00:01:59 -!- ZChris13 has joined ##crawl-dev 00:20:28 Crashes when loading saved characters (https://crawl.develz.org/mantis/view.php?id=3649) by Mister44 00:35:02 Manjusri (L8 DDNe) (D:7) 00:37:56 Windows development builds on CDO updated to: 0.8.0-a0-5914-gdfc6cae 00:42:07 Manjusri (L8 DDNe) (D:7) 00:45:35 Berserk spell uses resources before checking hunger (https://crawl.develz.org/mantis/view.php?id=3650) by Milgwyn 00:56:50 -!- jbenedetto has quit [Read error: Operation timed out] 01:03:01 -!- jbenedetto has joined ##crawl-dev 01:07:05 -!- jbenedetto has quit [Read error: Operation timed out] 01:08:31 -!- jbenedetto has joined ##crawl-dev 01:16:34 -!- jbenedetto has quit [Read error: Operation timed out] 01:19:00 -!- ZChris13 has quit [Ping timeout: 264 seconds] 01:22:59 -!- jbenedetto has joined ##crawl-dev 01:29:35 -!- Twinge_ has joined ##crawl-dev 01:33:24 -!- Twinge has quit [Ping timeout: 264 seconds] 01:34:13 -!- jbenedetto has quit [Read error: Operation timed out] 01:35:35 -!- OG17 has quit [Ping timeout: 255 seconds] 01:39:40 -!- jbenedetto has joined ##crawl-dev 01:48:54 -!- valrus has quit [Remote host closed the connection] 01:50:40 -!- elliptic has quit [Ping timeout: 246 seconds] 02:15:59 -!- ekiM has joined ##crawl-dev 02:25:32 -!- ekiM has quit [Ping timeout: 255 seconds] 02:38:34 -!- ekiM has joined ##crawl-dev 02:48:02 -!- ekiM has quit [Ping timeout: 255 seconds] 02:54:14 -!- ekiM has joined ##crawl-dev 02:59:49 moin 03:02:07 -!- galehar has joined ##crawl-dev 03:02:24 hi 03:28:52 Morning 03:29:37 Eronarn: Rather than mesmerize (for a viper-replacing snake), a petrifying gaze would be new, and would keep mesmerize more mermaid-specific. 03:30:21 (I'll admit that I have an idea to give imps mesmerize (but replace them from early game)) 03:33:25 galehar: I like "Simplify low spell success adjectives" very much! 03:33:40 :) 03:35:02 No internet at home at the moment. :/ 03:37:01 Maybe I do need to set up crawl dev tools etc on the work machine. :) 03:37:13 it might be good to change bad into very poor 03:37:46 I like the change (therer were too many similar adjectives), but I think it's still hard to know that "Bad" is worse than "Poor" 03:38:18 I recall many comments that "Good" is not very good etc; the adjectives can probably be adjusted in this way too, but anyway, that was a good change on its own! 03:43:43 LRD blasting out of LOS is sooo cheating. :P 03:48:23 -!- Kurper has quit [Remote host closed the connection] 03:49:30 -!- valrus has joined ##crawl-dev 03:50:25 Good is 70%-80%. 7/10 is a good chance. It describes the probability of success accurately in my opinion. 03:51:41 Hmm, I intended to disallow ice cave portals from branch ends (which have ending vaults, and that includes Orc now), but I guess allowing them is ok for the ends which have non-encompassing end maps. In other words, Orc:4, Elf:5 and Lair:8 are fine, but Vaults:8 should be excluded. 03:52:18 galehar: I guess the complaint is not about probablity but in-game usefulness :) 03:52:38 I haven't really paid attention myself.. 03:53:47 Good is usable for some spells, but for many even 80% success is quite bad 03:54:19 depends on how reliable you need the spell to be, and also how bad the miscasts can be (you can get some fairly nasty ones if you cast spells at Good) 03:54:53 We should use the same scale for all the spells of course, so it sounds tricky to improve from that viewpoint. 03:55:01 thats probably why the adjectives shouldn't be changed actually (it's rather subjective and varies a lot by spell) 03:55:12 Yeah. 03:58:23 Jordan7hm: Feedback for your spider maps (haven't played, just looking at the des now); there are chokepoints (esp. at entrance/exit), which should happen on Spider maps, so that the player gets swarmed. I can adjust the maps at some point, if that's ok with you. 03:59:29 I should put the layout design guidelines on top of the des file. 04:08:38 The adjective describe the probability. It's up to the player to translate it into in-game usefulness. 04:11:25 then perhaps it should use probability adjectives instead of usefulness adjectives 04:17:47 moin moin! 04:18:11 btw, did anyone see any bugs with the new wiki? 04:19:56 monky: the adjectives are neutral. And the column title is "Success", not "usefulness". 04:20:29 The [Header] thing in the changelist of a page looks good in the wiki update, no bugs I've noticed so far! 04:20:51 "Good chance of success", yeah, that makes sense. 04:20:55 great, thanks, Eino :) 04:21:43 I now agree that they adjectives are ok, though perhaps something like "Likely" instead of "Good" might be even better. 04:22:57 Last user test today; they just told me that not a single tester has managed to finish the tutorial in the testings. They've had from an hour to an hour and a half of time. 04:25:21 wow 04:25:46 where do they stop on average? 04:27:27 I don't know, and they wanted to hold on to the rest of it until they've finished the report to build up excitement. ;) 04:27:32 But I can ask if you really want! :) 04:27:56 Dispersal is the best invention ever, I just had a tricky situation trying to close in on a centaur warrior, and decided to back out. 04:30:54 The course deadline is 27.4. 04:41:38 It's ok, I can wait 04:47:41 It'll be late for 0.8, but that should be ok.. we'll hopefully get feedback from players as well, so we can look at those together 04:47:57 Napkin: could you update 4.1.2a? 04:48:38 Keskitalo: using mesm on more than one thing is fine imo... cobras have the reputation of hypnotizing their prey 04:48:49 and it'd be a good justification for a tmut/hex spell mimicing this effect 04:48:52 Napkin: you might notice a .git dir there (no changes, I half-heartedly tried to reproduce history by using the three versions lying around) 04:49:03 and two commits which are actual changes 04:50:32 Napkin: update-cdo-0.6.sh is expected to work, right? (I updated 0.7 some time ago and it worked, I guess it's just a copy) 04:50:34 chuck can walk out of his water cage in sprint3; perhaps change it to a lava cage 04:50:44 Eronarn: It is fine, it just dilutes "specialness" of Shoals. 04:50:50 monqy: nice 04:51:00 Eronarn: What would player mesmerize be good against? 04:51:04 galehar: i disagree that 'good' is an accurate assessment of 70-80%; there are lots of times where we would not say someone is 'good' at doing something if they fail utterly 1/3 of the time 04:51:17 A mesmerizing summon would be quite useful. 04:51:36 Keskitalo: believe it prevents blink/tele, yeah? 04:51:43 or maybe just blink 04:51:51 also rage 04:51:54 some other random stuff 04:52:15 & would prevent enemies swapping with each other 04:52:21 or enemies running when hurt 04:52:39 Sounds a bit narrow. Of course, it'd prevent yaks and stuff from fleeing but that would be kind of a convenience tool for collection the xp. :P 04:52:51 Swapping prevention sounds the most interesting. 04:53:29 Not trying to shoot the idea down, just questioning if it would be useful. 04:53:44 re 04:53:48 The flavour sounds cool though! 04:53:52 sorry, kilobyte - i don't understand.. what's up? 04:54:02 you need to update 4.1.2a & 0.6 on CDO? 04:54:03 we could also simply have player/monster work differently: monster = you cannot move away from; player = you are more likely to move closer to 04:54:05 And sure, we could add a spell for testing, to see if people find a use for it. 04:54:25 I don't think e.g. Golubria is considered very useful yet, but it's fine that it's in, at least in testing builds. 04:54:47 Eronarn: Yeah, having it force ranged characters move towards you could be useful. 04:54:59 Napkin: I can do what I need on DGL's end, but this way is much easier, even if a dirty hack. 04:55:34 It could also be a god ability, I sure could use that on my FeBe in Vaults. :P 04:55:37 kilobyte: I'm still lost 04:55:54 kilobyte: tell me, what you need me to do :) 04:56:16 Napkin: the changes for 4.1 are in source/crawl-anc, ready for compilation/install, if you could update it would be great 04:57:10 and for 0.6, I guess the script should work -- it just never hurts to ask since you have more clue than me about it 04:58:08 query 04:58:31 03kilobyte * r10ceba3cda27 10/crawl-ref/source/dat/des/sprint/sprint_mu.des: Don't let chuck catch a cold by wading in water. 04:58:40 03kilobyte * rb13ecea585c0 10/crawl-ref/source/itemname.cc: Mark rings of levitation useless when appropriate. 05:00:56 -!- monqy has quit [Quit: hello] 05:08:37 Keskitalo: here's a thought wrt Be: trog capstone ability to berserk everything on screen 05:08:56 i think that would fit with trog well, where each of his abilities has multiple uses 05:29:27 -!- Twinge_ has quit [Read error: Connection reset by peer] 05:30:26 -!- Twinge has joined ##crawl-dev 05:30:56 -!- Twinge has quit [Read error: Connection reset by peer] 05:31:31 -!- Twinge has joined ##crawl-dev 05:32:30 -!