00:05:19 -!- Xiberia has joined ##crawl-dev 00:17:12 -!- Pacra_ has joined ##crawl-dev 00:17:43 -!- Pacra has quit [Read error: Connection reset by peer] 00:17:59 -!- Adeon has quit [Ping timeout: 244 seconds] 00:18:17 -!- Adeon has joined ##crawl-dev 00:27:36 Visible features: a greater mummy, a lich, a lich, a guardian mummy, a mummy, a staircase back to the Crypt. 00:27:46 This might do for hardmode Tomb for now. 00:39:32 -!- blabber has quit [Quit: leaving] 00:40:15 -!- danharaj has joined ##crawl-dev 00:44:53 -!- MarvinPA_ has joined ##crawl-dev 00:48:45 -!- MarvinPA has quit [Ping timeout: 250 seconds] 00:48:45 -!- MarvinPA_ is now known as MarvinPA 00:52:55 -!- nrook has quit [Quit: ChatZilla 0.9.88.2 [Firefox 12.0/20120420145725]] 00:53:18 -!- evilmike has quit [Read error: Connection reset by peer] 00:54:47 -!- casmith789 has joined ##crawl-dev 01:03:18 Description of arrows still refers to enchantment (https://crawl.develz.org/mantis/view.php?id=5668) by nubinia 01:05:02 -!- Xiberia has quit [Ping timeout: 252 seconds] 01:18:18 03|amethyst * rdc39ea48bec1 10/crawl-ref/source/itemname.cc: Omit enchantment from missile base/qual/db name. 01:19:37 <|amethyst> err 01:19:50 <|amethyst> "enchantment" meaning "brand" here 01:20:07 <|amethyst> /nick derpsichore 01:20:09 -!- Pacra_ has quit [Ping timeout: 272 seconds] 01:58:53 -!- danharaj has left ##crawl-dev 02:01:02 -!- SkaryMonk1 has joined ##crawl-dev 02:01:52 -!- SkaryMonk1 is now known as SkaryWonk 02:05:42 -!- SkaryMonk has quit [Read error: Connection reset by peer] 02:06:14 -!- SkaryMonk has joined ##crawl-dev 02:07:43 Hey, is there someone can approve my mail posted to mailing list? 02:08:54 -!- edlothiol has joined ##crawl-dev 02:26:47 -!- ainsophyao has joined ##crawl-dev 02:39:22 -!- Textmode has quit [Ping timeout: 255 seconds] 02:45:01 -!- edlothiol has quit [Ping timeout: 272 seconds] 03:09:29 -!- dtsund has quit [Quit: dtsund] 03:20:47 -!- XnMojo has quit [Quit: XnMojo] 03:21:19 -!- danharaj has joined ##crawl-dev 03:27:19 the one you sent 3 days ago? Or a new one? 03:33:29 hey, I'm trying to resolve #5576, about prompt spam you get when you try to unwield a disto weapon, and I can see where the problem is but it seems too messy to cleanly resolve. 03:34:47 -!- absolutego has joined ##crawl-dev 03:39:20 The issue is that wield_weapon relies on prompt_invent_item to check for warnings when swapping manualy (i.e. not auto-wield). prompt_invent_item doesn't know about weapon wielding very much so it produces extraneous prompts that shouldn't be given. 03:40:36 it seems like a bad idea to duplicate prompt_invent_item's functionality elsewhere with less prompts. any guidance? 03:42:54 Really the issue is that check_warning_inscriptions not only checks the inventory item for warnings, but also checks the old item, i.e. the currently wielded item. That functionality probably should be split up because, as in the case with weapons, you will want more logic between the two. 03:44:41 also it occurs to me that this is an awkward time to bring this up. 5 am EST :P 03:46:07 galehar: 3 days ago. 03:48:05 danharaj: well, some of us live in europe :) 03:48:14 danharaj: looking at it 03:48:30 blmarket: I think someone did, I received your email. 03:49:25 galehar: it should be lunch time in many places over there :) 03:49:36 or perhaps late breakfast. 03:49:37 -!- absolutego has quit [Quit: Lost terminal] 03:53:56 -!- Pingas has joined ##crawl-dev 03:54:20 it's 11 AM in France 04:00:29 danharaj: this seems to work http://pastebin.com/rW5nZNAH 04:01:03 but it also breaks out of the prompt loop if you answer no to a wield prompt 04:01:18 which isn't a big deal I'd say 04:02:57 otherwise, we need check_warning_inscriptions to return 3 values. -1 for don't unwield (which breaks), 0 for don't wield(which stays in the loop) 1 for ok 04:05:20 blmarket: I haven't answered your email because I didn't really understand it. Not because of your english writing skills, but because of my ignorance of gettext 04:06:17 blmarket: you're using gettext for the korean version of crawl? kilobyte seems to think it's not really appropriate and that we shouldn'ty use it 04:06:52 hopefully he'll answer you, he's more knowledgeable of translation implementation 04:11:12 galehar, blmarket: gettext cannot do grammar at all. I see that Korean is almost gramar-free (static word order, no grammatic gender, article mess, etc), but there's still a bunch of special cases. 04:11:44 -!- gnsh has joined ##crawl-dev 04:11:51 -!- __duncan has quit [Remote host closed the connection] 04:12:52 gettext itself is useful for table lookup, though -- there's a great number of tables of strings, word classification, etc 04:12:55 korean is also an agglutinative language. such languages use affixes to express grammar stuff, but don't change the stem when adding that affixes 04:13:10 unlike german or polish for example 04:13:47 I bet it does have exceptions though :p 04:13:52 i don't know korean in detail, but gettext could be useful for an agglutinative language 04:14:00 of course, there are ALWAYS exceptions :) 04:14:14 but agglutinative languages in general are quite regular 04:16:26 yeah, but that just means there won't be almost any grammar functions and most things will go straight to a table 04:17:06 there are three _implementation_ problems with gettext: 04:17:40 1. it can't be told to use an encoding other than the system one; Crawl wants UTF-8 internally in all cases 04:18:40 2. you can't select a locale that is not installed on the system. Typical Unix boxes have source files for all locales/encodings but have only one or a few "generated". 