00:03:02 -!- [1]GrimmSweeper has joined ##crawl-dev 00:03:02 -!- GrimmSweeper has quit [Disconnected by services] 00:03:03 -!- [1]GrimmSweeper is now known as GrimmSweeper 00:03:46 Unstable branch on crawl.develz.org updated to: 0.11-a0-359-g3882fbc (32) 00:21:26 Windows builds of master branch on crawl.develz.org updated to: 0.11-a0-359-g3882fbc 01:28:07 eino_curves_02 and 04 are surprisingly deadly vaults it seems.. it'd probably be too much hassle to make it possible to turn off reporting furniture/feature vaults. 01:29:04 at least "furniture/feature" tag would be a lot of work. But it might be feasible to check if the vault has non-firewood MONS lines etc. 01:37:14 03Keskitalo * rb0117f371bcf 10/crawl-ref/source/mon-info.cc: Correct spacing. 01:38:34 Hehehe, KILLED_BY_WINNING 01:51:34 Meh, maybe later. 02:34:52 -!- ais523 has quit [Ping timeout: 276 seconds] 02:48:45 -!- jeanjacques has joined ##crawl-dev 02:55:54 nfogravity: I can certainly help with skills.cc. I rewrote a good part of it. I also made a few fixes in fsim, but nothing significant. 02:56:01 -!- heteroy has quit [Quit: ChatZilla 0.9.88 [Firefox 3.6.27/20120216013254]] 02:59:16 oh thank you 02:59:26 i'm going to completely rewrite fsim 03:00:09 i fixed the error that was making it not display melee damage correctly, but it could be a lot more useful 03:01:59 i'm going to have 'f' work while targeting a monster in vision and have crawl output the information directly 03:02:18 because right now it makes a lot of silly arbitrary assumptions 03:03:19 and if you want to see the effect of, say, different levels of Fighting at a fixed level of other data 03:03:38 you just wizmode your character to that data, then ctrl-f the monster, then hit fighting 03:03:59 and it outputs the data for all 27 levels of skill to an output file, like what it does now 03:07:27 and i realized i didn't use your name in this ramble so here it is galehar 03:10:56 anyway sleeptime 03:11:02 -!- nfogravity has quit [Quit: Leaving] 03:13:25 -!- [1]GrimmSweeper has joined ##crawl-dev 03:13:25 -!- GrimmSweeper has quit [Disconnected by services] 03:13:26 -!- [1]GrimmSweeper is now known as GrimmSweeper 03:15:18 !tell nfogravity I have plans for fsim too. We should try to discuss it some other time. 03:15:18 galehar: OK, I'll let nfogravity know. 03:26:47 about the db: currently, descriptions need to be written in a single line if they want to be wrapped. This makes them very hard to read. 03:27:18 what about requiring explicit newlines instead? 03:27:37 Orphaned games in WebTiles (https://crawl.develz.org/mantis/view.php?id=5393) by Eldray 03:35:11 kilobyte: that's what we have in the docs and it's a pain to edit 03:35:31 can't you enable autowrap in your editor? 03:35:45 Couldn't it just remove all newlines and then wrap? 03:36:48 ghallberg: as I said, manually wrapped text are a PITA to edit 03:37:02 galehar: after autowrapping, you have to remove newlines before saving 03:37:09 which is exactly what I want to avoid 03:37:37 galehar: My point was that you could add whatever newlines you wanted while writing and then the program would strip them out and wrap as needed. 03:37:46 galehar: not if it's done HTML-like, by newlines acting like spaces 03:37:58 don't you have "soft" autowrap? Text is wrapped on screen, but no newline is inserted. 03:38:01 I think vi can do this 03:39:06 for example, art-data.txt does this the way I and ghallberg are talking about 03:39:47 ie, newlines go away, and one description uses explicit newlines (marked as "\n") 03:40:37 I'm evicting descs out of art-data.txt to the database, so it can be translated, so it'd get the wrapping problem 03:42:03 art-data is manually wrapped like docs, and it's a pain to edit also. The only difference is you do it incorrectly, it will be fixed. 03:45:28 what do you mean by "do it incorrectly"? 03:46:02 if you don't respect the 80 col wrapping 03:46:45 s/is you/if you/ 03:47:21 yeah, for item descriptions you HAVE to do that incorrectly (by that definition) 03:47:43 while art-data accepts both wrapped and non-wrapped input, and shows it nicely 03:50:27 editing manually wrapped text is really terrible. Add a single word and you have to manually rewrap the whole thing. I don't see the point of it. Let the editor do the work ("softly") and each person can have the text width of his preference. 03:51:26 galehar: Our proposal would only allow you to manually wrap whiel editing and letting the parser strip and wrap like you suggested. 03:51:32 Manual wrapping is the devil. 03:53:15 I see. So it only works if you use an editor that does auto wrap by actually inserting newlines and everyone use the same text width. I find it... inelegant 03:53:53 I... no... 03:54:00 Uhm. I dunno. 03:54:10 I don't know of an editor which wraps text behind your back... and I'd consider it broken as it'd hide the text's structure from you 03:54:36 :set wm=0 03:54:49 text is visually wrapped, but no newline inserted 03:55:17 -!- ais523 has joined ##crawl-dev 03:55:21 you can display line numbers if you need to distinguish wrapped line from actual new lines 03:55:47 heh, even the official docs for vi say it's "generally a bad idea" 03:56:39 whatever 03:56:52 * galehar grumbles something about old bearded hackers 03:57:37 It's a nice idea :) 03:57:47 is 34 that old? And trimming the beard once in three months instead of shaving daily saves a lot of time and makes me look hot :p 03:57:49 Depending on what you're writing of course. 03:58:05 I'm on ly 25... 03:58:13 And I obviously can't write >.< 03:58:14 kilobyte: I'll be 34 in 5 months ;) 03:58:19 Oldies. 03:58:42 ghallberg: you're slightly older than me, then 03:58:49 ais523: :) 03:58:58 ais523: how much? 03:59:16 Btw I hate UPS. 03:59:18 I'll be 25 in about a month 03:59:24 so it rather depends on when your birthday is 03:59:35 Ah, I turned 25 four days ago :D 04:00:05 How can they think it's reasonable to have dwelivery "sometime between 9.00 and 19.00" 04:01:46 I... haven't even turned 24 yet :( 04:01:52 ghallberg: they think hard how to make you miserable no matter if they deliver to your home or work 04:02:14 kilobyte: Luckily Iäm still in school so I can sit at home and study for midterms... 04:02:20 But dammit, I want that computer. 04:02:21 Does anyone know what the state of audio output is on tiles? And... potentially even console? 04:02:45 Because I will definitely be trying to compose stuff for it; it would have to be a separate pack, obviously, to reduce size, but it would be nice to have background music for each branches. 04:02:46 due: basically, nearly utterly unmaintained. 04:02:54 Hmm. 04:03:12 you can't do audio over plain telnet/ssh, which is probably the more common way to play console 04:03:12 due: I fixed a security bug some time ago, it turned out it broke audio for a couple of releases and no one complained. 04:03:38 -!- [1]GrimmSweeper has joined ##crawl-dev 04:03:38 -!- GrimmSweeper has quit [Disconnected by services] 04:03:39 -!- [1]GrimmSweeper is now known as GrimmSweeper 04:03:50 I looked into it, looks like sdl_sound is straightforward to use and would do just The Right Thing 04:04:22 Okay 04:04:25 ais523: Yes, I know 04:04:30 there's one issue: it'd be great to get stereo and/or surround sound, except, it'd leak information out of LOS 04:04:48 kilobyte: limit it to LOS only? 04:04:51 due: Did you happen to read the TIGsource article/review? Paul Eres wanted music in the game. (I'm in favour, I like music in games. You can always turn it off. If the filesize gets too big, the download can be made optional.) 04:04:54 Well, branch-specific musioc (with differening themes for depths of the dungeon maybe). 04:04:57 LOS in Crawl is abstract, no reason LOH shouldn't be too 04:04:59 Keskitalo: Yes! 04:05:03 like in DoomRL where you know where enemies come from -- both angle and distance 04:05:07 Keskitalo: That's what I meant by having a separate pack 04:05:12 Ah, you already mentioned separate packing. 04:05:23 there is music for NetHack, but it doesn't really vary by branch 04:05:24 Keskitalo: That's what reminded me about getting music into crawl. 04:05:29 except for the endgame 04:05:31 Unfortunately I am a mediocre musician and an even worse composer sooo. 04:05:32 (it's used in Vulture) 04:05:37 due: What do you use for composing? I use trackers, Renoise most recently. 04:05:47 Keskitalo: A ... piano :) 04:05:49 due: I can give feedback. 04:05:50 Cool! 04:05:50 I use a physical piano for composing 04:05:55 and Rosegarden to enter it into a computer 04:05:59 Plugged into pianoteq <4 04:06:04 ais523: Do you quantize? 04:06:26 due: when I'm entering music into a computer, I do it turn-based not real-time 04:06:30 so quantizing would be a no-op 04:06:43 (Procedural music would be nice though!) 04:06:44 oooh 04:06:50 turn-based in rosegarden? 04:06:54 due: I can't try to get sdl_sound working right now, I can try it in the evening if you wish 04:07:00 (LÃLinl is, incidentally, very good with procedural music.) 04:07:09 due: it's pretty much a case of choosing note length, then pressing relevant key 04:07:12 Linley I mean. 04:07:16 kilobyte: Oh, god, no rush 04:07:20 ais523: Rosegarden does this? 04:07:26 would work great everywhere but online console 04:07:27 yes 04:07:30 kilobyte: Like I said, very very mediocre composer here :) 04:07:31 most music editors do 04:07:37 ais523: I need to get more used to rosegarden. 04:07:42 ais523: Do you use pianoteq? 04:07:45 it's in the notation editor 04:07:49 and no, just vanilla Rosegarden 04:08:00 Pianoteq is so amazing. The dynamic power <3. 04:08:25 due: actually, I guess sound effects would be better to look into first 04:09:00 There's no reason to hold back on music before sound effects. 04:09:39 they're pretty much completely independent, apart from (sometimes) code-wise 04:09:48 Well, what about style? 