- Twinge has quit [Read error: Connection reset by peer] 05:32:38 -!- Twinge has joined ##crawl-dev 05:34:05 -!- Twinge has quit [Read error: Connection reset by peer] 05:35:10 -!- Twinge has joined ##crawl-dev 05:37:17 -!- Mu_ has joined ##crawl-dev 05:37:19 -!- Twinge has quit [Read error: Connection reset by peer] 05:39:08 -!- Twinge has joined ##crawl-dev 05:39:09 03kilobyte * r854ddbee73dc 10/crawl-ref/source/feature.cc: Correct the colour of slime walls. 06:04:19 -!- evilmike has quit [] 06:05:55 -!- valrus has quit [Remote host closed the connection] 06:08:17 -!- galehar has quit [] 06:17:39 -!- OG17 has joined ##crawl-dev 06:22:14 03kilobyte * r96e0e3f36a54 10/crawl-ref/source/food.cc: A quick hack to disable hunger in ZotDef for non-Vp. 06:25:58 -!- galehar has joined ##crawl-dev 06:26:14 03galehar * rb29fac960595 10/crawl-ref/source/spl-cast.cc: Don't spent turn/MP when aborting the berserk spell. 06:26:25 03galehar * r50abf5739d0a 10/crawl-ref/source/ (6 files in 2 dirs): Ashenzari: add flavour messages for the new passive boost to low skills. 06:29:08 Eronarn: That'd be fun against casters 06:31:50 Sounds a little marginal in use though; your allies arrive berserk as it is, you can cheaply berserk yourself, berserking monsters would usually make them more dangerous... ok, they wouldn't shoot or use spells (and probably other special abilities)... 06:32:01 okay, that would be useful against tormenters/hellfire-users 06:33:01 and increase Trog usefulness in post-endgame 06:33:20 So, it's an interesting thought 07:07:31 -!- syllogism has joined ##crawl-dev 07:08:26 03j-p-e-g * r7861bd20fa65 10/crawl-ref/source/tilepick.cc: Fix #3643: Kraken simulacra causing crashes in Tiles. 07:09:37 "Magic restored" doesn't work as a trigger in Runrest_messages.txt (https://crawl.develz.org/mantis/view.php?id=3651) by SinsI 07:21:11 -!- Sequell has quit [Ping timeout: 255 seconds] 07:21:52 -!- Sequell has joined ##crawl-dev 07:46:29 "You have 2042 points of unallocated experience (cost lvl 15; total 28900)." 07:46:35 is this meant to be debug info? 07:46:44 it's showing up on cdo 07:48:35 -!- Zaba has quit [Ping timeout: 252 seconds] 07:50:14 -!- upsy has joined ##crawl-dev 07:52:24 yes it is 07:55:01 -!- Zaba has joined ##crawl-dev 07:59:53 oops 08:02:31 03galehar * rb5f1625a5e39 10/crawl-ref/source/ (skills2.cc skills2.h): Remove debug informations from non-debug builds. 08:03:55 -!- galehar has quit [Remote host closed the connection] 08:06:32 -!- valrus has joined ##crawl-dev 08:06:53 Morning 08:09:35 -!- Zaba has quit [Ping timeout: 252 seconds] 08:11:41 !tell doomunicorn Yes, I'm rewriting fight mechanics (or at least starting to). The code is too complex for a definitive "this is how its going to work" (whether in code or in pseudo-code), I'm more or less writing it as I go with a general plan and objective, consulting other devs for their input and perspective, periodically submitting the code for others to review and improve. 08:11:42 Maximum message length is 300 characters. Eschew verbosity, Gladys! 08:11:56 !tell doomunicorn Yes, I'm rewriting fight mechanics (or at least starting to). The code is too complex for a definitive "this is how its going to work" (whether in code or in pseudo-code), I'm more or less writing it as I go with a general plan and objective, 08:11:57 Cryp71c: OK, I'll let doomunicorn know. 08:12:04 Cryp71c: OK, I'll let doomunicorn know. 08:12:04 !tell doomunicorn consulting other devs for their input and perspective, periodically submitting the code for others to review and improve. 08:13:58 hehe, galehar 08:14:12 since rebuild on CDO required? 08:18:35 -!- galehar has joined ##crawl-dev 08:20:23 -!- Zaba has joined ##crawl-dev 08:35:18 -!- eith has joined ##crawl-dev 08:40:05 Mimics can be identified by targeting them with non-damaging spells (https://crawl.develz.org/mantis/view.php?id=3652) by MarvinPA 08:53:46 Muwahahaha, my combat branch compiles (who knows if it actually does what I want it to do) 08:54:45 -!- valrus has quit [Remote host closed the connection] 09:07:11 -!- galehar has quit [Ping timeout: 252 seconds] 09:12:27 03Cryp71c 07unified_combat_control * r1b78dd536d04 10/crawl-ref/source/ (attack.cc attack.h fight.cc makefile.obj melee_attack.h): Add attack to makelist, compile fixes 09:12:37 03Cryp71c 07unified_combat_control * r76882184f8ba 10/crawl-ref/source/ (7 files): Relocate melee_attack class to own cc file 09:21:40 I feel obliged to make a cool ossuary. 09:33:50 -!- elliptic has joined ##crawl-dev 09:37:34 Napkin: I'm making a few changes for CAO trunk update scripts :) 09:38:03 Napkin: I'm trying to keep every CAO config file + dgl script into one repository 09:38:06 *in 09:48:44 hmm 09:48:46 ok, sure 09:49:04 don't understand completely yet, but let me know later how it went :) 09:50:09 I'll put up the scripts on github once i have it going :) 09:50:29 Your merge-base magic with the crawl-git repo made my head explode, so I wanted something simpler :P 09:51:52 Napkin: Btw, do you automatically copy the user's 0.7 config when 0.8 is released? 09:52:07 greensnark, is there an equivalent for dgn.in_bounds but for checking mapgrd bounds in the API? 09:52:09 i.e. when 0.8 is released, if I start a game, will I get the stock 0.8 config and macros, or will I get the 0.7 config 09:52:28 Zaba: No idea, you should check :) 09:52:40 I don't see it but I might be missing something subtle. 09:52:58 If it's missing, go ahead and add it 09:54:28 greensnark, it should be quite easy to implement in lua, but what to call it? in_bounds is already taken.. 09:54:40 mapgrd_bounds maybe 09:55:09 We really need some kind of documentation for the Lua api 09:55:34 We have a million functions in different files, including Lua functions, and I keep reinventing my own wheels, to say nothing of other folks' wheels :P 09:56:06 and I have an awesome ossuary idea, I think 09:56:21 but it might turn out to be a little too deadly, too 09:56:25 Looks like the autobots are regularly getting to Lair:8 with FeBe 09:56:37 !hs phybot 09:56:37 46. phybot the Eviscerator (L20 FeBe), worshipper of Trog, shot by a yaktaur (runed bolt) (led by a yaktaur captain) on D:25 on 2011-03-23, with 268493 points after 71102 turns and 2:05:57. 09:56:41 yes, FeBe appears to be rather ridiculous 09:56:46 If that's a bot, that's pretty good 09:56:52 that was with some manual control 09:57:00 stock config, greensnark - there were too many changes and incompatible settings in the past 09:57:03 i was about to cite that in here as another argument for removing extra lives, though, yeah :P 09:57:05 Napkin: ah 09:57:33 Napkin: So nobody complains? 09:57:33 trunk config is never overwritten - hence the Reset Config option in DGL 09:57:45 MMMMMMMM 8 bog mummies 09:57:47 a bit too deadly 09:57:57 CAO shared the same config for every version, but I switched to per-version configs, and now I'm wondering when people will start complaining :P 09:57:59 nobody has complained yet, greensnark :) 09:58:06 might need some tuning... 09:58:13 i don't have per-version 09:58:20 No? 09:58:22 cdo has per branch config 09:58:27 0.6 and 0.7 have separate configs, right 09:58:30 so to speak 09:58:34 Hairsplitting :P 09:58:33 yes 09:58:36 ;> 09:58:37 That's what I meant :) 09:58:40 good :) 09:58:44 just wanted to make sure 09:59:17 MarvinPA: that'd be like removing slowness from nagas or swimming from merfolk... but all of KoBe, HaBe and FeBe are in a dire need of a nerf, yeah 09:59:33 i'm really curious on what your result will be, greensnark :) 09:59:49 wanna copy&paste your new nifty features! 10:00:45 -!- greensnark has quit [Read error: Operation timed out] 10:00:50 -!- greensnark has joined ##crawl-dev 10:00:50 Or at least leave the cat as close as possible to the death spot instead of teleporting 10:01:00 the item restriction is a more interesting gimmick than the extra lives to me, and there were some decent proposals on how to replace extra lives in here a while ago 10:01:02 -!- ais523 has joined ##crawl-dev 10:01:11 i'm really curious on what your result will be, greensnark :) wanna copy&paste your new nifty features! 10:01:22 but right now it's pretty ridiculous and the extra lives make it significantly worse than habe and kobe 10:01:33 I'm still thinking about whether to just copy or to reinstall/reimplement/optimize the new server 10:02:02 being able to get a rune and make it to d:25 with mostly bot control and minimal human interaction sort of proves that 10:02:47 Oh, someone got a rune? 10:02:52 yes, that same run 10:02:53 Which rune, out of curiosity? 10:03:01 I hope it wasn't Shoals :P 10:03:06 snake 10:03:08 !lm phybot type=rune 10:03:09 1. [2011-03-23] phybot the Ripper (L18 FeBe) found a serpentine rune of Zot on turn 52472. (Snake:5) 10:03:12 Ah, that's ok then 10:03:21 but he took over from the bot a decent amount of the time later on I think 10:03:23 Does the bot pick up runes? 