00:03:15 Unstable branch on crawl.develz.org updated to: 0.9-a1-802-g3aff67e (32) 00:08:32 Unstable branch on crawl.akrasiac.org updated to: 0.9-a1-802-g3aff67e (32) 00:11:53 !tell dpeg No-tele spider vaults works for me! 00:11:53 Keskitalo: OK, I'll let dpeg know. 00:18:21 Windows builds of master branch on crawl.develz.org updated to: 0.9-a1-802-g3aff67e 00:23:10 -!- Zaba has quit [Ping timeout: 258 seconds] 00:27:00 -!- Zaba has joined ##crawl-dev 00:49:06 -!- ainsophyao has joined ##crawl-dev 02:13:36 -!- monqy has quit [Quit: hello] 02:17:58 -!- galehar has joined ##crawl-dev 03:07:36 -!- gustaf_arg has joined ##crawl-dev 03:10:23 -!- ghallberg has quit [Ping timeout: 240 seconds] 03:16:52 -!- gustaf_arg is now known as ghallberg 03:25:37 -!- ais523 has quit [Remote host closed the connection] 03:40:04 -!- syllogism has joined ##crawl-dev 03:44:34 -!- ixtli has quit [Quit: Computer has gone to sleep.] 03:45:15 -!- ixtli has joined ##crawl-dev 04:05:39 -!- RichardHawk has joined ##crawl-dev 04:11:55 -!- mumra has quit [Read error: Connection reset by peer] 04:13:13 -!- mumra has joined ##crawl-dev 04:17:13 -!- Wensley_ has joined ##crawl-dev 04:20:36 -!- Wensley has quit [Ping timeout: 276 seconds] 04:20:48 -!- Wensley_ is now known as Wensley 04:57:18 -!- ainsophyao has quit [Remote host closed the connection] 05:20:40 -!- cesium has joined ##crawl-dev 05:25:42 -!- edlothiol has joined ##crawl-dev 05:50:46 monster pathfinding gets confused by traps (https://crawl.develz.org/mantis/view.php?id=4199) by KiloByte 06:15:56 Followers won't pick up sling ammo for sling launchers (https://crawl.develz.org/mantis/view.php?id=4200) by cesium 06:22:50 AecTalek (L27 NaWz) ERROR in 'dungeon.cc' at line 2444: No maps or tags named 'bazaar', and no backup either. (Bazaar) 06:23:26 AecTalek (L27 NaWz) ERROR in 'dungeon.cc' at line 2444: No maps or tags named 'bazaar', and no backup either. (Bazaar) 06:24:00 AecTalek (L27 NaWz) ERROR in 'dungeon.cc' at line 2444: No maps or tags named 'bazaar', and no backup either. (Bazaar) 06:24:58 AecTalek (L27 NaWz) ERROR in 'dungeon.cc' at line 2444: No maps or tags named 'bazaar', and no backup either. (Bazaar) 06:27:38 AecTalek (L27 NaWz) ERROR in 'dungeon.cc' at line 2444: No maps or tags named 'bazaar', and no backup either. (Bazaar) 06:28:58 AecTalek (L27 NaWz) ERROR in 'dungeon.cc' at line 2444: No maps or tags named 'bazaar', and no backup either. (Bazaar) 06:29:02 !lm * type=crash -log 06:29:03 1185. AecTalek, XL27 NaWz, T:470779 (milestone): http://crawl.develz.org/morgues/0.8/AecTalek/crash-AecTalek-20110701-112735.txt 06:29:54 0.8? Interesting... 06:30:27 AecTalek (L27 NaWz) ERROR in 'dungeon.cc' at line 2444: No maps or tags named 'bazaar', and no backup either. (Bazaar) 06:32:11 AecTalek (L27 NaWz) ERROR in 'dungeon.cc' at line 2444: No maps or tags named 'bazaar', and no backup either. (Bazaar) 06:36:02 Zot traps triggered by hostiles harming the player cannot be considered intentional behaviour (https://crawl.develz.org/mantis/view.php?id=4201) by Tonfa 06:36:16 kilobyte: can the game run out of bazaars? he had visited 39 already 06:37:18 !lm syllogism br.enter=bazaar game_id=595250 06:37:19 242. [2010-05-09] syllogism the Gelid (L26 HEIE) entered a Bazaar on turn 125511. (Bazaar) 06:37:58 of course it could have been broken since then 06:38:35 if none have allow_dup, you'd get that error when they're all used up 06:39:00 at least, i saw that error when i was messing around with subvaults, after i'd tested the same portal a few times in the same game 06:39:27 and i can't see any allow_dup on the bazaars 06:40:29 yup, that sounds like the issue 06:40:49 actually no 06:40:59 oh wait 06:41:15 yeah, only some bazaar entries have allow_dup 06:41:31 none of the actual bazaar maps 06:41:35 so that error is actually quite obvious 06:42:25 kilobyte, one or more bazaar maps need to get allow_dup 06:42:48 mumra: The new spider map looks cool, thanks 06:43:04 or better, Trowel could be capped somehow 06:43:06 thanks! 06:43:16 kilobyte, I guess 06:43:52 !tell dpeg Could also fix strandedness by providing portals out where appropriate. 06:43:53 Keskitalo: OK, I'll let dpeg know. 06:46:06 yeah i'm not so sure about stopping tele scrolls working. players might suddenly need to get away from that ghost moth :) 06:47:58 kilobyte, on an unrelated note, is player::level_type_ext still needed? 06:48:39 ghosts use it AFAIK 06:48:52 kilobyte, it only affects the extension of ghost files in portal vaults 06:49:32 kilobyte, ghosts elsewhere use level_id::current().describe(), which actually works for portal vaults, too 06:51:32 oh ok... got to leave right now, though 06:51:57 -!- edlothiol has quit [Ping timeout: 255 seconds] 07:00:15 -!- MarvinPA has quit [Ping timeout: 246 seconds] 07:03:27 -!- dpeg has joined ##crawl-dev 07:05:10 rough call graph for dungeon.cc: http://thorium.homeunix.org/~zaba/dungeon.cc.png 07:05:17 I wonder whether splitting it further is feasible 07:08:44 Zaba: absolutely no idea, but this is beautiful! 