04:18:55 1. wtf? that can't be true? 04:19:10 (I mean, type "locale -a" to see available ones, or "dpkg-reconfigure locales" to change the election) 04:19:57 bhaak: you'd need to change, say, "de_DE" (an alias to de_DE.ISO-8859-1) to "de_DE.UTF-8", but that will fail if the latter is not generated 04:20:22 galehar: That's a good solution but I think having such behavior for saying no to wield is probably inconsistent with other similar circumstances. Making check_warning_inscriptions return a 3-valued type seems inconsistent with the rest of the code. I will explore merging check_old_item_warning's functionality into the relevant callers. 04:20:27 2. yes. that's part of how locale works. i would like to read why they designed the c locale stuff like they did. some parts look quite braind-damaged and limiting (for example, you can't work with 2 locale concurrently) 04:21:20 kilobyte: bind_textdomain_codeset() changes the output encoding http://www.gnu.org/software/libc/manual/html_node/Charset-conversion-in-gettext.html 04:23:15 bhaak: except that this is a GNU extensions 04:24:18 danharaj: considering how messy those functions are, I'd say we should either do a quick and simple fix, or refactor/clean/simplify. Don't add to the mess! 04:26:05 galehar: that's sensible! :) I think a refactor will have to be done then, because the quick fix breaks the interface in another way, and I don't think anyone appreciates interface surprises. 04:26:28 I'll put my head to it. The equipment code seems rather messy as it is. 04:26:48 danharaj: thanks for looking into it! 04:27:40 galehar: I only do this because the last time I wielded a distortion weapon I fumbled 10-20 keys because of this confusion ;) 04:27:47 3. there appears to be no way to find out whether a translation for a string exists 04:28:15 gettext will return the query string instead, which for most programs is a good idea 04:28:17 kilobyte: gah, and it's not even documented that it is. 04:29:47 yeah, you would have to check the returned string. if it is the same as the query string there is no translation. 04:29:49 not sure who uses non-GNU gettexts. Sun's implementation is ancient, but knowing BSDs and their hatred for GPL, they might be. 04:30:02 or there is a translation and it is the same as the query string :) 04:31:21 with languages copying from each other, and being all forks of each other in the first place, there's a lot of identical translations, yeah 04:32:00 except, if there's any grammar involved, Crawl won't know whether he functions mentioned should be ran in english or in the language in question 04:32:50 reinventing this part of the wheel is not a big problem, though. Heck, we can even do 100% compatible .mo files :p 04:33:17 -!- st_ has quit [] 04:33:59 mo files? what for if you don't use gettext? 04:34:17 i'd understand .po files for using all the po-editors out there 04:35:05 O(1) lookups 04:35:57 ah, needless pre-optimizations ;-) 04:36:06 can of course parse the whole set of strings on every startup, but since we already have a cache for .des and the db, it's an obvious thing to do 04:37:46 not entirely needless, adding a MB or so (a 0-th degree appriximation, and std::map<> has a big waste ratio) per game would be meaningful for public servers 04:38:24 I mean memory usage here, not speed 04:44:01 if you're concerned about memory, i think .mo isn't the smallest possible representation. doesn't it save the query strings in the file? going from a string lookup key to an int lookup key would save most of the used space for the lookup keys 04:45:20 uhm, no. Gettext uses mmap for .mo, going key->int int->value would be an actual extra waste. 04:46:28 no point in over-optimizing here, what I mean is that .mo files are shared between all currently running games 04:46:35 no, i meant a preprocessing step that translates key->value to int->value. but that would also involve source code modifications 04:47:09 ugh, having a Microsoft-style string table would make the source totally unreadable 04:47:28 (I mean the Windows 2.0 way) 04:49:01 hehe, learning from the 800 pound gorilla 04:55:05 -!- ToBeFree has joined ##crawl-dev 04:55:13 -!- XnMojo has joined ##crawl-dev 04:57:49 -!- ark____ has joined ##crawl-dev 04:58:39 -!- Pingas has quit [Ping timeout: 244 seconds] 05:00:06 OSX builds of master branch on crawl.develz.org updated to: 0.11-a0-1818-gdc39ea4 05:00:25 you're right though, if this part of gettext has to be reimplemented, not creating a .mo in first versions is a good idea 05:04:47 oh .mo! 05:11:38 -!- absolutego has joined ##crawl-dev 05:15:45 -!- oberstein has quit [Ping timeout: 272 seconds] 05:16:18 -!- oberstein has joined ##crawl-dev 05:30:13 -!- PatashuXantheres has joined ##crawl-dev 05:31:31 -!- Patashu has quit [Ping timeout: 256 seconds] 05:33:32 the removal of ammo enchantment is probably not effective on cao-trunk yet, right? 05:36:10 -!