04:10:19 I really like the Catacomb Snatch soundtrack 04:10:34 hmm, there's some similarity between various roguelike musics, I think 04:10:48 Well, maybe you want to make sure that music doesn't overlap sound effects.. for instance, you may want to reserve some frequencies for them and not use them so much in music. (Or something like that anyway.) 04:10:52 Having a basic, musical theme with instruments, which then varies with electronic stuff :D 04:10:59 although Pokémon Mystery Dungeon breaks the mold there, it can be remarkably upbeat 04:11:06 Minor-key music <# 04:11:14 "is less than sharp"? 04:12:20 Listening to the Catacom Snatch soundtrack now (http://www.youtube.com/watch?v=OxI5iqNxSto) 04:12:24 let's not get into Microsoft's lack of Unicode, you want ♯ 04:12:35 <3, actually :D 04:12:44 There was that burrrnning hearrrrt againnn 04:12:58 Keskitalo: I really like it! But my musical taste, at least towards piano/composition, is quiet minor-key :) 04:13:17 due: I like music to be minor-key but not sad 04:13:32 background music is supposed to be background 04:13:35 although I've come up with the occasional joyful upbeat stuff instead 04:13:43 happy minor-key music can still be awesome :D 04:15:38 I like it to be, umm, "adventurous" 04:16:20 I don't know about style, but it (naturally) needs to be considered that any given piece may be played for long stretches. 04:16:45 So it should either be long or not so boring that repetition sucks? 04:17:07 (tangent) One idea that I've had was that you could have two tracks; theme for D:1, theme for D:27, and the cross-fade each according how deep you are in D :) 04:17:17 due: Yeah. 04:17:52 It would be ... interesting to create a melodic theme for each branch, D:1, D:27, and the Orb (of course). 04:18:05 I think Wesnoth has a playlist of things to play; there's actually different pieces, as one ends, another takes off. I'm not sure if it even has music according to what happens in the game, just a playlist that plays throughout. I may be wrong though. 04:18:07 Keskitalo: Vulture deals with it by having several music tracks for areas that are more than a single level long, and playing them at random 04:18:22 And then use that as the basis for a variety of different pieces. 04:18:22 Keskitalo: oh, I know loads about how music in Wesnoth works 04:18:45 each cutscene is typically assigned one piece of music (it's technically possible to do more IIRC but nobody ever does); each level is assigned a playlist 04:18:50 due: Yes, that would be great in my opinion. People could contribute tracks! 04:18:56 We could pimp it over at ocremix.org :) 04:18:59 which might be {DEFAULT_MUSIC_PLAYLIST}, but more commonly is tailored to the level, and one or more tracks long 04:19:15 ais523: Yeah, that sounds goodCool, thanks for the intel 04:19:40 and selection from the playlist is randomized, but designed so that repeats of the same track don't happen too close to each other 04:19:41 heh, was retyping that, accidentally pushed enter 04:19:49 Keskitalo: However, my melodic capacity -- non-existent! 04:19:55 ais523: That is a great idea 04:20:16 Vulture does something similar, but not as strongly thought through 04:20:32 oh, and scenarios can override the music for specific events, too 04:20:50 there's one level where there are two enemy armies fighting each other, and you have to sneak through the middle 04:21:01 and each of the armies has its own theme, that plays whenever it gets into LOS 04:21:09 and after a while, they're having a tug-of-war with the music 04:22:39 due: It might be cool, when entering the branch, to have the basic leitmotif play first, and then start playing the variations randomly 04:23:16 it's also worth mentioning that many (most) of the tracks in Wesnoth are very long, so that it doesn't get tedious if a level plays only one track in a loop 04:23:30 (apart from one track which is deliberately tedious, for use in levels which want to convey the feeling of tedium) 04:26:01 I wonder how well would the music be cached on webtiles 04:26:50 would be pretty well, I imagine; you'd be sending the music over HTTP, and you'd expect the browser to respect caching headers for that as it has no reason not to 04:26:57 it'd just be like caching an image 04:27:42 -!- alefury has joined ##crawl-dev 04:31:19 pivotal the Bringer of Life (L15 CeHe) ASSERT(smc) in 'mon-util.cc' at line 1052 failed on turn 40759. (Vault:1) 04:40:41 So, only send the music once, when you opt-in to it? 04:43:56 -!- monqy has quit [Quit: hello] 04:44:41 Could also do it OoT overworld-style, if you want to get ambitious 04:45:32 shackes (L14 HOFi) ASSERT(smc) in 'mon-util.cc' at line 1388 failed. (Lair:4) 04:47:28 That was done by having pre-composed segments which sort of flow into one another, generally around 15 seconds long (longer might be called for here); the game selects them according to what's going on in your surroundings. Tense music when you're being attacked, serene music for peaceful sunsets, and so forth. 04:49:04 That would indeed be awesome. 04:50:03 We could use tension. 04:50:08 dtsund: that actually gets on my nerves, especially in Majora's Mask 04:50:19 partly because the music for Termina Field is so much better than the battle music 04:51:10 ais523: Majora's Mask didn't do procedural generation, though, not really. 04:51:26 indeed 04:51:29 It just had two tracks, and faded from one to the other relatively ungracefully 04:51:51 I'm not really a fan of procedurally generated music in games 04:52:04 And yeah, as much as I love Majora's Mask, its battle music was not a strong point. 04:52:12 Portal 2 pretty much has the state of the art in terms of technology in that, and the result is that the music's a jumbled mess that's hard to concentrate on 04:52:33 I think you're not meant to concentrate on Portal 2's music. It's more for ambience. 04:52:48 Does the Twilight Princess overworld strike a better balance for you? 04:53:53 haven't heard it 04:54:13 It's more or less the two-tracks approach of Majora's Mask, but the battle music is better, and the game waits for a good transition point to switch gracefully. 05:00:06 OSX builds of master branch on crawl.develz.org updated to: 0.11-a0-360-gb0117f3 05:01:37 hmm, I don't like music that you can't concentrate on even if you want to 05:01:44 it's good for it to fade to the background most of the time 05:01:48 but I like being able to listen to it 05:05:01 -!- cosh has joined ##crawl-dev 05:05:20 -!- dtsund has quit [Quit: dtsund] 05:24:07 -!- [1]GrimmSweeper has joined ##crawl-dev 05:24:07 -!- GrimmSweeper has quit [Disconnected by services] 05:24:08 -!- [1]GrimmSweeper is now known as GrimmSweeper 05:52:28 -!- ainsophyao has joined ##crawl-dev 05:53:44 Tutankham (L14 DEAE) ERROR in 'tileweb.cc' at line 204: Socket write error: Bad file descriptor (Lair:1) 05:53:47 -!- ais523 has quit [Remote host closed the connection] 05:56:07 -!- ainsophyao has quit [Remote host closed the connection] 06:04:38 -!- BlastHardcheese has quit [Ping timeout: 240 seconds] 06:08:09 -!- BlastHardcheese has joined ##crawl-dev 06:47:54 -!- GrimmSweeper has quit [Ping timeout: 272 seconds] 06:56:03 Several bugs in the fight simulator. (https://crawl.develz.org/mantis/view.php?id=5394) by CommanderC 06:57:59 ttants: fight simulator, flight simulator 07:00:38 -!- BlastHardcheese has quit [Ping timeout: 240 seconds] 07:01:47 -!- BlastHardcheese has joined ##crawl-dev 07:15:07 -!- syllogism has joined ##crawl-dev 07:22:30 03galehar * r6e5f30cf97de 10/crawl-ref/source/shopping.cc: Clear last_pickup when entering a shop. 07:23:04 galehar: re tavern abyss stuff: i know you didnt mean burst speed. i also didnt mean burst speed. the overall speed change makes the abyss feel very different, and i think having low and high difficulty versions of both would be more interesting than tying difficulty to speed. 07:26:34 s/both/high and low speed abyss/ 07:27:29 ??badforum 07:27:31 badforum[1/30]: Make skill levels in magic skills (spellcasting, fire magic, hexes...) give a penalty to HP. 07:27:37 sorry, wrong channel 07:28:13 The thing is I don't think scaling difficulty with banishment depth or XL would work. I don't see any other way to scale it than with time spent. 07:28:37 and time spent increase speed, so it's simpler to just tie the 2 07:29:00 mhh. what about one of dpeg's proposals: toughest monster seen 07:29:14 it doesnt make sense thematically, but it might work quite well mechanically 07:29:44 one effect would be that it encourages getting the abyssal rune early 07:29:50 i dont know if thats good or bad 07:30:26 not early, but it discourages postponing it 07:30:27 OOD can be very random. 07:30:33 yeah, so? 07:31:04 if you get shafted and run into a unique or an OOD that then banish you, you end up in a very hard abyss 07:31:11 yes 07:31:17 :) 07:31:31 or if you're not banished right away, you must be extra careful until you catch up 07:31:40 sometimes its hard, sometimes not so hard. also abyss should still be fairly random, so would get tougher over time as you meet stronger monsters 07:31:49 it's a very weird and spoilery mechanism 07:31:56 spoilery is a problem, yes 07:33:09 some combination of several things might work 07:33:15 -!- GrimmSweeper has joined ##crawl-dev 07:33:26 increase over time, starting difficulty determined by something else 07:34:00 -!- casmith_789 is now known as casmith789 07:34:05 imo thats better than adjusting scaling difficulty, because then early abyss would always be boring 07:34:18 but what? That's the question? What could determine starting difficulty that wouldn't lead to degenerate behaviour? 07:36:43 randomly use one of max exp reached, toughest monster seen, lowest absdepth reached :) 07:38:15 if we really want to scale the difficulty with player's level, we can use the new you.total_experience I added for skill cost 07:39:32 -!