00:02:27 Unstable branch on crawl.akrasiac.org updated to: 0.9-a1-1014-gfb4a452 (32) 00:03:12 Unstable branch on crawl.develz.org updated to: 0.9-a1-1014-gfb4a452 (32) 00:18:26 Windows builds of master branch on crawl.develz.org updated to: 0.9-a1-1014-gfb4a452 00:19:38 LuckyNed the Skirmisher (L2 DSMo) (D:1) 00:43:02 -!- Guest13344 has quit [Read error: Connection reset by peer] 00:52:25 -!- Twilight has joined ##crawl-dev 00:52:51 -!- Twilight is now known as Guest37064 01:14:33 @??giant slug 01:14:33 giant slug (03j) | Speed: 6 | HD: 10 | Health: 54-77 | AC/EV: 0/2 | Damage: 23 | Flags: amphibious | Res: 06magic(40), 12drown | Chunks: 09poisonous | XP: 96. 01:19:32 @??gila monster 01:19:32 gila monster (06l) | Speed: 10 | HD: 5 | Health: 21-39 | AC/EV: 3/12 | Damage: 20 | Flags: cold-blooded | Res: 06magic(20) | Chunks: clean/none/unknown | XP: 110. 01:20:40 -!- monqy has quit [Quit: hello] 01:22:32 @??elephant slug 01:22:32 elephant slug (15j) | Speed: 4 | HD: 20 | Health: 113-146 | AC/EV: 2/1 | Damage: 40 | Res: 06magic(80) | Vul: 09poison | Chunks: 09poisonous | XP: 462. 01:37:19 Catoblepae and Petrifying Clouds Implemented (https://crawl.develz.org/mantis/view.php?id=4240) by Wensleydale 01:47:49 -!- blackpenguin has quit [Ping timeout: 258 seconds] 01:50:02 -!- blackpenguin has joined ##crawl-dev 01:54:00 -!- ais523 has quit [Remote host closed the connection] 02:06:13 -!- blabber has joined ##crawl-dev 02:11:54 -!- Zaba has quit [Ping timeout: 252 seconds] 02:14:25 hi. might be the wrong channel but i'll ask anyway... if i participate in the tourney as captain of a team, will i get statistics about my team members when the tourney is over? 02:15:37 -!- Zaba has joined ##crawl-dev 02:15:54 i am thinking about leading some of the folks in our local hackerspace into the tourney and would like some internal competition. 02:16:28 so i think it would be nice if a team gets also some kind of internal scoring and statistics. 02:17:00 of courxe this feature may already exist an i just didn't see it... 02:17:09 s/courxe/course/ 02:28:27 blabber: you mean like this? http://crawl11.dyndns.org/clans/casmith789.html 02:29:52 Wensley: yes, exactly 02:30:02 so this already exists 02:30:33 yep, tourney stat pages are the best 02:31:18 you'll have to go to each player's individual page though if you want to brag about which uniques you've killed or how many combo high scores you've gotten or banners you've attained 02:32:47 yeah, it would be nice to have this kind of stuff on the clan stat page too. but the existing data seems to be enough for me. 02:33:09 and honestly i don't expect to gather too much people :) 02:35:17 maximum of six per clan, in any case 02:38:10 right 02:40:51 -!- Wensley has quit [Ping timeout: 255 seconds] 02:48:32 -!- hoody_ has quit [Remote host closed the connection] 02:51:32 -!- hoody has joined ##crawl-dev 02:51:34 -!- ainsophyao has joined ##crawl-dev 02:52:23 -!- hoody has quit [Remote host closed the connection] 03:07:37 -!- ainsophyao has quit [Remote host closed the connection] 03:12:21 -!- ainsophyao has joined ##crawl-dev 03:28:13 -!- syllogism has joined ##crawl-dev 03:29:22 -!- edlothiol has joined ##crawl-dev 03:33:34 -!- ainsophyao has quit [Remote host closed the connection] 03:35:52 -!- ainsophyao has joined ##crawl-dev 03:41:02 !tell galehar (today evening, might be already fixed): 1. if the only enabled skill reaches 27, all other xp is being lost, 2. in automatic mode, you lose all xp if none of enabled skills has been exercised recently 03:41:02 kilobyte: OK, I'll let galehar know. 03:46:18 -!- edlothiol has quit [Quit: edlothiol] 03:55:21 -!- ainsophyao has quit [Remote host closed the connection] 04:02:05 03kilobyte * rca1a2be01329 10/crawl-ref/source/ (10 files in 4 dirs): ZotDef: replace electric eels by lightning spires. 04:09:10 -!- bmh_ has joined ##crawl-dev 04:09:15 -!- bmh_ has quit [Changing host] 04:09:15 -!- bmh_ has joined ##crawl-dev 04:09:42 -!- bmh_ is now known as bmh 04:12:59 -!- ainsophyao has joined ##crawl-dev 04:22:37 !tell dpeg I really like your suggestion for special effects around overflow altars. Perhaps the effects could be timed or one-shot (upon discovery) to thwart scumming. 04:22:37 bmh: OK, I'll let dpeg know. 04:55:54 -!- ainsophyao has quit [Remote host closed the connection] 05:11:35 -!- upsy has joined ##crawl-dev 05:26:25 -!- ainsophyao has joined ##crawl-dev 05:34:55 -!- RichardHawk has quit [Read error: Connection reset by peer] 05:35:04 -!- RichardHawk has joined ##crawl-dev 05:46:29 -!- Ashenzari has quit [Ping timeout: 252 seconds] 05:47:16 -!- Ashenzari has joined ##crawl-dev 05:47:16 -!- 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. 05:51:23 -!- upsy has quit [Read error: Connection reset by peer] 05:51:40 -!- Elynae has joined ##crawl-dev 05:55:06 Napkin: could you please cancel my game in progress? the game froze, and after exiting my terminal, I got the "Another game is already in progress using this save!" message since I didn't force termination 05:55:10 -!- ainsophyao has quit [Remote host closed the connection] 05:55:35 -!- upsy has joined ##crawl-dev 05:58:38 yes, Elynae 05:58:38 Napkin: You have 10 messages. Use !messages to read them. 05:58:50 Napkin: thanks 05:58:58 any idea what happened? 05:59:05 the game crashed at 99% cpu usage 05:59:16 Elynae (L11 DrCj) (Lair:1) 05:59:23 i was playing pretty fast, and then suddenly it froze 05:59:53 if some developer would have time to look into it, that would be great - because there have been a few 99% crashes lately 06:00:08 kilobyte, Zaba, due 06:00:13 try now, Elynae 06:00:40 works perfectly, thanks 06:04:05 Napkin: do we have a core? 06:04:42 most if not all skills-related crashes that happened before should be fixed 06:10:54 beh, no cores from today :( 06:14:20 i killed with sig15 06:14:26 doesn't that create a core? 06:14:56 true - no core from today 06:15:13 and too little space is not the problem this time ;) 06:15:23 I just booked my vacation! 06:15:44 Wisła - here I come! :D 06:16:53 Napkin: sweet, drop by Kilobyte's place while you're in Poland :-D 06:17:27 he probably lives in Warsaw ;) which is about 500km north ;) 06:18:04 Napkin: are you hiking the Tatras? 06:18:52 no way! I want to sleep in a nice and fluffy bed ;> 06:20:06 actually, add another 500km north 06:20:12 I'm doing Nepal for my holiday. No fluffy beds for miles. 06:21:01 yak fur sheets? 06:21:23 sleeping bag and lodges 06:21:57 cripes, crawl has made me afraid of Yaks :) 06:24:47 -!- ainsophyao has joined ##crawl-dev 06:25:11 kilobyte: have you looked at gameplay with my "don't thrash the abyss" patch? 06:29:02 -!- Elynae has quit [Quit: I'll teach my children to say "sudo" instead of "please"] 06:29:57 bmh: yeah, it merely slows down the flips but doesn't remove them :( 06:30:30 like, you have a patch of three squares of lava before you -- when the flip comes, they turn into stone, then on the next flip into rock, and so on 06:30:54 kilobyte: it *really* slows them down. The step size is 0.25, so the flip step is 0.025. The mean distance to a voronoi point is 1 06:31:21 So the minor octave should roll over about once every 200 turns 06:31:46 if you're seeing other behavior, it's possible that my math is off. 06:35:49 I might have done some mistake while #ifdefing things, I merely glanced at the original version... lemme rebuild it and test again. 06:36:51 I'm not seeing anything as bothersome as you describe, but I'm happy to admit to bias :) 06:37:21 there's a distinct possibility I'm on crack, too 06:37:38 Given ...that you've been smoking crack? 06:39:33 (actually, in RL, not a single time, but I often make errors so stupid like I did) 06:40:06 hrm, three turns one after another: stone->metal->rock 06:40:34 I'm going to test it a bit more on my end by disabling the major octave. 