07:08:45 dpeg: You have 3 messages. Use !messages to read them. 07:09:22 thanks :P 07:29:00 -!- cesium has quit [Remote host closed the connection] 07:29:23 -!- Textmode has quit [Ping timeout: 250 seconds] 07:35:02 -!- cesium has joined ##crawl-dev 07:35:49 dpeg: what do you think about changing the d& behaviour to make it drop all useless items instead of rotten chunks, corpses and skeletons? 07:37:08 -!- st_ has joined ##crawl-dev 07:37:10 galehar: d, already does that 07:38:04 I guess the d help could say that better 07:38:31 galehar: and it has false positives (books for Trog, food while in lich form, etc) 07:38:56 kilobyte: oh right, I forgot about d, 07:39:16 yes, there are false positives, and they can be fixed. 07:39:24 I'd suggest to skip the filter option from d, and just make it "drop all useless stuff" 07:39:29 I'm thinking about auto_drop 07:39:29 and to keep d& 07:40:18 some of these are debatable: if your Necromutation is unreliable, you won't be in lich form most of the time, later on, you will 07:41:03 I think being able to select autopickup on \ is more improtant 07:41:10 because it prevents the pickup before it occurs 07:41:22 you still do hunger when the spell is not on -- like, if you have no mana to recast it; carrying a piece of food at all times is a good idea 07:41:26 auto_drop would help with chunk management 07:42:33 yeah, chunks are the only useless item you'd be dropping often 07:42:47 kilobyte: let's be conservative and mark as useless only things which are really useless. I think lich form should autopickup food by default. 07:43:13 there is also the problem of *when* to autodrop 07:43:31 when autopickup would burden you 07:43:36 and only then 07:43:38 or you have 52 items 07:43:39 for chunks, a good point is when picking up new ones: automating to drop older ones right then would be good 07:43:53 no, I mean regarding to monsters around 07:44:09 if you can afford the time to autopickup, you can also auto_drop 07:44:20 right 07:45:12 I'll write an implementable for it. 07:45:13 also, you'll often want to autodrop chunks which are not rotten yet but will be soon 07:45:38 kilobyte, I wonder whether further splitting of dungeon.cc is feasible: http://thorium.homeunix.org/~zaba/dungeon.cc.png 07:46:30 carrying only X freshest chunks could be nice, but I'm not sure how to obtain X. A fixed value won't be good for both halflings of Chei and trolls of Trog... 07:46:33 seems to be quite intertwined for anything to be obviously splittable 07:46:54 the auto_drop function takes a parameter. It is the item your trying to autopickup. It start by dropping all useless items, starting be heavier ones. If you still can't pick up your chunk, you start dropping your oldest chunks of same or lower quality. 07:50:23 -!- edlothiol has joined ##crawl-dev 07:50:32 Zaba: placement of monsters is one part, but it's tiny :( 07:51:02 kilobyte: so, how is the variable radius tornado going on? 07:53:42 galehar: a complete disaster :( 07:54:21 I implemented all of the changes you guys suggested... and the spell is pretty worthless 07:55:12 tested in a Zig, with the nerfs it not only loses parity with Firestorm but even Refrigeration or Freezing Cloud 07:55:41 losing parity with firestom is ok. It's the point actually. 07:56:33 being exactly five times worse than Refrig (on empirical data on a couple of Zig levels) is not 07:56:50 true. 07:56:53 we can always adjust the numbers. How about you push it so we can test it? 07:57:28 as in, 1.5 times the damage as a single cast of Refrig, at 1.5 times the mana but 7.5 times as slow 07:57:55 you changed duration? 07:58:05 push THAT nonsense? Onto a branch, and even then, only if it never lands on trunk without a thorough rethinking 07:58:25 *4/3, like we talked about 07:59:12 *3/4 damage, but AC and partial radius penalty cause a more than proportional decrease 07:59:48 ok, push it into a branch if you like. I'm just saying that a growing radius is the way to go. Balance can be achieve by adjusting growing speed, damage, max radius and duration. 07:59:55 this, plus decreased damage due to growing radius not affecting everyone 08:00:37 also, testing this on Zig I realized the data on that spreadsheet fits only perfectly open area 08:00:57 true. 08:01:17 and Ziggurats often have lots of furniture. This is actually rolled once per Ziggurat... the one I happened to be testing it on had a lot. 08:02:03 not to mention that we shouldn't be balancing it for Ziggurats but for regular levels 08:02:06 kilobyte: sure? Wall colour and shape (diamond, ellipse, square) should be rolled only once, furniture should be per storey. 08:02:12 like, say, Cerebov's vault 08:02:59 dpeg: I did not look closely, at least several levels in a row had plenty of it 08:06:37 -!- edlothiol has quit [Quit: edlothiol] 08:06:58 -!- edlothiol has joined ##crawl-dev 08:07:15 the effect of AC makes calculations pretty hard :( 08:09:06 it's quite easy to deal with AC on spells that do damage in a single chunk, ones that have multiple damage calls get unproportionately weaker when the damage goes down 08:09:07 indeed. It's pretty hard to express average damage. 