00:03:10 Unstable branch on crawl.develz.org updated to: 0.10-a0-641-g506fcf9 (32) 00:21:06 -!- syllogism has joined ##crawl-dev 00:26:31 -!- Zaba has quit [Ping timeout: 260 seconds] 00:28:54 -!- Twilight-1 has quit [Read error: Connection reset by peer] 00:30:21 -!- Zaba has joined ##crawl-dev 00:34:17 -!- valrus has quit [Remote host closed the connection] 00:36:39 -!- RichardHawk has joined ##crawl-dev 00:43:09 -!- ZChris13 has quit [Remote host closed the connection] 00:55:20 -!- Twilight-1 has joined ##crawl-dev 00:57:10 moin Keskitalo :) 01:00:48 -!- hoody_ has joined ##crawl-dev 01:06:17 !tell galehar uh oh, every enemy now runs from you, including intelligent ones 01:06:17 kilobyte: OK, I'll let galehar know. 01:12:00 !tell galehar precisely: if they're not adjacent, they start running instead of trying to approach you 01:12:00 kilobyte: OK, I'll let galehar know. 01:14:41 CXX tilepick-p.o 01:14:42 tilepick.cc: In function 'tileidx_t _tileidx_monster_base(int, bool, int, int, int)': 01:14:45 tilepick.cc:1640: error: 'TILEP_MONS_BLIZZARD_DEMON' was not declared in this scope 01:14:48 --- 01:14:54 moin kilobyte 01:15:05 someone forgot to add a tile for the blizzard demon :) 01:15:08 @??blizzard demon 01:15:08 blizzard demon (122) | Speed: 11 | HD: 12 | Health: 49-86 | AC/EV: 10/10 | Damage: 20, 20 | Flags: 05demonic, see invisible, fly, !sil | Res: 06magic(144), 02cold++, 10elec++, 03poison, 04rot, 13neg+++, 13torm | Vul: 04fire, 08holy++ | XP: 1525 | Sp: b.lightning (3d18), freezing cloud (2d22), airstrike (0-34), b.cold (3d21), teleport other. 01:15:14 @??ice devil 01:15:14 ice devil (163) | Speed: 10 | HD: 11 | Health: 41-80 | AC/EV: 12/10 | Damage: 1612(cold:11-32) | Flags: 05demonic | Res: 06magic(88), 12cold+++, 03poison, 04rot, 13neg+++, 13torm | Vul: 04fire, 08holy++ | XP: 788. 01:18:01 we can use the blue death tile for repainted balrugs for now 01:19:15 Unstable branch on crawl.akrasiac.org updated to: 0.10-a0-641-g506fcf9 (32) 01:27:32 03kilobyte * r7141d17aba9c 10/crawl-ref/source/rltiles/dc-mon.txt: Temporarily use the blue death tile for repainted balrugs. 01:32:50 uh oh, the fleeing bug is nastier: applies to all ranged attacks, including conjurators. Ie, it's tedious but you're invulnerable. 01:38:30 -!- dtsund has quit [Quit: dtsund] 01:44:19 -!- ainsophyao has joined ##crawl-dev 01:45:48 -!- syraine has joined ##crawl-dev 01:55:13 -!- hoody_ has quit [Remote host closed the connection] 02:03:00 ... 02:06:56 What is the chance of getting a specific amulet, if jewellery is chosen as the loot for a vault? Are all amulets equally likely to appear on the ground as loot? 02:07:54 uh 02:07:56 no 02:08:07 item generation is a fall-through thing. 02:09:07 You're saying that it starts with a certain amulet then goes to the next amulet? 02:09:23 basically, yes. 02:09:37 Alright, what's the first GOTO? 02:09:39 it starts with someothing then has a chanc eof rpelacing it with something else, and thenanother chance of replacing it with something else. 02:09:49 Okay. 02:09:58 I'd check the item generation code :) 02:10:08 Yes, I realized that as I gave the question 02:10:17 I just couldn't find the piece of code I was searching for 02:10:19 =3 02:10:26 What exactly do you want to know? 02:10:34 Or rather, why? 02:11:04 I understand your question, but it's really difficult to work out the exact chances of something showing up, in my experience, so the data might not be too useful if you need it for a certain reason/purpose. 02:11:41 I want to determine the process by which amulets are derived so that I can create a minivault group containing one amulet in each vault without having identical amulets in all of them. 02:12:30 Each amulet positioned such that it is impossible to acquire without defeating the guardian like in the Ossuary map I got the idea from. 02:13:05 I just wanted to make sure that there wasn't one kind that was especially common, making the whole venture a bit unrewarding. 02:13:41 you can place specific amulets :) 02:13:59 there are only a few 02:15:11 That was the whole thing; I don't want to have the player defeat like twelve monsters through a powerful build then get one amulet of each type. I was hoping that the usual random distribution gave let's say, a good chance of having each type with a smaller chance of getting certain types, so you'd have two of one amulet, two of another and maybe one of a couple of types after that. 02:15:32 So I guess I'll just put that in by myself (= 02:16:23 thank you for your help 02:16:53 -!