- ainsophyao has joined ##crawl-dev 07:39:52 but use it to make late banishment harder, not early one easier! :) 07:40:23 imo have it be based on how annoyed lugonu is at you; increment this by 1 per level you gain, 1 per rune, 1 per x time spent in the abyss 07:40:36 -!- MarvinPA_ has joined ##crawl-dev 07:41:17 discourage early rune. Can drain yourself for easier abyssal rune hunt. 07:41:30 galehar: per level you gain, not per level you are 07:41:58 Draining yourself sounds really spoilery. 07:42:27 fast leveling species get harder abyss. I'd rather use total_experience 07:42:58 sure, same idea though 07:43:12 the point of having it be lugonu flavored is that you can have messages relaying abyss difficulty 07:43:22 and have them actually make sense thematically 07:44:03 dpeg suggested that lugonu altar's seen but ignored could also increase difficulty (you had an opportunity to escape that you dismissed) 07:44:34 -!- MarvinPA has quit [Ping timeout: 260 seconds] 07:44:34 -!- MarvinPA_ is now known as MarvinPA 07:45:04 galehar: that's a good idea; the only problem might be someone not noticing an altar just barely in los 07:50:32 One idea I had for abyss is that it should be filled with things of the typoe you have seen 07:50:55 or possibly a little harder 07:51:03 but this would also mean no rune unless ylou already had one 07:51:28 The abyss basically being built from the horrors of your own imagination ... 07:51:46 (as you look into the abyss, it looks into you ... that kinda stuff) 07:53:53 that would be extremely boring 07:55:32 -!- ncampion has quit [Quit: Linkinus - http://linkinus.com] 08:11:06 melee_attack.cc:1580, shouldn't we use div_rand_round instead? That would slighlty increase the effect of stats on melee, which can't be a bad thing. 08:12:08 -!- ainsophyao has quit [Remote host closed the connection] 08:14:10 -!- st_ has quit [Ping timeout: 265 seconds] 08:18:52 -!- dtsund has joined ##crawl-dev 08:20:20 also, interesting how stat values of 9 and 11 are special-cases to be exactly equal to 10 08:21:08 <|amethyst> in two different ways 08:22:31 same at 1619. We should use skill_rdiv somehow 08:27:02 obviously, weapon weighting itself sudden steps as well 08:27:59 although, with stats having 1+(x-C)/78 effect, it's negligible 08:42:46 -!- ainsophyao has joined ##crawl-dev 08:44:06 -!- edlothiol has quit [Ping timeout: 248 seconds] 08:49:39 -!- cosh has quit [] 08:51:07 yeah, we should probably start by fixing the formula before looking at its implementation :) 08:52:53 BTW, str's effect on carrying capacity is strange 08:53:28 crippling when <8, no real effect above 08:54:53 a, say, HaEE, is encumbered with a single dagger, robe/shoes/buckler, two bread rations, 5 wands, a few scrolls and a couple potions 09:00:10 now, is the intended effect of a possible change to educate players (increase effect above, motivation to not pick up and carry all the stuff, possibly leads to some players just doing more stash work and being frustrated) or to allow the low-str chars to carry more things (they have low str for a reason, how much balancing is behind this?) 09:07:30 -!- Ragdoll has joined ##crawl-dev 09:22:25 i seem to be unable to haunt things behind glass 09:22:35 how peculiar 09:25:11 Ragdoll: I think it's on purpose. We're trying to avoid any form of attack through glass. 09:28:35 so instead of turning it into a tactical (dis)advatange it will now be used merely for gimmickvaults 09:28:50 galehar: yes, Ragdoll just can't cope with it and brings it up at every opportunity, you can safely ignore him :P 09:29:24 re glass: fucking trees, they are stupid, and how they are handled is even worse 09:29:45 but i think i have said everything i wanted to on this topic at some point 09:29:50 i rarely bring this up, and if i do it´s only because i encounter new exciting situations regarding glass 09:29:55 "rarely" 09:39:25 I suggested that glass block all attacks, and tree/grates none. Then edit vaults to use the proper one to avoid abuses. 09:39:48 seemed to reach a consensus, but everyone is too busy with the tournament to implement it yet :) 09:39:49 what about pillars? 09:39:57 what about trees? 09:39:58 galehar: tree blocking none doesn't work 09:40:00 what about tree pillars? 09:40:08 yeah, trees blocking none doesnt work at all 09:40:18 there are far too many times that would be abusable with current tree use 09:40:18 its a tree 09:40:26 right, grates block none, we haven't actually discussed trees ;) 09:40:47 IMO trees should behave like plants 09:40:57 i like the stuff trees to to los 09:41:00 and be actual monsters 09:41:05 it looks very cool 09:41:12 very foresty 09:41:22 elliptic: interesting idea 09:41:34 Durkon would agree with it 09:41:38 or like bushes maybe 09:41:58 I don't really understand what the idea is with the plant/bush/tree triad though 09:42:11 currently they behave in completely different ways from each other 09:42:31 probably they have been implemented by different persons at different time 09:42:36 it is very non-obvious that you can shoot through bushes, for instance 09:43:08 uniformising sounds like a good idea 09:43:28 but shooting through plants should stay a fedhas exclusive, right? 09:44:10 galehar: yes, IMO only fedhas worshippers should be able to shoot through bushes too 09:44:58 that would probably make more sense than how they work currently, yeah 09:45:01 what about monsters? this would kill that stupid centaur vault. 09:45:11 it could be a centaur of fedhas! 09:45:19 then it could walk through the bushes 09:45:41 we can also kill the vault if needed 09:45:57 or use grates instead 09:48:23 fr glass trees 09:48:25 -!- dtsund has quit [Quit: dtsund] 09:49:39 saga frontier has crystal trees, they are not to be messed with 10:11:23 -!- nfogravity has joined ##crawl-dev 10:14:28 i like the idea of being able to fire through P, but not bushes specifically 10:15:26 what about adding an 'obscured' status to firing paths 10:17:33 like, 0-6: '(not), lightly, somewhat, moderately, heavily, almost completely, (los blocked)'. plants are +1, bushes are +2, trees are +3 10:17:46 (well, maybe 0-7 to preserve current tree behavior) 10:18:44 fire through a plant/whatever and you have an obscure/max_obscure chance that it hits the appropriate intervening thing 10:19:03 (firing directly at a plant still hits it) 10:20:00 alefury: actually, there was a vault with crystal trees already, but was removed 10:20:30 -!- dtsund has joined ##crawl-dev 10:20:39 so if you're: @..PP7o for plant/bush/tree you are 'almost completely obstructed' firing at the orc, which means you have a 1/7 chance to hit the plant, 2/7 chance to hit the bush, 3/7 chance to hit the, 1/7 chance unimpeded 10:21:02 mostly because they were vulnerable to fire, couldn't shatter/LRD, etc 10:21:23 crystal trees sound pretty cool. i meant they are not to be messed with in saga frontier :) 10:21:33 iirc they had oscillation and magnetic storm 10:21:41 not sure though :P 10:21:45 er, what they do? 10:21:49 trees are, uhm, trees 10:21:54 just regular monsters in saga frontier :) 10:21:58 its a weird game 10:22:02 crystal trees would be cool if trees became monsters 10:22:14 very shatter vulnerable, not at all vulnerable to fire 10:22:35 for the trees to be attacking, you need to apply an axe and fool the victim into shouting "Timber!" 10:22:45 then they'll be hit by the falling tree 10:23:04 that sounds like a nethack thing 10:23:04 nfogravity: You have 1 message. Use !messages to read it. 10:23:10 !messages 10:23:11 (1/1) galehar said (7h 7m 52s ago): I have plans for fsim too. We should try to discuss it some other time. 10:23:20 the real question would be what happens when players can attack trees in melee... i don't think it would be a big deal, personally 10:23:37 Eronarn: very tedious digging 10:24:09 kilobyte: yes... i think it'd be good to make them only about as strong as old bushes 10:24:24 don't want it to be like sawing through an eldritch tentacle 10:24:26 just give them infinite damage reduction except by stuff that makes them burn imo 10:24:59 -!- Zao_ has joined ##crawl-dev 10:25:07 choppable trees would break so many vaults 10:25:15 about monster trees: one of my branches has this: http://sprunge.us/PfUW 10:25:16 and it would just be annoying 10:25:25 -!- Zao_ is now known as Zao 10:25:43 -!- Zao has quit [Changing host] 10:25:43 -!- Zao has joined ##crawl-dev 10:26:05 i don't know about that... are there really any vaults that use trees in a way that matters like that? 10:26:14 fireball, fire, lightning will all destroy trees 10:26:36 Eronarn: all the foresty vaults 10:26:36 well there's lol_no_vehumet_altar_4_u_overflow 10:26:44 the "animated tree" is only a dummy for the player form, they'd be useless as monsters. Good only in specially designed vaults, and we already have those with "warrior statues". 10:26:47 they just use them as flavorful walls 10:27:55 alefury: if we're already letting the walls be destroyed by anyone with one of several items, it doesn't seem so bad to let them be destroyed by anyone with a powerful enough melee attack 10:29:07 Eronarn: wands of digging, LRD, ... 10:29:08 alefury: when you lick them they taste good? 10:29:16 rwbarton: Is it actually called that? 10:29:35 I would assume not :) 10:29:39 kilobyte: i believe that fire/fireball/lightning are all more common than digging 10:29:58 Eronarn: Did they change the wand generation probabilities, then? 10:30:20 spell level 4 vs 6 10:30:24 i might be wrong on that, of course 10:30:47 There's basically three tiers of wand probability, IIRC, and two of them are very small... 10:31:00 kilobyte: well, the spell is a higher level for a very different reason :P 10:31:08 also lightning bolt works and that's 5 10:31:47 -!- Zao has left ##crawl-dev 10:32:03 kilobyte: you should make the jelly attack be 'Pseudopod' 10:32:06 rather than 'Body' 10:44:57 03galehar * r2fa9559feeb8 10/crawl-ref/source/wiz-fsim.cc: Fix ranged combat fight simulator not displaying skill level properly. 