- Pingas has joined ##crawl-dev 05:36:15 absolutego: it is 05:36:49 then d:5 curare <3 05:36:53 some description fixes were made since the last update 05:37:13 yeah, i saw the post on the tavern, which is why i asked 05:37:25 only if you got some throwing 05:37:56 kobold apts! 05:38:06 !apt throwing 05:38:07 Throw: Ce: 3!, DD: -1, DE: 1, Dg: -1, Dr: -1, Dr[black]: -1, Dr[green]: -1, Dr[grey]: 0, Dr[mottled]: -1, Dr[pale]: -1, Dr[purple]: -1, Dr[red]: -1, Dr[white]: -1, Dr[yellow]: -1, Ds: -1, Fe: -99*, Gh: -1, Ha: 3!, HE: 1, HO: 0, Hu: 0, Ko: 3!, Mf: 0, Mi: 0, Mu: -2, Na: -1, Og: 1, Op: 0, SE: 2, Sp: 1, Te: 1, Tr: -1, Vp: -2 05:38:37 (it's a kock though. last one god abyssed at xl8) 05:38:41 Hemzell shouldn't consider -99 to be the worst aptitude, I guess 05:38:54 so...? 05:39:03 oh doh, CK not AK 05:39:25 i'll try not to get attached 05:40:03 ("Your Throwing skill gained 3 levels and is now at level 3!") 05:41:01 -!- syllogism has joined ##crawl-dev 05:41:07 -!- XnMojo has quit [Quit: XnMojo] 05:43:18 kilobyte: about database aliases, maybe they should be resolved when building database instead of when doing lookups. What do you think? 05:44:28 not a big deal either way: a bit of wasted space for one lookup less in a non-critical path 05:47:16 in the conversation about .mo a page above, speed matters because item/monster names are resolved inside loops up to O(n^2) and memory matters because of CAO/CDO, but the only bit where the db could possibly care is searching by contents 05:50:59 -!- alefury has joined ##crawl-dev 06:00:36 -!- MarvinPA_ has joined ##crawl-dev 06:04:31 -!- MarvinPA has quit [Ping timeout: 272 seconds] 06:04:31 -!- MarvinPA_ is now known as MarvinPA 06:11:01 03kilobyte * r10519d1b5a9e 10/crawl-ref/source/skills2.cc: Adjust formatting. 06:11:08 03kilobyte * rff7d79b693d6 10/crawl-ref/source/ (dgn-overview.cc dgn-overview.h files.cc): Keep unique annotations on portal levels until the level is deleted. 06:11:08 03kilobyte * r8c4ae6daeaef 10/crawl-ref/source/ (item_use.h shopping.cc throw.cc throw.h transform.cc): Make two functions static, one gone. 06:11:08 03kilobyte * r83d9845a1c98 10/crawl-ref/source/ (travel.cc travel.h viewmap.cc): Purge a strange quirk of sorting the player's god's altar first on 'X'. 06:11:09 03kilobyte * r919270b38deb 10/crawl-ref/source/ (dgn-overview.cc dgn-overview.h files.cc): Allow annotating portal levels. 06:11:09 03kilobyte * r4ced95433dc9 10/crawl-ref/source/abyss.cc: Make a note about corrupted levels. 06:11:09 03kilobyte * r0bb39f3841ed 10/crawl-ref/source/tileweb.cc: Fix a compiler warning. 06:11:09 03kilobyte * rd77cb243f562 10/crawl-ref/source/dat/descript/spells.txt: Rewrite misleading descriptions of RMsl/DMsl. 06:11:54 -!- ainsophyao has quit [Remote host closed the connection] 06:18:29 -!- mikee_ has quit [Quit: HydraIRC -> http://www.hydrairc.com <- Po-ta-to, boil em, mash em, stick em in a stew.] 06:35:44 -!- Sabaki has quit [Read error: Operation timed out] 06:38:34 -!- syllogism- has joined ##crawl-dev 06:39:24 -!- syllogism has quit [Ping timeout: 244 seconds] 06:39:27 kilobyte: so, what about line length in descriptions? 76 means we'll have to do a lot of rewrap of manual edits. Like the recent moth of suppression edit. 06:40:04 80 is safest. 79 is a bit nicer for git diff (leaves room for +/-) 06:47:36 I don't see a big difference, I'd be ok with any length. 06:47:42 i like that we have hexcrawl, squarelos, *and* new_squarelos branches 06:48:08 Eronarn: time to merge in hyperrogue into Crawl as well 06:48:45 is gitorious acting weird for anyone else? i can't do a pull 06:49:02 -!- monqy has quit [Quit: hello] 06:49:17 works for me 06:49:34 hmm, i just get 'fatal: The remote end hung up unexpectedly' 06:50:42 Would it be a lot of work to have it so every scroll variety seen no earlier than D:5 is annotated {possibly immolation}? 06:50:58 Immolation is the only scroll that is instantly devastating for certain builds (being a mummy, being a hopr) 06:51:42 immolation is a pretty bad scroll but that's not a good solution to it 06:52:14 oh oops i didn't specify a branch 06:52:55 would torment also piss off all your allies, btw? 06:57:37 -!- absolutego has quit [Quit: leaving] 07:00:53 -!- Ragdoll has joined ##crawl-dev 07:01:05 aw crap, git choked on the throwing.cc change 07:04:33 PatashuXantheres: are you a recklessly diving FeWz? If not, immolation isn't exactly going to kill you. 07:05:33 kilobyte: in the scenario 'I am a hopr, I am surrounded by my orcs, I read immolation' you are pretty much dead 07:05:37 since now all your orcs are enemies 07:05:37 I'd increase the scroll's effect so it's not useless. 07:05:41 ah heh :) 07:06:11 beogh doesn't get mad at you, at least - that part is explicitly coded 07:07:42 oh whaaaaaaaat 07:07:45 i found part of silly hat patch 07:08:45 cloak, cape, mantle, tabard, shawl, scarf, stole :D 07:09:27 added more helmet types too 07:09:52 helmet, helm, horned, crested, plumed, visored, mail coif, greathelm, winged, spiked, golden, crown, tiara, coronet, diadem 07:10:30 changing the noun would unnecessarily muddle things 07:10:49 kilobyte: people like cosmetic stuff (and the people who don't can disable it) 07:12:59 I was afk... I agree gettext sucks because I had to use so many of hacks to translate crawl. so many word reorderings and pgettext for special contexts... 07:15:58 galehar: i disagree with the closing of the short questions thread on the tavern. it trailed off a lot more than it used to in the past few weeks, but generally it was quite good. also, not everyone knows about or uses IRC, so telling people to ask stuff in ##crawl instead doesnt work very well. 07:16:03 I just realized... can't check right now, but does the existing Korean translation handle the display of text in the message area in local tiles? 