- Pacra has quit [Ping timeout: 260 seconds] 02:17:35 uh 02:17:52 you can still specify every amulet and give a weight for each one :) 02:18:01 which allows you to influence which one is placed in the vault 02:18:03 That's what I meant I would do 02:18:06 Yes, I will do that 02:18:19 Thank you 02:18:50 which places the choice of the item in the vault definition rather than in the item generatoin code, so it is uninfluenced by thegeneric chances 02:31:52 -!- Zaba has quit [Ping timeout: 260 seconds] 02:35:51 -!- Zaba has joined ##crawl-dev 02:54:54 -!- syraine has quit [Read error: Connection reset by peer] 03:06:58 -!- galehar has joined ##crawl-dev 03:16:05 -!- evilmike has quit [Quit: something happened] 03:19:02 hi galehar! I'm afraid the retreating AI pretty much breaks the game :( 03:21:32 how? 03:21:32 galehar: You have 2 messages. Use !messages to read them. 03:21:36 !messages 03:21:37 (1/2) kilobyte said (2h 15m 19s ago): uh oh, every enemy now runs from you, including intelligent ones 03:21:44 !messages 03:21:45 (1/1) kilobyte said (2h 9m 44s ago): precisely: if they're not adjacent, they start running instead of trying to approach you 03:21:56 damn 03:23:00 this includes conjurers: you can keep enemies at bay by plinking them with a spell 03:24:37 ok, I think I know what's wrong 03:28:38 sorry for poor testing 03:28:44 is it on CDO already? 03:30:06 03galehar * r2db25626906d 10/crawl-ref/source/mon-behv.cc: Fix the new retreating behaviour. 03:31:15 Unstable branch on crawl.develz.org updated to: 0.10-a0-643-g2db2562 (32) 03:35:02 -!- syraine has joined ##crawl-dev 03:36:50 <3 03:43:46 -!- syraine_ has joined ##crawl-dev 03:43:55 -!- syraine has quit [Ping timeout: 252 seconds] 03:43:59 -!- syraine_ is now known as syraine 03:51:05 -!- syraine has quit [Ping timeout: 260 seconds] 03:56:27 -!- syraine_ has joined ##crawl-dev 03:57:29 -!- Pacra has joined ##crawl-dev 03:58:09 -!- syraine_ is now known as syraine 04:00:33 -!- syraine has quit [Read error: Connection reset by peer] 04:00:55 -!- syraine_ has joined ##crawl-dev 04:00:57 -!- syraine_ is now known as syraine 04:08:09 -!- syraine has quit [Ping timeout: 250 seconds] 04:08:52 -!- syraine has joined ##crawl-dev 04:18:44 -!- Textmode has quit [Ping timeout: 276 seconds] 04:21:10 -!- Pacra has quit [Read error: Connection reset by peer] 04:21:20 -!- Pacra has joined ##crawl-dev 04:31:38 -!- Pacra has quit [Ping timeout: 260 seconds] 04:50:51 -!- elliptic has joined ##crawl-dev 04:51:22 -!- Pacra has joined ##crawl-dev 05:00:04 OSX builds of master branch on crawl.develz.org updated to: 05:15:01 The butterfly flaps its wings in an oddly mesmeric pattern. 05:15:11 _You feel entranced by its movement. 05:15:41 butterflies were buffed? 05:15:42 oy vey the spam... and guess how loud the *plonk* of it hitting the bottom of my mute: list was 05:15:54 no, made @#$%^&* spammy like hell :p 05:16:07 hehe 05:16:14 ah, so that's like cantrip 05:16:20 the problem is not with spam mostly, but with lightred normally meaning vital messages 05:18:37 aaaah so much better now :) 05:19:01 commit, kilobyte? :D 05:19:48 second episode, btw: http://roguelikeradio.blogspot.com/ 05:20:53 Napkin: my own config, whoever added that might take offense in me deleting flavour or such 05:21:19 but.. as you said.. isn't lightred inconsistent? 05:22:09 lightred? 05:22:14 naughty :o 05:22:47 commit 2c28e582: MarvinPA, over a month ago 05:23:02 added roguelike radio in "related" link-section 05:23:43 everyone was busy with the tournament I guess, and I didn't run into a big pile of butterflies until now 05:24:29 there were some complaints back then 05:24:40 big piles of butterflies aren't that common though 05:27:00 good, so it's just not me whining then 05:27:10 would lightwhite be enough? 05:27:23 yes 05:27:24 iirc the complaints then were not about the color though... they were about the messages stopping autoexplore 05:33:35 Kilobyte: IIRC previously that had a pink color 05:41:09 -!- syraine has quit [Quit: ChatZilla 0.9.87 [Firefox 6.0.1/20110830092941]] 05:42:03 pink is bright but doesn't match an existing convention 05:45:13 galehar: another thing: monsters with ranged attacks shouldn't flee either. Here, I've got an orc knight with a polearm, I'm duking it with him through a river of lava in a bailey, he can't run as other orcies surround him, but he's keeps begging for mercy :p 05:52:25 monsters with ranged attacks shouldn't flee, but I didn't handle reaching 05:52:34 thanks for noticing, will fix 05:58:33 -!- monqy has joined ##crawl-dev 06:05:06 another minor issue: should this retreat be considered to be fear or a tactical decision? Currently you get stab bonuses, and they use fear-related speech. 