08:09:51 AC working better on weak blows _does_ make sense thematically, though 08:11:37 galehar: do you have an idea how big a part AC takes away for different ranges of blows? 08:12:19 I mean, especially between different sources: a 30 hp hit with an axe, a 100 hp crystal spear, ... 08:13:05 no, I'm not sure. 08:14:05 I wonder if we ignored the "AC is better for weak blows" principle for a moment (at least for the sake of the discussion), it might make things easier 08:14:15 when I tried to fix the AC calculation in the tornado spreadsheet, I did this: 08:14:16 https://spreadsheets.google.com/spreadsheet/pub?hl=en_US&key=0AqvhLOPFHpiMdE1oRUx5UzhUNXRMX1ZRU1ppZGpkcFE&hl=en_US&gid=2 08:15:01 to take into account the fact that a high AC doesn't help more than a low AC when the damage roll is low. 08:15:51 if we increase tornado's duration, it makes sense to reduce the AC used too. We used AC/3 so that it equals to 2*AC over 6 turns. 08:16:05 maybe AC/4 or AC/5 would be better. 08:16:06 -!- edlothiol has quit [Ping timeout: 255 seconds] 08:16:07 a formula I devised for shotgun bolt of magma is: process every 1 hp pellet on their own; on that pellet, every point of AC has an independent 1/81 chance of stopping the pellet 08:16:42 "81" taken by scientifically rigorous rectal extraction, being of the same order of magnitude as what we want :p 08:17:38 :) 08:17:47 so AC of fourty-something halves damage 08:18:47 I'm not against upping the damage if necessary, to compensate for the lower radius at beginning. I think the basic mechanism of growing radius is good (both thematically and gameplay) and I believe the spell can be balanced this way. 08:19:49 galehar: this is obvious, my observation is that the set of proposed nerfs together lead to ridiculous results 08:20:19 also, I'm not sure about increased duration. Instead, we can keep the duration, but use a somewhat logarithmic speed for radius growth. Fast at start, slower at the end. 08:21:18 like, monsters in this particular experiment took damage per turn: Refrig ~33, Freezing Cloud ~15, Tornado ~6 08:22:00 (not even averaged, merely picking typical numbers that appeared, we'd have to calculate this in a more rigorous way) 08:22:08 kilobyte: I believe you. 08:22:58 I don't mind having a worthless tornado in trunk (for a change) for a couple of days while we tweak the numbers. But having it in a branch is fine too. 08:23:01 refrigeration is a single cast with a bigger radius, freezing cloud is damage over time like Tornado but with a smaller effect radius (and more affected by resistances / AC) 08:23:34 ??refrigeration 08:23:35 ozocubus refrigeration[1/2]: A fun lv5 (6 in 0.9) Ice spell that hits everything in LOS (including invisible/submerged things. and YOU, so have cold res) with AC-bypassing cold damage. 50% chance to slow cold-blooded monsters. This can shatter your potions if damage is >0, so drop them or at least wear conservation. Works nicely in many places, e.g. Slime, Snake, Elf, Zig. 08:23:57 it was level 5 until recently, too... 08:24:04 whoa 08:24:12 level 6 in 0.9? 08:24:13 rC nearly totally stops it, though 08:24:13 :\ 08:24:36 is it really worth casting on the same level as cloud spells 08:24:40 I think it's still considered very powerful. 08:24:45 on of the best ice spell 08:24:58 Pacra: it's single school 08:25:05 hrmmmm 08:25:06 okay 08:25:10 good vs low-hp high-ev monsters like in Elf 08:25:17 that's a plus for the change I guess 08:25:55 kilobyte: comparing tornado with shatter could be interesting too. 08:26:02 and can win you easier Ziggurats if you don't get many undead 08:26:39 Spider portal fix and tweaks (https://crawl.develz.org/mantis/view.php?id=4202) by mumra 08:28:00 I wonder if replacing AC calculations by that pellet formula could be a good experiment (even if it loses theme somewhat) 08:28:27 kilobyte: pellet formula? 08:29:26 dpeg: by the way, my math skills are pathethic. Having studied at a department that includes "Mathematics" in its name, I don't remember things good enough to even understand blasted Wikipedia article :( 08:29:41 kilobyte: can you explain how a cooldown timer is gameable? 