07:16:28 usually TILES_local but I see webtiles works too 07:16:56 but it needs font hack... to expand crawl's limit to render only alphabet characters 07:17:17 * kilobyte is REALLY interested what that "font hack" is. 07:17:19 alefury: I don't see what's unreasonable about asking people to create threads in a forum. 07:17:43 'only alphabet characters'? huh? 07:17:48 well, most of the questions just need a quick one line answer, and theres nothing to discuss 07:17:56 current crawl 'caches' 256 characters, so I needed to expand these characters to include korean characters, which is almost 5000+ characters 07:18:17 03kilobyte * rb006b6920088 10/crawl-ref/source/dat/descript/spells.txt: Purge the descs of ex-spells. 07:18:19 just simple hack, one big string contains all available korean characters, and let it cache all of them 07:18:30 recently the thread tended to trail off a bit, but it didnt use to be like that. the question really didnt need their own thread. 07:18:43 alefury: except the ones who don't. And there's nothing wrong with a thread with a single one line answer. It's easy to seaarch for and to link to. 07:19:50 alefury: you never know when asking a question if it's going to spill over. And when it does, the burden is on the mods to sort it out. 07:20:53 fair enough i guess. i just think most of these short questions will just end up never being asked, instead of getting a 2 post thread. which is kind of bad. 07:20:55 'behaviour_event(monster*, mon_event_type, const actor*, coord_def, bool)' - the const actor* is new? is that supposed to be whatever it's targeting, or what? 07:21:18 i have from my old code behaviour_event(mons, ME_DISTURB, MHITNOT, mons->pos()); 07:24:44 alefury: when asking help from a community, it's good practise to make the effort of thinking up a good thread title, so the answers can be useful to others too. 07:24:50 -!- ainsophyao has joined ##crawl-dev 07:24:58 oh, nevermind, i see what happened 07:25:50 well yeah. but thats a lot more work than just asking some one-off question in a thread designated for it. I actually find the short questions thread a lot more convenient to read than the tavern in general. 07:26:18 ...i think, anyways. MHITNOT is just not used in that context anymore? 07:27:07 creating a thread takes 2 secs, it's not a lot more work 07:28:06 well, you also have to think of a proper title 07:28:21 is there something i can do to get 'both modified: contrib/sqlite' to stop showing up, without messing up my local 07:28:52 alefury: which is a good thing. 07:28:56 i don't get anything if i try to diff it, not sure what is going on with it 07:29:49 galehar: it has good and bad sides. i dont really know how much the tavern gets searched anyway, so even if people still asked these questions and just put a proper title on them, i dont know if it would help at all. 07:31:02 <|amethyst> !apt evasion 07:31:04 Could not understand "evasion" 07:31:20 alefury: now, answers have a chance to be found. Before they didn't. 07:31:21 galehar: a lot of those questions arent repeat questions anyway, but very specific stuff. 07:32:00 galehar: are you not listening, or do you disagree? im saying there wont even be questions, so how will there be answers to find? 07:32:20 <|amethyst> !apt dodging 07:32:20 Dodge: Ce: -3*, DD: 1, DE: 2, Dg: -1, Dr: -1, Dr[black]: -1, Dr[green]: -1, Dr[grey]: -1, Dr[mottled]: -1, Dr[pale]: -1, Dr[purple]: -1, Dr[red]: -1, Dr[white]: -1, Dr[yellow]: -1, Ds: -1, Fe: 3, Gh: -1, Ha: 2, HE: 1, HO: -2, Hu: 0, Ko: 2, Mf: 3, Mi: 1, Mu: -2, Na: -2, Og: -1, Op: 0, SE: 2, Sp: 4!, Te: 1, Tr: -2, Vp: 1 07:32:51 alefury: why wouldn't there be questions? If someone can't take 2 secs to create a thread, why anyone should bother answering? 07:33:16 alefury: I don't think people will stop asking questions on the tavern because I locked this thread 07:33:17 <|amethyst> kilobyte: should "drunk plate-clad dwarf" maybe be a race with worse dodging aptitude? 07:33:53 galehar: im pretty sure many of the questions that were asked and answered in that thread would never have been asked without it. 07:34:45 alefury: if you think about it "too small to need their own thread" doesn't mean anything 07:35:20 galehar: it doesnt mean anything to you. and the meaning differs when you ask different people. 07:36:27 blmarket: it'd be quite unplausible to load all of Korean, Chinese and so on at once. It looks like there's some paging needed. 07:36:52 blmarket: do you load it all onto a single giant texture, or do you have multiple pages already? 07:37:08 galehar: anyway, ill shut up. if nobody else asks for it to be reopened its probably fine to keep it closed. 07:37:56 I did one giant, but It seems we need SDL 1.3 for at least IM 07:38:14 IM? 07:38:15 and SDL 1.3 also offers ttf drawing, so we may not need to cache in one texture 07:38:22 Input Method... we can't input in Korean yet 07:38:46 One korean characters are combine of 2~5 small characters 07:39:02 for example.. 