06:40:48 you could just be seeing major octave changes 06:41:27 there's also a screen flash every once in a while, but that's probably unrelated 06:41:35 it does linger in tiles, though 06:41:44 I think I forgot to tell it to recompute LOS 06:42:26 in tiles, sometimes after a change most of the map shows something resembling magic mapping 06:51:32 Ragdoll (L13 MuCK) ASSERT(you.skills[sk] || you.wizard) in 'skills.cc' at line 424 failed. (Orc:4) 06:52:01 kilobyte: even without the abyss changes, I think my mapping changes are worth pushing into trunk. I think they make labs far less annoying. 06:54:13 90db7f2 06:55:02 yeah, good point 06:55:22 I can push them #ifdefed out, except for the labs 06:58:06 03kilobyte * r250da486f9d4 10/crawl-ref/source/ (beam.cc enum.h zap-data.h): Remove removed beams. 06:58:06 03bstriegs * r33066b2b5765 10/crawl-ref/source/ (14 files in 3 dirs): Catoblepas. 06:58:07 03kilobyte * raf4fe250d682 10/crawl-ref/source/ (rltiles/dc-mon.txt tilepick.cc): Correct the tile for lightning spires. 06:58:07 03kilobyte * re61acb6ba7dc 10/crawl-ref/source/ (3 files in 3 dirs): Denzi's tile for catoblepi. 06:58:17 03kilobyte * ree6da548d5da 10/crawl-ref/source/rltiles/dc-mon/statues/lightning_spire.png: Hand-antialias the lightning spire tile somewhat. 06:58:17 03kilobyte * r1d9ee3131edf 10/crawl-ref/source/ (newgame.cc util/gen_ver.pl version.cc version.h): Determine release type (alpha / beta / final) in only one place. 06:58:18 03kilobyte * r8a26538ac4f0 10/crawl-ref/source/mon-pick.cc: Bump catoblepas rarity in alpha releases, lower it in betas and finals. 06:59:22 if I failed to mention that, your map rot kicks ass 06:59:43 oh, thanks 07:00:02 using math to make crawl cool ;-) 07:00:58 Is there any change for new portal vaults to make it in? 07:03:30 hardly more distruptive than a regular encompass vault, I'd say 07:04:06 03kilobyte * rd71797255ee5 10/crawl-ref/source/ (11 files): Merge branch 'abyss' 07:04:08 What I had in mind is either a vault with clouds instead of walls, or sand dunes 07:05:04 Unbreathing characters would be kept in line with TELE_CLOUDS to hurl them to the abyss for bad behavior. 07:05:29 my idea: Perlin noise, applied without any regards for connectivity, but with a tiny time scale 07:05:53 what for? 07:06:36 such a portal vault would often require waiting a bit to be able to move forward, and you'd often be pushed back by walls, so it's unfit for any regular branch or even Abyss 07:09:20 That's similar to what I was thinking for the Mists portal vault 07:09:30 with tele clouds rather than solid walls 07:10:07 bmh: did you make any changes to your abyss branch? Due to #ifdefing, merging changes is hard -- if you have any pending changes, please send them my way so I can merge them in somehow, and then you could switch to what's currently in trunk. 07:10:35 kilobyte: I haven't. I'll dump my branch and track trunk. 07:18:10 kilobyte: for the sand dunes, I was thinking endless dunes with varying elevation (conveyed through color) 07:18:24 and sandstorms that occasionally wipe the map 07:24:33 Given enough floating vaults inside, I think it would be interesting 07:26:02 aah... this kind of resembles the mutable part of the Forest branch I didn't yet get to implementing 07:26:20 perhaps we should have a common codebase for this sort of behavior? 07:26:38 I confess that I know nothing about the existing level generators 07:27:09 the problem there is that I want a regular dungeon builder than can also regenerate parts out of LOS... it would be hard to use noise for that 07:27:31 there are two algorithms I've pondered so far: 07:30:17 algorithm A: 1. start with a graph, a regular grid. I think a hex grid with edges ~10 long should be good. 2. pick a node, randomly move it without crossing any other edge, repeat a good number of times. 3. delete everything but a minimal spanning tree, plus a percentage of remaining nodes, ... 07:30:56 Beta skeletons might give a better result than MST 07:31:21 4. render the graph on the map with a fuzzied version of Bresenham (if it breaches another edge, no big loss), 5. add "clearings" around some nodes 07:32:24 might work, I didn't even try implementing that idea yet... that could make them more realistic 07:33:10 when I have some time I need to finish up a planar graph implementation. I want to support Delauney, Voronoi and Beta Skeletons 07:33:28 not sure if it would be appropriate for use in crawl 07:33:57 an important part is: if the player is on the level, preserve any nodes and edges within a radius, exempting them from shaking and use exact same rendering as before 07:34:17 I think you can do this with noise. 07:34:44 about Voronoi: I tried some naive generators based on straight and biased Voronoi diagrams, and the results sucked 07:34:49 Perlin noise gets its gradients from a spatial hash 07:35:09 Were you using Lloyd's algorithm? 07:35:36 so Perlin is a spatial hash, if you put a bounding box around the area you want to preserve, you can swap the hash function outside of the boundary 07:36:12 no, I didn't 07:36:37 might give better results, depending on your complaint. Run Voronoi, move each point to the centroid of the enclosing poly and re-run Voronoi 07:36:44 with Perlin, the problem is that there will be a sharp difference between the preserved and non-preserved part 07:37:12 Sure, that's the idea of bounding the preserved area 07:37:21 I'll sketch something. 07:39:42 sample result of my Voronoi generator (without a smart handling of the boundary): http://sprunge.us/XQBb 07:40:10 What's wrong with that? 07:40:41 oh that's nice 07:40:52 isthis for a Snake generator? 07:41:15 due: no, just an old attempt of mine 07:41:38 it's nice though 07:44:40 http://imgur.com/ijIBR 07:45:13 That approach will avoid discontinuities. 07:47:34 forest algorithm B: 1. pick a place, roll a "major direction" and a "minor direction" (initially close to the major one). 2. Go a step in the minor direction, do some random wandering (Brownian? noise?) on both the major direction (slowly) and the minor (faster), bias the minor toward the major. Repeat until you meet an existing path. 3. unless {end condition}, go to 1 07:48:50 when re-generating the map, find all paths intersecting the boundary of the preserved area, take their major direction (might be stored, might be calculated from the map) and continue, never allowing to wander back into preserved area 07:49:10 that sounds painful 07:50:06 probably too naive, but might work 07:50:41 about the diagram you just linked: I kind of fail to see what's the difference between using Perlin or other noise, and pure randomness 07:50:58 -!- ainsophyao has quit [Remote host closed the connection] 07:51:01 How about this: Place some points at random, compute Voronoi with Lloyd's relaxation, and generate a beta skeleton 07:51:12 the goal is to have the transition between old and new areas unnoticeable 07:51:15 When you want to regenerate, perturb some of the points and re-generate the skeleton 07:51:32 kilobyte: That's the point of this. I just drew the random vectors at the lattice points 07:52:04 The idea is that you want to preserve the upper right hand corner, so you preserve all vectors that contribute to the corner and randomize the rest 07:52:15 -!- ainsophyao has joined ##crawl-dev 07:52:59 It will be continuous 07:54:19 I'd be happy to implement it if you can provide a Perlin or Simplex noise implementation 07:56:23 -!- bmh is now known as bmh_away 07:57:19 perlin is quite simple unless you care about computation speed 07:58:09 in the main Crawl repository, there's a branch "layout" which contains some extracted parts of Crawl with my experiments 07:58:13 right, in which case Simplex Noise is the right way to go 07:59:43 -!- galehar has joined ##crawl-dev 07:59:52 !lm * crash -log 07:59:52 galehar: You have 1 message. Use !messages to read it. 07:59:53 1252. Ragdoll, XL13 MuCK, T:26229 (milestone): http://crawl.develz.org/morgues/trunk/Ragdoll/crash-Ragdoll-20110709-115129.txt 08:00:10 !messages 08:00:11 (1/1) kilobyte said (4h 19m 9s ago): (today evening, might be already fixed): 1. if the only enabled skill reaches 27, all other xp is being lost, 2. in automatic mode, you lose all xp if none of enabled skills has been exercised recently 08:01:23 kilobyte: thanks, I'll look into it 08:03:06 2. can be trivially avoided by switching to manual when training a skill you currently don't use, but it can be surprising 08:03:17 -!