08:30:04 dpeg: if we have n Bernoulli trials each with chance p, how to roll it in one go? 08:30:26 ghallberg: [15:16] <+kilobyte> a formula I devised for shotgun bolt of magma is: process every 1 hp pellet on their own; on that pellet, every point of AC has an independent 1/81 chance of stopping the pellet 08:30:30 getting just "was there any success" is trivial: 1-(1-p)^n 08:31:26 dpeg: but I want to roll the number of successes, having a random variable with uniform distribution at the input 08:31:41 with, say, range [0..1) to make it simpler 08:32:39 galehar: if you know the duration, you can take a swift action to manipulate it so you can recast as soon as possible 08:33:04 kilobyte: but we should always use aut not turns 08:33:48 such actions are for example swapping weapons (5 aut), removing jewelry (5 aut), moving (6 aut with swiftfly); haste gives a better resolution, too 08:33:56 kilobyte: re Bernoulli: there is no faster way than rolling n times (at least I am not aware of one) 08:34:46 kilobyte: how about randomising the cooldown timer? random_range(25, 35) 08:34:52 dpeg: I mean: if the duration is 54 aut, taking full 10 aut turns means you lose 6 aut (more than 10%) 08:34:58 ah, I see 08:35:09 randomisation would help with that 08:35:22 galehar: you'd still benefit from doing something fast and trying every round 08:36:05 dpeg: I know at least one way, if rolling random variables is slow (and it is) but calculations are free (they are not) 08:38:34 you roll the number in [0..1), then subtract (1-p)^n, see if it becomes negative -- if no, you subtract (1-p)^(n-1)*(p)^1*n -- all the binomials or whatever the numbers on the bottom row of a Pascal triangle are called 08:38:35 kilobyte: I'm not sure it's really a problem. The point of the nerf is to make tornado a tool you have to use in conjunction with others, instead of being a big gun you keep spamming. You should have better things to do with your time than optimise at which point you can recast it. 08:39:22 it is slow and is ridiculous, but I bet there's a formula for that, just my math skills after all those years of not using them are gone 08:40:00 galehar: swapping weapons is a waste of time, movement is not 08:40:15 kilobyte: if you give me a link or so, I can help you (gotta prepare an exam right now) 08:41:27 dpeg: if I had a link that explains it well, I guess I'd be able to figure it out (I used to have a clue as a kid...) 08:42:33 the formula I'm looking for is probably in the plain sight, I just don't know how it is named 08:43:12 moving fast (0.6) instead of casting gains you 4 aut. Make that 3 aut when hasted: fast movement (0.35) against casting (0.65). 08:43:51 I think the choice between moving or acting has more important factors than gaining 4 aut. 08:44:14 -!- Twilight-1 has quit [Read error: Connection reset by peer] 08:44:28 -!- Twilight-1 has joined ##crawl-dev 08:46:10 dpeg: oh, I see: we have a cumulative distribution function, and the roll is the y on that function, I need x 08:47:07 it's more like choosing the ordering of moving and casting rather than whether to do one or the other 09:59:34 so, trunk on CAO seems to be in a strange state 09:59:35 https://crawl.develz.org/tavern/viewtopic.php?p=22469#p22469 10:06:02 -!- mumra has quit [Read error: Connection reset by peer] 10:07:47 -!- mumra has joined ##crawl-dev 10:11:38 any feedback on the skill cost change? How is training magic skills in the early game? 10:34:21 I haven't noticed a difference really 10:40:52 so what are the biggest features of 0.9 that are actually making it in? 10:54:07 Oc 10:54:24 new skill training (let's hope I finish in time) 10:54:49 but 0.9 has a short time frame. It's more a balance release than a feature release. 10:55:03 Skalds 10:55:05 galehar: octos aren't getting into .9 10:55:15 really? 10:55:43 probably better. Let's take the time to polish it. 10:56:38 yeah it was discussed yesterday, we'll be able to do constriction or reaching for the initial release of them this way 10:57:59 by the way, we still have no idea on the details of the constriction mechanism. Or is it somewhere on the wiki? 10:58:20 it used to be on sourceforge, since it was originally a proposal for nagas 10:58:57 the idea is that you have some combination of restricting enemy movement and combat cpaability, size and strength based, occurs passively rather than being an activated ability, has few to no drawbacks to using it (so you don't get annoyed if you constrict something you don't want to) 10:59:01 well, if you can dig it up, clean it, update it and put it on the wiki it would help. 10:59:51 https://crawl.develz.org/wiki/doku.php?id=dcss:brainstorm:effect:held_and_restrained&s[]=constriction 10:59:55 looks like somoene did it for me 11:00:14 cool, thanks :) 11:03:04 -!- galehar has quit [] 11:08:20 -!- Wensley has quit [Ping timeout: 260 seconds] 11:14:39 -!- Zannick has quit [Ping timeout: 276 seconds] 11:15:32 -!- Cryp71c has joined ##crawl-dev 11:15:59 -!- Zannick has joined ##crawl-dev 11:19:55 Eino was it 11:21:23 -!- Zannick has quit [Ping timeout: 244 seconds] 11:23:40 -!- Zannick has joined ##crawl-dev 11:25:55 -!- ainsophyao has joined ##crawl-dev 12:01:22 -!- mumra has quit [Read error: Connection reset by peer] 12:02:51 -!- mumra has joined ##crawl-dev 12:07:50 Cure mutation had no effects (https://crawl.develz.org/mantis/view.php?id=4203) by BirdoPrey 12:10:57 -!