밥 is combine of ㅂㅏㅂ 07:40:32 so we need IM to make a input. but SDL 1.2 doesn't offer so... we doesn't offer korean input yet. 07:41:07 not sure about other distributions, Debian has sdl1.3 in unstable already, and Crawl ships contribs 07:41:53 kilobyte: i finally got what |amethyst meant with his drunk plate-clad dwarf comment. it really doesnt make much sense without MD in the game. 07:42:07 alefury: hmm right 07:42:17 sdl 1.3 also has proper multi-monitor support. Currently, local tiles defaults to using the full desktop width 07:42:24 I think supporting other charsets can be done with branch other than master... most of crawl players don't need that big texture or unstable SDL 07:42:26 alefury: care to rewrite it? 07:42:43 no 07:42:48 i cant write 07:42:55 everything i write sounds like shit 07:43:14 blmarket: in 0.10, they don't. 0.11 has translations, and I already sprinkled fancy typography in a bunch of places. 07:43:29 (ok, badly unfinished translations, but still) 07:43:50 and this all works perfectly in console and webtiles already, only local tiles trail behind 07:44:08 alefury: welcome to the club 07:44:17 <|amethyst> blmarket: I don't like that idea... if we need newer SDL for proper i18n, we need newer SDL for proper i18n. 07:44:38 except russian it seems crawl fit in less than 512 characters 07:44:53 <|amethyst> russian? 07:44:59 but we need to think if newer SDL makes some old machine user unplayable 07:45:06 russian characters 07:45:08 <|amethyst> why would russian need more than 512 characters? 07:45:11 i don't know how many is 07:45:12 just add a line about how wiglaf is the last mountain dwarf in the dungeon because he has always been slow 07:45:29 only CJK charset needs more than 1024 characters i think 07:45:58 alefury: my trick is to write something factually correct, then wait for a native speaker like Dolorous to make it sound like non-shit :p 07:46:11 (perhaps even |amethyst could be fooled into correcting this one :p) 07:46:23 <|amethyst> kilobyte: what about "drunk plate-clad centaur" ? 07:46:31 Russian is pretty fine with 256 (KOIR) 07:47:02 <|amethyst> why do we have to cache characters anyway? 07:47:09 someone who hasn't tried a centaur yet might not be aware they can't dodge well 07:47:13 |amethyst: GL issues... 07:47:23 <|amethyst> why are we using GL? 07:47:32 |amethyst: because... uhm... 07:47:34 for speed? seems using freetype2 library directly to render all characters should be far slower... 07:47:53 <|amethyst> isn't that what every other program does? 07:48:11 <|amethyst> my web browser doesn't cache character images AFAIK 07:48:24 well, there were SDL-ttf for SDL 1.2 but if we move to SDL 1.3 we better just move to 1.3 I think 07:48:27 |amethyst: to ensure many platforms and/or drivers can't run it! To ensure we have a good selection of driver bugs! To make old machines without fancy hardware work unplayably slow! 07:48:30 <|amethyst> or, if it does, it's deep inside the pango library 07:49:19 <|amethyst> orcs aren't great at dodging, and are definitely more likely than a centaur to be drunk 07:49:24 <|amethyst> maybe orc 07:49:41 <|amethyst> (why, yes, my multitasking is preemptive) 07:49:43 you see, you don't need to make a comparison there... 07:49:55 its funny, though 07:51:43 sdl 1.3 (or 2.0, as it's now called) still misses some pretty obvious features (relative mouse movement on linux, for one), but I'm not sure if there are any that crawl would need 07:53:37 and yeah, why does crawl use opengl? Doesn't look like there is much to accelerate... 07:53:50 ...AK in sprint start in the abyss? what 07:54:08 I think so... changing SDL is so unstable... I'd prefer just translating first. then isolate korean output code in some branches. 07:54:36 because render code resides only in fontwrapper-ft.cc, so it doesn't need much care 07:54:44 <|amethyst> if it's in a branch it will never be used 07:54:59 <|amethyst> maybe it won't be mainline before 0.11, but it should be eventually 07:55:25 well, maybe I'll make my own build for koreans, so that would be enough for a while 07:55:48 <|amethyst> how could we omit the world's most logical writing system? 07:55:51 <|amethyst> :) 07:56:39 haha :) 07:57:45 unintentional, but i just noticed this: reducing lorc radius to 1, due to the way it's implemented (have not checked this for other auras), means that stuff is not affected by it if you run from it 07:58:07 it'll be in the radius, you step, it's out of the radius during its action when damage is checked, it steps, repeat 07:58:38 keep or fix? i think the effect is probably good even if counterintuitive 07:59:16 <|amethyst> accidental anti-kiting? sounds good for this purpose 07:59:20 <|amethyst> not sure about the other auras 08:00:23 lorc? 08:00:24 there are no other damaging auras, right 08:00:34 the only similar thing is ring of flames and that uses clouds 08:00:35 <|amethyst> PatashuXantheres: lava orcs 08:00:39 oh, that thing 08:03:04 -!- absolutego has joined ##crawl-dev 08:03:43 true story: movespeed += and movespeed -= are not the same 08:06:15 03|amethyst * r6b48cd236379 10/crawl-ref/source/dat/descript/spells.txt: Improve DMsl description. 08:16:37 Pacified Boris shows up again. (https://crawl.develz.org/mantis/view.php?id=5669) by elliptic 08:16:53 fr pacified boris shows up again pacified 08:18:12 -!- Pacra has joined ##crawl-dev 08:18:57 did he have two instances of boris in that game, then? 08:19:02 could he have lead them to each other and have them duel? 08:19:06 because that would make -excellent- crawl tv 08:21:13 I assume boris left the level in the first game 08:21:17 er, first level 08:25:32 -!- Mottie has joined ##crawl-dev 08:29:32 -!- magistern has joined ##crawl-dev 08:32:59 -!- jeanjacques has joined ##crawl-dev 08:44:17 -!- absolutego has quit [Quit: leaving] 08:44:46 -!- Xiberia has joined ##crawl-dev 08:48:59 !!! 