10:03:31 "being able" != guaranteed 10:03:34 Oh, this is a bot+human hybrid? :( 10:03:41 * greensnark wants a pure undiluted bot 10:03:42 MarvinPA: you run autoexplore+autofight most of the time on melee chars... with short breaks for running away 10:04:02 the bot on its own can do all of lair quite consistently i think 10:04:05 I'm still working on making the bot as good as possible for getting to lair:8 10:04:20 it can sometimes but usually has no lives or very few lives by then 10:04:29 elliptic: You could make yesno() try a Lua callback to handle the prompt before using the default prompt action 10:04:55 You could also make the message printer call a Lua function for each message so Lua can read messages 10:05:28 Btw, does it handle level-up stat gain yet? 10:05:46 Napkin: Why do you prefer test to [ ] and [[ ]] ? 10:06:23 sorry, I don't actually know much about Lua... are you suggesting changes to crawl code to make the Lua interface better, changes to my script, or both? :) 10:06:31 can we ditch extra lives on felids? I think the item restriction is good enough. Counteract it by buffing felid hp to kobold levels or similar. 10:07:09 elliptic: Both :) 10:07:23 This bot is a great opportunity to improve the user Lua api 10:07:37 Since otherwise nobody uses it 10:07:40 it handles level-up stat gain by using 'd' as the macro key :P I would love it if there was a better way of doing that sort of thing... 10:07:40 Well, autofight apart 10:08:07 I wonder what kind of CPU usage the bots cause :) 10:08:11 hmm, does this prove my point that Crawl bots are easier to write than NetHack bots? 10:08:36 I'd say it proves Fe are breaking Crawl :P 10:08:37 there was some anecdotal evidence that using the bot in sprint is a bad idea 10:08:40 anyway, something I'd like in the Lua API (or just as an in-game command) for bot use: the ability to inscribe monsters 10:08:58 elliptic: that's unsurprising, Sprint's a lot more tactical than regular Crawl 10:08:59 Naming monsters like NH, you mean? 10:09:02 yep 10:09:07 ais523: no, I mean the cpu usage 10:09:07 just to keep track of which is which between turns 10:09:12 humans can normally guess 10:09:12 because clouds clouds clouds 10:09:14 elliptic: ah 10:09:42 ofc, the current Crawl bots aren't trying to use tactics, and I prefer to try to make tactical-type bots 10:10:06 casmith789: uhm... are you seriously thinking about that? With no gear, 1.5 ogre hp would make them a challenge species. 10:10:26 1.5 ogre hp sorry? 10:11:02 what? they're fast, they have good apts, they have carnivore 10:11:07 no armour isn't that big a drawback with their intrinsics, yeah 10:11:12 they're easy and extra lives just let you play carelessly 10:11:13 they do get jewellery after all 10:11:16 felids are good enough without extra lives 10:11:26 and a win-rate in the middle of most species 10:11:35 (without greatplayers) 10:11:36 people speedrun them 10:12:24 do you seriously say a majority of players try speed runs every time? 10:12:26 so, what is it about felids, other than the extra lives, which make them decent despite the crippling slot and item use restrictions? 10:12:29 03kilobyte * rf37324e74f1b 10/crawl-ref/source/ (aptitudes.h player.cc): Another FeBe nerf. 10:12:30 I think that they would certainly be decent with no extra lives and higher hp... maybe not that interesting a race though 10:12:38 ais523: they aren't crippling 10:12:40 03kilobyte * rb111cea2eb17 10/crawl-ref/source/player.cc: Make kitty xp aptitude more godly -- ie, less lives. 10:12:42 kilobyte: what did you just nerf? 10:12:47 apts? 10:13:28 Fighting and UC, xp apt 10:14:22 are you just going to keep nerfing felid apts forever :P 10:14:27 ais523: EV, speed 10:14:35 tweaking apts doesn't really change much 10:14:58 casmith789: EV is smaller than kobold's 10:15:01 they're still fast which counts for way more 10:15:52 spriggan fast or centaur fast? 10:16:01 no 10:16:01 kilobyte: I thought they were kobold-size? 10:16:05 try a naked kobold... you'll have +2 apt in most apts, a good deal more hp, wands and so on 10:16:06 slower than centaurs but faster than normal races 10:16:21 kilobyte: yes I like kobolds 10:16:23 ais523: Fast 1 10:16:34 hmm, so why is febe so much better than kobe? 10:16:38 KoBe are broken beyond comparison 10:16:41 it is worse 10:16:42 try them... 10:16:43 naked kobold doesn't get free resists and swiftness and a bit of AC 10:16:47 kilobyte: ah 10:17:11 elliptic: one level of rC 10:17:12 kobe aren't "broken beyond comparison". But they are good 10:17:16 (be is one of the best classes for bots, by the way, because what Trog does is the sort of thing a bot can exploit very easily) 10:17:28 (not necessarily because it's inherently overpowered as a class) 10:17:49 not many characters can just enter Vault:8 and press autofight until everything is dead... KoBe can 10:17:58 no it can't 10:18:05 kilobyte: uh, have you tried that or are you just making it up? 10:18:23 kilobyte: with which buffs active? 10:18:31 ais523: with none 10:18:33 I hope you were at least berserk 10:18:37 what char was this 10:18:39 elliptic: what for? 10:18:45 killing things 10:19:07 seriously, answer the question of whether you have actually done this :P 10:19:47 at one moment I got a bit afraid and thought to do it in a paranoid way (no rN), and didn't finish purely by autofight 10:20:02 kobe are definitely great but so are most berserkers 10:20:13 was this character online? can we see how overpowered it was? 10:20:14 and most berserkers are great without having extra lives and swiftness 10:20:42 sorry, you just held down autofight on all of v:8 and cleared it did you 10:20:54 I am having trouble with what you mean because you can't have meant that 10:21:38 casmith789: no, not all, I decided to eat my hubris and play the usual way, but I retreated with more than half of hp when less than 1/3 guards were still alive 10:21:50 oh. The starting area 10:22:00 uh, that doesn't sound *at all* like "just enter Vault:8 and press autofight until everything is dead..." 10:22:02 I've killed every vault guard with MDBe easily 10:22:07 so please stop misleading us 10:22:16 I've also killed every vault guard with shatter without them moving 10:22:19 auto_fight_. Not auto_explore_. 10:22:30 autofight can move 10:22:31 "I retreated with more than half of hp when less than 1/3 guards were still alive" 10:22:34 but you said all of v:8 10:22:41 not 2/3 of the initial area 10:22:55 seriously, admit that you were exaggerating terribly and let's move on 10:22:56 all of v:8 you can fight with no walking 10:23:02 but you didn't, did you 10:23:37 there was one yaktaur and two shadow dragons IIRC 10:23:38 and often more stuff wakes up and comes in, but apparently you didn't wait for that 10:23:56 the yaktaur is peanuts, but shadow dragons would take several XLs 10:24:14 seriously, if you didn't "just enter Vault:8 and press autofight until everything is dead", don't say you did 10:24:52 you CAN NOT win by autofight alone -- it's pretty obvious what I meant 10:25:08 even shield of the gong can't bring everything to you 10:25:34 "everything is dead" != "less than 1/3 guards were still alive" 10:25:45 what about using tt as well? or repeated projected noise on your own square? that wouldn't be a large change to make 10:25:46 and shield of the gong does a pretty good job of bringing stuff to you 10:26:02 getting back to the point somewhat, i think felids could still be an interesting race with a different mechanic than extra lives, i think greensnark suggested that they could have something more like lifesaving, or get boosts when they're low on health 10:26:22 btw, when it says "Noise: Deafening" in projected noise's description, it's rather unclear to an unspoiled player whether that's at the aimed location, or the current location 10:26:27 whether or not kobe is overpowered is kinda beside the point 10:26:40 it's obvious I could have killed all of them but remember that EV works only stochastically. A couple bad rolls -- kobold splatters. 10:27:05 hmm, felids without extra lives is a bit like lava orcs without swimming in lava 10:27:07 MarvinPA: +1 10:27:14 but I'm not sure if that's an issue 10:27:14 my claim is, FeBe are far weaker than some other chars 10:31:53 kilobyte: well, I'll agree that felids in general aren't the right choice if you want a really overpowered character late game 10:33:14 the value of being able to wear randart armour and use wands of healing increases as the game progresses because you actually have such items :) 10:33:56 but I'm not really sure why this is relevant to the question of whether extra lives is a good mechanic 10:34:24 so we have a race that is strong early on and weak later 10:34:39 personally I think it is a more interesting and flavorful mechanic than most of the alternatives I've heard suggested 10:34:59 extra lives are only abusable finitely many times 10:35:06 multiply that by berserkers who are strong early on and medium later 10:35:25 really, they have much the same effect as extra HP, in that they let you play more riskily without dying 10:35:25 well that finitely many number of times is plenty to get 15 runes 10:35:34 just they're more obvious 10:35:38 felids are really mainly strong early on after you get one life, too 10:35:49 getting to L4 can be rather annoying with non-berserkers 10:35:59 since low HP 10:36:19 and unarmed at low skill isn't as impressive 10:36:33 hmm, it might be insightful to graph at what levels most deaths occur 10:36:40 got to go though 10:36:50 overall? XL 1 and 2 10:37:03 !lg * ktyp!