- Cryp71c has quit [Quit: Leaving] 12:12:32 dpeg: Yeah 12:14:06 !tell galehar re: held & restricted page, I once tried to look into generalizing throwing net stuff, so that it could be handily used for constriction, webs and slug goo spit, I guess I lacked the skill and stamina to get far enough :) 12:14:07 Keskitalo: OK, I'll let galehar know. 12:15:05 !tell galehar anyway, if you are looking at implementing, those might be good to keep in mind 12:15:05 Keskitalo: OK, I'll let galehar know. 12:16:00 coming up with a consistent set of possible effects of being constricted would be a good start 12:25:48 Well, looks like I have listed basically three things on the page: inability to move, inability to physically interact with squares other than your own, and if teleportation frees you or not. 12:26:06 Dropping EV to zero sounds sensible as well. 12:59:30 -!- monqy has joined ##crawl-dev 13:03:38 Ah cool, forest wyrm branch <3 13:06:19 -!- ainsophyao has quit [Remote host closed the connection] 13:16:44 -!- MarvinPA has joined ##crawl-dev 13:22:40 yay 13:23:18 there an email or something with details? dunno what if anything got changed 13:44:48 -!- galehar has joined ##crawl-dev 13:51:26 No details at all in the email.. some of the branch stuff don't have diffs 13:51:26 ruthinator the Poker (L6 MfSt) (D:5) 13:53:13 http://gitorious.org/crawl/crawl/commits/forest_wyrm 13:53:37 I tried looking at the top diff there, but it says the "git object is too large to be displayed in the browser" 13:59:10 -!- golgepapaz has joined ##crawl-dev 14:02:19 -!- MarvinPA_ has joined ##crawl-dev 14:02:54 -!- MarvinPA has quit [Ping timeout: 252 seconds] 14:04:33 -!- MarvinPA_ is now known as MarvinPA 14:08:27 -!- golgepapaz has quit [Quit: ChatZilla 0.9.87 [Firefox 4.0/20110318052756]] 14:20:02 -!- blabber has joined ##crawl-dev 14:34:11 -!- RichardHawk has quit [Ping timeout: 240 seconds] 14:42:52 -!- mumra has quit [] 14:54:09 03galehar * r49bc7b7dec12 10/crawl-ref/source/itemname.cc: Also let Mummy Stalkers get their starting potions. 14:54:13 03galehar * r58a5468c609b 10/crawl-ref/source/item_use.cc: Fix auto_switch not handling time properly. 14:54:13 03galehar * r741e87be1f24 10/crawl-ref/source/fight.cc: Don't auto_switch to melee weapon if said weapon isn't in slot a or b. 14:54:13 03galehar * r31905a829b88 10/crawl-ref/source/itemname.cc: No scroll is useless if you don't know spellcasting. 15:04:01 hey 15:04:01 galehar: You have 2 messages. Use !messages to read them. 15:04:05 !messages 15:04:05 (1/2) Keskitalo said (2h 49m 58s ago): re: held & restricted page, I once tried to look into generalizing throwing net stuff, so that it could be handily used for constriction, webs and slug goo spit, I guess I lacked the skill and stamina to get far enough :) 15:04:16 !messages 15:04:17 (1/1) Keskitalo said (2h 49m 12s ago): anyway, if you are looking at implementing, those might be good to keep in mind 15:04:45 Keskitalo: ok thanks. 15:05:01 but I'm not implementing, I'm just curious. 15:05:38 anyway, I think we can officially reflavour the ench split as buff/debuff instead of self/others. 15:06:22 self/others never really made sense with all the spells that affect either :) 15:06:22 2 reasons: first, gameplay. We haven't been able to make self/other work. And with skalds and the plan for new AM, we see that buff/debuff works better. 15:06:47 i thought that was the current flavour anyway :P 15:07:03 second: theme. Hexes is strongly linked with debuff. Players don't understand if you tell them to hexes their weapon. 15:07:24 MarvinPA: in practise yes. 15:07:34 aha :) 15:08:15 but due wrote some stuff about how magic school work. And hexes is "others". Should be rewritten for "debuff". 15:08:26 ah, yeah 15:08:50 in docs/develop/spells.txt 15:12:34 galehar: sure, go ahead. 15:12:58 This is stuff that might warrant a note to c-r-d. Not a question -- there'll be hostile feedback anyway. 15:13:03 I'm better at writting code than english :P 15:13:30 dpeg: ok, that I can write :) 15:14:14 I need to start folloind c-r-d 15:14:16 btw, I forgot to reply about the skald description. Did anyone compiled the different proposal into a good desc? 15:14:19 following* 15:14:35 ghallberg: it's very low traffic. 15:15:17 That's a good thing I guess. Seems some discussion still happens ther 15:15:19 e 15:17:21 galehar: yes, due compiled something out of Eronarn's, Adam's and my proposal. Don't worry about it. 15:17:56 ok, good. 15:18:17 The sole point of c-r-d (imo) is to reach active and not so active developers (dolorous, jpeg,... ) -- i.e. for announcements like this. 15:19:09 mmk 15:19:37 I think the wiki discussions are a bit hard to follow :/ 15:20:57 I'm thinking a bit about the 'D' behaviour. 15:21:10 How often is a "queue" really interesting? 15:21:18 not so often 15:21:34 ghallberg: wiki is not good for discussion, but perfect for presentation 15:21:37 I'm tihnking it should only remember the last item/bunch of items. 15:21:45 and you can make a clean plate after a while 15:21:47 ghallberg: yes I agree. 15:21:50 dpeg: Hard to navigate even for presentation though, and hard to manage 15:21:56 a queue would be complicated to code and to use 15:22:16 So if you pick up several items you can drop all of them, but it's erased if you get something else. 15:22:26 ghallberg: you haven't had the pleasure of doing it all with the SF tracker. Trust me, the wiki is much better for certain things like brainstorming a god and then polishing the result. 