08:49:02 i found ALL of silly hat patch 08:49:11 apparently it was on my github this whole time 08:51:53 https://crawl.develz.org/mantis/view.php?id=5154 <-- anyone want to check out lava orcs? 08:52:12 they are definitely ready for some testing, they're quite different than last time around 08:53:32 is there a serious hat patch? 08:54:17 no, but silly hats are serious business! 08:54:54 well, it includes both silly and serious hats, at least in the context of crawl :) 08:55:16 -!- edlothiol has joined ##crawl-dev 08:59:28 I noticed blink frogs have !sil while prince ribbit does not 08:59:30 Is that intentional? 09:00:07 doubt it 09:02:18 <|amethyst> he has spells, they don't 09:02:54 <|amethyst> (teleport self, specifically) 09:03:17 <|amethyst> also why Trog rewards you for killing him 09:05:11 oh, true. he's explicitly a spellcaster 09:06:18 |amethyst: more reason to do the thing we talked about a few days ago :P 09:06:27 <|amethyst> which thing? 09:06:44 having spell status (arcane, innate, etc.) be able to be set per-spell-in-spellbook 09:07:36 <|amethyst> ah, yeah 09:07:46 <|amethyst> that's his only spell, though 09:07:56 oh, is blink frog blink handled elsewhere? 09:08:06 ~crawl code~ 09:08:11 his blinking ability wouldn't be stopped by silence? 09:08:26 Prince Ribbit (11F) | Spd: 14 (swim: 60%) | Int: normal | HD: 6 | HP: 40 | AC/EV: 0/16 | Dam: 20 | amphibious, spellcaster, cold-blooded | Res: 06magic(40), 12drown | Chunks: 07contam | XP: 302 | Sp: blink; teleport self. 09:08:26 %??prince ribbit 09:08:30 also, i'm starting work on the evaporate patch. remove fulsome entirely, y/n 09:08:37 blink frog (09F) | Spd: 14 (swim: 60%) | Int: insect | HD: 6 | HP: 21-45 | AC/EV: 0/16 | Dam: 2006(blink) | amphibious, cold-blooded, !sil | Res: 06magic(40), 12drown | XP: 265 | Sp: blink. 09:08:37 %??blink frog 09:08:52 blink as an attack flag is so dumb 09:08:57 you'd expect it to blink the thing it hits 09:08:58 Eronarn: please explain, what is "the evaporate patch"? 09:09:13 alefury: make the evaporate spell used on a corpse to generate vials, which you throw 09:09:23 vials decay fairly quickly, like blood 09:11:30 <|amethyst> cut out the middleman and turn it into a more flexible version of corpse rot :) 09:12:07 ...oh man 09:12:12 Evaporate => vials 09:12:17 Eronarn: who is the "you" who would expect it to blink the thing it hits? 09:12:18 Freeze Distillation => snowglobes 09:12:42 elliptic: all the other attack flags work like that (barring chaos which is weird) 09:13:13 MI_NONE // was MI_EGGPLANT... used for launch type detection 09:13:14 <|amethyst> maybe monster should write it as "(then blink)"? 09:13:15 what 09:13:42 or (blink self) 09:14:02 Eronarn: players don't usually even know that it is an attack flag, so I don't see how they would be bothered 09:14:16 they just see the monster blinking a lot 09:14:25 |amethyst: I like (blink self) 09:14:27 elliptic: no reason not to change it though, at least for henzell 09:14:52 would be good to change monster to be a little clearer, yes 09:19:16 Prince Ribbit (11F) | Spd: 14 (swim: 60%) | Int: normal | HD: 6 | HP: 40 | AC/EV: 0/16 | Dam: 20 | amphibious, spellcaster, cold-blooded | Res: 06magic(40), 12drown | Chunks: 07contam | XP: 302 | Sp: blink; teleport self. 09:19:16 <|amethyst> %??prince ribbit 09:19:25 blink frog (09F) | Spd: 14 (swim: 60%) | Int: insect | HD: 6 | HP: 21-45 | AC/EV: 0/16 | Dam: 2006(blink self) | amphibious, cold-blooded, !sil | Res: 06magic(40), 12drown | XP: 265 | Sp: blink. 09:19:25 <|amethyst> %??blink frog 09:19:36 -!- atrodo has joined ##crawl-dev 09:19:39 blink frog (09F) | Spd: 14 (swim: 60%) | Int: insect | HD: 6 | HP: 21-47 | AC/EV: 0/16 | Dam: 2006(blink self) | amphibious, cold-blooded, !sil | Res: 06magic(40), 12drown | XP: 265 | Sp: blink. 09:19:39 <|amethyst> %?blink frog 09:19:50 <|amethyst> updated in my repo 09:20:24 case SPELL_DEMONIC_HORDE: 09:20:24 return (SPELL_CALL_IMP); 09:20:25 heh 09:20:47 <|amethyst> is that in ghost spell determination? 09:20:57 yes 09:21:14 <|amethyst> monster call imp is more like demonic horde than player call imp anyway :) 09:21:39 oh god i forgot that monster evap is a thing 09:21:52 just going to not touch that at all 09:22:07 <|amethyst> %git 1d0f57cb 09:22:09 dshaligram * r1d0f57cbceb7: Merged stone_soup r15:451 into trunk. (5 years ago, 180 files, 39016+ 25829-) 09:22:17 <|amethyst> That is the commit that changed MI_EGGPLANT to MI_NONE 09:22:50 <|amethyst> description += "A purple vegetable. The presence of this object in the game indicates a bug (or some kind of cheating on your part). "; 09:25:11 <|amethyst> Eronarn: huh? monster evap gets changed to meph, no? 09:25:22 |amethyst: no, there is actually code for it 09:25:24 from paracelsus 09:25:31 nothing uses it right now 09:25:49 <|amethyst> axe it 09:25:51 look in mon-cast it's hilaribad 09:27:11 <|amethyst> FR: give fire cloud and miasma cloud to some monsters 09:27:26 hm, are throwing nets still supposed to be enchanted 09:27:29 please, not more miasma 09:27:40 <|amethyst> Pacra: they have minuses indicating damage 09:27:46 <|amethyst> Pacra: they can't be repaired, though 09:27:55 strange 09:27:56 |amethyst: here is a thought: a high level nec/tmut that targets and is mr-resisted 09:28:03 |amethyst: but not efficacy? 