=winning s=level 10:37:05 berserkers have a different distribution, of course 10:37:09 ais523: xl 10:37:12 is the Sequell's database downloadable somehow? 10:37:13 thanks 10:37:16 !lg * ktyp!=winning s=xl 10:37:22 !lg * class=be ktyp!=winning s=xl 10:37:30 elliptic: of course, I meant relatively to other races 10:37:34 also sequell is really slow currently for some reason 10:37:54 can't be used for advanced queries 10:37:54 kilobyte: yeah, it is an interesting question 10:38:05 hmm, all interesting data will be on CDO 10:38:48 the logfile is only 73MB 10:38:54 Unknown selector level 10:38:54 959056 games for * (ktyp!=winning): 444678x 1, 108037x 3, 107150x 2, 80735x 4, 51781x 5, 35363x 6, 27955x 7, 23633x 8, 20751x 9, 16910x 10, 12084x 11, 7268x 12, 4793x 13, 3906x 14, 3049x 15, 2333x 16, 1623x 17, 1461x 27, 1232x 18, 962x 19, 763x 20, 570x 21, 491x 22, 456x 23, 400x 24, 340x 25, 325x 26, 7x 0 10:38:57 milestones 142MB 10:38:59 54318 games for * (class=be ktyp!=winning): 12937x 1, 6715x 3, 6346x 2, 5044x 4, 4122x 5, 3378x 6, 3097x 7, 2992x 8, 2595x 9, 2092x 10, 1463x 11, 865x 12, 540x 13, 463x 14, 368x 15, 284x 16, 191x 17, 146x 18, 126x 19, 118x 27, 79x 22, 79x 20, 68x 21, 64x 24, 64x 23, 47x 25, 35x 26 10:39:02 Unfortunately the machine Sequell is running on has been pretty loaded recently 10:39:39 !lg * crace=fe ktyp!=winning ktyp!=quitting ktyp!=leaving s=xl 10:39:40 3047 games for * (crace=fe ktyp!=winning ktyp!=quitting ktyp!=leaving): 946x 1, 685x 2, 510x 3, 153x 5, 126x 4, 111x 6, 82x 8, 81x 9, 66x 11, 50x 10, 49x 12, 42x 7, 27x 14, 25x 15, 17x 17, 16x 13, 14x 18, 12x 16, 6x 23, 5x 19, 5x 21, 5x 24, 4x 20, 4x 27, 3x 26, 2x 22, 1x 25 10:39:48 it seems that even berserkers have a tendency to mostly die early 10:39:53 -!- galehar has joined ##crawl-dev 10:40:02 !lm * crace=fe type=death s=xl 10:40:03 Napkin: Why do Debian sudos not have -n option :( 10:40:04 1817 milestones for * (crace=fe type=death): 286x 4, 187x 10, 170x 8, 158x 7, 149x 5, 138x 11, 122x 9, 101x 6, 72x 12, 68x 14, 67x 13, 51x 16, 50x 17, 39x 15, 29x 18, 24x 19, 20x 20, 18x 27, 14x 23, 13x 21, 12x 22, 11x 25, 10x 26, 7x 24, 1x 3 10:40:19 -!- valrus has joined ##crawl-dev 10:40:22 honestly? when I started writing I never understood the difference between [] and [[]] and forgot the closing brackets too often 10:40:23 ais523: yeah... just lemme toss them gnuplot's way 10:40:25 ;) 10:40:29 oh? -n does what? 10:40:34 greensnark: Looks like CAO is only running 0.7.1. Is that intentional? dgl says 0.7.2 10:40:34 greensnark: Ubuntu sudo seems to have -n, at least 10:40:39 greensnark: why does sudo exist at all? 10:40:40 !lg * crace=ko cv>=0.8 ktyp!=winning ktyp!=quitting ktyp!=leaving s=xl 10:40:41 2606 games for * (crace=ko cv>=0.8 ktyp!=winning ktyp!=quitting ktyp!=leaving): 522x 1, 404x 2, 371x 3, 291x 4, 207x 5, 184x 6, 119x 7, 114x 8, 79x 9, 79x 10, 68x 11, 39x 12, 18x 27, 18x 15, 17x 14, 16x 13, 11x 20, 10x 17, 9x 16, 8x 18, 8x 19, 3x 25, 3x 24, 3x 23, 3x 22, 2x 21 10:40:46 and -n makes it error out rather than ask for a password if what you're doing requires a password 10:40:48 don't tell me it's a bsd extention? ROFL! 10:40:58 I don't see an 0.7.2 binary; should we generate one, or does that version not exist yet and the text in DGL is wrong? 10:41:01 Napkin: No, it isn't, it's a non-motheaten-sudo thing :P 10:41:13 so I suspect it's just a very old version of sudo 10:41:15 rax: Sorry, I ripped off CDO's menus and I haven't installed 0.7.2 on cao yet 10:41:31 cao is somewhere betwen 0.7.1. and 0.7.2 currently 10:41:34 I can build it if you remind me how to update the source tree with git ^^;;; 10:41:35 greensnark: I guess it's a patch specific to Ubuntu. They use it everywhere. 10:41:47 kilobyte: My Mac also has -n 10:41:52 * rax still has not learned git and thus has completely lost track of crawl source. :( 10:41:54 greensnark: ah 10:42:05 rax: git pull should update it from the place it's set to pull from by default 10:42:09 !lg * cv>=0.8 ktyp!=winning ktyp!=quitting ktyp!=leaving s=crace / xl>3 o=-% 10:42:11 22196/48955 games for * (cv>=0.8 ktyp!=winning ktyp!=quitting ktyp!=leaving): 906/3047x Felid [29.73%], 563/1748x Vampire [32.21%], 2170/6596x Demonspawn [32.90%], 1300/3447x Mummy [37.71%], 596/1508x Naga [39.52%], 740/1820x Kenku [40.66%], 1076/2579x Sludge Elf [41.72%], 810/1940x High Elf [41.75%], 1739/4144x Deep Elf [41.96%], 317/682x Ghoul [46.48%], 311/669x Demigod [46.49%], 625/1323x Ogre ... 10:42:16 CAO's sudo is astoundingly old, because CAO's everything is astoundingly old. 10:42:37 ais523: Cool, trying that. 10:42:40 I'd be quite hard pressed to name a single case where sudo is a good idea. 10:42:48 rax: One moment, I'm trying to remember where I put the source :) 10:42:49 hopefully it's pointing at gitorious rather than sourceforge 10:42:58 /home/crawl/git-sources/ 10:42:59 rax: etch has had no security support for a long time already 10:43:04 felids unsurprisingly are the race with the highest proportion of deaths at xl <= 3 10:43:15 but not by a huge margin or anything 10:43:16 rax: and you do not want that on a public server 10:43:41 -!- upsy has quit [Ping timeout: 255 seconds] 10:43:55 kilobyte: anyway, dgl run stuff in chroot, fwiw 10:43:58 runs* 10:44:01 kilobyte: Thanks, mom. <3 10:44:06 in full, it would be "git pull git://gitorious.org/crawl/crawl master" 10:44:10 (I know, it's just not a high priority for me) 10:44:15 ...well, it pulled from sourceforge. 10:44:23 (I jsut ran "git pull") 10:44:27 try "git pull git://gitorious.org/crawl/crawl master" 10:44:44 Not sure that's the right thing to do in a branch directory? 10:44:54 branches don't have directories in git 10:44:57 but.. honestly.. you only need -n if you don't manage to configure sudo correctly ;-P 10:44:58 Perhaps I will wait for greensnark. 10:44:59 but right, you wouldn't want latest dev source 10:45:04 rawr! The first step in fight rewrite is done! Attack class implemented, melee_class in its own file...now fight.cc is a pathetic 500 lines. 10:45:10 paxed: 1. chroot does quite little for security, 2. I doubt there are DGL-specific attacks, I'd be afraid about Apache and such. 10:45:11 rax: You can edit .git/config and point at the gitorious URL there 10:45:19 After which a git pull will do the right thing 10:45:45 Non-Verbal Item Descriptions (https://crawl.develz.org/mantis/view.php?id=3653) by golan2072 10:45:47 omg, so noisy in here! 10:45:51 rax: I can do it, unless you're trying to get familiar with the git setup :) 10:45:52 nonetheless, CAO probably doesn't want the latest source, but an older branch 10:46:05 greensnark: I will leave this to you for now. :) 10:46:09 * Napkin heads out to continue burning in the sun * 10:46:21 Ok, I'll do a quick upgrade to 0.7.2 :) 10:46:28 rax - do it now, if anything goes wrong you have that greensnark guy to fix it for you ;) 10:46:33 I was teached that chroot was never designed to be a security tool and mistaking it for one is quite suicidal. 10:46:36 Napkin: <3 10:46:37 kilobyte: chroot works quite well when there's nothing to run in it except the game binary :P 10:47:13 kilobyte: AS it turns out I track Debian security updates for my job. :P There are a couple of things that are an issue, but nothing I care deeply about. Upgrading the machine to lenny is on my tasklist, but it's not a super high priority. "CAO's users are getting service commensurate with its cost to them" 10:47:38 (And yes, lenny is oldstable, but there are issues with squeeze and xen right now) 10:47:47 (at least on my machine, which is where CAO is, so) 10:47:51 paxed: I can count 4 Turing-complete tools with full access to the filesystem on CDO 10:48:08 squeeze+xen+nvidia seems borked 10:48:17 kilobyte: well, it ain't my problem you people pollute the chroot :P 10:48:30 wtf it works 10:48:40 er. 10:48:40 Zaba: ? 10:48:58 I've written some vault lua code and it works. 10:49:04 to my own surprise 10:49:16 ah 10:49:16 paxed: you do not need helper tools if you get a buffer overflow 10:49:31 or, can write to any place on the filesystem and execute that 10:49:42 -!