10:44:57 03galehar * r0fc7e7326f9d 10/crawl-ref/source/melee_attack.cc: Replace grid_distance() == 1 by adjacent() in a few places. 10:44:57 03galehar * rc231866995d5 10/crawl-ref/source/wiz-fsim.cc: No stabbing during fsim. 10:44:58 03galehar * r6b03a9fd0f10 10/crawl-ref/source/wiz-fsim.cc: Move the monster next to the player during fsim. 10:44:58 03galehar * r7770cf633833 10/crawl-ref/source/ (fight.cc fight.h wiz-fsim.cc): Fix the accuracy calculation of fsim melee. 10:47:22 -!- ainsophyao has quit [Remote host closed the connection] 10:50:44 amplicon (L6 MiBe) ERROR in 'tileweb.cc' at line 204: Socket write error: Bad file descriptor (D:5) 10:50:53 Bushes were implemented for shoals and not really used as planned (maybe it didn't work out); I guess not much would be lost if they were removed. 10:52:39 -!- dtsund has quit [Quit: dtsund] 11:02:14 03galehar * r1d2dbd5c7c05 10/crawl-ref/source/attack.cc: Show damage numbers in wizard mode. 11:02:14 -!- blabber has joined ##crawl-dev 11:03:47 -!- alefury has quit [Quit: ChatZilla 0.9.85 [Iceape 2.0.11/20111108220737]] 11:07:25 ugh damnit galehar stop changing useful shit while i'm hacking on it 11:08:22 nfogravity: I'm done with fsim for now, go ahead and hack. 11:08:40 oh you are here cool. wanna talk about fsim? nice changes btw 11:08:59 don't really have much time, have to go. 11:09:03 okay 11:09:25 'later 11:10:31 Unstable branch on crawl.akrasiac.org updated to: 0.11-a0-361-g6e5f30c (32) 11:19:56 -!- ToBeFree has joined ##crawl-dev 11:35:01 Eronarn: the whole jelly form idea appears to suck orange halfling balls, I uncovered enough problems during first stages of implementation that it's going out 11:35:22 -!- alefury has joined ##crawl-dev 11:35:37 even if I'll actually resume the bad forms project; I'd go more for hellspider, portal_branches and translations 11:41:37 portal_branches? 11:43:56 ripping away a good part of portal vault code and rewriting it sanely 11:44:03 -!- ortoslon has joined ##crawl-dev 11:50:45 Overloaded Status maintained on Felid Resurrection (https://crawl.develz.org/mantis/view.php?id=5395) by XuaXua 11:51:43 -!- dtsund has joined ##crawl-dev 11:54:40 speaking of portal vaults, is there a way to temporarily create an arena, move the player there, do some stuff, and then put him back where he was and destroy the arena? 11:55:55 i know there's code for entering portal vaults, but this is more for debugging purposes 12:04:03 nfogravity: do you mean Arena arena, or just an arena-shaped map 12:04:14 i don't know about the former... the latter, you'd have to write yourself, for sure 12:08:27 you could make a wizard-only portal vault 12:08:28 03kilobyte * r2a01b23ff772 10/crawl-ref/source/misc.cc: Reset burden on felid revival (in case of stat death). 12:08:38 03kilobyte * ree6c366fd479 10/crawl-ref/source/ (7 files): A wizmode command, &^P, to transform yourself. 12:10:28 -!- dtsund has quit [Quit: dtsund] 12:27:18 -!- MarvinPA_ has joined ##crawl-dev 12:30:09 -!- MarvinPA has quit [Ping timeout: 245 seconds] 12:30:09 -!- MarvinPA_ is now known as MarvinPA 12:40:33 well for now i'm putting it in wizlab, just to see if it works 12:41:05 < magistern> you get red glow from sticky flmae 12:41:14 i think i might have even added it, but it seems potentially bad on second thought 12:44:36 well, it makes you glow. and fire is red. contam is contam, glow is something different (now). 12:46:36 "an alligator (moving slowly, moving somewhat quickly)" (https://crawl.develz.org/mantis/view.php?id=5396) by ortoslon 12:52:45 -!- MarvinPA_ has joined ##crawl-dev 12:53:13 alefury: i know, i'm the one who submitted that patch :P just thinking that it might have been ill-advised to either a) have stayed with glow b) not adjusted the color 12:56:13 -!- MarvinPA has quit [Ping timeout: 260 seconds] 12:56:13 -!- MarvinPA_ is now known as MarvinPA 13:00:07 the Glow status light is basically never relevant 13:00:21 and usually it comes from some other status ailment which is listed 13:00:59 Fire or Contam 13:02:26 at first it was quite distressing considering what Glow used to mean, now it's just unnecessary 13:05:41 rwbarton: well, it wasn't at all clear that sticky flame and having contamination made you easier to hit 13:06:01 that's definitely worth noting 13:06:06 who cares 13:06:12 those are much bigger issues in their own right 13:06:21 that could be shown as EV decrease 13:06:36 it's not like to-hit chance is generally transparent 13:06:36 ortoslon: not with the way it works right now, unfortunately 13:06:49 that would make more sense, though 13:06:58 it'd mean making it an EV penalty rather than an accuracy penalty 13:07:05 (also making it non-randomized) 13:07:56 rwbarton: those are bigger issues, sure, but it matters more than a lot of things that are shown 13:08:09 In the status area? 13:14:08 -!- blabber has quit [Quit: reboot] 13:16:55 -!- blabber has joined ##crawl-dev 13:27:24 it hints players at what stuff nullifies invis 13:27:56 it's relevant when you're invisible, yeah 13:28:07 then Invis turns grey already 13:28:30 really I would prefer only to have stuff that actually matters in the status area... 13:28:36 this stuff about status lights reminds me 13:28:52 can we get rid of the 'You are standing in water' messages, and have a Water status light, at least as an option 13:29:26 imo stuff that is in your tile really shouldn't be printed to a log; the only reason it is is that it's important that you are currently in water 13:30:05 stuff like altars, it makes sense to just print, but water comes up very often so a status light makes sense 13:31:06 i simply muted those messages, but i play tiles, so i know whats on my tile 13:31:16 because i can see it 13:31:30 * alefury points and laughs at console players 13:32:22 i can also see polearms btw! (something should definitely be done about that for console players) 13:32:48 -!- ToBeFree has quit [Remote host closed the connection] 13:33:08 also, bows etc 13:33:49 i wonder if you can colourze the borders around an ascii leter 13:34:13 basically, you cant do shit 13:34:23 apparently you cant even easily set the background color of a glyph 13:34:38 but what about the wandering/allied/sleeping backgrounds 13:34:49 Ragdoll: there are two colors allowed per glyph: foreground and background 13:34:51 note the "easily" 13:35:07 glyphs are almost exclusively 1 column wide 13:35:22 you can add combining characters to glyphs to augment what's there, but that is poorly supported 13:35:44 nfogravity implemented a "display health" toggle using glyph backgrounds recently, he would know more about this i guess 13:35:52 but iirc it was a major hassle 13:35:55 you can use multiple fonts, apparently, but it's REALLY poorly supported 13:36:03 it hasn't been committed yet 13:36:05 :( 13:36:16 what about some sort of blinking/colour swapping, like simulacrum 13:36:17 and you can use a partial font with fallback, for custom glyphs, but that requires getting users to download a special font 13:36:43 Ragdoll: the problem with that is that we're so tight on colors that almost anything is already something 13:36:44 but yeah it is a major hassle, the glyph code is extraordinarily minimal 13:37:05 also, it isn't proper blinking. it has a chance to change each time the screen is drawn 13:37:17 which means it won't show up unless you press keys 13:37:22 right now my code piggybacks off COLFLAG_ITEM_HEAP actually 13:37:22 maybe just a simple toggle button, like for example that button that shows terrain only 13:37:26 maybe make a button that shows ´weapons´ only 13:38:01 Ragdoll: if you want to know exactly how minimal it is 13:39:19 -!- ToBeFree has joined ##crawl-dev 13:39:30 sure 13:41:14 there's exactly one byte allocated for colour. the first three bits give you 8 basic colours, and the next is a flag that toggles brightness. so that's 16 colours. 13:41:30 what about a button that, when you press it, draws: 1) whether you can move to a square 2) whether that's in melee range of anything 3) whether it's in range of anything 4) whether anything you can't see might be in los of that tile 13:41:34 the next four bits are all hardcoded to flags 13:41:46 so for a sleeping monster, i believe the flag is 1000 13:42:03 -!- ToBeFree has quit [Read error: Connection reset by peer] 13:42:09 er, 3) is in ranged range of anything 13:42:51 and when a monster is asleep, the colour output code in colour.cc just ORs the colour in the first four bits with a colour that's set in user options for sleeping monsters 13:43:17 the default colour is fi:cyan i think 13:44:33 so, as Eronarn explained to me when i was trying to get it working, there simply isn't any room in the colour byte for more information 13:44:55 heh, sorry i couldn't give a better explanation of why it was a problem 13:45:08 it's okay, i spent a while with it just last week 13:45:29 the way i worked around the problem was by changing the code that actually sets the colour of a monster glyph 13:46:09 nfogravity: while you're at it, you should make elemental colors work for the player glyph :) 13:46:14 so there's a toggle now, and if that toggle is on the code that returns monster glyph colour instead returns the colour of their hp pool 13:46:34 it'd be cool to have a robe that made your @ be discocolors 13:46:48 ahahah 13:46:57 real_colour() strips elementals 13:47:08 so you could just wrap any call in that and see if it works 13:47:08 is that all? just remove that? 13:47:15 no, add that 13:47:32 Talk-Attack-Target should not be blocked by Translucent Walls (https://crawl.develz.org/mantis/view.php?id=5397) by XuaXua 13:47:33 it converts an elemental colour to one that's not fucked up 13:47:45 the issue i was having last i tried was that any elemental color just got set to green 13:47:51 does xuaxua do any coding or does he just report issues 13:48:08 i think xuaxua does some maps or tiles or something? 