ has quit [Ping timeout: 260 seconds] 17:38:01 -!- Textmode has joined ##crawl-dev 17:48:27 -!- due has quit [Ping timeout: 260 seconds] 17:51:12 -!- greatzebu has quit [Quit: greatzebu] 18:04:31 -!- due has joined ##crawl-dev 18:04:31 -!- due has quit [Changing host] 18:04:31 -!- due has joined ##crawl-dev 18:20:11 -!- dpeg has quit [Quit: zzz] 18:23:06 -!- galehar has quit [Remote host closed the connection] 18:24:52 -!- hoody_ has joined ##crawl-dev 18:40:27 -!- evilmike has joined ##crawl-dev 18:47:51 is anyone around who can fix ##crawl's topic? 18:48:17 -!- MarvinPA_ has joined ##crawl-dev 18:48:29 lol 18:50:24 lol netsplit topic conflicts 18:51:35 -!- MarvinPA has quit [Ping timeout: 260 seconds] 18:55:20 -!- greatzebu has joined ##crawl-dev 19:06:12 Is there a listing of monsters other than mon-data.h or should I just parse that to create a list? 19:11:42 -!- valrus has joined ##crawl-dev 19:16:05 -!- valrus has quit [Ping timeout: 260 seconds] 19:18:43 Fail_Bot: enum.h 19:18:49 Fail_Bot: what exactly do you want to do? 19:25:40 -!- SamB has quit [Read error: Connection reset by peer] 19:25:51 -!- elly has quit [Ping timeout: 248 seconds] 19:27:55 I'm playing with the idea of creating a vault IDE, and am going to parse the monsters into a gui application that allows drag and drop. I'm essentially discovering the Crawl codebase 19:28:16 welcome to pain! 19:28:33 Your best option for a vault IDE would actually by javascript/jquery based, using the webtiles back-end. 19:29:08 (that was my initial thought on the matter) 19:29:16 i'm presuming you mean tile-based design? 19:29:18 I was writing it in python, because I sort of know what I'm doing there. 19:29:34 what are you using for the interface, then? 19:29:45 Well they are equivalent, although the intention of the project is to lower the learning curve for vault design to Nill, and allow people to quickly get their feet wet 19:29:46 wxpython? pyqt? pygtk? 19:29:53 wx 19:29:56 ahh 19:29:59 hm, it would be interesting. 19:30:35 * due uses both of these quite regularly, but has never mixxed crawl into it... much 19:30:39 I wrote half of something similar before as a backend for an rpg I was designing, but it never really got off the ground 19:31:26 hm, that sounds as though you don't have /quite/ as much work as i envisoned :) 19:32:18 Basically its just parsing the files, dropping the tiles into a grid format, and then a generator code that spits out vaults... only its not that basic 19:32:30 * due nod. 19:32:30 rright. 19:32:41 well, your issues are ... 19:32:57 1) you need the monster definitions; enum.h and mon-data.h would be enough to give you names, etc. 19:33:14 2) you need the monster tiles; this means you need tile-pick.cc and the relevant tiles definition files 19:33:32 3) you need the feature definitions, same issue as 19:33:40 and 4) you need the feature tiles, same issue as 2. 19:34:15 Thats just monsters, I also need the enviroment tiles as well 19:34:34 yes, feature definitions :) 19:37:01 The monster def's in enum.h should match what I can pull from mon-data.h right? Is there a prefrence of which to parse, or should I just do both, and parity check them? 19:37:29 well, enum.h just gives you the monster enums 19:37:36 mon-data.h gives you the names, etc. 19:38:34 you might want to talk to edlothiol, because the webtiles backend does use a bit of python 19:38:43 it's possible he's already written a parser for these 19:39:28 do you plan on having an editor that can let you build the entire vault, or just the map part? the headers can get pretty huge and complicated if the vault is big enough, and pretty scary looking if you don't understand the syntax... 19:39:32 the map part is pretty easy in comparison :P 19:40:19 if this editor can somehow let me preview fog generators it would be awesome because i have no idea wtf any of those variables do 19:40:26 haha me too 19:40:41 I just look at already existing fog vaults and pick one that behaves the way I want 19:41:38 -!- SamB has joined ##crawl-dev 19:42:09 In theory the entire vault, although I feel that initally a map generator is the bones of it, dpeg suggest a console way of quickly testing your vaults as well, as the current testing implemntation is a bit... lacking 19:42:27 -!- valrus has joined ##crawl-dev 19:43:32 -!- valrus has quit [Remote host closed the connection] 19:43:33 best way to quickly test a vault is to name it L so you can just place it in wizmode by typing &LL 19:43:39 hmm, it looks like the way things are now isn't very conducive to the listing of 0-star god benefits on the ^ screen... 