06:05:06 (I'm not sure if the former is good or not) 06:05:06 the latter could be better done with a different set: "come here, you coward" and so on 06:05:10 good point 06:06:56 yeah, I think it would be better to consider it as a tactical manoeuvre 06:15:15 a yet different question, this time more for my side: is there any reason berserk would shorten your weapon? Currently you can use polearms only at range 1. 06:15:16 -!- Ashenzari has quit [Ping timeout: 252 seconds] 06:15:59 -!- Ashenzari has joined ##crawl-dev 06:15:59 -!- The topic of ##crawl-dev is: DCSS Development | Logs: http://tozt.net/crawl | People with +v have commit access. | Please keep general crawl-related chit-chat to ##crawl. | Dev wiki: http://crawl.develz.org/wiki | Long stuff to pastebin.ca please | Immortal Warwalrus and Crazy Yiuf forever. 06:18:38 yeah, that's a good point too. I think berserk prevents using reaching just because it uses the evoke interface. 06:18:50 it doesn't make much sense that you wouldn't be able to use a polearm properly just because you are berserk 06:19:04 so monsters with polearms getting reaching is pretty annoying... because it means that you care what weapon an orc is carrying much more than you used to 06:19:24 so you spend more time checking out the weapons of orc warriors, knights, etc 06:20:30 elliptic: play tiles, you'll see the weapon ;) 06:22:54 galehar: cut your legs off so you'll always have a chair with you so you can sit down when tired ;) 06:22:54 still, elliptic has a valid point 06:22:54 it might help a bit if we could display the visible weapon on the monster list 06:22:54 I used to have a --more-- on being hit by reaching; in fact, I didn't remove it so I wonder why it doesn't trigger 06:23:33 kilobyte: unlike with reaching weapons, here you actually know for sure that the orc warrior has reaching before it hits you the first time... if you can be bothered to x over to it 06:24:12 more reaching on monsters is tactically interesting, but I'm worried about this part of the interface 06:25:10 there's the "comes into view. it is wielding a (polearm)" message but I'm afraid that's probably not good enough 06:25:43 it doesn't tell the weapon if there's more than one monster coming into view. 06:25:46 we can use a slashed o as the glyph :p 06:26:10 a good part of orcs, and like half of gnolls use polearms 06:26:11 galehar: at which point you x or ^X / hence the not good enough 06:26:12 if only this were just about orcs, I would approve ;P 06:27:50 enhance monster list +1 06:31:24 the main issue I see with making the monster list say stuff like "o orc warrior (runed trident)" is what do you do when there are two orc warriors in sight 06:31:24 Is there a reason you can't berserk while under silence field, by the way? 06:31:24 still it might help 06:31:24 what about when it doesn't show up on mlist do to too many monsters, or is that okay 06:31:24 RichardHawk: invocations require speaking, which isn't possible under silence 06:31:24 you can still berserk with an amulet of rage or a potion, just not with the trog ability 06:31:24 you can drink potions 06:31:24 two orc warriors is one case, more than 4 or 5 different monsters is another 06:31:24 soft monster list by distance? stop summing them up? key to switch between summary and detailed&distance based mlist? 06:31:24 *sort 06:31:24 detailed mlist key exists 06:31:25 it's ^X 06:31:25 well, ctrl-x or whatever is what you really do if you want all the details about monsters in sight 06:31:30 admittedly, it also shows items and features?? 06:31:49 and that much information wouldn't fit in the regular mlist 06:33:30 one option is to show the glyph only for the monsters' weapons in the regular mlist... so oo (( 2 orc warriors 06:33:37 or )), I forget which is weapons 06:33:48 and you could tell a little bit from the colors 06:33:58 (also whether a monster is wielding a weapon) 06:36:11 showing weapon colour would only be useful informationwise in the case of polearms, artefacts, and demon weapons, I think. is that acceptable? 06:41:19 well, you'd also know whether a monster has a weapon this way... very important early on 06:41:20 ah, right 06:41:20 -!- Ashenzari has quit [Ping timeout: 260 seconds] 06:42:10 -!- Ashenzari has joined ##crawl-dev 06:42:10 -!- The topic of ##crawl-dev is: DCSS Development | Logs: http://tozt.net/crawl | People with +v have commit access. | Please keep general crawl-related chit-chat to ##crawl. | Dev wiki: http://crawl.develz.org/wiki | Long stuff to pastebin.ca please | Immortal Warwalrus and Crazy Yiuf forever. 