13:48:43 quick question 13:48:48 is webtiles project looking for devsa 13:48:54 i'll happily try to get disco colours working, it sounds fun 13:49:26 with that said, https://crawl.develz.org/mantis/view.php?id=5376 13:49:44 3-2 was confirmed by pivotal/faze to work on linux 13:49:49 Dixbert: more work on webtiles would be great 13:49:55 for example, there's some browser specific bugs 13:50:08 some might be fixable on our end 13:50:47 so it might be nice if it could maybe be committed possibly? 13:51:18 nfogravity: your best bet if you want something committed is to find a dev you haven't annoyed yet and bother them directly 13:51:55 stuff on mantis can languish for a really really long time 13:52:00 evilmike's been watching this one, i guess i'll bug him the next time i see him 13:52:09 that was my plan but it's been a few days since he's been on 13:52:44 leave people messages on henzell 13:54:16 (via !tell) 13:54:50 !tell evilmike anything more you'd like me to change about the colour patch? if not, could you commit it? 13:54:51 nfogravity: OK, I'll let evilmike know. 13:55:01 thank you :D 14:01:58 Eronarn, do you know where the player glyph is set now? 14:04:59 webtiles looks pretty well-tasked 14:05:52 -!- dtsund has joined ##crawl-dev 14:07:53 Action Count layout messed up by longer texts (https://crawl.develz.org/mantis/view.php?id=5398) by XuaXua 14:08:46 no, sorry 14:10:08 nfogravity: xuaxua reports issues, did (at least) one tile (the constriction one), cleaned up the dev wiki, and talks a lot of bull on the forums 14:11:50 Eronarn, the answer is that MONS_PLAYER is a monster type 14:28:18 -!- syllogism has quit [Read error: Connection reset by peer] 14:28:25 -!- syllogism has joined ##crawl-dev 14:32:53 -!- ussdefiant has quit [Ping timeout: 248 seconds] 14:43:53 -!- ussdefiant has joined ##crawl-dev 14:44:07 -!- monqy has joined ##crawl-dev 14:46:18 -!- nrook has joined ##crawl-dev 14:46:36 -!- ortoslon has quit [Read error: Connection reset by peer] 14:48:03 Dixbert: webtiles can definitely use some coders 14:48:45 you'll have to talk to edlothiol. He coded it and for now, he's the only contributor. 14:48:58 mouse support would be great for example :) 14:49:42 nfogravity: about fsim. The current implementation scale the weapon skill level from 0 to 27 and keep fighting at half that level. 14:49:57 yeah i don't like that 14:50:02 it's half-way between a pure synthetic test and something a bit more realistic 14:50:28 yeah having more than one contributor may be helpful 14:50:38 here's what i recommend, galehar 14:50:45 i figured if crawl needs help anywhere it's on webtiles 14:50:46 My idea is to implement an edit mode in the skill menu to allow setting the training percentages manually 14:51:01 then start fsim and scale XL from 0 to 27, while training skills 14:51:12 (or higher than 27 since skills continue to train) 14:51:23 that's okay, but how about this 14:51:39 and change the current implementation so it changes only one skill 14:52:15 if you hit &f, it runs the simulation once at your current skill levels and outputs the result in the game 14:53:07 if you hit &^f, it prompts you for a skill to scale, then leaves everything else alone while running from 0 to 27 in the selected skill 14:53:18 and outputs the result to fight.stat 14:53:48 if we're using it for debugging purposes, it's really annoying for multiple variables to change simultaneously 14:54:00 especially since you can easily set specific skills with the wizmode already 14:54:17 well, seems like our ideas are complementary :) 14:54:57 anyway, a lot of things can be done to fsim. It has a lot of potential. And we really need it if we want to rebalance weapon stats, melee and ranged combat 14:55:04 yeah 14:55:10 how do you feel about my suggestion? 14:55:35 another thing which could be nice is to allow the player in the arena, but that's yet another thing 14:55:37 I like it 14:55:39 brb 14:55:40 -!- heteroy has joined ##crawl-dev 14:55:49 i was also looking at that code this morning 14:55:55 i like the error-checking you did 14:56:07 i think the best way to implement that is to make the arena a portal vault 15:01:43 ah HA 15:01:49 i think i found the error, Eronarn 15:02:17 galehar: my recommendation is that you should be able to import a test case for a skill 15:02:35 i agree 15:02:49 it would be good to have, say, 'hill orc fighter.skills' 15:03:27 some batch of 10 or so test cases ranging from XL 1 to 27 each, which we can run any time we're messing with fight code, new monsters, etc. 15:03:50 definitely, that's easy to code given the other stuff 15:03:55 Eronarn: yeah, that could work too 15:04:31 do you mind if i do it, galehar? i can work pretty quickly :D 15:04:46 nfogravity: about the HP toggle patch, my guess is that evilmike is busy playing for the tournament. Bug him next monday :) 15:04:55 haha okay 15:05:01 hm... i see a way we could give it an in-game interface, actually 15:05:01 nfogravity: I don't mind at all 15:05:11 fsim isn't at the top of my todo 15:05:18 despite my earlier fixes 15:08:11 you could have the test case be something like: hill orc, fighter, axes 15:08:43 since XL isn't defined, you vary on XL. since skill isn't defined, you vary on skill 15:08:55 but you don't vary race, or weapon skill 15:10:14 then have XL be the base, skills be defined as a range based on racial apt, weapons as a probability-based range based on xl 15:10:14 -!- blabber has quit [Quit: leaving] 15:11:53 so if you load up the hill orc fighter with an axe, it might run: axe skill expected value -2, -1, +0, +1, +2 for XL1. then the same for XL2. etc. 15:12:33 this would let the simulations cover anything from a huge range of data, to your current character with a few tweaks 15:14:08 then in-game, you could have each variable display with a bar. test cases would just be presets we use across multiple versions 15:25:17 something very deep is making you a lightgrey @ 15:29:07 ahhh found it 15:29:37 is this a balance tuning tool 15:30:13 yes 15:30:53 the idea is to make mathematical parts of balance testing as automated as possible so that people actually do them 15:31:02 since they're a huge chore otherwise 15:31:13 yeah automation sounds like a very good thing 15:39:59 -!- rejuxst has joined ##crawl-dev 15:46:27 -!- gnsh has joined ##crawl-dev 15:54:18 jeanjacques (L27 DDGl) ASSERT(you.can_pass_through_feat(grd(p))) in 'player.cc' at line 435 failed. (Geh:7) 15:54:35 still the blink into malign gateway bug 15:56:59 -!- rejuxst has quit [Quit: Leaving.] 15:59:31 jeanjacques: doesnt seem to be on mantis, at least i dont see it when i search for blink among unresolved stuff 15:59:36 so maybe report it? 16:02:03 Eronarn: To avoid 0.6 heavy armor all over again? 16:04:20 Eronarn: i was right, one of the calls was missing a real_colour 16:04:37 it's one line to fix, lemme return to head and patch it there 16:06:55 dtsund: yes :) 16:07:05 nfogravity: that's great! i had wanted it for lava orc glyphs, for when they're on fire (elemental fire) 16:07:23 but it can be used elsewhere too, for sure 16:08:00 i'm surprised the player is treated completely differently from a monster 16:08:29 it's crawl, why would you be surprised at that 16:08:52 i dunno 16:09:05 next step is to see if there are any armors, cloaks, spells, etc. that justify changing player glyph to elemental :D the fun part! 16:09:38 if we really wanted to unify combat (and output) we would make the player a monster with the player's stats that lives at you.pos() 16:10:37 i imagine that would be an EXTRAORDINARY pain in the ass to implement though 16:11:46 also i have a suggestion eronarn 16:12:29 you should be disco-coloured whenever you're xom's teddy bear 16:12:52 or really just whenever involving xom 16:13:15 or while wielding anything branded as xom's 16:14:21 nfogravity: we want as much monster/player unification for implementation, but not for balance 16:14:31 nfogravity: well you could make a shim monster, actually. just return nothing any time a function tries to access the monster at that position 16:14:41 (or return the player instead, depending) 16:14:43 the whole fighting code has been recently rewritten for player/monster unification 16:15:15 as for disco colored... that strikes me as something that should be used VERY sparingly, but in a cool place 16:15:15 well that's what i heard, but i think it was marvin who was complaining that it's a giant if/then 16:15:15 more monsters should have gods 16:15:15 Dixbert: code a monster that has a god 16:15:15 like berserkers that can invoke brothers in arms 16:15:23 which was one of the source of bugs in fsim (the other being the skill rewrite) 16:15:29 @??spriggan berserker 16:15:29 spriggan berserker (04i) | Speed: 16 | HD: 12 | Health: 27-45 | AC/EV: 2/25 | Damage: 21 | Flags: see invisible | Res: 06magic(112) | Chunks: 07contaminated | XP: 1094 | Sp: brothers in arms, hand, berserker rage. 16:15:47 :< 16:15:59 the problem is that some monsters have obvious, flashy, non-complicated effects 16:16:02 and others don't 16:16:12 er, some gods 16:16:12 okay disco doesn't work 16:16:15 but ETC_FIRE does 16:16:33 disco might not work because it's based on cell position or something? 16:16:39 it's not truly random like the others 16:17:02 ETC_RANDOM or whatever probably would work 16:17:57 indeed it does! 16:18:14 god that's so distracting 16:18:17 haha 16:18:35 ooh 16:18:41 what about making GLOW cause ETC_RANDOM? 16:18:51 it's never been specified what color you're glowing, maybe the answer is all of them 16:19:25 if that's the only easy way to be a rainbow i guarantee players will do it 16:23:10 weirdly, the only things that can make you change colour right now: 16:23:19 if you're swimming, it's hard-coded into the function 16:23:48 and if you're berserk, EVERYTHING in sight is forced to be red 16:28:02 -!- ainsophyao has joined ##crawl-dev 16:28:29 what about olgreb's toxic radiance 16:29:08 -!