15:22:30 ghallberg: yes, that's good behaviour 15:22:39 ghallberg: +1 15:22:51 Cool, then a map conatining the contents of your last pickup. 15:22:58 it's a PITA to organise pages into folders on the wiki 15:23:17 I did some cleaned up and organisation and it was really painful. 15:23:18 If you move any of the items around in inventory ('=') the queue is nulled. 15:23:48 or only remove that item from the bunch? 15:23:55 hmm, I'm gonna thinka some more on it. 15:24:06 ghallberg: would be better if '=' didn't interfere with the feature, but it's a small drawback. Which can be fixed later. 15:24:35 galehar: I'm thinking if you're taking the time to move the item around you probably won't want to drop it immediatley 15:24:55 you could update the map when switching inventory slots to keep it up to date. 15:25:02 hmm, true. 15:25:09 And it's really cumbersome to keep track of the item when you change it's slot. 15:25:46 ok. If it's a pain to code, don't worry about it and empty the map when = is used. 15:26:00 -!- Zannick has quit [Ping timeout: 246 seconds] 15:26:15 I think I'm gonna implement it so that any item you move around is removed from the map. Gives you a way to drop all ecept one of the items too. (Clumsy I know) 15:26:46 galehar: It's not hard, but it's not worth it for the small possible gain. 15:27:04 true. 15:27:05 So yeah, that's my plan now. Gonna learn how to handle maps and grab a glass of wine now I think. 15:27:06 -!- Zannick has joined ##crawl-dev 15:27:28 about the wiki, it would be good if we had more editors. People use it like a forum and rarely edit other people's comment. Many pages need to be cleaned up and summarised. 15:27:55 http://www.cplusplus.com/reference/stl/map/ 15:28:28 * dpeg hands ghallberg an auxiliary coke to go with the wine. 15:28:43 galehar: yes, wiki editors would be awesome. How to recruit them? 15:28:56 * galehar load is shotgun and aim at the savages. 15:29:03 Eino and I were supposed to be them, but we're both shaky at best. 15:29:14 No coke today :D Just water. 15:29:15 galehar: hey, I do neither wine nor coke! 15:29:49 I really need a third monitor! 15:29:49 dpeg: I really don't know. I'm just saying it would be good to have, but I don't think anyone will volunteered. 15:30:07 I asked on the tavern a while ago and nobody answered 15:30:48 I think it is much harder to say "hey, I'll do that" than to react to us asking someone personally 15:31:29 could look at some folks who've been using the wiki for some time and in a reasonable manner 15:34:35 dpeg: true. 15:35:08 * galehar hands dpeg his HR badge. 15:37:03 the map could just be couldn't it? With the key being the slot. 15:37:42 ghallberg: sure, whatever is simpler. 15:38:19 What's the difference between item_def.slot and item_def.link? 15:38:29 They seem to be realated and quite similar? 15:38:49 for items in inventory, none. They are supposed to be equal. 15:39:07 link is used for items on the ground 15:39:09 Ok, cool 15:52:18 vaults time, yay 15:53:05 mumra updated the spider vault 15:54:10 auto_drop feature (https://crawl.develz.org/mantis/view.php?id=4204) by galehar 15:54:11 I saw 15:59:36 sorry for a really stupid question: I have his patch in my crawl source folder. How to apply it? (I do this once every four months, and I always forget how :O) 16:05:54 Isn't it just "patch 'patchname'" 16:06:00 somthing simple like that I think 16:07:03 yes, but I never get the file where it ought to be 16:07:16 and then I am fighting with this level option 16:07:47 * dpeg longs for the days when compilation was a cycle of "assembler, linker, run program, debugger" 16:08:15 Ok, compiling with the last_pickup map now. This is so gonna blow up 16:10:15 -!- Pacra has quit [Ping timeout: 250 seconds] 16:14:31 now that I have his patch applied, how do I commit it such that his name and message are kept? (and yes, I am looking at git help commit -- I am just too stupid for this) 16:15:05 you are not stupid, git's manual really sucks 16:16:19 i think applying it with "git am patchname" preserves the author and commit message and such 16:17:28 yes, i think so too. but you should be able to manually override it with 'git commit --author=... --date=...' 16:19:13 MarvinPA: what do I use as "patchname"? 16:19:36 I applied the patch already. 16:19:42 the file name, 0001-blablabla.patch usually i guess? 16:19:45 ah, hmm 16:20:10 i'm not that proficient with git either i'm afraid :P 16:20:44 MarvinPA: just to be sure ... I could have used "git am ..." before I did "patch"? 