- ais523 has quit [Remote host closed the connection] 10:49:49 ld means no execute bit is not a problem 10:51:13 Actually, while I'm here and thinking about it, does anyone have a book or reference for git that they recommend? Preferably one target at someone who wants conceptual understanding rather than a list of commands. 10:51:16 kilobyte: sure. as the chroot doesn't help at all, i'll remove it from the next dgl then, right? :D 10:51:26 ok, going home. I'll try to get the nice graphs later on. 10:52:27 rax: http://www.eecs.harvard.edu/~cduan/technical/git/ 10:53:45 greensnark: Holy wow that is _exactly what I have wanted all of this time_. Or at least it appears to be. Bookmarked, thank you so much! 10:54:49 rax: I'm not sure if it's actually good, I haven't read it myself :) 10:55:22 -!- valrus_ has joined ##crawl-dev 10:55:37 "you can only really use Git if you understand how Git works" <-- that. 10:56:20 Well, taking the time to do that is like taking the time to learn a decent editor instead of using that notepad of a VCS that is svn :P 10:56:35 I mean, I feel that way about most things; until I seriously understood SVN, it scared me and I used RCS isntead. 10:56:40 Which was _terrible_ 10:56:49 Some folks find mercurial nicer to learn that git, which is also fine, they're both great for source control 10:57:02 What you really don't want to do is to keep punishing yourself with svn :P 10:57:27 I use svk for personal stuff. It works fine. If I started a new repository I would force myself to learn something new, but it doesn't seem worth migrating away from. 10:57:45 rax: svk has some seriously wrinkly aspects, though 10:57:57 just like you, greensnark :D 10:58:16 paxed: Which comment was that a reply to? :P 10:58:22 guess ;) 10:58:46 Damn inscrutable finns :P 11:00:04 -!- upsy has joined ##crawl-dev 11:00:30 greensnark: For better or for worse, I actually understand it (I did some doc work for it a while back), and so it feels comfortable. 11:02:09 git really helps me a lot keeping local changes checked for software installed manually (wordpress/mantis/wiki/forum) 11:02:23 that's probably exactly the reason you used SVK? 11:02:41 *use 11:03:18 Yes. 11:03:54 Especially back when I made changes to the crawl on CAO before they existed upstream. 11:04:02 I used svk a fair amount for crawl in the early days, and I think git is a lot simpler to use 11:04:07 svk is scarily complex :) 11:04:28 See, I feel the opposite. You are probably _right_, so hopefully git won't be too hard. ;) 11:04:30 probably doesn't matter if you understand how it works 11:08:12 greensnark, ok, there is dgn.is_valid_coord which checks for validity within mapgrd 11:10:58 Zaba: Aha 11:11:11 Might be nice to reference mapgrd in the name though 11:11:39 ...now I need to find a way to pick a random cell with a given glyph and get its coordinates... 11:12:08 maybe I should just place a marker (but, if all I need is the coordinates, what kind of a marker would it be?) 11:12:43 Any props marker 11:12:50 You can find a marker with a given property 11:13:08 off home o/ 11:13:10 But if you're searching for a specific feature, why use a marker 11:13:35 I need the coordinates 11:16:18 easiest would be to simply iterate over the whole map and choose one, but there's bound to be a better way, right? 11:16:51 A marker might be somewhat faster, yeah 11:16:55 But I have no idea how much faster :P 11:17:12 The game will still traverse all markers and call out to Lua to query the marker's properties 11:17:41 greensnark, also.. when are subst/shuffle/etc. applied, exactly? 11:17:50 When they're executed 11:17:51 will the values in mapgrd differ before and after calls to them? 11:17:55 Yes 11:18:01 ok.. kinda messy. 11:29:05 Ok, cao has 0.7.2. 11:29:17 Although the help menu wants to show... 0.6 highlights, ugh 11:29:24 What file is that :P 11:31:00 What file is which? 11:32:43 I was wondering which file crawl was finding "0.6 Highlights" from, given this is a 0.7 install 11:33:23 And it turns out command.cc is doing some really weird stuff with hardcoded version names 11:33:32 Hmm. They should point to entirely different data directories --- ahh. 11:33:43 Apparently it looks for a section titled "Highlights" 11:35:49 Ok, fixed! 11:35:58 I'd carelessly neglected to copy the 0.7 changelog in :P 11:47:47 I've just made a really horrible ossuary that is really a cross between an ossuary and a sewer 11:47:56 and it has an overengineered lua chunk in it 11:47:58 am I becoming due? 11:56:28 greensnark, isn't this quite horrible: http://sprunge.us/IUGD 11:56:52 at least I didn't copy and paste any of that :P 11:58:31 Utility functions shouldn't be in des files :P 11:58:48 greensnark, all of that is of quite.. specific purpose 11:59:06 Looks like you could extract a utility function that takes suitable parameters, though 11:59:09 although I'm sure it can be adequately converted into something more reusable 11:59:36 Don't forget you can pass functions to other functions :P 11:59:44 flood_single should be passed to flood 12:00:24 'intensity' is really a misnomer.. but I'm not sure what it should be instead 12:00:58 or indeed how to make it into something less specific to flooding 12:05:46 greensnark, seems like 'flooded' is not needed, since keeping data in 'queued' is enough to prevent visiting one cell more than once 12:15:13 intensity should be called range or distance.. and if I want to make this reusable, I need a better way to handle it, too 12:26:27 greensnark, Napkin: The password updating instructions I got yesterday return errors; the command "mkpasswd" is not found, and "Error: unable to open database "passwd.db3": unable to open database file 12:26:41 Am I doing something crazy? 12:27:40 rax: passwd db is called dgamelaunch.db under /dgldir/ for cao 12:27:55 The mkpasswd thing is probably not installed, let me check 12:28:26 > dpkg -S /usr/bin/mkpasswd 12:28:26 whois: /usr/bin/mkpasswd 12:29:10 Do I need to install that, then? 12:29:19 paxed - how about adding a small script to reset user passwords? just as a convenience tool 12:29:35 you are running.. etch, right? 12:29:44 rax: You can use perl -le 'print crypt($ARGV[0], substr($ARGV[0], 0, 2))' instead 12:30:05 geek! 12:30:06 I am currently running etch, yes. :P 12:30:13 both of you! ;) 12:30:18 03galehar * r94a54098fec8 10/crawl-ref/source/map_knowledge.cc: Ashenzari: scale passive mapping density with piety. 12:30:31 03galehar * rbdb70b7b5e00 10/crawl-ref/source/ (godpassive.cc godpassive.h player.cc skills2.cc): New Ashenzari passive skill boost. 12:30:31 Wait, someone still using etch is obviously trying to *not* look like a geek :P 12:30:32 03galehar * r6854920e05d0 10/crawl-ref/source/ (godpassive.cc tilepick-p.cc transform.cc): Meld cursed weapons on transform. 12:30:33 03galehar * r7832d4d99ca3 10/crawl-ref/source/ (skills2.cc skills2.h): Add a function to compare skill levels. 12:30:52 hmm, packages.debian.org doesn't allow me to see package content of etch packages anymore.. grrr 12:31:13 OK, now I get "Error: near line 1: attempt to write a readonly database 12:31:19 i'd say go with greensnark's perl suggestion 12:31:30 rax: sudo -u games? 12:31:36 no unix write permission? 12:31:55 there is nothing else that sqlite cares about 12:32:12 Let me quickly take a backup copy of the db now :P 12:34:08 I used the perl thing. 12:34:13 I tried it as _root_ 12:34:42 I will wait to try anything else until you say it is OK. 12:34:54 rax: Can you PM me the command-line you used? 12:38:21 galehar: liking the look of those ash changes 12:38:57 hopefully i'll manage not to screw up my next decent character and actually give them a try :P 12:41:16 greensnark, I wonder how to handle that intensity/distance/range/whatever so that it's the best API-wise. 12:43:00 greensnark, also, when I move the function from the vault definition, I can't call is_valid_coord anymore... how should I deal with that? 12:50:09 Miaow 12:50:19 Oh, is_valid_coord presumably wants the map as first argument? 12:50:30 yes 12:50:46 I guess I want the function to be passed _G from the vault 12:51:10 That's one way of doing it 12:51:27 The other way is to put the function in a namespace like 'dgn' that is auto-setfenved 12:51:40 If you get what I mean :) 12:52:17 well, just defining it in the dgn namespace doesn't seem to be enough for that 12:52:28 You have to call it unqualified 12:52:33 ah. 12:52:48 i.e. any dgn.foo(, ...) will be bound to be callable as foo(...) from a map def 12:52:55 I see 12:53:14 That's how maps can have Lua calls like tag('blah') without specifying the map: it's implied 12:53:25 And when you pass _G, you're passive the environment that contains those function defs 12:53:43 how do I call is_valid_coord then? It tells me I can't index an userdata when I try to call it on the first argument I get (the map) 12:53:44 So an external function can use _G.tags('blah') and it'll use the wrapped map 12:53:53 Can you show me your code 12:54:16 http://sprunge.us/JiNV 12:58:33 How are you calling flood 12:58:51 Are you guys familiar with mounting a server drive locally (linux server, linux workstation) via fstab? 