- gnsh has quit [] 08:06:13 bmh_away: to get the layout I posted on sprunge, replace in main.cc layout_nolithius(); with layout_voronoi(100, true);. The code looks terrible because of many experiments going back and forth. 08:07:28 the algorithm is: 1. generate voronoi (with metric: euclidean distance^2 + a random constant that's a property of every beacon), 08:08:49 2. random walk on the regions, turning regions into floor, repeated until half the regions are visited 08:12:12 in an "uncompact" mode, walks that are detected as spiralling are undone 09:01:42 -!- hoody has joined ##crawl-dev 09:03:16 antiskillrobin (L18 SpVM) ERROR in 'skill_menu.cc' at line 407: Invalid view state. (Vault:5) 09:06:07 03kilobyte * ra2059cf71638 10/crawl-ref/source/ (cloud.cc enum.h): Let the calcifying dust block vision. 09:06:10 03kilobyte * re194c60c91c8 10/crawl-ref/source/ (6 files in 3 dirs): Weak-ass tiles for calcifying dust. 09:10:15 Napkin: there's some permission problem when trying to access your git cache from CDO: ~/source/monster> make clean -> R access for crawl-ref DENIED to crawl 09:11:24 there is no crawl-ref anymore 09:11:44 could you change it to crawl.git, please? 09:12:22 since you guys moved to gitorious, i added crawl.git additionally - but kept both for a while 09:12:35 and i removed crawl-ref recently 09:15:14 -!- hoody has quit [Read error: Connection reset by peer] 09:15:28 also, any changes you do for monster.git, please push it 09:15:38 yay, it didn't explode yet 09:15:56 as the knowledge bot website is no longer using the binaries from ~crawl 09:16:09 it's not in git, the monster/crawl-ref/ subdirectory is checked out manually 09:16:43 you mean "make install-trunk" won't work anymore? 09:16:43 monster is in git too 09:16:50 it will, kilobyte 09:17:07 but i'll have to do that separately again for the knowledge bot website 09:17:28 i removed www-data's access to crawl's home 09:17:51 ~crawl/bin/monster-trunk is used by gretell 09:18:13 you get my point, kilobyte? 09:19:19 m'kay 09:19:28 just lemme fix all bit rot, there is some 09:19:39 thank you :) 09:21:18 -!- hoody has joined ##crawl-dev 09:21:31 -!- upsy has quit [Remote host closed the connection] 09:23:06 @??catoblepas 09:23:06 catoblepas (06Y) | Speed: 8 | HD: 14 | Health: 58-102 | AC/EV: 10/2 | Damage: 36 | Flags: !sil | Res: 06magic(93) | Chunks: 07contaminated | XP: 898 | Sp: petrifying cloud. 09:23:20 omgwtfbbq, it works again! 09:23:31 hehe 09:24:45 pushed 09:25:03 you sure? 09:25:13 35bf47d..d05e032 bleeding-edge-crawl -> bleeding-edge-crawl 09:25:43 To ssh://git@git.develz.org/monster.git 09:25:53 ah, stupid me - thanks! 09:28:29 Use of uninitialized value $host in concatenation (.) or string at ./util/gen-cflg.pl line 21. 09:28:37 help? 09:28:59 just a warning though, of course 09:32:17 http://crawl.develz.org/info/index.php?q=catoblepas - omgwtfbbq, too! 09:32:40 thanks :) 09:34:41 yay! 09:35:40 -!- upsy has joined ##crawl-dev 09:46:31 !lm * crash -log 09:46:31 1253. antiskillrobin, XL18 SpVM, T:52574 (milestone): http://crawl.develz.org/morgues/trunk/antiskillrobin/crash-antiskillrobin-20110709-140313.txt 09:54:39 paplaukes (L27 MiFi) (Pan) 10:00:58 Tonight is wine&code-night at school :D 10:01:51 !lm * crash -log 10:01:52 1254. paplaukes, XL27 MiFi, T:179904 (milestone): http://crawl.develz.org/morgues/0.8/paplaukes/crash-paplaukes-20110709-145434.txt 10:02:36 oh, this one's mine 10:02:53 hrm, 0.8 10:03:53 -!- edlothiol has joined ##crawl-dev 10:04:33 -!- Pacra has joined ##crawl-dev 10:16:44 -!- casmith_789 has joined ##crawl-dev 10:17:21 -!- casmith_789 is now known as casmith789 10:25:51 -!- monqy has joined ##crawl-dev 10:26:58 -!- bmh_away is now known as bmh 10:28:11 kilobyte: I can't seem to find layout_nolithius 10:28:16 ghallberg: your school sounds awesome. 10:28:52 In skill descriptions interface (m?), selecting skills still toggle their training (https://crawl.develz.org/mantis/view.php?id=4241) by rriegs 10:32:09 bmh: branch "layout" (it doesn't have full Crawl, just a small subset), main.cc 10:32:20 that's the invocation, the actual code is elsewhere 10:32:44 ok -- how do I pull down all the branches from gitorious?? 10:33:20 any fetch or pull will get them all unless you explicitely configured it not to 10:33:44 neat 10:33:52 "git checkout -b layout origin/layout" (where "origin" is how you named gitorious) 10:34:09 in new gits you don't need to specify the last argument, too 10:34:18 (if it's unique) 10:34:30 sure, got it. Just `make`? 10:35:29 hrm -- Cannot find data file 'dlua/macro.lua' anywhere, aborting -- I seem to have gotten something wrong 10:36:27 it doesn't even have any lua in 10:36:51 I'll cleanup 10:37:06 it's a small subset -- the RNG (not updated to ASG but this shouldn't matter), iterators, env 10:39:44 oh -- doh :-) 10:39:50 I was running ./crawl not ./layout 10:40:13 ah ok 10:41:03 kilobyte: this is way neat :-) 10:41:30 -!- Pacra has quit [Ping timeout: 250 seconds] 10:42:35 I really dislike the tile '♣' 10:42:50 It gets printed as two characters in my terminal 10:43:20 that's the tree glyph? Try pilcrows 10:43:20 monqy: You have 1 message. Use !messages to read it. 10:43:37 pilcrows? 10:43:52 bmh: then any non-ASCII character will be broken for you 10:44:02 ¶ 10:44:19 they're good trees imo 10:44:28 kilobyte: most non-ASCII characters are fine, it's just the spades symbol that's too fat 10:44:46 It overflows its box, doesn't disrupt character width though 10:46:35 http://i.imgur.com/VSX1z.png 10:48:35 I see that most other characters also have this problem -- '~' slightly, '≈' a lot 10:49:40 something is broken on this display; horizontal advance between characters is too small, vertical nearly doubled 10:49:42 It's much more mild with them. They don't obscure adjacent monsters or equipment 10:49:56 (judging by '#') 10:50:38 -!- MarvinPA has joined ##crawl-dev 10:52:45 -!- Pacra has joined ##crawl-dev 10:53:07 you can adjust it in main.cc 10:54:12 what is layout_nolithius for? 10:54:13 it looks nice 10:54:31 -!- nrook has joined ##crawl-dev 10:54:40 your forest experiment? 10:55:21 no, dpeg fell in love with an example in an article, so I adapted that algorithm for Crawl 10:55:24 ??layout 10:55:25 layout[1/1]: For some beautiful islands, see http://www.nolithius.com/images/dod-worldgen.png 10:55:54 oh, amitp's article? 10:56:00 it looks good but would currently fit well only in Shoals, and it's probably redundant with the existing shoals generator 10:57:02 how do you feel about infinitely large levels? 10:57:47 i like those islands very much 11:00:12 bmh: it would be damn hard to implement without a sliding window, but a window can work fine 11:00:16 -!- RichardHawk has quit [Ping timeout: 252 seconds] 11:01:46 this algorithm is overcomplicated -- it multiplies two types of noise 11:02:08 That doesn't mean it's overly complicated :-) 11:02:29 the other, "roiling particle", gives little randomness but provides shape 11:02:31 I've been thinking of combining Worley noise with Perlin 11:04:00 I got about the same result by just multiplying Perlin by the distance from the center 11:04:16 interesting 11:04:38 (not Euclidean x^2+y^2, x^4+y^4 or x^6+y^6 gives better corners) 11:05:33 scaled of course, the point is to reduce the border while not affecting the center 11:06:44 Perlin+shape can really kick ass 11:07:10 The rolling particle algorithm is neat 11:07:22 like, in a paper by Ken Perlin: http://www.noisemachine.com/talk1/8.html 11:09:36 too bad, rolling(roiling?) particle is not stable: unlike perlin or similar noise, you can't perturb it a little and get a similar result 11:10:05 right, you can't evaluate it at a point 11:10:26 same for every algorithm for post-processing perlin to get connected shape :( 11:10:39 (s/every/every I thought of so far/) 11:11:06 is your day job in game programming? 