16:20:49 that'd be much easier indeed 16:20:54 s/before/instead 16:21:01 i think instead of 'git patch' 16:21:01 yeah 16:21:38 oh, I did "patch ..." not "git patch ..." -- traps everywhere :) 16:23:13 git commit -a 16:23:17 sorry 16:23:24 dpeg: to clean your tree: git reset HEAD --hard 16:23:48 then apply his patch with git am 0001-patch_bla_bla.patch -s 16:24:05 too late, I just did the bad way 16:24:08 -s is signoff. It adds "signoff by dpeg" in the commit message 16:24:17 I really hate this, every time I want to add a patch I lose an hour :( 16:24:26 git reset HEAD --hard 16:24:35 pushed before you were back :( 16:24:35 it restores a clean tree 16:24:43 you need to take notes :) 16:24:49 yes, that occurred to me too 16:25:31 made a notes file! 16:25:35 pushed? you sure? I don't see your commit 16:25:49 this always takes a while 16:28:08 03dpeg * r11ab53c1e849 10/crawl-ref/source/dat/des/portals/spider.des: Spiderweb vault patch (by Pete Hurst) 16:28:51 :) 16:32:14 If you pick up runes, gold, orb the 'D' should remain the same, any other cases? 16:34:17 not the orb. It's a normal object. 16:34:45 and no, I don't think there's any other special case outside of gold and runes. 16:35:34 how does 'D' currently handle picking up an item that stacks with something in your inventory already? 16:36:09 it would be nice if it dropped the same number of that item that you picked up 16:36:19 elliptic: That's the plan. 16:36:27 okay, cool :) 16:36:31 I'm working really slowly tonight. 16:36:45 It's galehars fault for not allowing me to drink coke with my wine. 16:36:52 :P 16:37:59 elliptic: tavern people says that skalds are very hard to get going with the slower magic skills training. Have you tried? 16:39:02 I just played a HuSk and skill training was not a problem 16:39:16 ok cool 16:39:19 MuSk was harder but that's to be expected with -2 apts :) 16:39:31 it's just that they have to get used to it. 16:40:11 well, the change makes magic aptitudes much more relevant in the early game. Which is good I guess. 16:40:38 yes! 16:41:02 the difference between 0 and -1 apts is fairly large in early game (not just for skald) because the skills start at 2.0 or 3.0 or whatever for apt 0 16:41:16 so with apt -1, it will always be one lower 16:41:23 btw, anyone happens to know if I can collapse two lines "FTILE: x = foo \\ FTILE: y = foo" with "FTILE: xy = foo" 16:42:55 seems not 16:43:01 elliptic: that can be fixed if we want to. Start apt 0 at 2.3 for example. But maybe we don't :) 16:43:32 dpeg: I am completely ignorant of vault syntax, sorry. 16:47:31 * ghallberg just made a meal wholly consisting of eggs and cheese. 16:47:46 I feel like such a stereotypical student sometimes. 16:48:43 -!- blabber has quit [Quit: leaving] 16:48:50 -!- blabber has joined ##crawl-dev 16:51:11 ghallberg: the students I know of would replace the wine by beer. 16:51:36 dpeg: there's beer in the fridge 16:51:44 dpeg: obviously. You're german. 16:52:05 * dpeg drinks milk. 16:54:45 -!- blabber has quit [Quit: leaving] 16:54:51 -!- blabber has joined ##crawl-dev 17:00:40 I'm out of milk or I would've made a real omelette. 17:01:44 -!- blabber has quit [Quit: leaving] 17:02:44 it seems there is no documentation for serial vaults at all :O 17:11:56 What's the best way to print debug info while testing? 17:13:15 dprf + make debug 17:13:33 Eronarn: Thanks 17:14:09 03dpeg * r37850e752426 10/crawl-ref/ (2 files in 2 dirs): New serial vault (ponds) by minmay. 17:14:16 03dpeg * r00b03b334880 10/crawl-ref/source/dat/des/ (branches/orc.des variable/altar.des): Altar vaults by Brendan (bmh). #4010 17:26:50 Woah, it actually works! 17:28:20 !schnaps ghallberg 17:28:27 galehar: link and slot do not seem to be equal, slot refers to the ascii code of the item I think? 17:40:41 oh right. Link is index, slot is the letter. 17:40:58 I tried to use slot for dropping, didn't go so well ;) 17:41:29 Trying to think of a solution for gold/runes now. 17:41:41 03dpeg * r09ba6268b135 10/crawl-ref/source/dat/des/branches/elf.des: Elf minivaults (minmay) 17:43:35 minmay is producing extremely good vaults... I would give him free reign on that front. (He has clashing views on some other aspects, for example numbers.) 17:46:55 Where can I find wizard mode commands? 17:47:03 &? 17:47:15 More specifically, how do I spawn a rune or stack of gold on the floor? 17:47:37 &o 17:47:58 &o} 17:48:00 rune 17:48:03 then choose 17:48:23 cool thanks 17:48:26 for gold &o$ 17:48:35 Hum, now I managed to drop everything and then crash... 17:48:37 Not so good. 17:50:35 dpeg: btw, I had an idea for an Ashenzari overflow vault. Just an altar with a cursed mace and a book of cantrips. 17:50:48 As an hint about using animate dead for butchering. 17:50:55 animate skeleton 17:50:57 -!- ainsophyao has joined ##crawl-dev 17:55:46 galehar: nice, will make one 17:55:56 thanks :) 17:56:12 Ugh I'm getting tired. 17:56:20 Crashes don't make sense anymore :P 17:57:16 should have a cursed trident half the time 17:57:25 fine 18:08:09 03dpeg * r4c1b49797e77 10/crawl-ref/source/dat/des/variable/altar.des: Ashenzari overflow vault (galehar). 18:09:04 :) 18:09:49 at your service :) 18:16:46 As my workflow goes right now there will be a few intermediary patches which will all be overwritten by other stuff. Can I remove those intermediary commits somehow? 