06:44:20 how about: by default, mlist shows monster's weapon and merge entries only if the description strings match. When it hit the display limits, it starts loosening up merge restriction and you start using ^X more 06:45:24 galehar: sounds reasonable to me 06:47:00 actually, this could be extended: merging genuses, even friendliness 06:52:12 yeah, the monster list algorithm can be improved 06:52:12 and it can be used with monsters coming into sight too 06:52:12 and with the tiles' monster panel which I'd like to change it to make it like the monsters' list 06:52:12 -!- Pingas has joined ##crawl-dev 06:52:12 -!- edlothiol has joined ##crawl-dev 06:52:12 anyway, have to go 06:52:12 'later 06:52:12 -!- Ashenzari has quit [Ping timeout: 260 seconds] 06:53:14 -!- Ashenzari has joined ##crawl-dev 06:53:14 -!- The topic of ##crawl-dev is: DCSS Development | Logs: http://tozt.net/crawl | People with +v have commit access. | Please keep general crawl-related chit-chat to ##crawl. | Dev wiki: http://crawl.develz.org/wiki | Long stuff to pastebin.ca please | Immortal Warwalrus and Crazy Yiuf forever. 06:58:32 03edlothiol * rd6cc18c08500 10/crawl-ref/source/libweb.cc: Fix Webtiles crashing on startup after 58b79301. 07:02:26 03edlothiol * r71b91416cd02 10/crawl-ref/source/webserver/server.py: Improve Webtiles server logging. 07:02:26 -!- Ashenzari has quit [Ping timeout: 276 seconds] 07:03:07 -!- Ashenzari has joined ##crawl-dev 07:03:07 -!- The topic of ##crawl-dev is: DCSS Development | Logs: http://tozt.net/crawl | People with +v have commit access. | Please keep general crawl-related chit-chat to ##crawl. | Dev wiki: http://crawl.develz.org/wiki | Long stuff to pastebin.ca please | Immortal Warwalrus and Crazy Yiuf forever. 07:09:50 -!- edlothiol has quit [Remote host closed the connection] 07:31:05 -!- Mu_ has joined ##crawl-dev 07:31:32 -!- Euphoria has quit [Quit: Leaving] 08:09:27 -!- hoody_ has joined ##crawl-dev 09:21:42 -!- Pingas_ has joined ##crawl-dev 09:22:39 -!- Pingas has quit [Ping timeout: 276 seconds] 09:31:59 are wights supposed to use drain attack from a distance if they have a polearm of reaching? 09:34:08 03kilobyte * r850e6fdc2188 10/crawl-ref/source/ (dat/lua/autofight.lua evoke.cc): Fix polearms/whips of reaching magically shortening when you're berserk. 09:34:08 03kilobyte * re1c006c99a9b 10/crawl-ref/source/ (dat/des/builder/layout.des l_dgnbld.cc): A new layout type: twisting cavern. 09:34:08 03kilobyte * rfaf8ebd7c0e2 10/crawl-ref/source/dat/database/monspeak.txt: Don't allow butterflies to use the "important warning" colour. 09:35:41 Unstable branch on crawl.develz.org updated to: 0.10-a0-648-gfaf8ebd (32) 09:44:35 Am I unable to create new pages on the DCSS wiki or just missing something blatantly obvious? Had a suggestion for humans I wanted to put there but they have no page. 09:46:26 RichardHawk: Go to the url where you want the page; then click "Create Page" 09:46:40 (you have to sign in first, with you cdo account) 09:47:21 mumra: Ah, I see. Much appreciated. 09:51:16 -!- Pingas has joined ##crawl-dev 09:53:13 -!- Pingas_ has quit [Ping timeout: 260 seconds] 09:58:12 -!- evilmike has joined ##crawl-dev 10:27:05 right now hellions and balrugs are both red 2's. The hellions flash, but it's still fairly close, and one of the colours hellions flash to is the same as balrugs 10:35:58 -!- Wensley has joined ##crawl-dev 10:36:10 evilmike: right 10:36:38 rwbarton proposed in ##crawl to remove the balrug colour from the hellions flashing, seems like a reasonable idea 10:39:46 galehar: What's all this about abyss morphing no longer destroying exits? >:( 10:39:47 Wensley: You have 1 message. Use !messages to read it. 10:41:24 -!- st_ has joined ##crawl-dev 10:51:33 what colours should they be? I see only lightred is the only free fitting one. 10:52:01 cacos, hell beasts 10:52:45 hm, i guess that much was true even before the tier changes since there were still sun demons on yellow 10:53:02 yeah, it wasnt a big problem with sun demons though, because they were sun demons 10:54:08 03kilobyte * rfd30c6b1312b 10/crawl-ref/source/dat/des/builder/layout.des: Undo a testing weight. 10:54:18 03kilobyte * r77465a5a808a 10/crawl-ref/source/dat/des/builder/layout.des: Don't use brickwork in layout_twisted_cavern, pick some natural tiles. 10:54:18 03kilobyte * re4c7c19fe847 10/crawl-ref/source/ (10 files in 3 dirs): Often, make the cavern passage crystal. With a new set of tiles! 