- ainsophyao has quit [Remote host closed the connection] 16:29:11 Dixbert: that flashes the screen, which is different 16:29:23 ah 16:29:47 nfogravity: btw, that reminds me of one improvement to output that i don't think anyone has done yet: right now, OTR lights up the entire screen 16:29:51 even though it might not hit all of it 16:29:54 because of walls/etc. 16:30:02 refrig and ignite have this problem too 16:30:47 it would be good to have flash_screen take an optional LOS parameter 16:31:18 -!- rawrmage is now known as ponymage 16:31:22 (right now, i think nothing that uses it is monster-castable, but if it were, you'd also need to flash_screen from a specific actor, rather than defaulting to the player) 16:33:51 -!- ponymage is now known as rawrmage 16:36:32 fleeing monsters attack you with reaching weapon. Those with ranged weapons have only 1 chance in 8 to do. 16:37:25 galehar: should be the same chance for both, I'd say... 16:39:07 also, when they reach attack, they keep fleeing. When they fire, they stop. 16:39:32 I think they can shoot wands too... 16:40:36 i think a 1/8 chance is good 16:41:05 1/8 chance sounds fine, not sure whether they should keep fleeing or not... maybe 1/2 chance of either or something? 16:41:17 however, they should keep the fear status effect imo. and if you get close again they will try to run again 16:42:20 sorry, how do i get git to tell me what version i'm on again? 16:42:38 git status? or git branch 16:42:47 no not that 16:42:52 the version of crawl 16:43:30 <|amethyst> git describe --tags 16:43:32 if you do git branch -t you can get tags i think 16:44:03 ah there it is thanks |amethyst ' 16:49:16 Eronarn: https://crawl.develz.org/mantis/view.php?id=5399 16:49:26 patch it from master and go nuts 16:50:19 nfogravity: cool! hopefully it'll get into master before i even need to patch it 16:50:22 since it's such a small tweak 16:51:02 player colour display fix (https://crawl.develz.org/mantis/view.php?id=5399) by nfogravity 16:56:22 <|amethyst> I didn't fully test, just compiled view.cc, so hopefully you did :) 16:56:43 i did 16:57:03 i've played a lot with real_colour, there's no way this will ever break anything 16:57:17 congrats, you're probably the only person who has :P 16:57:41 03|amethyst * r16c1eaefb04c 10/crawl-ref/source/ (showsymb.cc view.cc): Allow players to be elemental-coloured (nfogravity). 16:58:01 hoorah 16:58:16 03|amethyst * re5c3cec02490 10/crawl-ref/source/view.cc: Whitespace fix. 16:58:27 -!- HangedMan has joined ##crawl-dev 16:58:45 <|amethyst> I should have noticed the blank line before pushing the first commit, oh well 16:58:48 <|amethyst> later 16:59:10 how did i fuck up the whitespace this time? 16:59:57 oh i see a blank line shouldn't have any spaces on it 17:00:01 nfogravity: are you aware of the util/checkwhite utility, by the way? 17:00:31 kinda. it broke my msysgit last time i tried to use it 17:00:40 oh 17:01:01 -!- dtsund has quit [Quit: dtsund] 17:01:03 if I am reading these logs right the means to taking down trees were discussed without remembering conjure flame can do it too 17:01:16 <|amethyst> btw, if you prepare your patches with git format-patch instead of just diff, your commit messages and names will (usually) be included when someone commits your change 17:01:19 <|amethyst> s/names/name/ 17:01:23 which is something accomplishable by the end of d:2 17:01:31 ...if you're a fe or wz 17:01:53 conjure flame can destroy trees? 17:01:54 ah great thank you 17:02:00 this is news to me :) 17:02:05 <|amethyst> (assuming the committer uses git am) 17:03:18 oh checkwhite works now, thank you 17:03:28 in which cases is casting conjure flame on an occupied tile successful 17:03:33 just plants? 17:04:17 it doesn't work on plants 17:04:51 it does work on trees though 17:04:55 why does conjure flame work on trees 17:05:22 I had no idea this was possible, since there is no reason to expect it to be... 17:05:26 it was news to me that it did, i didn't expect it 17:05:27 but indeed it is 17:05:29 clojure flame? 17:05:40 shocking revelations today 17:05:42 that seems sort of unfair 17:05:56 you can't even cast projected noise on a tree 17:06:15 you can't even cast *fire storm* on a tree 17:06:20 projected noise more like "tt" 17:07:02 please don't tell me that this is going to lead to reverting conjure flame working on trees, it was great fun doing the lair:8 forest vaults as hofe just from trivial forest fire setting 17:07:15 -!- alefury has quit [Read error: Connection reset by peer] 17:07:26 it sounds wrong to me too, yeah 17:07:45 aahahhahaha you can't firestorm a tree 17:08:12 I thought you can't fire storm trees just because the targetting and such would be awkward and also those stupid glass changes things 17:08:28 firestorm in front of trees doesn't ignite them though 17:08:41 then again fireballs can be stopped by trees without lighting them unless it's the targetted one or whatever so who knows how the hell trees work 17:08:43 you can't firestorm trees because whoever added trees didn't enable it 17:11:47 apparently conjure flame working on trees is a recent thing 17:13:45 first thing when loading up internet being open git and mantis paying off 17:14:46 it isn't even how one would expect conjure flame to be able to ignite trees (casting it on an adjacent square and waiting for the flames to spread 17:15:32 protip: this would all be solved if we made trees monsters 17:15:47 plants are monsters 17:15:49 bushes are monsters 17:16:29 walking through trees would probably screw up a notable amount of vaults 17:17:16 HangedMan: not really, there are already several ways to bring down trees 17:17:33 multiple starting classes get spells that can do it 17:17:38 actively, on the monster side, with tree los working the way it does? 17:17:56 maybe something besides the lernaean hydra could trample trees 17:18:07 (I was thinking *fedhas*) 17:19:05 fedhas doesn't need to allow walking through trees 17:19:20 -!- MarvinPA_ has joined ##crawl-dev 17:19:39 am i allowed to think that xuaxua is sort of rude in his bug reports 17:19:41 is that okay 17:19:49 my proposal: firing directly at a P/7 hits it as normal. if you miss, or fire through it, it has a self.cover/max_cover chance of intercepting the attack. cover = 0 to 7; if cover = 7 los is blocked; trees = +3, bushes = +2, plants = +1, toadstools = +0 17:20:32 gameplay effects: toadstools never block ranged attacks; single plants block them 1/7 of the time; two trees block 6/7 ranged attacks and anything else added will block LOS too 17:20:35 considering how he got up in arms over "targetting", probably 17:21:07 !vault oklob_temple 17:21:10 Couldn't find oklob_temple in the Crawl source tree 17:21:39 some vaults might need to be changed a bit to account for that 17:21:52 but it would make ranged combat around trees/plants a lot more fun, and give more LOS options 17:22:28 I do *not* anticipate trying to still keep hangedman_tree_tricks as small as possible while dealing with tht change, sigh 17:22:30 -!- MarvinPA has quit [Ping timeout: 272 seconds] 17:22:30 -!- MarvinPA_ is now known as MarvinPA 17:22:31 might be worth looking into how many vaults depend on current tree behavior 17:22:41 Dixbert: not many since trees are relatively new 17:22:48 certainly less than use glass or rock 17:22:53 like it was done with the glass change? scnr 17:23:10 what do fungi count as, eronarn 17:23:17 HangedMan: i'd make them a 1 17:23:25 fair enough 17:23:40 -!- Wensley has joined ##crawl-dev 17:23:55 -!- alefury has joined ##crawl-dev 17:24:26 some 'real' monsters would want it - oklobs would be a 2, wandering shrooms a 1, i guess 17:24:34 -!- dtsund has joined ##crawl-dev 17:24:40 this would never come up firing at them directly, except for an increased chance you'd hit them even if you missed based on their EV 17:25:04 ask somebody who's played fedhas if this works out 17:25:27 you mean the bush/oklob turret stuff? or what 17:25:33 now the question is why can't you fire through a cockroach tile but through a tree tile 17:25:46 and where to stop thinking about this 17:26:01 how turret stuff works out with this system, yeah 17:26:14 because cockroaches fill the entire 1x1x1 square whereas trees are smaller :P 17:26:19 better to just make it a monster property, which only applies to monsters with branches and the like 17:26:38 perhaps treefolk would eventually get it 17:26:38 squareroach 17:26:40 why is penetration so magical 17:26:46 squareghost 17:27:05 when is squarelos getting reimplemented 17:27:15 effectively it'd be 'it's harder to aim past this creature than it is other creatures' 17:27:31 -!- Ragdoll has left ##crawl-dev 17:27:39 i feel this is a good way to represent plants (and grates, we'd want to do this for grates too. maybe statues) 17:27:42 and an excuse to work on forest wyrms again? ;P 17:27:54 A cockroach. It is as big as several trees. 17:27:56 whoa ^ 17:28:02 careful with grates 17:28:15 yeah, they're already precarious as is 17:28:30 "There is a chance your spell will hit the grate instead of slipping through it." 17:28:43 perhaps 'grate' (small openings) vs. 'portcullis' (large openings) 17:35:56 -!- syllogism has quit [] 17:41:40 -!- evilmike has joined ##crawl-dev 17:43:36 so, uh 17:43:57 can the pain mirror colours not be made to involve darkgray and purple if there are purple monsters usually present with them 17:44:17 HangedMan: +1 17:45:03 also, evilmike, can you help me with a bit of syntax I can't manage to comprehend 17:45:09 alright 17:45:10 evilmike: You have 1 message. Use !messages to read it. 17:45:39 hey, here's an idea 17:45:46 while you have pain mirror up, gain umbra 17:46:00 http://pastebin.com/07e0xZmV 17:46:08 i guess etc_unholy works, that's yred colours too 17:46:16 halos are even harder to distinguish in console 17:46:25 seems flavorful, and would help with the piety spam issues (since umbra = get hit less) 17:46:34 except berserkers will be red if they're berserk 17:46:56 and those are probably a high priority to kill so maybe even worse 17:47:05 fr: more colours :( 17:47:21 fr: fewer monsters! 