18:31:48 yes 18:32:02 git reset HEAD^n 18:32:07 then commit again 18:32:25 with n being the number of commits you want to undo. 18:32:37 It doesn't touch the source tree. 18:34:08 -!- Pacra has joined ##crawl-dev 18:35:46 sorry no, it's: git reset HEAD~n 18:35:56 -!- galehar has quit [Quit: zzz] 18:51:11 cool 18:51:31 I'm a bit stuck now and really tired. Gonna have to fix it tomorrow. 18:51:42 morning 18:51:55 ^ = -1, ~n = -n. 18:52:00 HEAD^^^ = HEAD - 3 18:52:09 HEAD^^^ == HEAD~3 in theory. 18:52:09 cool 18:52:15 due: morning? 18:52:22 ghallberg: yes, it's 9am. 18:52:28 I'm just about to go to sleep :D 18:52:30 2am here. 18:52:53 So yeah. morning to you, and good night :D 18:53:08 sleep tight! don't let the boulder beetles bite! 19:00:44 -!- RichardHawk has joined ##crawl-dev 19:05:26 -!- Zannick has quit [Ping timeout: 252 seconds] 19:06:52 -!- ainsophyao has quit [Remote host closed the connection] 19:07:15 -!- Zannick has joined ##crawl-dev 19:10:34 -!- Textmode has joined ##crawl-dev 19:13:06 -!- Pacra has quit [Ping timeout: 255 seconds] 19:16:42 -!- RichardHawk has quit [] 19:27:32 Anyone around? 19:28:52 -!- MarvinPA has quit [Ping timeout: 252 seconds] 19:45:23 Isn't it rather late in Germany 19:45:52 * dpeg passes out 19:46:28 greensnark: is there a lua-usable function which checks whether a monster is in LOS? 19:46:45 hey, greensnark, are you aware that CAO trunk doesn't seem to be working quite right? 19:46:59 dpeg: There was, I think due knows more, I've forgotten 19:47:05 What's up with cao trunk 19:47:25 the bots don't seem to know about it and everyone is just seeing their own ghosts and high scores 19:47:31 Whuh 19:47:45 Did something change in how crawl finds its directories 19:48:10 !tell due You're our local lua expert these days. Is there a function to check whether a monster is in LOS? 19:48:11 dpeg: OK, I'll let due know. 19:48:59 -!- Zannick has quit [Ping timeout: 276 seconds] 19:50:04 greensnark: I cannot say, as usual :( 19:50:25 -!- Zannick has joined ##crawl-dev 19:50:43 Ah, interesting. Even if I transfer a save it disappears 19:51:04 Although reloading saves works... 19:55:16 -!- Zannick has quit [Read error: Connection reset by peer] 19:55:36 Oh, it looks like Crawl's handling of the -dir option has changed 19:56:18 I wonder if I can just dispense with the -dir without torching the setup 19:58:19 03dpeg * rd4837a454a29 10/crawl-ref/source/dat/des/variable/xom_trap.des: Baroque Xom trap vault (7hm) 19:58:29 03dpeg * rdb6067f82e68 10/crawl-ref/source/dat/des/branches/temple.des: Temple map (roderic) 19:58:30 03dpeg * r5d12b9c741d3 10/crawl-ref/source/dat/des/entry/entry_simple.des: Fix to tetris entry vault (Chamrin). #4016 20:01:08 -!- Zannick has joined ##crawl-dev 20:04:44 Ok, I think cao trunk should be fixed now, thanks for reporting that 20:04:59 Looks like the -dir option was making crawl dump saves and logfiles to each player's ttyrec directory 20:05:22 greensnark: I still can't access my save... 20:05:30 Which save was this 20:05:36 https://crawl.akrasiac.org/saves/dumps/Shiren-crawl-git-3aff67ebd4-110701-2105.tar.bz2 20:05:52 backup works, but starting a new game ignores it.. 20:06:49 (at least I think backup works - it gives a URL, but I obviously can't access the file..) 20:08:12 The backup link is correct 20:08:50 Your save is also in the right place, I'm not sure why the game isn't finding it 20:08:52 If you just logged into dgl, killing your ssh session and restarting may help 20:09:07 (i.e. you may be using a stale dgl session) 20:09:08 Yep 20:09:13 You're right - thanks! 20:09:17 Cool 20:28:25 -!- syllogism has quit [] 20:32:54 Yunor the Tortoise (L16 MDBe) (Swamp:2) 20:34:10 -!- RichardHawk has joined ##crawl-dev 20:36:21 03dpeg * r61dc04a68120 10/crawl-ref/ (3 files in 2 dirs): Lemuel's castle randomised and sourced out (carwin), #3794. 20:44:12 -!- Twilight-1 has quit [Read error: Connection reset by peer] 20:44:19 03dpeg * r1e507a6a3c8b 10/crawl-ref/source/dat/des/entry/entry_small.des: New entry vault (oiseaux), #3926. 20:44:29 -!- Twilight-1 has joined ##crawl-dev 20:50:27 paplaukes (L5 DSMo) (D:4) 21:12:27 03dpeg * r61aaa4442aa0 10/crawl-ref/source/dat/des/branches/temple.des: Slight change to autumnal temple (oiseaux), #3925. 21:12:29 03dpeg * r484861f4f7dc 10/crawl-ref/source/dat/des/entry/ (entry_large.des entry_simple.des entry_small.des): Entry vaults (co), #3008. 21:13:35 greensnark: can you repair CAO? 21:35:02 ? 21:35:10 I thought it is broken :) 21:35:24 (no idea what is really going on there, I only ever use CDO.) 21:40:36 git on CAO was a bit broken earlier... 21:40:45 now it seems fine 21:47:03 -!- neunon has quit [Ping timeout: 240 seconds] 21:48:19 03dpeg * r525e5dc15a91 10/crawl-ref/source/dat/des/entry/ (entry_large.des entry_simple.des): Entry vaults (roderic), #4062. 22:07:56 -!- Zannick has quit [Ping timeout: 264 seconds] 22:09:45 -!- Zannick has joined ##crawl-dev 22:27:52 -!- evilmike has joined ##crawl-dev 23:30:51 -!- petete has quit [Remote host closed the connection] 23:59:03 Unstable branch on crawl.akrasiac.org updated to: 0.9-a1-819-g525e5dc (32)