19:45:43 -!- ixtli has joined ##crawl-dev 19:49:03 am I correct in thinking that the other array referred to in the comments before god_abilities[] is god_gain_power_messages[] ? 19:51:16 Fail_Bot: well, at the very least, some kind of convenient editor for making maps would be nice to have 19:51:31 Anyway, it seems like those ought to have entries for 0-star piety... 19:51:33 doing it in a text editor (even one good for this sort of thing) isn't too nice 19:52:20 on the other hand, I cant really imagine how you'd convey maps that use a lot of transforms... 19:55:04 It would have to be stative, and assign a "trasform" object flag that would allow you to define transformations on objects? 20:03:17 btw evilmike, elliptic mentioned that it's pretty easy to just miss reverse minivault9 20:04:06 -!- elly has joined ##crawl-dev 20:04:24 was thinking of leaving the doors linked and making one of them a regular door so that the player will open the door and automatically drop all three layers of secret doors 20:04:42 but i guess it's not really minivault9 with linked doors 20:07:11 -!- Galefury has quit [Read error: Connection reset by peer] 20:07:39 Actually, vault testing shouldn't be /that/ difficult. 20:09:40 -!- ainsophyao has joined ##crawl-dev 20:13:37 -!- Galefury has joined ##crawl-dev 20:16:06 Mu_: seems reasonable to make a change like that 20:16:27 i've also seen that you can get "rock" secret doors in places like dis, and it really looks weird... but i don't know how to fix that 20:18:10 with that option, the best thing to do might be to just give it a 50% chance or so of being like that... should be fairly trivial 20:22:43 -!- Twilight13 has quit [Read error: Connection reset by peer] 20:22:56 -!- Twilight13 has joined ##crawl-dev 20:34:10 -!- Galefury has quit [Read error: Connection reset by peer] 20:40:08 -!- Pingas has quit [Read error: Connection reset by peer] 20:40:23 -!- elliptic has quit [Ping timeout: 260 seconds] 20:42:58 -!- elliptic has joined ##crawl-dev 20:44:15 -!- Leissi_ has quit [Read error: Connection reset by peer] 20:46:26 -!- Leissi has joined ##crawl-dev 20:51:38 -!- MarvinPA has joined ##crawl-dev 20:55:08 -!- MarvinPA_ has quit [Ping timeout: 252 seconds] 20:58:52 std::string buf = "You can call upon " 20:58:52 + god_name(which_god) 20:58:52 + " to make pieces of armour ponderous."; 20:59:42 is that good wording, or not? 21:00:31 -!- ainsophyao has quit [Remote host closed the connection] 21:01:30 sounds good to me 21:26:07 -!- hoody_ has quit [Remote host closed the connection] 21:26:31 -!- hoody_ has joined ##crawl-dev 21:27:28 -!- elly has quit [Ping timeout: 248 seconds] 21:31:43 -!- hoody_ has quit [Ping timeout: 276 seconds] 21:31:46 -!- MakMorn has quit [Ping timeout: 260 seconds] 21:32:07 -!- MakMorn has joined ##crawl-dev 21:34:15 Cheibriados' "Make Ponderous" ability not listed on religion (^) screen. (https://crawl.develz.org/mantis/view.php?id=4547) by SamB 21:39:02 -!- purge has joined ##crawl-dev 21:43:14 -!- Mu_ has quit [Quit: Defecator, may everything turn out okay so that you can leave this place.] 21:44:05 Changes and are provided --^ 21:54:18 -!- ainsophyao has joined ##crawl-dev 22:04:40 -!- Fail_Bot has quit [Ping timeout: 260 seconds] 22:16:09 -!- MarvinPA_ has joined ##crawl-dev 22:17:28 -!- ZorbaBeta is now known as ZorbaTHut 22:19:40 -!- MarvinPA_ has quit [Client Quit] 22:20:11 -!- MarvinPA has quit [Ping timeout: 260 seconds] 22:23:02 -!- ussdefiant has joined ##crawl-dev 22:32:33 -!- hoody_ has joined ##crawl-dev 22:33:34 -!- hoody_ has quit [Remote host closed the connection] 22:34:00 -!- hoody_ has joined ##crawl-dev 22:38:12 -!- hoody_ has quit [Ping timeout: 240 seconds] 22:56:29 -!- MarvinPA has joined ##crawl-dev 22:59:43 -!- Wensley has joined ##crawl-dev 22:59:56 -!- hoody_ has joined ##crawl-dev 23:17:22 -!- valrus has joined ##crawl-dev 23:19:47 -!- Fail_Bot has joined ##crawl-dev 23:36:38 -!- Wensley has quit [Ping timeout: 260 seconds] 23:40:10 -!- ainsophyao has quit [Remote host closed the connection] 23:41:43 -!- ainsophyao has joined ##crawl-dev 23:51:50 -!- hoody_ has quit [Remote host closed the connection] 23:52:18 -!- hoody_ has joined ##crawl-dev 23:53:15 -!- ainsophyao has quit [Remote host closed the connection] 23:56:36 -!- hoody_ has quit [Ping timeout: 252 seconds]