10:54:40 Unstable branch on crawl.develz.org updated to: 0.10-a0-651-ge4c7c19 (32) 10:54:53 need to rebuild CDO, every floor gets the cavern passage now :p 10:55:12 heh, I see ccache works well 10:56:38 did you try the crystal floor tile for crystal passages? maybe the tiled look is too "unnatural" 11:01:03 that "crystal floor" is thoroughly manufactured, doesn't fit there at all 11:02:44 oh well 11:06:34 why does the "default_manual_training" option exist by default in the 0.9 rc, but not in the trunk rc? 11:06:44 or maybe I am looking at the wrong trunk rc? 11:06:52 http://crawl.akrasiac.org/rcfiles/crawl-git/Wensley.rc 11:09:17 it probably would if you reset the trunk rc 11:10:10 -!- cx1213 has quit [Ping timeout: 252 seconds] 11:10:19 i think it just doesn't overwrite existing configs 11:11:04 really need a way to automatically do diffs on old and new rcfiles and import the merged version 11:11:37 -!- Twilight-1 has quit [Ping timeout: 272 seconds] 11:12:33 -!- Twilight-1 has joined ##crawl-dev 11:12:54 -!- FaMott has joined ##crawl-dev 11:13:01 Friends, I have a bug to report: http://i.imgur.com/ikPZZ.png 11:14:23 awesome, how did you do that? 11:14:38 Watching someone's game 11:14:40 anyone's game 11:14:42 on webtiles 11:14:45 new xom effect 11:17:29 -!- cx1213 has joined ##crawl-dev 11:30:38 -!- ais523 has joined ##crawl-dev 11:33:19 trippy 11:34:34 -!- dtsund has joined ##crawl-dev 11:38:08 03MarvinPA * r291262a008d7 10/crawl-ref/source/dat/des/entry/large.des: Fix an entry vault with no exit 11:38:27 impressive. have you zoomed the page perhaps ? 11:41:19 mumra, no, no zoom or anything 11:41:58 FaMott: Which browser? 11:42:04 Chrome 11:42:22 13.0 most recent if you need to know version 11:43:54 Oh! 11:43:57 it was a zoom glitch 11:43:59 how about that. 11:44:02 oh wait 11:44:03 Heh. 11:44:05 still breaks. 11:44:14 I dunno, I'm updating Chrome to try. 11:44:25 But with my older 13 it's fine for me. Also tried Firefox. 11:46:21 Still fine for me after updating. 11:47:27 odd. 11:47:48 -!- valrus has joined ##crawl-dev 11:48:31 great, now everything is stacking vertically 11:49:44 FaMott: Try ctrl-shift-del and clear your cache? Dunno. 11:50:50 imma try it on firefox , sec 11:51:19 I can't even log on to webtiles 11:51:24 just says the connection is closed 11:54:58 It did take a while to start to spec a game after I logged in, but it worked. 11:56:55 I think some other people said they were having problems, earlier. 11:56:58 But I see one of them playing now. 12:23:19 -!- Pandoora has joined ##crawl-dev 12:23:22 -!- Pandoora has quit [Remote host closed the connection] 12:25:39 -!- Wensley has quit [Quit: ChatZilla 0.9.87-rdmsoft [XULRunner 1.9.0.17/2009122204]] 12:28:39 -!- Twilight-1 has quit [Read error: Connection reset by peer] 12:29:44 -!- Wensley has joined ##crawl-dev 12:49:02 when did the bog mummy description stop mentioning canopic jars? 12:51:49 -!- Twilight-1 has joined ##crawl-dev 12:52:52 Wensley: how could they have one? A bog mummy is natural. 12:53:43 kilobyte: it used to say something like "instead of canopic jars and , this mummy was preserved by the brackish waters of the swamp" 12:54:23 I thought the old description was actually one of the more well-written monster descriptions, which is why I'm a little curious why it was changed :P 12:58:46 uhm no, the description is the same as the one due initially put there 13:02:43 Wensley: You're thinking of bog mummy[2]? 13:03:17 maybe I was 13:03:26 (that entry is better) 13:03:57 ??bog mummy[2] 13:03:57 bog mummy[2/4]: A brackish-water variant of your average mummy. Instead of canopic jars and copious fluttering bandages, this creature was pushed face first into a bog and left to mummify naturally over hundreds of years. 13:04:57 anyway, on the tavern they have been compiling a huge spreadsheet with improved monster descriptions, and dpeg's approved it so I am compiling that all into a patch right now 13:05:07 this is how I came to notice 13:13:43 -!- Pacra has quit [Ping timeout: 260 seconds] 13:14:57 When trying to set up a henzell bot, where does the milestones file come from? 13:23:13 -!- valrus has quit [Remote host closed the connection] 13:26:25 galehar: I just realized how much more dangerous gnoll packs got now! 14:04:57 -!- dtsund has quit [Quit: Changing OS...] 14:10:52 -!- ainsophyao has quit [Remote host closed the connection] 14:14:38 -!- dtsund has joined ##crawl-dev 14:22:35 -!- Twilight-1 has quit [Ping timeout: 276 seconds] 14:27:48 Bah. I cannot grasp the Henzell syntax. 14:29:35 considering that I think henzell is written in perl, that sounds reasonable 14:32:45 True enough! But I meant for viewing his statistics. 14:33:23 Also, it's only about 2/5 Perl. The rest is an amalgamation of C++, Ruby, and Python. 