17:47:22 brocrawl 17:47:34 even worse, ETC_NECRO is #ifdef TAG_MAJOR==32 17:47:57 hmm? why shouldn't it be? 17:48:19 it was removed since it's identical to etc_death i thought 17:48:25 HangedMan: that looks weird, I'll try to figure out what's going on 17:48:48 ah, I was looking at 4283068bb8b4cf242b326a49deee9999851cfd5f, not current code 17:49:15 aha right 17:49:18 MarvinPA: more colours won't solve the issue since we need fallbacks for 16 colour mode anyway 17:49:43 yeah 17:50:07 there's just only so many blood/death/necromancy themed colours, and considering almost all the deep dwarves have that theme... :P 17:50:28 fr deep dwarf blood knights 17:51:02 also, adding 256 colour support is blocked by the code using an 8-bit type for colours in a thousand places, and in others, it assumes higher bits are free as well 17:52:38 solution: get rid of DD monsters 17:58:27 03galehar * rae7027bf5b35 10/crawl-ref/source/ (main.cc spl-book.cc wiz-you.cc wiz-you.h): New wizard command: &^M to memorise a spell. 17:58:27 03galehar * rd79679de2d88 10/crawl-ref/source/skills.cc: Fix an off by one error in set_skill_level. 17:59:45 galehar for mvp 18:00:09 mvp? 18:00:20 most valuable person 18:00:26 HangedMan: i'm not having a problem with your subvault thing 18:00:38 HangedMan: put the vault + its subvaults in its own des file, and it works 18:01:18 most valuable dev 18:01:34 in its own des file, what 18:01:56 I couldn't get putting in foreign .des files into local copies of crawl to work either 18:02:02 as in make a file like "subvault_test.des", paste that in to there, and try it. it works fine 18:03:49 evilmike: only if you're running from the source dir 18:04:01 ah 18:04:22 evilmike: installed builds list .des files to include explicitely, as otherwise there's no way to avoid unclean installs 18:04:40 is that list anywhere I can edit 18:04:56 strange thing about that subvault example on pastebin, is I did get an error pasting that into an existing des file. but creating a new one works fine 18:05:07 dunno why, seems to be like the former ought to work just as well as the latter 18:08:32 need to make a web-based vault designer tool 18:08:45 oh right 18:08:56 Wensley: it would be useful for laying out the map but not much else 18:08:59 also remembering not being able to find a vim thing listed in the documentation 18:09:06 show examples of randomization? 18:09:09 HangedMan: what d oyou need vim for? 18:09:18 not *need* 18:09:22 evilmike: it would be useful for *me* because HangedMan is always telling me to look at his maps and I am like "wtf am I seeing here" 18:09:27 hahaha 18:09:54 Wensley: copy-paste the stuff into one of the des files (like test.des) and then place it with &L 18:09:57 there's your vault viewer 18:10:03 ugh that takes so long 18:10:04 "levdes.vim - If you use vim, the levdes.vim syntax highlighting script (provided in the dat directory) can make level-editing far more pleasant by colouring different features in maps and syntax-highlighting .des-file syntax. vim is available for nearly all operating systems, including Windows." 18:10:04 -!- jeanjacques has quit [] 18:10:37 oh hey there google search 18:11:18 I'm guessing support for this was dropped because I've never seen source/dat/vim/ 18:11:46 you can perhaps check old versions of the git repo 18:12:32 if you get to go "ugggggggghhhh" about reading vaults I get to go "ugggggghhhh" about trying to actually use any substanial bits of git 18:15:48 oh, nice. 18:16:01 i should add levdes.vim to my vim setup 18:16:05 evilmike: disagree, there could definitely be useful online level editing 18:16:42 for example, a modified crawl client, and if you press a button it tries to load the map and either spits out errors or lets you play on it 18:16:42 Eronarn: I didnt say it wouldn't be useful, just that it would be limtied to editing the map itself 18:16:59 ... 18:17:09 so this does exist but isn't packaged with default things or something? 18:17:12 whatever 18:17:39 evilmike: well, what else is there? :P edit a map and have a 'submit to devs' button 18:18:06 Wensley: there might still be an abortive attempt at using the geshi syntax highlighter to do vault syntax on the wiki 18:18:19 Eronarn: if you want to make it randomized in a decent way, the header is just as important as the map itself 18:18:50 -!- ussdefiant has quit [Ping timeout: 260 seconds] 18:19:04 evilmike: sure, but you can have it [textfield] space [textfield] and have some buttons like 'insert shuffle' at the first textfield's cursor 18:19:56 it would be doable as a webapp, if someone cared enough 18:21:12 Just a quick bug, if some monster sets off a chaos pressure plate/grate trap, the message reads "Space ruptures around you!" 18:21:19 It would certainly be enough for simple vaults 18:22:46 simple vaults, what are those 18:22:50 for very complex stuff though... maybe I'm just not smart enough to imagine how a web app would handle that. Look at sprint2.des, for example 18:25:28 hrm, we really, really should get rid of cloud traps abusing the one-off trap thingy 18:25:28 guess who's looking for webapps experience 18:25:30 <--- 18:26:19 evilmike: well, you'd want to have at first, basic buttons like 'shuffle these things with these weights' or 'add an item with this property' 18:26:34 but you could add new things based on commonly used code 18:26:59 rotate foo 18:27:02 bring up resource 18:27:14 Eronarn: could work. Although, shuffle is only one way of randomizing features (and it has two distinct ways of working) 18:27:17 tiles list 18:27:26 there might be things you want to insert with a placeholder, which later gets replaced with the specific stuff 18:27:51 actually a picture of how it looks in the two different mode would be cool 18:27:54 dealing with tiles would be one nice feature of a program like that 18:28:21 right now if you want to define a floor tile, you have to go "FTILE: [every glyph used in the vault] = whatever" 18:29:10 heck, even just a web app that does: paste your map, press a button, it loads up a splitscreen game of crawl (shellinabox / webtiles) 18:29:29 FTILE: 1234567890cdIJKLM.`'"~-+ = FLOOR_ROUGH_LIGHTGRAY / FLOOR_ROUGH_WHITE / FLOOR_ROUGH_BLUE / FLOOR_ROUGH_LIGHTBLUE 18:29:33 even if it didn't have syntax highlighting or anything i'm sure people would use it 18:29:48 one thing that would make this web app a lot handier would be a built in text editor 18:30:09 also, a mode for displaying the map as raw text, since that will still be useful for some things 18:30:55 HangedMan: yes, something like: $ftile = 'STRING' 18:31:01 might as well have syntax highlighting too, at least for lua 18:31:08 and if $ftile[key] isn't set it defaults to STRING 18:31:25 hehehe 18:31:54 it'd require a minor check to avoid the 80 characters per line thing 18:32:16 that's just a style thing, it's not a technical limit 18:32:27 good text editors let you work with this easily anyway 18:32:38 bad ones let you resize the window to 80 columns 18:33:12 Dixbert: anyways if you want to work with this - it might be something few people would use, but that means you can focus specifically on what they need 18:33:33 another option, if you want to do a webapp, is writing a real website for crawl games 18:33:47 right now there's just text dumps. compare that to a character file on te4.org 18:34:50 http://www.tome.te4.org/characters/1842/tome/12bd7817-29a0-11e0-84b0-001cc000edf0 18:35:31 note the highlights over the item stats, abilities, etc. 18:35:47 crawl is actually a bit more limited than that even, you can't even get an up to date character dump unless you bug the person to make one 18:35:47 for a variety of reasons I think crawl has less things to list then this 18:36:23 HangedMan: sure, but that just means it will fit very nicely on one screen 18:36:47 the point is to make something that's nicer to read than a plain textfile 18:37:20 you can also do stuff like 'current character' images 18:37:23 that people can use for signatures 18:37:32 this is reminding me, is there a reason not to automatically make a character dump whenever a player saves? 18:38:03 that would be an awesome way to make badges relevant - since the crawl server would be generating the images, they can't modify it with it remaining dynamic 18:38:05 elliptic: that would be a nice feature, for sure 18:38:15 so you can have sigs that show up with the banners they've won so far in the tournament 18:38:15 somewhat interesting, if not something I'm not nearly as interested in compared to a vault editor 18:38:52 well, the vault editor would be nice, but much more complex 18:38:58 it would be more useful, but to a much smaller audience 18:38:58 banners are the coolest thing about the tournaments. the ones in this one are great 18:39:49 keeping all-time player stats would be cool, too 18:40:51 I don't like how almost nobody will get pious III 18:41:14 that one is ridiculous, but whatever. we have some ridiculous players 18:41:39 it demands less "pay one weird game" and more "go play a ridiculously varied number of games" 18:42:11 probably the easiest way of getting it is to play one or two weird games 18:42:18 -!- ussdefiant has joined ##crawl-dev 18:42:27 beoghfedhasjiyva 18:42:38 just need to scum pan/zigs for piety after creating an uberchar 18:42:42 ...hah 18:43:07 but the other weird games are at least restrictions or challenges instead of grinding 18:43:33 would be better to unify this with other god.maxpiety rewards 18:44:55 by the way, there's a glaring discrepancy between "win with god" and those 10 points per "champion": the latter doesn't trigger if you started with a shit god and renounced before ****** 18:45:14 kilobyte: neither one does 18:45:18 not sure what you mean 18:47:08 -!