12:59:27 Cryp71c: You want to mount a remote server's drive? 12:59:30 greensnark, I got it to work: I need to do dgn.is_valid_coord(map, { x = ..., y = ... }) 12:59:38 I call flood as just flood(...), with no dgn or _G 12:59:53 Ah, ok 12:59:57 greensnark: yeah, I use it to monitor the drive's full-ness 13:00:03 ok, now I just need to figure out how to make the interface of 'flood' better 13:00:05 Yeah, that's because flood() doesn't inherit the global environment 13:03:14 is there any chance of tiles ever being able to play on cao/cdo 13:03:24 discussion about it came up on SA 13:03:58 -!- Kurper has joined ##crawl-dev 13:04:27 Last I heard someone was making progress on nettiles, but that was 4 months ago. 13:04:57 ye it had a lot of bugs too 13:05:34 https://crawl.develz.org/wiki/doku.php?id=dcss:brainstorm:internal:clientserver 13:05:45 someone was working on that stuff i believe 13:06:13 does actor->attack(actor& target) seem like a reasonable replacement (given sufficiently elegant code) for fight.cc's 3 "you_attack(...)", "monster_attack(...)", and "monsters_fight(...)" 13:06:15 but i don't think anyone's actively working on it at the moment 13:06:24 so it's theoretically possible but a long way off 13:06:38 no chance of tiles players competing in the next tournament then? :P 13:06:49 they could compete by playing ascii :P 13:06:53 heh 13:07:34 heh, the irc hatred is pretty amusing 13:07:44 irc hatred? 13:08:45 somethingawful forum hating on ##crawl, just normal internet dumbness really 13:10:00 Mu_: Someone would need to follow up on Luca's work 13:10:21 is that the nettiles guy 13:10:26 indeed 13:10:32 I think there's about 1-2 months' full-time effort to build a real net-tiles that supports the full tile interface 13:10:42 Yeah, he wrote a page on the wiki outlining the steps involved 13:10:55 * 1-2 person months 13:11:27 Problem of course is that all the tiles devs are busy :P 13:11:37 And none of the *other* devs have an incentive 13:12:30 But there are some good programmers on SA, I think. Just need to get someone there interested :P 13:12:43 "After a first solution is working, the Crawl codebase should be refactored, so that the game, the clients and the servers are all easily instantiable C++ objects, the source code of which is in separate subdirectories." 13:12:47 pfft sounds totally easy 13:13:07 I recall some goons getting a tiles port up based on Mitsuhiro's work at about the same time Enne was working on it 13:13:23 So there's definitely ability on SA, just need to harness it :) 13:13:25 but yeah, link them to that wiki page and get them to contribute :P 13:14:29 ok 13:18:06 -!- monqy has joined ##crawl-dev 13:21:21 yeah, crawl should be refactored so that server, client, common and game code is separated, with all of the game code compiled into special RISC bytecode that gets just-in-time-recompiled into appropriate native code. 13:21:25 that would be grand! 13:22:15 :P 13:22:28 oh, did I mention lots of unportable hacks in the networking code? 13:22:38 that's absolutely a requirement, too 13:22:54 Are you referring to net-tiles or just being Mr. Grumpy Old Man? :P 13:23:42 grumpy would be suggesting to port crawl to QuakeC 13:24:01 Oh, you're looking at the fifth milestone stuff 13:24:18 I think that's just in there to include a suitably visionary apocalyptic feel to that page 13:24:53 crawl as a facebook app, the future. 13:24:58 I think everything after and including "third milestone" is an exercise in science fiction writing best left to the point in time when we have a working client/server model 13:25:10 yeah, it really looks like that... 13:28:05 Hey pay me for a year and I'll do it! 13:31:34 @??bog mummy 13:31:34 bog mummy (07M) | Speed: 8 (swim: 140%) | HD: 6 | Health: 31-47 | AC/EV: 1/9 | Damage: 25, 412(cold:6-17) | Flags: 07undead, amphibious, evil, spellcaster | Res: 06magic(40), 05fire, 02cold, 03poison, 12drown, 04rot+++, 13neg+++, 13torm | Vul: 08holy++ | XP: 273 | Sp: throw frost (3d6), slow, corpse rot. 13:31:45 @??mummy 13:31:45 mummy (15M) | Speed: 6 | HD: 3 | Health: 15-24 | AC/EV: 3/6 | Damage: 20 | Flags: 07undead, evil | Res: 06magic(20), 02cold, 03poison, 12drown, 04rot+++, 13neg+++, 13torm | Vul: 04fire, 08holy++ | XP: 18. 13:39:20 @??bog mummy hd:3 13:39:21 bog mummy (07M) | Speed: 8 (swim: 140%) | HD: 3 | Health: 15-24 | AC/EV: 1/9 | Damage: 25, 412(cold:3-8) | Flags: 07undead, amphibious, evil, spellcaster | Res: 06magic(40), 05fire, 02cold, 03poison, 12drown, 04rot+++, 13neg+++, 13torm | Vul: 08holy++ | XP: 51 | Sp: throw frost (3d5), slow, corpse rot. 13:39:31 -!- ZChris13 has joined ##crawl-dev 13:44:21 03Cryp71c 07unified_combat_control * r13ab76826679 10/crawl-ref/source/fight.cc: Remove unnecessary #includes from fight.cc 13:54:46 hmm, i have absolutely no idea what ashenzari actually does at each piety level now 13:55:02 should probably update the religion screen and such 13:56:15 03zaba * r3239e6458bef 10/crawl-ref/source/dat/des/branches/ (orc.des snake.des temple.des): Remove unnecessary DEPTHs from some branch entry vaults. 13:56:25 03zaba * r9dae1817f964 10/crawl-ref/source/dat/ (clua/dungeon.lua des/portals/ossuary.des): Add a new ossuary. 13:58:14 03MarvinPA * rda8ded80a5b8 10/crawl-ref/source/godpassive.cc: Fix message for Ash skill boost 14:04:59 xw (L6 FeBe) ERROR in 'fixedvector.h' at line 72: range check error (-1 / 52) (D:4) 14:16:32 -!- valrus has quit [Remote host closed the connection] 14:23:02 Added a bit more content, for anyone who cares to review it: https://crawl.develz.org/wiki/doku.php?id=dcss:brainstorm:fight_rewrite 14:23:04 afk for a bit 14:32:59 -!- ZChris13_ has joined ##crawl-dev 14:37:00 -!- ZChris13 has quit [Ping timeout: 264 seconds] 14:37:10 -!- ZChris13_ is now known as ZChris13 14:55:37 Zaba: quiet about Quake until we sneak Quad Damage into normal games :p 14:56:26 greensnark: do you have the code to parse and import logfile data into SQL? (Perhaps just 10 minutes of coding, but still...) 15:02:57 -!- Nexos has joined ##crawl-dev 15:04:11 I gained FireMagic skill from Sage card effect with 0 Spc/Fire... Is it normal that skill was selected (I'm a tank)? 15:07:26 -!- eith_ has joined ##crawl-dev 15:07:46 greensnark: was reading scrollback, at least fight.cc will be (mostly) instantiable C++ objects after I'm done with it!@ 15:07:58 (with regard to nettiles and Luca's quote) 15:08:25 -!- eith has quit [Ping timeout: 252 seconds] 15:09:18 instantiable objects would be very useful for multiplayer Crawl... for nettiles, not really 15:09:21 DO we have a weapon class anywhere, I'm not finding it in my grep. 15:09:38 it's about separating view from the model 15:09:50 Cryp71c: no, a weapon is just an item 15:10:03 kilobyte: right, I understand, was just referencing what marvin had pasted before when they were talking about luca 15:10:11 kilobyte: so we have an item class? 15:10:54 (I'm not seeing that either, in grep) 15:11:04 It seems like both should be a class, or at least item. 15:11:22 -!- eith_ is now known as eith 15:12:11 uhm... item_def... and it's probably the single most used class all around 15:12:52 (and it is passed over very often, so making it virtual is likely to lead to problems) 15:16:14 Nexos: yes, at low card power it's more random which skill you get saged 15:17:15 kilobyte: ah, my grep was insufficient and I wasn't thinking about item_def :) But just because a weapon class would inherit from item_def doesn't mean item_def would need to be pure virtual. 15:17:23 but that's irrelevant to my rewrite, was just curious. 15:18:50 MarvinPA: OK just seemed odd 15:20:07 this includes weapon skills for Felids and invocations for Demigods 15:20:10 -!- casmith_789 has joined ##crawl-dev 15:20:26 -!- casmith789 has quit [Ping timeout: 255 seconds] 15:21:16 -!- casmith_789 is now known as casmith789 15:22:24 do you guys think Sage should never pick totally useless skills? 15:23:33 kilobyte: it makes sense to me that at low card power its pretty random, so long as that randomness decreases sharply with each card / evo level 15:23:44 greensnark: poke poke 15:25:16 hrm, trying to identify functions that should be moved to the player/monster/actor classes. 15:26:12 I'm talking about totally useless ones, not just unlikely to be of use (like magic for a GDA-clad minotaur) 15:29:06 why is it that autopickup doesn't works when it can see sleeping harmless monsters -- like it works if it can see a fish behind a glass wall, but not if the fish is asleep 15:29:58 i wouldn't mind it not picking invoc on demigods and weapon skills on felids, not too bothered either way though 15:30:28 either way i liked the idea of making the status remain until you've gained a certain amount of xp 15:30:41 so that you can't just rest it out if you get a sage you don't want 15:31:39 add a cancel sage (a)bility? 