14:33:34 statistics? 14:33:42 you mean the !lg syntax? 14:33:44 You know, winrates and such. 14:33:45 Yeah. 14:33:47 ??listgame 14:33:48 listgame[1/6]: !lg command displays info about past games. The manual is available here: http://github.com/greensnark/dcss_henzell/raw/master/docs/listgame.txt 14:34:09 Yeah, I've read that entry, and the examples one. 14:34:12 you can do a lot of complex things that I've never bothered to learn 14:34:36 so what's the issue? 14:34:49 is there something specific you want to query? 14:35:19 I was interested in comparing winrates and playrates over versions 14:35:33 okay 14:36:14 !lg * s=cv / won 14:36:19 5425/1265732 games for *: 531/83485x 0.9-a [0.64%], 582/101623x 0.8-a [0.57%], 117/21247x 0.9 [0.55%], 435/84396x 0.3 [0.52%], 705/144565x 0.4 [0.49%], 636/142405x 0.8 [0.45%], 157/37709x 0.6-a [0.42%], 679/179420x 0.7 [0.38%], 68/18057x 0.10-a [0.38%], 838/227790x 0.5 [0.37%], 149/42390x 0.2 [0.35%], 62/17864x 0.1 [0.35%], 317/105430x 0.6 [0.30%], 106/36719x 0.7-a [0.29%], 22/9231x 0.4-a [0.24%],... 14:36:46 Oh, that's why it was never working. I was /msg'ing Henzell. 14:36:57 ha 14:36:59 :) 14:37:27 wow, that's an interesting ordering of versions 14:37:33 perhaps some have more startscummers than others 14:37:42 what's "max level reached"? 14:37:50 where? 14:38:30 in any case, either the game keeps getting easier or the players keep getting better 14:38:42 ah, I'll go by time instead 14:39:00 !lg * turns>5000 s=cv / won 14:39:04 5425/205011 games for * (turns>5000): 435/11248x 0.3 [3.87%], 582/16307x 0.8-a [3.57%], 531/16970x 0.9-a [3.13%], 705/24774x 0.4 [2.85%], 838/29654x 0.5 [2.83%], 22/856x 0.4-a [2.57%], 117/4643x 0.9 [2.52%], 68/2824x 0.10-a [2.41%], 636/26429x 0.8 [2.41%], 62/2737x 0.1 [2.27%], 157/7087x 0.6-a [2.22%], 679/31858x 0.7 [2.13%], 149/7055x 0.2 [2.11%], 317/15729x 0.6 [2.02%], 106/5281x 0.7-a [2.01%], ... 14:39:20 that shows something of a different trend 14:39:26 we can assume that alpha versions attract better players 14:39:31 oh right, I need to exclude quitters and leavers 14:39:57 !lg * s=cv ktype!=(quitting|leaving) / won 14:40:03 but to me, that shows the game getting harder over time, except that 0.6 and 0.7 were harder than the more recent versions 14:40:06 Yeah. Early versions would have more startscummers, because startscumming was possible. 14:40:08 5425/926512 games for * (ktype!=(quitting|leaving)): 582/65186x 0.8-a [0.89%], 531/65180x 0.9-a [0.81%], 435/54037x 0.3 [0.81%], 705/104763x 0.4 [0.67%], 68/10791x 0.10-a [0.63%], 22/3498x 0.4-a [0.63%], 117/18830x 0.9 [0.62%], 838/135781x 0.5 [0.62%], 157/29589x 0.6-a [0.53%], 636/126534x 0.8 [0.50%], 106/23060x 0.7-a [0.46%], 679/155266x 0.7 [0.44%], 149/34917x 0.2 [0.43%], 317/77023x 0.6 [0.41%... 14:40:25 (And good riddance.) 14:40:40 0.3, 0.4, 0.9, 0.5, 0.8, 0.7, 0.2 14:40:53 , 0.6 14:41:18 It doesn't surprise me to see 0.6 bringing up the rear 14:41:31 ais523: by that same logic, if good players prefer to play the alpha versions, that means that the stable versions see a disproportionate amount of poor players 14:41:37 With all the heavily-armored berserkers going "why can't I win this bloody game" 14:41:37 Wensley: I think so, too 14:41:55 I'd expect poor players to have less of a tendency to improve over time than good players 14:42:15 I'd also like to be able to specify "not during a tournament", but I'm not sure how possible that is 14:42:24 as tournaments probably skew statistics too 14:42:37 really we need some way to track success on a per-player basis, to see average rates of improvement 14:42:41 we can exclude tournaments 14:42:49 hm, I think 14:42:51 Wensley: that's not very hard to do, per-player 14:42:56 but for all players 14:43:16 And I wanted to take a look at it for some specific classes 14:43:24 I want to see how many games on average it takes to get a win, and from there how many games it takes to get an allrune win 14:43:50 I can practically guarantee the first number is a lot smaller than the second. 14:43:51 those stats would be utterly meaningless 14:43:56 a lot of people play offline before they go online 14:44:01 yes, sadly 14:44:16 I like playing offline, on the tram etc. 14:44:22 for instance, i didn't really start online until i had about 6 wins 14:44:22 (we just need global usernames and reporting of stats from local games, with drm to ensure no cheaters :P ) 14:44:26 I like playing offline too. 14:44:39 Minimap! 14:44:44 and my first online win was just a single tournament win, then i went back to offline for a good while 14:44:59 Yeah, I think my first online win was allrune. 