11:13:16 -!- hoody__ has joined ##crawl-dev 11:14:48 -!- hoody has quit [Ping timeout: 276 seconds] 11:16:19 no, I hardly learned anything about noise since playing with it at home in high school :( 11:16:55 so unlike you, my knowledge is not only very dated but also without a solid base 11:17:16 mine is mostly without a base. I just read Perlin's book 11:17:36 Musgrave has some good stuff too 11:20:14 -!- moocowpong1 has joined ##crawl-dev 11:24:22 -!- elliptic has joined ##crawl-dev 11:29:24 -!- MarvinPA has quit [Ping timeout: 246 seconds] 11:30:01 -!- gnsh has joined ##crawl-dev 11:50:02 !lg * killer=ignacio 11:50:02 No games for * (killer=ignacio). 11:50:09 !lm * uniq=ignacio 11:50:10 21. [2011-07-09] splat the Farming Ninja (L27 HaBe) killed Ignacio on turn 200284. (Pan) 11:50:12 :( 11:52:10 -!- MarvinPA has joined ##crawl-dev 12:11:23 -!- MarvinPA_ has joined ##crawl-dev 12:14:24 -!- MarvinPA has quit [Ping timeout: 240 seconds] 12:23:38 -!- MarvinPA_ is now known as MarvinPA 12:47:05 -!- Guest37064 has quit [Read error: Connection reset by peer] 12:54:02 ??ignacio 12:54:03 ignacio[1/1]: Unique executioner found in Pandemonium in 0.9. Comes with executioner's axe of pain, casts torment and agony (basically torment that checks MR). Pandemonium Prince Ribbit. 12:57:55 -!- Twilight has joined ##crawl-dev 12:58:21 -!- Twilight is now known as Guest78501 13:01:57 kilobyte: Is there an easy way with layout to make it redraw in place? (I've never used curses) 13:12:40 -!- MarvinPA_ has joined ##crawl-dev 13:13:49 -!- jeanjacques has joined ##crawl-dev 13:13:57 -!- MarvinPA has quit [Ping timeout: 264 seconds] 13:23:25 -!- hoody__ has quit [Remote host closed the connection] 13:56:10 03elliptic * ra079eaff909e 10/crawl-ref/source/files.cc: Make time taken to go up/down stairs more consistent. 14:05:03 -!- MarvinPA_ is now known as MarvinPA 14:16:51 -!- Mu_ has joined ##crawl-dev 14:26:07 damn, edlothiol, when is the tourney? 14:26:23 I'm again totally not motivated to upgrade webtiles :-$ 14:26:35 and you've done so much work! 14:28:22 august 13th, it seems 14:28:39 it would be nice to have the new features tested before that, though ;) 14:29:09 ok, i'm going on vacation on 25th 14:29:55 me on the 1st 14:29:59 well, i'm hoping for tomorrow! 14:30:09 would be nice ;) 14:30:16 are you in tomorrow? 14:30:19 yes 14:30:38 ok - then i hope i'm not too lazy and will try to catch you ;) 14:31:40 ok 14:32:01 people should see those awesome new features you wrote! 14:32:02 :) 14:52:09 what are the new features? 14:52:19 skills, octopodes? 14:53:00 finally adding the nethack branch, where you have to fetch a rune in the dungeon of doom? 14:53:05 ... 14:53:48 bhaak: no but we're adding a full level castle vault on D:14. With liches. 14:54:17 galehar: neat! 14:54:40 oh wait, that can't be real. Liches on D:14 would kill everyone. 14:55:22 galehar: that's a step into the right direction! keep going! 14:55:51 bmh: yeah it's a joke. But a full level vault on D:14 with a portal needing a rune to open is an idea that's been itching the back of my brain the last few days. 14:56:43 galehar: I've been thinking of a sands portal vault, infinitely large and chock full of floating vaults 14:59:19 though scorpions and mummies are sort of taken by Oussary 15:10:41 -!- evilmike has joined ##crawl-dev 15:24:36 -!- moocowpong1 has quit [Quit: moocowpong1] 15:41:14 -!- blabber has quit [Quit: leaving] 15:41:34 -!- moocowpong1 has joined ##crawl-dev 15:46:18 -!- ainsophyao has quit [Remote host closed the connection] 15:48:45 -!- Wensley has joined ##crawl-dev 15:54:48 bmh: a long time ago I had the idea of a hunting portal vault, where the player has to track some creature across a very big landscape (with no reason to explore), but I never entertained it seriously 15:59:21 nrook: in some ways that sounds more fun than labyrinths, honestly :P 15:59:55 I don't think it's a bad idea, I just never seriously thought about putting in the work to implement it 16:02:33 I compare it to labyrinths because it's a place where autoexplore would (hopefully) not be optimal 16:02:55 you would just need to find a way to guide players towards their prey 16:03:06 which I'm guessing could be something like a lamb made of gold or somesuch 16:03:49 yes, or a big boss monster like the Lernaean hydra 16:17:36 -!- dpeg has joined ##crawl-dev 16:21:06 Wensley: awesome! 16:21:07 dpeg: You have 1 message. Use !messages to read it. 16:21:10 !messages 16:21:10 (1/1) bmh said (11h 58m 33s ago): I really like your suggestion for special effects around overflow altars. Perhaps the effects could be timed or one-shot (upon discovery) to thwart scumming. 16:21:27 bmh: need more context :) 16:21:37 dpeg: I as well :) 16:22:12 Wensley: catopeblas in 0.9! Congrats! Doubly so on creating an interesting slow monster. 16:22:24 dpeg: thank you! :D I am very excited. 16:22:38 I hope that everyone else will find them interesting as well 16:22:39 Wensley: was it your very own idea= 16:22:52 Wensley: expect outcries from players who die to them! 16:23:20 dpeg: apparently it has been on the devwiki for a long time, but I unknowingly conceived it independently, using myth as a guide 16:23:25 Once you feel proud about those (as opposed to annoyed or sorry), you're one of us :) 16:23:29 are you going to direct everyone at poor Wensley? :) 16:23:36 I hope so! 16:23:44 I have, in fact, already done !lg * killer=catoblepas 16:23:56 but I don't think CDO/CAO have been compiled yet today :P 16:32:05 dpeg: have you had a chance to try them out for yourself? 16:35:49 Wensley: no just in, left 16 hours ago. Will have a look at nrook's prices and that's it for today. 16:36:10 no problem, I hope that you will lose a few characters to them in time :) 16:36:58 oh, were you traveling? 16:37:11 wensley: what sort of stats did you end up giving them? 16:37:29 @??catoblepas 16:37:29 catoblepas (06Y) | Speed: 8 | HD: 14 | Health: 58-102 | AC/EV: 10/2 | Damage: 36 | Flags: !sil | Res: 06magic(93) | Chunks: 07contaminated | XP: 898 | Sp: petrifying cloud. 16:37:36 oh, @?? got updated! 16:37:37 wow 16:37:43 yes :) 16:37:58 elliptic: I detailed my reasons for their stats in the mantis issue I filed: https://crawl.develz.org/mantis/view.php?id=4240 16:38:36 what depth do they generate at? 16:38:42 same as hydras 16:38:53 which is represented as +=5 16:38:57 in the lair branch 16:39:16 even though I have definitely seen hydras on lair:1 so I'm not really sure how that influences things 16:39:56 but you and kilobyte definitely hit the nail on the head, the _one_ thing that I was unsure about with these guys was what their depth and rarity should be like 16:40:12 I think that 5 is only where they are most common or something, but they are still in-depth earlier 16:40:40 i think it's not considered out of depth until 7 or something, yeah 16:41:07 I think I need to play more. 16:41:23 they do look significantly nastier than hydras, but we can work out stuff like depth and rarity with some testing :) 16:42:14 I was aiming for hydra-level dangerousness, but they are really unlike hydras so it's a little hard to judge 16:42:56 it is good if they're overly dangerous now, better than the other way around 16:43:12 dpeg: my sentiments exactly :) 16:43:14 (easier to spot an overperforming monster than an underachiever) 16:43:17 also, I saw that basilisks spawn in the dungeon, so I put catoblepae in the dungeon at the exact same depth and rarity 16:43:21 dpeg: yeah 16:43:40 wait, catoblepas at the same depth as basilisks? 