09:28:08 if it lands, they take horrible damage and start spewing miasma too 09:28:15 so it's like 'miasma cloud' but way more flavorful 09:28:18 most people think that once a throwing net is -3 or somesuch, it's useless 09:28:18 <|amethyst> Pacra: this is going to change---possibly just changing -3 to "quite frayed", or possibly eliminating damage altogether (increasing the chance of destruction to compensate) 09:28:32 I agree with that direction 100%. 09:28:37 I like that it frays. it's something unnique to nets 09:28:51 net golems 09:28:59 ...hey that would actually be pretty cool 09:29:18 net golems entangle you on hit? 09:29:21 like ocnstriction but more netty 09:29:23 and less damagey 09:29:30 PatashuXantheres: we just gave jumping spiders that ability and it seems to be fun 09:29:37 would be a good contrast for it to only otherwise show up on a slow monster 09:29:54 also, "You hack the net golem." 09:30:53 hah 09:31:18 almost on par with toenail golems in crazyness 09:34:09 ...ooh, i have an idea 09:34:31 what about not getting rid of fulsome, raising it in level, and having it do the same thing as evaporate (making throwable vials) 09:34:37 so you have a necro version and a fire alchemy version 09:40:09 <|amethyst> meh 09:40:17 <|amethyst> duplicate spells aren't really interesting 09:40:52 |amethyst: bringing back throwable miasma vials as a L6 spell or something sounds cool 09:41:06 definitely not the poison clouds though 09:41:32 <|amethyst> oh, you're removing miasma? 09:41:43 <|amethyst> so what do you get when you evaporate a necrophage corpse? 09:42:00 can't cast it on one, as of yet. i am still open to changing this of course 09:42:36 'vial of (steam, noxious gas, poison vapor)' is my current plan 09:44:24 <|amethyst> no more MR-ignoring polymorph? 09:45:02 sadly, no :P 09:45:17 higher level spells could make more interesting vials, but just being able to make a few sounds good for a L2 spell 09:45:36 no miasma please, overlaps too much with corpse rot 09:46:05 Xom says, "Here, take this!" Some vials of seething chaos appear at your feet. 09:46:37 <|amethyst> Eronarn: you should add disarming of gas traps to get vials :) 09:46:46 <|amethyst> (maybe not) 09:47:39 <|amethyst> so how does cloud size work? 09:47:48 <|amethyst> since you can't use spell power for that 09:48:04 <|amethyst> unless you make it an invisible enchantment on the vials 09:48:31 actually having to throw the vials sounds sort of annoying with the current throwing interface 09:48:35 i was thinking that there would be a few tiers, and they'd be visible, with decay moving it down tiers 09:48:43 so the strongest = high power and fresh 09:49:02 <|amethyst> then initial freshness should depend on spell power 09:49:03 a few tiers? we just removed missile enchantment for this reason 09:49:21 elliptic: missiles don't rot :P 09:49:25 having three different types of vials with three different tiers, all of which you have to quiver to throw, sounds... awful 09:49:38 <|amethyst> elliptic: I figured it'd be more like blood freshness 09:49:48 <|amethyst> which now that I think about it is also bad 09:50:09 perhaps we could just ignore power for it, give it a reasonable one 09:50:12 again, L2 spell 09:50:13 also, does throwing skill have anything to do with this 09:50:43 -!- elliptic has left ##crawl-dev 09:51:47 <|amethyst> I'd imagine not, since neither accuracy nor damage is an issue 09:52:00 there's something in exercise.cc for training from potions 09:52:05 i haven't looked into that too much though 09:52:22 -!- elliptic has joined ##crawl-dev 09:52:44 as far as i can tell throwing anything trains throwing though 09:53:01 <|amethyst> EX_WILL_THROW_POTION seems to be unused 09:54:52 <|amethyst> (putting together a commit to remove it now) 09:55:26 imo all vials of the same type should stack, and you should always throw the oldest one. no difference in effect from oldness. 09:55:31 <|amethyst> Eronarn: you could special-case that away in throw.cc; or not, leaving it at the 1/20 chance from junk-throwing 09:56:01 what about letting vials actually do some damage on a hit 09:56:06 <|amethyst> no 09:56:08 spellpower could give a chance to get two vials from one corpse 09:56:09 so you can aim it at their feet, or throw it directly at them 09:56:12 <|amethyst> needles don't, vials shouldn't either 09:56:24 |amethyst: needles don't explode when they hit, and aren't made of glass :P 09:56:35 <|amethyst> needles stick into you when they hit 09:56:42 (or lack of spellpower could give a chance to just waste the corpse, but i dont like that) 09:57:18 <|amethyst> The reason I bring up the spell power thing is because I wouldn't want evaporate to have a guaranteed 3x3 cloud when meph (1 level higher) does not 09:57:24 |amethyst: think of it like steam dragon breath 09:57:39 ball of compressed steam that deals damage if it hits you, then explodes into a cloud that hits you even if it 'missed' 09:57:54 confusion vials should at most have the effect of minimal power meph cloud 09:58:00 because throwing them doesnt cost mana 09:58:10 |amethyst: if we are comparing it to meph, giving poisonous cloud at L2 is ridiculous too 09:58:32 yes, but i think the ship sailed on giving evap any real nerfs forever ago 09:58:33 alefury: yeah, I was about to say something about that too 09:58:50 Eronarn: many of us want to give it the ultimate nerf 09:59:07 i gave up on that a long time ago :( 09:59:29 (i still think distill/evap is fucking stupid, for the record) 09:59:45 Eronarn: seriously though, being able to throw poisonous cloud for 0 MP at XL 2 is ridiculous 09:59:50 it won't happen 10:00:03 