14:46:01 yes, hard to get anything very conclusive from these statistics 14:46:24 one issue with looking at winrates is that it doesn't tell you where people are dying 14:47:04 that would be cool 14:47:05 a death heatmap 14:47:08 if more people are dying on d:1, this brings winrate down... but doesn't necessarily have anything to do with how hard the game is to win, just how hard it is to start 14:47:32 you can do stuff like this: 14:48:14 !lg * ckiller!=quitting ckiller!=leaving xl>=10 s=cv / won 14:48:15 5425/87480 games for * (ckiller!=quitting ckiller!=leaving xl>=10): 435/4959x 0.3 [8.77%], 582/7297x 0.8-a [7.98%], 531/7341x 0.9-a [7.23%], 705/9838x 0.4 [7.17%], 838/12542x 0.5 [6.68%], 22/352x 0.4-a [6.25%], 118/2000x 0.9 [5.90%], 68/1172x 0.10-a [5.80%], 62/1081x 0.1 [5.74%], 149/2629x 0.2 [5.67%], 636/12182x 0.8 [5.22%], 156/3043x 0.6-a [5.13%], 679/13520x 0.7 [5.02%], 317/6543x 0.6 [4.84%], ... 14:48:31 (winrate among games that reach XL 10) 14:48:55 Curious that 0.4 ranks higher than 0.4-a 14:49:00 but really tourneys distort such things a lot 14:49:01 oh, I was using ktyp before, not ckiller 14:49:03 was cBlink still a low level spell in 0.3? 14:49:10 dtsund: trunk didn't really get popular until 0.6-a 14:49:25 Was 0.4 the one with the bug that sometimes gave you hilarious amounts of skill increase? 14:49:26 also note 0.4-a has so few games 14:49:36 Ah, yeah. 14:49:57 pretty sure there have been such bugs at multiple times in crawl history :P 14:50:14 Yeah, but I think one made it into a stable build at some point. 14:50:20 one of the early 0.4 versions multipled xp by several times 14:50:25 it might have been as much as 10x 14:51:28 what in the world is csplatco 14:51:34 !lg * ckiller!=quitting ckiller!=leaving cv=~0.9 when!=t11 when!=t11a / won 14:51:39 56971 games for * (ckiller!=quitting ckiller!=leaving cv=~0.9 when!=t11 when!=t11a): N=417/56971 (0.73%) 14:51:50 rawrmage: C-splat is a termcast thing continually broadcasting splat deaths. 14:51:58 how can I turn it off from henzell... 14:52:03 besides removing the code 14:52:05 !lg * ckiller!=quitting ckiller!=leaving cv=~0.5 when!=t09 / won 14:52:09 113239 games for * (ckiller!=quitting ckiller!=leaving cv=~0.5 when!=t09): N=573/113239 (0.51%) 14:52:57 !lg * ckiller!=quitting ckiller!=leaving cv=~0.5 xl>=10 when!=t09 / won 14:52:58 10537 games for * (ckiller!=quitting ckiller!=leaving cv=~0.5 xl>=10 when!=t09): N=573/10537 (5.44%) 14:53:03 elliptic: I don't think the t11a tourney would distort 0.9-a's stats 14:53:10 wensley: oh right :) 14:53:13 Yeah, tourney skewed that up. I thought that'd happen. 14:53:14 :P 14:53:30 !lg * ckiller!=quitting ckiller!=leaving cv=~0.9 when!=t11 xl>=10 / won 14:53:31 6494 games for * (ckiller!=quitting ckiller!=leaving cv=~0.9 when!=t11 xl>=10): N=417/6494 (6.42%) 14:54:06 Does this include alphas? 14:54:13 !lg * ckiller!=quitting ckiller!=leaving cv=~0.7 when!=t10 xl>=10 / won 14:54:14 yes 14:54:14 11680 games for * (ckiller!=quitting ckiller!=leaving cv=~0.7 when!=t10 xl>=10): N=464/11680 (3.97%) 14:54:32 !lg * ckiller!=quitting ckiller!=leaving cv=0.9 when!=t11 xl>=10 / won 14:54:33 513 games for * (ckiller!=quitting ckiller!=leaving cv=0.9 when!=t11 xl>=10): N=27/513 (5.26%) 14:54:36 you can do s=cv to those to see each version 14:54:46 what is the difference between cv and v? 14:55:11 v gives stuff like 0.9.0 14:56:01 elliptic? 14:56:15 could you send me the morgue logic of the scoring scripts? 14:56:16 !lg * ckiller!=quitting ckiller!=leaving cv=~0.8 when!=t11a xl>=10 / won 14:56:17 16363 games for * (ckiller!=quitting ckiller!=leaving cv=~0.8 when!=t11a xl>=10): N=985/16363 (6.02%) 14:56:34 -!- galehar has quit [Remote host closed the connection] 14:56:39 napkin: oh, right, I forgot about looking at that... 14:56:45 me too ;) 15:00:51 https://github.com/elliptic/dcss_tourney/commit/44f432bff9e34a2f57b7f45e0d1a3a2d00ed09a9 15:00:58 is the quick hack I put into the tourney scripts 15:02:48 -!- valrus has joined ##crawl-dev 15:08:18 hm, getting sequell to believe that these CDO games aren't trunk will require some rewriting 15:09:05 since currently it looks like sequell determines the morgue directory based solely on the location of the logfile 15:10:47 elliptic: what's the git command to make a patch that consists of one commit? 15:12:32 goit format-patch -1 15:12:42 last one commit: -1 15:13:09 yeah, git format-patch -1 should work if it is the most recent commit 15:13:20 thanks 15:14:29 thanks, elliptic 15:14:43 napkin: are you looking at fixing the morgues for the scoring scripts or for sequell or both? 15:15:31 i don't have access to sequell - it's running on greensnark's root server, right? 