16:43:42 Wensley: might not be necessary, but this game has a tendency to put everything into D -- not your fault :) 16:44:18 I like to think of D as the grab-bag of crawl... you never know what you might pull out 16:46:16 MarvinPA: petrifying clouds are in! :D I defined them as sanely as I was able, but if you take a look and see that I did anything poorly please let me know. In particular, I don't think I made evaporate ever produce 3x3 petrifying clouds, although it will sometimes produce a single petrifying cloud square when it places clouds randomly 16:46:44 yus, i spotted :) 16:46:44 hm, basilisks in D are at the same depth as hydras and catoblepas? that seems clearly wrong 16:47:00 i'm fighting with git to get it set up again and then i'll take a look 16:47:20 !lg * killer=basilisk s=place 16:47:21 5 games for * (killer=basilisk): 2x D:8, 1x D:2, 1x Lair:2, 1x D:1 16:47:47 !lg * killer=gila monster s=place 16:47:47 390 games for * (killer=gila monster): 84x Lair:1, 50x Lair:2, 45x D:8, 29x D:9, 29x Lair:3, 17x D:11, 17x Lair:4, 17x D:10, 15x D:5, 13x D:4, 13x Lair:5, 12x D:7, 11x D:6, 7x D:12, 7x D:3, 6x Lair:6, 5x D:13, 4x Lair:7, 4x D:2, 2x Lair:8, 1x Vault:1, 1x D:1, 1x D:16 16:47:48 !lg * killer=gila monster cv=0.8 s=place 16:47:48 35 games for * (killer=gila monster cv=0.8): 8x Lair:1, 5x Lair:2, 4x D:7, 3x D:5, 3x D:9, 3x D:8, 3x Lair:3, 2x D:4, 1x D:3, 1x D:11, 1x D:6, 1x D:10 16:48:00 i'm going to assume those D:1 and 2 deaths are from xomscumming 16:48:13 !lg * killer=basilisk s=place,xl 16:48:14 5 games for * (killer=basilisk): 2x D:8 (2x 8), 1x D:1 (1x 1), 1x Lair:2 (1x 5), 1x D:2 (1x 3) 16:48:35 since their native depth is much deeper and they're reasonably rare in D, i think 16:48:35 and the lair:2 death was probably a SpAr diver or something 16:48:52 @??basilisk 16:48:52 basilisk (06l) | Speed: 10 | HD: 5 | Health: 21-39 | AC/EV: 3/12 | Damage: 20 | Flags: cold-blooded, !sil | Res: 06magic(20) | Chunks: clean/none/unknown | XP: 153 | Sp: petrify. 16:48:58 -!- Pacra has quit [Ping timeout: 250 seconds] 16:49:03 they're literally just gila monsters with petrify :P 16:49:14 I wonder whether anyone will notice if I increase their damage to 25 :P 16:49:20 go for it 16:49:40 actually i was wondering if they could do with a bit higher HD or something, since they don't seem to succeed in petrifying very often 16:49:58 that might be better, yeah... 16:49:59 although i'm fine with them not being hugely dangerous alone 16:50:34 @??basilisk hd:6 16:50:34 basilisk (06l) | Speed: 10 | HD: 6 | Health: 26-46 | AC/EV: 3/12 | Damage: 20 | Flags: cold-blooded, !sil | Res: 06magic(20) | Chunks: clean/none/unknown | XP: 234 | Sp: petrify. 16:50:39 I'm happy to report that catoblepae can petrify other monsters, although it's very hard to make this happen intentionally 16:50:58 dpeg: my altar comment is in response to a post you made on the tavern 16:51:09 bmh: just browsing through forum 16:51:17 suppose when you come across an altar of TSO and you worship a good god -- blam, TSO gives you a mini-might effect 16:51:55 you can't scum it because you didn't know the altar was there, it's temporary, and not really game changing 16:52:24 bmh: I had an idea where overflow altars would sometimes randomly have lugonu's corruption occur as the player approaches... 16:52:24 yes, that is the correct spirit 16:52:39 I made a cryptic remark about bad effects, this one is much better 16:52:58 Xom could do bad things :-) 16:53:09 heh -- we should make a Xom altar that chases the player. 16:53:16 haha 16:53:23 xom altar mimics 16:53:32 :-D 16:53:40 bmh: sounds like a good idea for a monty hall xom vault... 16:53:44 xom has an effect where he makes the stairs run away from you 16:53:51 three altars, choose the one that isn't the mimic 16:53:56 nah, having the altar be a mimic would be silly 16:54:01 it should be a real altar 16:54:03 just a moving one 16:54:14 03elliptic * raf52982a8c23 10/crawl-ref/source/mon-data.h: Buff basilisks slightly. 16:54:19 Is this a job for lua? 16:54:20 it wants you to pray on it. 16:54:22 FR: wanderlust god, altar moves about randomly, never found in the temple 16:55:22 a stalker god, whose altars follow you like wandering mushrooms :P 16:56:39 dpeg: my thoughts are: Good god worshippers get from TSO +str, Ely +a few hp, Zin a chance to cure negative status effects 16:57:32 Chousuke: my thief god proposal had the idea of camouflage altar (use the colour of most of adjacent cells, e.g. blue near water :) 16:57:51 Are Sif and Vehumet getting merged into Vehumuna? 16:58:40 No. 16:58:47 It's Sifumet. 16:59:06 Slightly less silly. Good, they're being merged. 16:59:38 bmh: no, I'm joking. The merge is far from being agreed on. It's just a proposal for now. 16:59:59 I'd love to see them either merged or differentiated. 17:00:09 03galehar * r51e7c5ea1b7f 10/crawl-ref/source/ (10 files): New feature: focus skill. Double the training weight of focus skills. 17:00:16 03galehar * r58acbd46f0fa 10/crawl-ref/source/ (skills.cc skills.h tags.cc wiz-you.cc): Make sure that there is always one skill selected. 17:00:55 bmh: IMO they already feel very different 17:01:46 they could be differentiated more of course, but I really don't like the merge option very much 17:02:36 Any thoughts on splitting them by school a bit more, so we'd have Kiku for Necro, Veh for blasting, and Sif for something 17:02:38 I always refer to sif as just sif because "muna" happens to mean "egg" in Finnish and it feels pretty awkward. :) 17:03:05 bmh: I think having a generic caster god is fine... it would be good if veh didn't have the strange conj/summ alliance though 17:03:19 well, to be honest it also has a colloquial more vulgar meaning. 17:03:25 elliptic: Sif can be more interesting, too 17:03:43 personally i prefer the proposals for taking away veh's summoning stuff and supporting other forms of ranged combat too, instead 17:03:43 dpeg: sure 17:04:07 IMO, Sif Muna's theme is great, abilities could use some work. Vehumet theme is a bit lacking. Abilities are simple but powerful and efficient. 17:05:00 they both aren't perfectly designed, but mashing them together doesn't seem right to me 17:05:18 * dpeg likes the idea of removing things so much. 17:05:32 But for the sake of playability, they should perhaps stay (individual gods). 17:05:42 there are a bunch of monsters i keep meaning to suggest removing, if that helps :P 17:06:15 gila monsters were one but they got changed to basilisks, i think giant toads and one out of wolves/wargs were the others 17:06:26 sif gets a bit boring after you get enough books to have all the spells in the game .P 17:07:51 galehar: sif's abilities: book gifts, miscast protection, channeling, amnesia... which of those do you not like? 17:08:12 MarvinPA: yes, it does! 17:08:37 MarvinPA: although "Removed [species]" or "Removed [god]" carries more ooomph in the changelog :) 17:08:48 heh :) 17:08:55 elliptic: access to spells could be a lot more interesting 17:08:59 I know dpeg doesn't like the channeling ;) and I'd agree that it isn't a very exciting ability, but I think the others work pretty well 17:09:21 changes to the book gifts could be interesting, but they are already much better than kiku or vehumet's gifting 17:09:37 'killed all gnomes' 17:09:54 elliptic: well, I'm not sure. Seems I have spoken without really thinking. 17:10:10 'nailed grey elves to the sky' 17:11:06 Eronarn: vault idea: A bunch of dead gnomes in cages. 17:11:53 I really like vehumet, in that he gives your abilities a noticable improvement but is completely passive in it 17:12:00 I dont think any other god is quite so "hands off" 17:12:12 about Veh, for spells for which he doesn't boost range, maybe he should boost power instead. 17:12:58 Isn't Ash support almost better than Veh for tornado for example? Since the skill boost does boost power. 