- gnsh has quit [Quit: leaving] 18:47:17 !lm . t type=god.maxpiety s=god 18:47:18 6 milestones for kilobyte (t type=god.maxpiety): Trog, Cheibriados, Nemelex Xobeh, Ashenzari, Fedhas, Okawaru 18:47:55 "For purposes of these points we say that a player wins with a god if she reaches full (******) piety with that god without worshipping any other god first." 18:48:02 no two in the same game, yet "champion" counts only two because most were done as HuPr 18:48:16 yeah, that's what I'm pointing at 18:48:19 "10 points the first time you reach full (******) piety with each god without having worshipped any other god first that game." 18:48:23 it is in the rules 18:48:44 it is, and works as described 18:48:48 not sure what the discrepancy is between the two things you list given they use the same rules, then :P 18:49:21 are you just complaining that changing gods early on disqualifies you from god points? 18:49:35 I agree that isn't great, but without that you have various abuses 18:50:07 huh, the '1x's are elided now 18:50:08 neat 18:51:12 if you take the first god.maxpiety milestone in each game, then there is incentive to carefully avoid getting to ****** with one god so that you can do a different god later 18:52:00 if you take all god.maxpiety milestones in each game, then scumming pan/zigs for piety gets you points 18:52:15 etc 18:54:07 (I don't care *that* much about scumming for 10 points per god.maxpiety, but assigning points for winning with a god would still be a problem) 18:54:40 -!- st_ has joined ##crawl-dev 18:55:14 kilobyte: if you have another suggestion though, I'd love to hear it... handling gods is definitely one of the most awkward parts of the rules, I just don't know how to do it better 19:00:16 what about points for getting maxpiety with a god you don't abandon? 19:01:00 that would end up counting only the last one, of course 19:01:18 really liking the idea of crawl badges 19:01:21 and you'd still need to specify "without ever worshipping a god" for the no god case 19:01:42 it's the first real god that matters 19:02:00 ie, one you had in early and mid game 19:02:12 under the current rules 19:02:15 zannick: the problem with counting just the last one is that people will do stuff like clear zot -> worship chei -> get max piety -> pick up orb and leave 19:02:34 people can do that anyway 19:02:57 not unless they are atheist the entire game through clearing zot 19:03:21 oh right 19:03:32 that's why I'm suggesting the first ****** god in a game 19:03:40 kilobyte: see above 19:03:50 it's already that way, kilobyte 19:03:59 kilobyte: encourages people to very carefully avoid getting ****** with their first god in some cases 19:04:17 aside from the "don't switch" clause 19:04:20 elliptic: what's the reason it doesn't just count for any time you reach maxpiety 19:05:07 Wensley: lategame switches (like after clearing zot) and also that doesn't work for the "win with god" points 19:05:15 why not? 19:05:20 could split the points awarded evenly 19:05:35 chei worth 60 and zin worth 40, you get maxpiety with each, you get 50 19:06:01 Wensley: it's easy to get maxpiety by sitting in Pan for that 500 kills 19:06:08 wensley: in that case it again encourages carefully not getting beyond ***** with okawaru 19:06:29 5.5 stars but yeah 19:06:44 kilobyte: judging from the rate at which ely 1 was awarded I'm guessing it's even easier to just start a game and get to champion the natural way 19:07:06 that "extremely 5*" vs "exalted 5*" thing is silly 19:07:07 Wensley: it isn't just champion 19:07:22 Wensley: also points for winning games 19:08:05 all I'm saying is that the current setup seems needlessly complicated and still doesn't entirely discourage gaming the system 19:08:32 Wensley: how are you gaming the current system? 19:09:01 going athiest past endgame and then grinding with chei :P 19:09:23 note that you're giving up the No God points by doing this 19:09:25 seriously chei isn't so hard that this makes much sense unless you are also streaking :P 19:10:34 rwbarton: only if you don't already have the no god points 19:10:40 true 19:11:09 chei = shatter at XL 13, vehumet or ashenzari can't even dream of that 19:12:10 something about movement being important 19:13:17 something about killing all opposition 19:13:55 chei has his downsides (putting it lightly), but he's nowhere among weaker gods 19:14:03 kilobyte: I've gotten shatter at XL 16 with Makhleb 19:14:27 it's like claiming it's hard to win with Trog because you can't cast 19:15:06 kilobyte: also I got tornado at XL 13 with vehumet once 19:15:38 kenku? doable 19:15:43 indeed 19:16:02 (I'm speaking mostly about 0 apt) 19:16:26 doesn't 15 int just give you 2.25 levels of school skill, or do I forget the formula 19:16:37 plus more power of course, but 19:16:48 for casting success rate purposes 19:16:53 *2.5 19:17:17 it is just 2.5 levels of school skill, yes 19:18:37 pretty sure that is smaller than vehumet boost; maybe kilobyte was cheating and using brilliance? 19:18:47 (vehumet doesn't stack with brilliance) 19:19:20 elliptic: you're ignoring the stepdown 19:19:27 kilobyte: huh? 19:21:30 skills are stepped down, int is not (spell_fail() uses spell power for skills but not int) 19:22:51 only at really high skills will you get into the stepdown region here 19:23:06 but I guess you might, which would make it 5 levels of school skill 19:23:28 so I tried with spc 13 earth 19.5 and 35 int, same as spc 13 earth 22.5 and 20 int 19:23:33 50% fail 19:24:15 kilobyte: you have to remember that calc_spell_power() isn't getting multiplied by enhancers or intelligence here, so it is quite low 19:26:35 03evilmike * rff9fb87a6f8b 10/crawl-ref/source/ (14 files): New ctrl+V command to show monster HP, and other HP colouring improvements. (nfogravity) 19:26:44 kilobyte: anyway, it looks to me like the vehumet and chei boosts are pretty comparable 19:27:12 the exact relationship between the two is complicated because of the chance_breaks[][2] stuff 19:27:37 yeah, and chei gives you armoured casting plus great EV... while Vehumet allows haste and walking away 19:27:46 since the int boost decreases chance by 30 before that and the veh boost multiplies chance by 2/3 after that 19:28:39 we just wasted half an hour due to a remark that a Chei win needs to be scummed 19:28:46 let's deal with Xom instead 19:29:32 i think xom scum wins are more likely to be problematic 19:30:46 he 1. needs to be greatly buffed, but 2. should never bring you back to good terms just because he got bored 19:32:30 2. was an attempt to remove the need for scummy spoily acts like drinking bad potions, but instead resulted in Ragdoll sitting in the temple 19:37:23 preposterous 19:37:28 ragdoll sits on d:1 19:38:22 ragdoll sits on d:1 and dies most games 19:47:52 -!- HangedMan has quit [Ping timeout: 252 seconds] 19:50:10 -!- umrain has joined ##crawl-dev 19:50:26 -!- ainsophyao has joined ##crawl-dev 19:59:13 -!- ainsophyao has quit [Remote host closed the connection] 20:25:03 03evilmike * r81611bce62cc 10/crawl-ref/docs/options_guide.txt: Update options_guide.txt for enemy_hp_colour. (nfogravity) 20:29:47 I've seen 4.1a Xom a bit, he's more like what I'd want Xom to be... He basically just gives you random good stuff when nasty things are around 21:01:30 -!- alefury has quit [Remote host closed the connection] 21:22:47 -!- ainsophyao has joined ##crawl-dev 21:29:02 My Xom opinion is he should be able to react to how you respond to his acts, and any nasty act should have at least some slight compensation. 21:30:16 Like, say he sends demons to kill you. He might drop a scroll of blinking at your feet to let you get a headstart for an escape, and call you a coward (possibly losing a bit of favorability) if you use it. If you don't need it to survive, you can keep it. 21:37:59 that seems a bit... premeditative? 21:38:08 makes him sound like the guy from saw 21:39:09 This is Xom we're talking about. 21:39:11 <|amethyst> I think of Xom sending demons against the player as his form of Arena mode 21:49:52 -!- ainsophyao has quit [Remote host closed the connection] 21:50:26 the only xom effects I dont really like are random banishment and making your weapon dance 21:50:36 the latter is just because it encourages you to carry two weapons 21:50:59 or worse, unwield whenever you rest 21:51:21 banishment is the only thing that makes me actually avoid playing xom though (I think the rest of what he does is pretty entertaining) 21:52:12 chaos brand is also fun, but it has issues. since it can berserk monsters, it's best used on demons and undead, imo 21:53:24 yeah the weapon unwield on rest is not nice 21:53:51 but solutions like animate anything from the inventory will just mean that we'd need to drop things instead of not wielding them 21:56:34 xom should just cast tukima's ball or something when he decides to be a jerk 21:56:57 could be hostile or friendly, even. or both at the same time 22:10:43 -!- Wensley has quit [Ping timeout: 276 seconds] 22:11:32 -!- nfogravity has quit [Quit: Leaving] 22:15:36 -!- nfogravity has joined ##crawl-dev 22:16:18 -!- nrook has quit [Quit: ChatZilla 0.9.88 [Firefox 10.0.2/20120215223356]] 22:27:25 -!- ainsophyao has joined ##crawl-dev 22:29:25 -!- ainsophyao has quit [Remote host closed the connection] 22:31:41 Felids autopickup needles of curare (https://crawl.develz.org/mantis/view.php?id=5400) by LorrdErnie 22:36:32 03dolorous * rc989334bf4d4 10/crawl-ref/docs/options_guide.txt: Typo fix. 22:46:19 -!- Cryp71c has joined ##crawl-dev 22:48:05 -!- umrain has left ##crawl-dev 23:01:29 -!- mikee_ has joined ##crawl-dev 23:02:31 -!- ToBeFree has joined ##crawl-dev 23:02:40 -!- ToBeFree has quit [Changing host] 23:02:40 -!- ToBeFree has joined ##crawl-dev 23:30:22 figs (L3 DEWz) ERROR in 'tileweb.cc' at line 204: Socket write error: Bad file descriptor (D:2) 23:50:27 -!- ToBeFree has quit [Remote host closed the connection]