15:31:55 or make it depend on exp gained rather than time spent 15:32:09 or that, yes, but having it depend on xp would be more interesting i think 15:32:37 fits better with the riskiness of decks of wonders 15:35:58 -!- evilmike has joined ##crawl-dev 15:41:27 -!- Nexos has quit [Quit: Ex-Chat] 15:50:07 Eronarn: regarding monster melee_attacks, it would be better to treat each physical attack as a separate instantiation of attack (and thus, melee_attack), instead of using the which_attack variable that's currently being used. 15:50:39 Correct? I feel like i've mentioned this to you, snarky, or someone else earlier this week. 15:57:52 might be a good idea 15:58:32 03kilobyte * r7f9d474e6b7a 10/crawl-ref/source/cloud.cc: Make poison clouds all lightgreen. 15:58:33 03kilobyte * r0440419004af 10/crawl-ref/source/ (acquire.cc decks.cc skills2.cc skills2.h): Don't acquire manuals or sage to utterly useless skills. 15:58:34 03kilobyte * r833bced19427 10/crawl-ref/source/directn.cc: Don't include unknown mimics in beam targetting. 15:58:34 03kilobyte * rcd066b191156 10/crawl-ref/source/player-stats.cc: Cap stat loss at -100 (elliptic) 15:58:38 03kilobyte * rdf38e3068a52 10/crawl-ref/source/acquire.cc: Manual acquirement: mark more gods as not using Invocations (half the weight). 16:02:35 -!- galehar_ has joined ##crawl-dev 16:04:16 03galehar * rdf0302c5b7f3 10/crawl-ref/source/map_knowledge.cc: Unerf low piety passive mapping. 16:04:45 should Vehumet reduce food cost of big conjurations as well? 16:04:53 -!- Cryp71c has quit [Quit: Leaving] 16:05:02 kilobyte: about passive mapping, what you said was already needed anyway. Passive mapping range scaled with piety, I just scaled density. 16:05:55 hmm, right! Thanks for correcting me, I didn't realize it was already broken. 16:06:03 galehar: currently you get messages about becoming an avatar of ashenzari as soon as you join, should it be limited to a certain piety level? 16:06:44 Veh doesn't need buffing 16:07:06 and yeah the mp reduction is really useful, i think food reduction too would be a bit much 16:08:17 03Cryp71c 07unified_combat_control * r8cf1acf2f02e 10/crawl-ref/source/ (cloud.cc fight.cc): Cleanup unnecessary wrapper methods 16:08:27 03Cryp71c 07unified_combat_control * r9596e462eaa8 10/crawl-ref/source/ (attack.cc attack.h melee_attack.cc melee_attack.h): Move melee_attack properties to attack 16:08:46 doing Swamp with level 9 spells: even with gourmand but no rPois, you can use your big gun once in a large fight, and you still suffer food problems 16:09:48 swamp has a lot of edible corpses (especially since you probably have rPois), so food isn't much of a problem 16:10:04 you can't chain cast fire storm as easily, but it's not like you have to either 16:10:06 i don't see anything wrong with having to dip into permafood if you feel like unecessarily using level 9 spells to clear an easy branch :P 16:11:48 no other spellbooks :p 16:11:49 also there are staves of energy 16:12:07 the other biggest spells my current char got are IMB and freezing cloud 16:12:09 MarvinPA: thanks, I'll look 16:12:41 it could even just start at 1* or 2* 16:12:47 (and yeah, getting level 9 spells even with great apts and copious victory dancing is badly broken) 16:13:01 it's just that currently you get "you are no longer an avatar of Ashenzari." as soon as you join, which is obviously not quite right :P 16:13:22 st_: there are, just like there are rPois items -- not in every game. 16:13:49 swamp is easy even without rpois 16:14:50 what piety level does ash identification start at? 16:14:55 or does that just depend on boundedness? 16:15:24 I think veh granting easy access to high-level spells but with the normal (and high for early usage) food cost is nice 16:15:51 monqy: hey, I know I could kite everything with Swiftness+Shock, but it's just wrong if with wizardry you get Tornado at "great" :p It's just taking an edible chunk for every cast that leads to ridiculous results. 16:16:18 Change his welcoming to "Let it end in starvation!" 16:16:27 st: <3 16:16:57 then I'd argue that kiting everything with swiftness+shock is problematic 16:17:52 & I don't really see the problem with ridiculous food costs 16:18:15 unless there's some hidden thing that I'm missing, like instadeath by starvation 16:18:30 MarvinPA: I think id starts right at the beginning. 16:18:48 casting an extremely high hunger spell at the bottom end of near starving would be interesting to try... 16:19:26 -!- ZChris13_ has joined ##crawl-dev 16:19:29 monqy: it can mean scumming for Necromutation if you don't get lucky with a staff of energy 16:19:55 oh right that exists 16:20:24 and by that I'm guessing scumming for it is actually feasible? 16:20:56 that in itself is a bit worrisome 16:22:45 -!- ZChris13 has quit [Ping timeout: 252 seconds] 16:22:52 -!- ZChris13_ is now known as ZChris13 16:25:27 looks like weapon enchantments and jewellery is fully identified at 2*, and self-identifying stuff is shown from 0 piety 16:29:39 it's damn annoying you have to pick everything up 16:30:19 including going back to old levels you were on when you had the slot uncursed 16:30:37 agreed, yeah 16:36:27 -!- MarvinPA_ has joined ##crawl-dev 16:38:12 -!- MarvinPA has quit [Ping timeout: 264 seconds] 16:40:54 Just let Ash iD eveeeeryythhiiinngg seen for you. 16:43:11 being more useless than 0.5 Zin or current Xom, he could take that 16:51:53 -!- ZChris13 has quit [Quit: ChatZilla 0.9.86.1 [Firefox 3.6.15/20110303024726]] 17:08:28 03galehar * r22b8d9891fe7 10/crawl-ref/source/ (godpassive.cc religion.cc): Fix a few messages for Ash. 17:08:40 03galehar * r431acdf161c1 10/crawl-ref/source/ (abl-show.cc enum.h food.cc food.h godabil.cc godabil.h): Give Ashenzari a butcher ability. 17:09:00 kilobyte: any suggestion? 17:12:11 for identification? Retroactively iding everything you have seen once you get a given slot cursed (with enough piety) would fix this particular problem. 17:12:57 since piety and curses come and go, we'd probably have to assign every item the time when you last seen it 17:13:03 I meant for balance. 17:13:47 a lot of tweaks would be needed 17:14:50 the worst thing is, with so many boosts here and there, it might be hard to get it in time for 0.9 if it's going to be for July 17:15:53 Well, he's now feature complete for 0.8. 17:16:05 ... 17:17:03 don't tell me you're even thinking about using the current wild flurry of changing design for 0.8... 17:18:31 the version we had for half a year was roughly balanced even if lacking in a few places (picking up items, no wrath). A completely new god with no testing at all... 17:18:41 so, you'd rather go back to XP boost? 17:20:26 for 0.8, it's either that or no Ashenzari for now. For 0.9, we can think of something. The current list of boosts is pretty inconsistent and not _yet_ well thought out, but that's pretty natural for something that's mostly 1-2 days old. 17:20:49 the XP boost is just a bad crutch, like stats for Chei 17:22:59 the current skill boost sounds decent enough to me 17:23:19 it seems worth at least trying 17:23:24 -!- MarvinPA_ is now known as MarvinPA 17:24:10 sorry for the misunderstanding. We had several mail on c-r-d where we said we wanted to replace XP boost before 0.8 release. 17:25:08 i'm not at all a fan of god-given butchery, for what it's worth 17:25:22 would much rather ash just favour bladed weapons until a better solution is found 17:25:39 but other than that the new stuff seems a lot better than the xp boost 17:25:59 it's nowhere near finished 17:26:34 even that "final great design" of +6 skills that suddenly got turned into max(6,skill) -- ie, total uselessness 17:26:51 there's no way to skip long testing 17:27:50 like... I was sure my design for Forest is good. Fortunately I got the idea of placing that big vault in trunk so I know where glaring problems are. 17:29:06 MarvinPA: yes, the butcher ability is ugly, but I wanted it in so players can curse a magic staff and use the new passive boost to evoc or spellcasting. 17:29:33 I can't think of a better workaround 17:29:40 passive evo or spellcasting to 6 doesn't sound very useful at all, if the boost still works like that 17:29:49 (ie, melee players having insane problems while those with access to spells which ignore EV just roll over the monsters) 17:31:05 monqy: http://gitorious.org/crawl/crawl/commit/bdb70b7b5e00509415db6b7539134480edcba44a 17:31:11 monqy: it boosts skills based on boundedness 17:31:19 and yeah 17:31:20 thanks I'll check it out 17:31:22 that probably explains it better 17:32:52 this looks much nicer 17:34:55 -!