17:13:10 veh casting support is much greater 17:13:15 veh wizardry is a pretty big deal, likewise the mp cost reduction 17:13:17 and power doesn't matter much 17:13:26 or rather, it does but skills don't help it much 17:13:36 i definitely don't think he needs anything in the way of buffs, certainly don't disagree that he's lacking in theme though 17:13:45 Is it heresy to suggest that maybe xom should be better than GOD_NO_GOD? 17:13:49 I think the range boost is fine; I'm less happy about the MP reduction 17:13:59 L5 and higher is such an arbitrary cutoff point 17:14:02 bmh: indeed, xom needs work 17:14:15 it makes L5 spells unusually good for veh and L4 spells unusually bad 17:14:34 however I'm not sure there is a reasonable alternative 17:14:51 that works quite nicely with crawl's current spells though, since l4 conj is mostly overpowered and l5 conj mostly underpowered :P 17:15:07 but yeah, it's not very elegant 17:15:20 bmh: for Xom there is a plan, though: make him better tactically. Just needs someone to do it :) 17:15:34 dpeg: Not it. 17:15:35 (and the relevant l4/5 spells have been tweaked recently anyway i guess) 17:15:37 elliptic: a rebate instead of a cost reduction, with only a chance of recovering 1 MP? 17:16:29 bmh: not it? 17:16:43 dpeg: so you're away on holidays, but somehow you're still here? 17:16:44 chance could even depend on spell level to keep the behaviour reasonably similar (maybe veh likes overkill or something) 17:16:49 dpeg: http://www.urbandictionary.com/define.php?term=not%20it 17:17:02 eronarn: I've considered it, but it sounds annoying to play with 17:17:46 because you will often be in the situation of not knowing whether you have 2 more casts of 17:19:10 -!- hoody_ has joined ##crawl-dev 17:20:30 I'd sort of like to see veh gifts removed, actually 17:20:36 galehar: around today and tomorrow 17:21:17 -!- hoody__ has joined ##crawl-dev 17:21:39 -!- hoody__ has quit [Remote host closed the connection] 17:21:48 elliptic: veh should probably gift at least a book of 'starter' conj spells 17:22:03 even if the other gifts are removed 17:22:04 maybe he could gift a small fixed number of randart conj books 17:22:13 maybe remove veh gifts and the summoning boost and give him other abilities in place of that (like what marvinpa was saying with ranged combat) 17:22:21 eronarn: don't think that is necessary 17:22:22 or also that 17:22:40 currently plenty of people go veh even though they won't get a starter conj book until really late 17:22:47 one caster god could gift spells rather than books 17:22:52 elliptic: not being able to use a god's abilities unless you find a certain item sucks, and lair isn't 'really late' 17:23:11 dpeg: That would be interesting. 17:23:20 eronarn: nobody is forcing you to join the god on every char, and he could have appeal for non-conjurers too 17:23:58 eronarn: often one doesn't get veh gifts until after clearing lair 17:24:24 -!- hoody_ has quit [Ping timeout: 240 seconds] 17:24:34 they could be pushed earlier but I don't like deterministic gifts anyway, and I think there are more interesting things that can be done with the god 17:24:41 nobody is forcing anyone to do anything, so that is not a good argument 17:24:49 taking a god and not knowing when you will be able to use any of that god's abilities sucks 17:24:56 eronarn: then don't do it 17:25:06 kilobyte: This layout code is tremendous fun! 17:25:08 also the idea would be to give the god other abilities 17:25:37 elliptic: I toyed with the idea of making Vehumet good for all kinds of ranged attacks (not easy to do in a natural fashion, though) 17:25:38 https://crawl.develz.org/wiki/doku.php?id=dcss:brainstorm:god:vehumet#dpeg i like the idea of stealing some of those abilities to replace the book gifts and summ boosts 17:25:44 perhaps instead you should not do the thing where you remove the conj god giving you access to conj spells 17:26:15 eronarn: he doesn't have to be "the conj god" 17:26:24 you don't seem to be getting this 17:27:00 marvinpa: the explosion on kills thing there could be an inner flame invocation! 17:27:08 drumfire in particular, and maybe explosion on kills although i suppose inner flame already does that 17:27:09 heh 17:27:34 the other ideas might be more interesting though 17:28:03 not sure if it's on that page, i remember seeing a suggestion for attack wand uses to occasionally not use a charge based on veh piety though 17:28:35 MarvinPA: it should be there, it's my idea :) 17:28:49 :D 17:29:17 no book at all? No annihilation? 17:29:41 galehar: you could get a choice between annihilations and at ****** 17:29:52 -!- moocowpong1 has quit [Quit: moocowpong1] 17:30:17 explosion-brand a launcher :P 17:30:18 the god could gift you the spells, but not the book 17:30:25 (not that that would be very useful, sadly) 17:31:37 i can't think of a particularly good 6* alternative to the book gift, just directly gifting spells could work though (or that could be saved for sif?) 17:31:46 another random idea for a passive ability for this sort of ranged god would be reflecting some missiles 17:32:40 !tell kilobyte I found a (stupid) bug (that I introduced) in the worley code. I'll submit a one line^Wcharacter patch 17:32:41 bmh: OK, I'll let kilobyte know. 17:35:47 elliptic: i made a proposal a long time ago that made him something other than a pure conj god, so i have nothing against the idea, but 'we'll give him abilities' and 'he could eb good with ranged combat' are way vague 17:37:11 some things from that proposal... quiver a wand, and when you cast a destructive spell, it zaps the wand too 17:37:29 forcefully unlearn a spell, giving it increased power but de-memorizing it afterwards 17:37:56 activated ability to drain magical power from things in the area 17:38:16 i think the second has some promise if we tie it to one of the brain feed proposals, where it locks you out of using a spell for a while 17:38:54 actually forgetting it might be too much, but it would be cool to have an 'overcharge' ability (and it's easy to think of how the same ability could work on wands/rods) 17:39:12 eronarn: you mean lock you out of using the spell rather than actually forgetting it? that does sound like an interesting ability 17:39:42 (actually forgetting it sounds annoying) 17:40:04 oooooh 17:40:10 lock you out of using the spell 17:40:13 elliptic: yeah... the idea is that brain feed as an int drain is an OK start, but that it doesn't really hurt casters that much, since they have much more to start with. if brain feed attempts also had a chance to lock out spells, it would become much worse for casters than it is now, without being worse for non-casters 17:40:14 in return for QUAD DAMAGE 17:40:20 ?!!??! 17:40:27 marvinpa: haha 17:40:29 :P 17:40:53 MarvinPA: quad damage: too much. ALLCAPS: too much. everything explodes when it dies: not too much' 17:41:01 eronarn: yeah, I've heard those proposals and they sounds good 17:56:39 Footloop the Archmage (L27 SEEE) (Abyss) 17:58:17 -!- hoody_ has joined ##crawl-dev 18:00:06 Wensley: I think the prices are very good. For jewellery, stern looks much better than gentle. With the insanely high armour enchantments, we may need to tweak the formula. 18:11:51 dpeg: was that aimed at nrook? :P 18:12:01 yes! 18:12:04 nrook: ^ 18:18:39 -!- hoody_ has quit [Remote host closed the connection] 18:26:02 -!- upsy has quit [Quit: Leaving] 18:26:43 connectivity aside, I'm toying with cavern layouts: https://gist.github.com/1074047 18:30:01 anyone see the new Kharne? it's funny how much it's ripping off crawl 18:30:04 http://roguebasin.roguelikedevelopment.org/index.php/Kharne 18:30:15 the inferface is almost identical 18:30:55 good for him, it means we're doing something right 18:31:13 http://www.flickr.com/photos/41078995@N05/3786126066/ 18:31:19 stabwound: we should release fake features and see if he copies them 18:31:32 looks exactly like crawl 18:31:45 except it has nethack style shops. 18:31:53 well, yeah 18:32:06 i'm sure it's not a ripoff as far as gameplay but it looks very similar 18:32:37 Footloop the Archmage (L27 SEEE) (Abyss) 18:33:00 It looks like a cross between tiles (mini-map) and terminal? 18:33:50 his mini-map is cool 18:33:53 we we copy it? 18:34:42 whoa. He writes healthbars under monsters. That's overkill. 18:36:32 HUD doesn't update weapon/quiver letter upon remapping (https://crawl.develz.org/mantis/view.php?id=4242) by Core Xii 18:36:36 -!