15:16:12 so, only the scoring scripts 15:16:26 ah 15:18:10 does anyone know if god colours are defined anywhere in source, other than the altar symbol colours ? 15:18:14 the scoring scripts look easier to fix to me because they are simpler and because I know a lot more python than ruby :P 15:21:08 Huh. Seems like some 3/4 of all Wanderer games post-tournament were startscums. 15:21:49 !lg * wn rstart>=20110730 s=ckiller 15:21:49 6635 games for * (wn rstart>=20110730): 4141x leaving, 1962x quitting, 55x a kobold, 52x a hobgoblin, 46x a snake, 24x an orc wizard, 22x a player ghost, 19x an ogre, 18x a gnoll, 17x a giant gecko, 16x an orc, 15x a goblin, 15x an orc priest, 14x a jackal, 14x Sigmund, 13x a centaur, 9x a worker ant, 8x a worm, 8x an orc warrior, 8x Grinder, 7x a killer bee, 6x Terence, 6x cloud, 6x an ice beast,... 15:22:00 What is crawl coded in? 15:22:02 C? 15:22:04 c++ 15:22:06 Oh. 15:22:09 I'm learning C :B 15:22:09 C++, sadly. 15:22:10 !lg * wn rstart>=20110730 ckiller=leaving|quitting s=name 15:22:10 6103 games for * (wn rstart>=20110730 ckiller=leaving|quitting): 4309x Maurice, 714x Xenene, 502x Sebi, 219x DEATHRAY, 144x Xiberia, 66x flashdoom, 42x zogre, 22x Raelwyn, 21x moohaus, 11x Elynae, 9x Uamenti, 3x m1nced, 3x johnnyzero, 3x Nile, 2x JollyPong, 2x TheOptimist, 2x ryw, 2x Orshelack, 2x TheWanderer, 2x Deen, 1x morazor, 1x Karandash, 1x Palladion, 1x Pumpking, 1x Grildrak, 1x sirprim, 1... 15:22:11 its not surprising at all that wanderers are mostly startscums 15:22:28 dtsund: 2/3 of the startscumming is one person though :P 15:22:31 I could probably learn C++, I already know Java 15:22:44 what is surprising is that most of them aren't sebi 15:23:34 !lg * rstart>=20110730 ckiller=leaving|quitting xl=1 s=cls x=cdist(name) 15:23:34 8028 games for * (rstart>=20110730 ckiller=leaving|quitting xl=1): 142x Fighter [73], 91x Wizard [52], 75x Stalker [51], 70x Berserker [49], 56x Transmuter [44], 6059x Wanderer [40], 44x Conjurer [37], 56x Skald [37], 53x Necromancer [37], 147x Enchanter [36], 59x Hunter [26], 37x Monk [26], 29x Ice Elementalist [25], 28x Fire Elementalist [23], 27x Air Elementalist [23], 51x Warper [23], 33x Summ... 15:23:44 FaMott: C++ is horrifying in ways that Java isn't. Just a warning. 15:23:52 dtsund, how so? 15:24:13 http://yosefk.com/c++fqa/ 15:24:54 If you want software that is "easy to change", stay away from C++ classes. 15:24:56 ha ha ha 15:25:38 "There is no reason to use C++ for new projects. However, there are existing projects in C++ which might be worth working on. Weighting the positive aspects of such a project against the sad fact that C++ is involved is a matter of personal judgment." 15:25:52 DCSS Text Improvement Taskforce: Monster Descriptions, Round 1 (https://crawl.develz.org/mantis/view.php?id=4528) by Wensleydale 15:26:30 In other news: YOu guys improved monster AI to retreat if they're getting attacked by something they cant reach? 15:26:57 that was the idea, yes 15:29:13 Cool 15:30:34 -!- MarvinPA_ has joined ##crawl-dev 15:33:29 -!- st_ has quit [Remote host closed the connection] 15:33:39 -!- st_ has joined ##crawl-dev 15:33:43 -!- MarvinPA has quit [Ping timeout: 260 seconds] 15:39:30 -!- MarvinPA_ has quit [Read error: Connection reset by peer] 15:46:37 -!- MarvinPA has joined ##crawl-dev 16:25:31 -!- mumra has quit [Quit: Want to be different? Try HydraIRC -> http://www.hydrairc.com <-] 16:28:31 -!- Pacra has joined ##crawl-dev 16:47:04 -!- hoody_ has quit [Remote host closed the connection] 16:58:56 -!- MarvinPA has quit [Read error: Connection reset by peer] 17:00:10 -!- MarvinPA has joined ##crawl-dev 17:02:17 15:02:26 < Henzell> valrus the Spry (L23 HaGl) is cast into the Abyss! (deep elf demonologist) (Elf:5) 17:02:20 15:02:33 < valrus> Item buggily placed in feature at (45, 38). 17:09:31 -!- valrus has quit [Remote host closed the connection] 17:11:25 Gold in a wall in Abyss (https://crawl.develz.org/mantis/view.php?id=4529) by smock 17:38:10 -!- purge has joined ##crawl-dev 17:42:49 -!- Textmode has joined ##crawl-dev 17:49:29 right 17:49:36 who wrote _scale_spawn_parameter 17:49:38 i want names 17:49:52 no return statement should be seven lines of code 17:50:48 -!- Wensley has quit [Ping timeout: 260 seconds] 17:53:32 what's so bad about it? not that I wrote it or anything. I don't even know what it is. 17:54:24 It's a single statement with a ? branch within a ? branch. I would argue that anything like that really should just be written with if statements for readability. 17:57:09