- ZChris13 has joined ##crawl-dev 17:35:16 for blunt weapon users' food troubles, the only solutions I can think of would make bladed weapon users too good at food, since they would get both the crutch and butchery :( 17:35:43 unless ashenzari disliked eating chunks or something, which would be weird 17:38:14 is piety_rank the number of stars? 17:38:47 yes 17:38:52 one of the wiki proposals was something like just allowing ash worshippers to butcher with their knife if they have a cursed weapon, whenever implicit butchery is added 17:39:12 yes, I have thought about it, and it might not be so simple 17:39:48 I like the autoknife implicit butchery idea on mantis. 17:40:03 me too, yes 17:40:07 or were you accounting for that too 17:40:53 If ash worshippers can butcher with a curse blunt weapon, can they also do it with a distorsion or vampiric weapon? 17:41:59 if it's cursed, sure 17:43:07 just hit the corpse with your hammer strong enough :p 17:43:34 I use IOoD to butcher 17:45:32 how does ashenzari-butchery work flavourwise? I ask this because I noticed that it doesn't work while levitating because the player has to "reach" the corpse 17:46:07 that's a good question :) 17:46:26 although, there was a case when a wisent stood on a railroad and didn't yield to the train. The train derailed, the wisent was of course thoroughly dead but its skin held. A bag full of steak tartare, but still not open. 17:46:26 03galehar * red92f751d78f 10/crawl-ref/source/travel.cc: Ashenzari on-sight identification. 17:46:40 I didn't really think about the flavour because it is really a temporary shim. 17:48:00 still a week before code freeze. Maybe I can implement implicit butchering :) 17:57:37 -!- ZChris13_ has joined ##crawl-dev 18:01:01 -!- ZChris13 has quit [Ping timeout: 260 seconds] 18:01:04 -!- ZChris13_ is now known as ZChris13 18:25:43 -!- golgepapaz has joined ##crawl-dev 18:28:25 03galehar * r5831c72e8189 10/crawl-ref/source/ (mon-flags.h mutation.cc): Stop only once for sensed monsters. 18:29:11 Unstable branch on CDO updated to: 0.8.0-a0-5940-ged92f75 (32) 18:34:15 03MarvinPA * re38f22002fe1 10/crawl-ref/source/dat/descript/gods.txt: Update Ashenzari description 18:34:43 uhm hello, is it possible to build DCSS without openGL anymore, I've got the impression it's not 18:36:04 well, you should be able to make a console build 18:36:34 I mean tiles version of course.. 18:37:52 Has tiles ever not used GL? 18:38:08 I am building with MSVC, got the console part right.Now trying it with tiles 18:38:09 I think it has needed it since 0.6 at least (which was teh first time I built it) 18:39:47 alright I'll try it with openGL. 18:44:26 I should use mingw instead of msvc 18:47:44 well I've wanted a msvc port. I've modified the source to compile with it, and wrote a cmake based built system for it. I just have some linking problems because of Ogl 18:50:49 Windows development builds on CDO updated to: 0.8.0-a0-5942-ge38f220 18:57:38 -!- syllogism has quit [] 19:14:39 -!- Cryp71c has joined ##crawl-dev 19:17:06 -!- galehar_ has quit [Remote host closed the connection] 19:19:10 -!- ZChris13 has quit [Ping timeout: 250 seconds] 19:25:06 -!- ZChris13 has joined ##crawl-dev 19:35:01 -!- Cryp71c has quit [Remote host closed the connection] 19:49:32 -!- eith has quit [Ping timeout: 255 seconds] 19:50:39 -!- ZChris13_ has joined ##crawl-dev 19:52:09 -!- ZChris13__ has joined ##crawl-dev 19:52:43 -!- ZChris13__ is now known as ZChris13____ 19:55:01 -!- ZChris13 has quit [Ping timeout: 264 seconds] 19:55:17 -!- ZChris13_ has quit [Ping timeout: 276 seconds] 20:01:08 -!- ZChris13____ has quit [Ping timeout: 276 seconds] 20:08:21 -!- ZChris13____ has joined ##crawl-dev 20:08:52 -!- ZChris13 has joined ##crawl-dev 20:10:42 -!- ZChris13_ has joined ##crawl-dev 20:12:59 -!- ZChris13____ has quit [Ping timeout: 248 seconds] 20:14:35 -!- ZChris13 has quit [Ping timeout: 248 seconds] 20:14:38 -!- ZChris13_ is now known as ZChris13 20:41:50 -!- golgepapaz has quit [Quit: ChatZilla 0.9.86 [Firefox 3.6.15/20110303024726]] 20:44:31 -!- Mu_ has quit [Quit: Defecator, may everything turn out okay so that you can leave this place.] 20:55:14 -!- ZChris13 has quit [Remote host closed the connection] 20:56:03 -!- ZChris13 has joined ##crawl-dev 21:13:29 -!- ZChris13__ has joined ##crawl-dev 21:15:27 -!- ZChris13 has quit [Ping timeout: 246 seconds] 21:15:39 -!- ZChris13__ is now known as ZChris13 21:21:28 -!- Cryp71c has joined ##crawl-dev 21:24:31 Hrm, a long while back someone gave me either a link to the bash / terminal configuration so that you get a nice, pretty display of the branch you're currently in and whether there are uncommited changes, unpushed changes, or unpushed files (as well as the count of how far ahead / behind HEAD you are) 21:24:47 due, it might've been you, or perhaps someone else..your name is jumping out to me. 21:24:50 Let me know if it was indeed you. 21:35:48 -!- ZChris13_ has joined ##crawl-dev 21:38:27 -!- ZChris13 has quit [Ping timeout: 240 seconds] 21:38:28 -!- ZChris13_ is now known as ZChris13 21:41:44 -!- upsy has quit [Quit: Leaving] 21:46:05 -!- m1nced has joined ##crawl-dev 21:54:44 -!- elliptic has quit [*.net *.split] 21:54:45 -!- elly has quit [*.net *.split] 21:54:47 -!- varmin has quit [*.net *.split] 21:54:48 -!- RjY has quit [*.net *.split] 21:54:48 -!- kilobyte has quit [*.net *.split] 21:54:49 -!- jld has quit [*.net *.split] 21:54:49 -!- Zao has quit [*.net *.split] 21:54:53 -!- ZChris13 has quit [Ping timeout: 276 seconds] 21:56:46 -!- Cryp71c has quit [Quit: Leaving] 21:57:52 -!- elliptic has joined ##crawl-dev 21:57:52 -!- elly has joined ##crawl-dev 22:00:48 -!- varmin has joined ##crawl-dev 22:00:48 -!- RjY has joined ##crawl-dev 22:00:48 -!- kilobyte has joined ##crawl-dev 22:00:48 -!- jld has joined ##crawl-dev 22:00:48 -!- Zao has joined ##crawl-dev 22:16:37 -!- valrus_ has quit [Ping timeout: 264 seconds] 22:17:48 -!- ZChris13 has joined ##crawl-dev 22:20:15 !tell Cyrp71c I don't think it was me. There is tig, which is a curses-based frontend for git but... 22:20:16 due: OK, I'll let Cyrp71c know. 22:40:48 18:08 < CIA-45> galehar * r431acdf161c1 /crawl-ref/source/ (abl-show.cc enum.h food.cc food.h godabil.cc godabil.h): Give Ashenzari a butcher ability. 22:40:53 are you serious 22:41:06 Yes I think they are 22:41:32 flavor has been nailed to the sky and left up there for good, i am afraid 22:44:06 :'( 22:44:37 why not magical sustenance or sustenance for exploration or something? 22:45:02 why 22:45:12 this is so far removed from ash's awesome initial flavor 22:45:17 and completely unnecessary 22:45:58 i don't get why curses are supposed to be made less onerous 22:46:26 in my mind, ash should be *extra* happy if you have a cursed blunt weapon, or cursed mutagenic gear, or whatever 22:47:25 solution: cursed bladed weapons can't chop corpses; the curse dulls them 22:47:30 what, no, that's awful :| 22:47:37 monqy: THAT'S AWFUL TOO. 22:47:41 oh 22:47:42 :( 22:48:00 what is your solution -- I expect wonders 22:48:18 drop cursed weaponry and shields 22:48:28 they are't nearly as cool as cursed armour/jewellery are 22:48:44 and introduce all kinds of weirdness 22:49:09 that works 22:49:43 a ring you can't take off: pretty thematic 22:49:55 a sword your hands are glued to: not 22:50:19 and what theme is this 22:51:45 the theme of "ashenzari is not a comedy movie", of course 22:52:26 I don't watch comedy movies so I don't know what that means 22:52:55 monqy: just imagine your adventurer going around the dungeon with his hands stuck to a greatsword or whatever 22:53:30 now I'm imagining him hurting his fingers trying to pull off a ring 22:53:54 wield two cursed daggers -> blade hands? 22:54:42 ring/armour curses are more believeable being stuck to you since they're designed to stay on you without your involvement 22:55:20 there are weapon curses that work but the common roguelike one is maximum goofiness 22:56:16 I don't like sticky curses at all, really, but I haven't yet thought of anything more elegant than "remove them" 22:56:44 which would kill ashenzari :( 22:58:03 monqy: imo: remove ash caring about held items; remove layering effects of sticky gear; make the skill boost based on non-magical vs. magical and able to select a skill rather than having it chosen for you 22:58:06 stupid idea: what if ashenzari and cheibriados merged into a time/divinations god (on the point of curses/slowing) 22:58:29 i actually liked that idea but it won't happen because both chei and ash have their defenders for some reason 23:18:29 -!- m1nced has quit [Quit: Leaving] 23:20:06 -!- ZChris13_ has joined ##crawl-dev 23:20:49 -!- ZChris13__ has joined ##crawl-dev 23:22:57 -!- ZChris13 has quit [Ping timeout: 252 seconds] 23:23:00 -!- ZChris13__ is now known as ZChris13 23:24:41 -!- ZChris13_ has quit [Ping timeout: 246 seconds] 23:45:25 -!- ZChris13_ has joined ##crawl-dev 23:47:04 -!- ZChris13 has quit [Ping timeout: 246 seconds] 23:47:05 -!- ZChris13_ is now known as ZChris13