- bmh has quit [Quit: bmh] 18:37:37 -!- galehar has quit [Remote host closed the connection] 18:39:59 -!- bmh has joined ##crawl-dev 18:40:49 !tell kilobyte Here's the best layout I've been able to muster so far, it's based on worley noise with a threshold determined by the 4th root of distance from the center, fuzzed with some perlin noise: https://gist.github.com/1074061 18:40:49 bmh: OK, I'll let kilobyte know. 18:44:05 bmh: it suffers from being non-contiguous, but that's common to all noise 18:44:22 oh sure, but it doesn't suffer too badly 18:44:48 I see five or six isolated squares 18:45:07 er, maybe a couple more than that, but the main body is all connected 18:45:17 Can't you clean them up (e.g. fill them) in another passß 18:45:48 could you run more tries and see if this happened just in this run, or does most of the time? 18:46:07 it's producing consistently good looking results 18:46:35 Ashenzari transfer knowledge on lvl1 skill causes crash (https://crawl.develz.org/mantis/view.php?id=4243) by Footloop 18:47:20 https://github.com/bhickey/crawl/tree/layout 18:47:28 there are three approaches: 1. plugging all areas but the main one, 2. sliding them around until they all meet, 3. adding connectors 18:48:10 all are unstable with regard to perturbation, but this doesn't matter for one-shot generation 18:48:13 In this case, I would plug all the small islands and add connectors to the large ones 18:49:07 I should also clamp the floor probability toward the edges of the map to remove the straight borders 18:49:40 yeah, like in nolithius 18:50:16 that voronoi attempt of mine that I thought it sucks but due liked needs some kind of the same 18:51:52 I think the best bet for your forest idea is beta skeletons with 1-dimensional perlin walks along the edges 18:53:59 I was considering Bresenham with a random value added to the direction every step, but perlin might indeed work better 18:54:03 You'll be able to jiggle the graph as much as you'd like. It will only appear unstable when skeleton edges break 18:54:41 In the context of level generation, the RNG is your enemy. 18:55:44 -!- hoody_ has joined ##crawl-dev 18:56:31 -!- dpeg has quit [Quit: zzz] 18:56:45 -!- bmh has quit [Quit: bmh] 18:58:09 03kilobyte * rc351e77d4eb7 10/crawl-ref/source/rltiles/dc-mon/undead/skeletons/skeleton_toad.png: Robsoie's updates to the frog skeleton tile. 18:58:19 03kilobyte * r4417407b0926 10/crawl-ref/source/command.cc: Redraw weapon and quiver after adjusting item letters. 18:59:47 -!- Pacra has joined ##crawl-dev 19:06:21 -!- edlothiol has quit [Ping timeout: 255 seconds] 19:07:44 -!- hoody_ has quit [Remote host closed the connection] 19:14:35 -!- jlewis has quit [Ping timeout: 258 seconds] 19:16:24 -!- jlewis has joined ##crawl-dev 19:30:50 -!- moocowpong1 has joined ##crawl-dev 19:32:53 -!- nrook has quit [Quit: leaving] 19:51:47 -!- MarvinPA has quit [] 19:56:00 looking at those kharne screenshots makes me really wish crawl had more colors available :( 19:58:30 -!- Mu_ has quit [Quit: Defecator, may everything turn out okay so that you can leave this place.] 20:05:16 yeah 20:12:47 oh crap 20:12:51 reading the commit log 20:12:58 kilobyte made it so that calcifying dust blocks vision 20:13:05 I don't think that's going to work 20:33:03 -!- syllogism has quit [] 20:33:36 -!- hoody_ has joined ##crawl-dev 20:34:02 -!- hoody__ has joined ##crawl-dev 20:37:46 -!- hoody_ has quit [Ping timeout: 250 seconds] 20:38:46 wensley: yeah, I guess it means you often won't be able to see the catoblepas immediately after it breathes? 20:44:05 -!- casmith789 has quit [Quit: Leaving] 20:59:45 elliptic: I'm compiling now to see if it really changes anything 21:02:18 jeanjacques (L27 OpWr) ASSERT(!actor_at(newpos)) in 'spl-tornado.cc' at line 376 failed. (Zig:8) 21:12:13 03flodiebold * r820ee422d1fa 10/crawl-ref/ (90 files in 6 dirs): First prototype of webtiles. 21:12:13 03flodiebold * r8b15f4571447 10/crawl-ref/source/tileweb.cc: Just send the changed tiles. 21:12:13 03flodiebold * ref62d0fc2b31 10/crawl-ref/source/ (5 files in 4 dirs): Draw player and monsters. 21:12:14 03flodiebold * r8b9ed1298473 10/crawl-ref/source/ (9 files in 3 dirs): Rendering improvements; stats, messages 21:12:14 03flodiebold * r69f3a30f0713 10/crawl-ref/source/ (5 files in 3 dirs): Draw text areas with canvas, too (performance is better). 21:12:14 03flodiebold * raa9fe6c8a5b1 10/crawl-ref/source/ (6 files in 2 dirs): Handle more keys (specifically, the numpad and escape). 21:12:15 03flodiebold * r228770c10182 10/crawl-ref/source/ (12 files in 3 dirs): Move rendering logic to the client, make it more complete. 21:12:15 03flodiebold * r4b2434ad28f4 10/crawl-ref/source/ (4 files in 3 dirs): Send view region shifts to reduce the data sent; add and use packed_cell::==. 21:12:16 03flodiebold * r81a5812d15b1 10/crawl-ref/source/webserver/static/keyConversion.js: Send arrow keys as arrow keys instead of numpad (they have the same key codes in js). 21:12:16 03flodiebold * r13d737ecc79b 10/crawl-ref/source/ (16 files in 3 dirs): Yet another text area rewrite. 21:12:17 03flodiebold * r2b026e36b08d 10/crawl-ref/source/ (tileweb.cc webserver/server.py): Fix a nasty memory corruption bug. 21:12:20 03flodiebold * r2b52fd2ba264 10/crawl-ref/source/webserver/static/render.js: Fix a (repeated) bug in render.js 21:12:20 03flodiebold * r57e158722534 10/crawl-ref/source/ (tileweb-text.cc tileweb-text.h tileweb.cc): Only send texts if they were changed, and fix a bug with buggy redraws after teleports. 21:12:20 03flodiebold * r50c970002a82 10/crawl-ref/source/rltiles/Makefile: Add js file generation to the makefile. 21:12:20 03flodiebold * rf2cff4eaac49 10/crawl-ref/source/webserver/static/client.html: Add favicon. 21:12:20 03flodiebold * rc9548c7707af 10/crawl-ref/source/ (libweb.cc tilesdl.h tileweb.cc): Implement clear_to_end_of_line. 21:12:21 03flodiebold * r88a89d6e619e 10/crawl-ref/source/ (14 files in 2 dirs): Implement cursors and fix some bugs. 21:12:21 03flodiebold * r4919b8ba2aec 10/crawl-ref/source/ (libweb.cc webserver/static/client.js): Implement delay(). 21:12:24 03flodiebold * r485fd0a20447 10/crawl-ref/source/ (delay.cc tilesdl.h tileweb.cc webserver/static/render.js): Clean up tileweb.cc, fix a bug that would cause some tiles not to be sent. 21:12:24 03flodiebold * r7eda7557cf95 10/crawl-ref/source/webserver/static/keyConversion.js: Add numpad-5 (it seems to have keycode 0, which is weird). 21:12:24 03flodiebold * r6634199d1a88 10/crawl-ref/source/tileweb-text.cc: Fix unicode text output (hopefully) and encode special chars as HTML entities. 21:12:36 03flodiebold * reed99c9ec955 10/crawl-ref/source/webserver/static/render.js: Fix rendering of nagas and centaurs. 21:12:36 03flodiebold * r2d2aa6ec1a21 10/crawl-ref/source/webserver/ (client.html server.py static/client.html static/client.js): Fix the server not noticing that crawl has ended; autodetect the right websocket URL. 21:12:36 03flodiebold * rce681bebdd39 10/crawl-ref/source/webserver/static/render.js: Really use flv.special (don't set offset to 0 immediately afterwards...). 21:12:36 03flodiebold * r7d581c2a62de 10/crawl-ref/source/ (tileweb-text.cc tileweb.cc): Fix handling of colored space characters. 21:12:36 03flodiebold * rda96efdbd21b 10/crawl-ref/source/webserver/server.py: A bit of cleanup in server.py. 21:12:37 03flodiebold * rc92bf1b8c4eb 10/crawl-ref/source/webserver/static/key_conversion.js: Add ctrl+G to the allowed key combinations. 21:12:39 03flodiebold * r547d3685d299 10/crawl-ref/source/webserver/static/render.js: Fix a rendering bug. 21:12:39 03flodiebold * r8440f08efeaf 10/crawl-ref/source/ (tilecell.cc tilecell.h tileweb.cc webserver/static/render.js): Fix compilation after rebase. 21:12:39 03flodiebold * r82b49a24b7ed 10/crawl-ref/source/webserver/ (client.html server.py static/client.js): DGL authentication. 21:12:39 (43 lines omitted) 21:12:40 03flodiebold * rc0429f69783b 10/crawl-ref/source/ (5 files in 3 dirs): Implement watching games. 