and also 0 failure rate 10:00:09 elliptic: vials decay faster than blood 10:00:19 miscast should definitely destroy the corpse 10:00:24 evap can at least fail and you can run out of MP, etc (this actually happens early on) 10:00:24 <|amethyst> still, MP outside of combat is essentially free 10:00:31 i'm thinking that they won't even last for a full dlvl 10:01:24 definitely do not want people stockpiling up large numbers of vials, should be something they cast reasonable near to when they want to fight 10:01:38 it will just be something people cast on every corpse they see 10:01:43 unless they are hungry 10:01:57 <|amethyst> Eronarn: but kobold room means the rest of the level is essentially free to kill 10:02:03 |amethyst: as opposed to now where 10:02:11 Eronarn: i think its fine if the clouds are not at all guaranteed, even on the targeted square (maybe give 4 clouds in open areas, randomly distributed across the 3x3 thing, in corridors guarantee one cloud, but if the rest spawns in a wall thats fine) 10:02:16 <|amethyst> where you have to spend 2 MP to throw a vial 10:02:17 also they should last at most 3 turns or so 10:02:29 <|amethyst> meaning you can't spam 10 vials at a time 10:02:59 Eronarn: I also don't really like encouraging people to kite kobolds or whatever across the level before killing them 10:03:02 |amethyst: when do you wish you could cast evap 10 times in a row? 10:03:20 <|amethyst> Eronarn: bee rooms :) 10:03:40 <|amethyst> especially when they get out before I get to the room 10:05:25 anyways, the vial thing can be improved on, but it will still at least be better than potions which are terrible for a variety of reasons 10:05:36 and since evap isn't going to be removed... 10:05:39 -!- ussdefiant has joined ##crawl-dev 10:05:52 do you want me to go remove evap :P 10:06:05 elliptic: i don't even think it should be removed; it's a very flavorful spell 10:06:15 it just shouldn't be as terrible to use 10:06:54 what you are describing sounds more annoying to use in some ways 10:07:15 (e.g. needing to throw stuff) 10:07:45 (and needing to worry more about whether you still have vials of the right kind) 10:08:16 well, i'm going to code it anyways, so we can see how it works :) 10:08:23 hopefully it will be good enough 10:08:33 the main inconvenience with using evap is that you get a lot of potions and they weigh a ton 10:12:16 Unstable branch on crawl.akrasiac.org updated to: 0.11-a0-1828-g6b48cd2 (33) 10:14:17 03|amethyst * rdc8edf9d9f0c 10/crawl-ref/source/ (exercise.cc exercise.h): Remove the unused EX_WILL_THROW_POTION. 10:22:35 elliptic: having an extra type of "potion" that rots away just for evap is at least better than making all distilled potions rot, which is another evap proposal 10:23:02 alefury: yeah 10:23:13 generally distill is really stupid, and evap is just annoying 10:24:12 i think the vial proposal would benefit a lot from a limited number of vials 10:24:27 but that just doesnt fit in with crawl imo 10:24:55 maybe if it was flavored as a buff there could be a limit of one 10:25:26 like delayed fireball, except you get to chose the type of ball by expending a corpse 10:25:35 yeah, that sounds sort of nice to me 10:26:11 certainly I'd prefer having it be thrown using the 'a' menu than using throwing if it isn't going to use throwing skill 10:26:24 i just dont really see it. having to use a corpse implies using that corpse for *something*. and "extracting the innate poisonous energy" of kobolds is really shitty flavor. 10:26:50 you enter the dungeon with only one vial :P 10:27:01 and then miraculously it regenerates after you throw it! 10:27:01 vampires can bottle blood as much as they want, though 10:27:14 also the throwing thing, yes... 10:27:31 I agree it is hard to justify, unfortunately... from a gameplay point of view I do like it though 10:28:49 <|amethyst> so if you evaporate a lorc corpse do you get a potion of lava? 10:28:57 <|amethyst> s/potion/vial/ 10:29:06 probably meph vial 10:29:19 depends on whether lorc meat is clean or not 10:29:29 :P 10:30:17 <|amethyst> I like how corpses are, other than fountains, the only source of drinkable water in the dungeon 10:30:17 alefury: my best attempt at coming up with a justification is that you need to keep the spell active to preserve the vial from deteriorating 10:30:41 that doesnt sound so horrible 10:30:46 and it is only strong enough to preserve one at a time 10:30:50 still not great though 10:30:53 one other problem: it would entirely change the way the spell works 10:31:19 <|amethyst> I thought that was the point 10:31:25 currently the whole point of evap is that it costs more than just mp, and you can stockpile that "more", then spam the spell in times of need. 10:31:57 <|amethyst> hm 10:32:02 having throwable vials changes the timescale on the stockpiling a bit, but otherwise leaves that intact 10:32:40 also, in practice confusion corpses are common enough to not have to care about the limitation 10:32:51 its relevant for other types, though 10:33:21 < alefury> like delayed fireball, except you get to chose the type of ball by expending a corpse 10:33:23 alefury: I agree it would change the spell to be only able to use it once and then need to find another corpse 10:33:30 breathe bees was going to achieve this by making the spell trigger breath timer 10:33:34 I don't have a problem with this though 10:33:44 -!- Xiberia has quit [Ping timeout: 252 seconds] 10:33:51