21:12:40 03flodiebold * rf24c5584399a 10/crawl-ref/source/webserver/server.py: Don't allow watching or starting a game if you're already doing one of those things. 21:12:41 03flodiebold * rd79a19f8dc82 10/crawl-ref/source/ (4 files in 2 dirs): Fix the game exiting a menu if the window size changes or a watcher joins. 21:12:43 03flodiebold * rb199131950cd 10/crawl-ref/source/webserver/static/client.js: Fix layout changes if the window is resized while in a menu. 21:12:44 03flodiebold * rc8d7cebc430e 10/crawl-ref/source/webserver/server.py: Fix an exception if on_close is called before open. 21:12:44 03flodiebold * r38dbb53dc51e 10/crawl-ref/source/webserver/ (config.py lobby.html server.py): Implement the "where" display for DGL. 21:12:44 03flodiebold * rd22dac60f338 10/crawl-ref/source/tileweb.cc: Fix a really stupid display bug. 21:12:56 03flodiebold * rfe1c0c6b9090 10/crawl-ref/source/webserver/ (client.html server.py static/client.js static/style.css): Allow registering new users through the web interface. 21:13:01 03flodiebold * r079634fbcd61 10/crawl-ref/source/webserver/lobby.html: Use pre for the "where" column. 21:13:01 03flodiebold * rfa300dbbeb6e 10/crawl-ref/source/webserver/ (client.html static/client.js static/style.css): Make register cancelable. 21:13:01 03flodiebold * radcd3eeb7256 10/crawl-ref/source/webserver/ (config.py server.py): Allow only accepting ssl connections. 21:13:01 03flodiebold * rb7c2f2a1bb6d 10/crawl-ref/source/webserver/static/style.css: Make links more readable. 21:13:01 03flodiebold * rcdea1f61c682 10/crawl-ref/source/webserver/server.py: Add logging when a user is registered. 21:13:01 03flodiebold * r937e94e6f96e 10/crawl-ref/source/webserver/static/client.js: Hide the login form after registering. 21:13:02 03flodiebold * r7e56a04aa044 10/crawl-ref/source/webserver/lobby.html: Don't use pre, instead just replace spaces by  . 21:13:05 03flodiebold * r88f264bccb5b 10/crawl-ref/source/webserver/static/style.css: New style for links. 21:13:05 03flodiebold * rb093ba8bfd1f 10/crawl-ref/source/webserver/server.py: Dynamically reload page templates. 21:13:05 03flodiebold * r757b26b6cafa 10/crawl-ref/source/webserver/ (client.html footer.html): Add a footer. 21:13:05 03flodiebold * rb75470b563e0 10/crawl-ref/source/webserver/static/client.js: List the browsers that can run webtiles. 21:13:06 03napkin * ra2a82e7b38c4 10/crawl-ref/source/webserver/ (banner.html client.html static/client.js): Change name to WebTiles. 21:13:06 03flodiebold * r6ff3c4d80c7e 10/crawl-ref/source/webserver/server.py: Fix the keep-alive message being considered as activity for idle-time. 21:13:07 03flodiebold * r66ce9d66d99c 10/crawl-ref/source/webserver/server.py: Better logging. 21:13:07 (52 lines omitted) 21:13:11 03flodiebold * r1f770036d23c 10/crawl-ref/source/webserver/server.py: Handle exceptions when trying to write to a websocket. 21:13:11 03flodiebold * r612d21c86f30 10/crawl-ref/source/webserver/server.py: Fix the last commit. 21:13:11 03flodiebold * re934d9b30d46 10/crawl-ref/source/webserver/server.py: Fix an exception after killing crawl. 21:13:11 03flodiebold * re56f43601e97 10/crawl-ref/source/webserver/static/style.css: Highlight the game the mouse points at. 21:13:12 03flodiebold * r74967c60d0c6 10/crawl-ref/source/invent.cc: Fix an overlooked instance of the format tag bug. 21:13:27 03flodiebold * r0ba4044fca72 10/crawl-ref/source/webserver/ (config.py server.py): Add a non-dgl mode in which crawl is immediately started, without options. 21:13:27 03flodiebold * rd6db144fb18f 10/crawl-ref/source/webserver/static/style.css: Fix table row hover style. 21:13:27 03flodiebold * r13db276196a2 10/crawl-ref/source/webserver/static/key_conversion.js: Don't handle numpad keys with numlock twice in FF. 21:13:27 03flodiebold * r13dff7ee9b6a 10/crawl-ref/source/webserver/static/client.js: Disable alt key combinations, since it's currently not working and causes #4075. 21:13:27 03flodiebold * r20fc31cb8d54 10/crawl-ref/source/files.cc: Don't load an empty view buffer in webtiles. 21:13:27 03flodiebold * rbeb3a4e89104 10/crawl-ref/source/ (6 files in 3 dirs): Render and send the whole map. 21:13:28 03flodiebold * rd295bb173111 10/crawl-ref/source/ (tilesdl.h tileweb.cc webserver/static/game/game.js): Fix spectator joining. 21:13:39 03flodiebold * r1c9eff168e4b 10/crawl-ref/source/webserver/ (server.py static/client.js): Fix watchers joining from the lobby. 21:13:39 03flodiebold * r4fff6730bba6 10/crawl-ref/source/tilecell.cc: Don't put overlays on unseen tiles. 21:13:39 03flodiebold * rd2ea3127a7ef 10/crawl-ref/source/tileweb.cc: Don't send tile data for completely unseen tiles. 21:13:39 03flodiebold * r11b6330a41f0 10/crawl-ref/source/tileweb.cc: Fix the tile flavour not being sent. 21:13:40 03flodiebold * r5d8c2980b999 10/crawl-ref/source/tileweb.cc: Only send flavour for cells that need it. 21:13:40 03flodiebold * r7fdec98f3351 10/crawl-ref/source/webserver/server.py: Delay messages until the client is loaded. 21:13:41 03flodiebold * ra2c7c939e985 10/crawl-ref/source/webserver/static/game/render.js: Fix a bug when shifting the view area by too much. 21:14:39 03flodiebold * rc1da5432b6b8 10/crawl-ref/source/ (files.cc tileweb.cc webserver/static/game/render.js): Clear the map when changing levels. 21:14:39 03flodiebold * r7433dfc689c9 10/crawl-ref/source/webserver/static/game/render.js: Add a debug function. 21:14:39 03flodiebold * ree47c374aa2b 10/crawl-ref/source/ (6 files in 4 dirs): First minimap implementation. 21:14:39 03flodiebold * r2c98d5e3153d 10/crawl-ref/source/ (4 files in 2 dirs): Don't leak exact map coordinates; center the minimap on the known area. 21:14:40 03flodiebold * radd7ad768988 10/crawl-ref/source/tileview.cc: Don't leak information about unknown water tiles. 21:14:40 03flodiebold * r4489ce029ca2 10/crawl-ref/source/webserver/static/game/game.js: Update the minimap overlay when the view area changes (again). 21:14:41 03flodiebold * r028b8d79c2dc 10/crawl-ref/source/ (tilesdl.h tileweb-text.cc tileweb.cc): Abstract sending of messages into its own function(s). 21:15:39 (25 lines omitted) 21:17:49 !tell kilobyte Thanks so much for putting in catoblepae! :D Just two things: 1) you used the wrong one of the two tiles that Denzi made for the catoblepas (the cloud tiles look fine) and 2) why make the clouds obscure vision? 21:17:50 Wensley: OK, I'll let kilobyte know. 21:43:45 -!- ainsophyao has joined ##crawl-dev 21:44:49 -!- ainsophyao has quit [Remote host closed the connection] 21:48:05 -!- ainsophyao has joined ##crawl-dev 21:55:33 -!- Wensley has quit [Ping timeout: 255 seconds] 22:01:13 -!- Wensley has joined ##crawl-dev 22:01:40 -!- ainsophyao has quit [Remote host closed the connection] 22:07:58 -!- ainsophyao has joined ##crawl-dev 22:11:01 Wensley: add stoneskin potions so there can be evap petrify clouds 22:12:40 -!- Wensley has quit [Ping timeout: 250 seconds] 22:13:42 -!- hoody__ has quit [Remote host closed the connection] 22:25:36 -!- jeanjacques has quit [Ping timeout: 276 seconds] 22:32:51 -!- hoody_ has joined ##crawl-dev 22:36:16 -!- st__ is now known as st_ 22:37:16 -!- jeanjacques has joined ##crawl-dev 22:45:40 -!- ainsophyao has quit [Remote host closed the connection] 23:02:10 03elliptic * r11ecce8b9b3b 10/crawl-ref/source/spl-cast.cc: Fix Tornado cooldown affecting non-tornado spells. 23:17:50 -!- ainsophyao has joined ##crawl-dev 23:18:00 -!- ainsophyao has quit [Remote host closed the connection] 23:20:51 !tell dpeg Not in the sense of "we should go and implement this now", more like "this was a well thought-out proposal and I liked it". Of course there is a mass of stuff, it's actually an Eronarn proposal! :) 23:20:52 Keskitalo: You have 1 message. Use !messages to read it. 23:20:52 Keskitalo: OK, I'll let dpeg know. 23:23:38 -!- GrimmSweeper has quit [Read error: Connection reset by peer] 23:24:18 -!- ainsophyao has joined ##crawl-dev 23:27:20 which proposal is this? 23:39:40 weapon reform 23:50:15 Oooh, weapon reform.