00:01:20 <|amethyst> that vault doesn't place his band 00:02:14 <|amethyst> minmay_rotating_lava looks suspicious 00:03:10 <|amethyst> everything else looks fine 00:17:14 -!- xnmojo_ has joined ##crawl-dev 00:17:47 playing today I was thinking it would be nice to easily see if your stats were modified by items 00:17:58 so where it says STR 23 00:18:10 if you had +4 from a ring it could say STR 23 [+4] 00:18:23 to show you that thats not your base number 00:18:35 or when berserked etc 00:26:45 -!- Wensley has quit [Ping timeout: 248 seconds] 00:43:44 -!- MakMorn has quit [Ping timeout: 246 seconds] 01:16:30 -!- Sequell has joined ##crawl-dev 01:20:14 !tell kilobyte Recent Yred army sighted: 6 bone dragons and 7 profane servitors following an XL19 char around. Didn't we try to make this sort of thing not happen? 01:20:15 elliptic: OK, I'll let kilobyte know. 01:29:07 -!- dtsund has quit [Ping timeout: 244 seconds] 01:33:24 -!- dtsund has joined ##crawl-dev 01:40:03 -!- MakMorn has joined ##crawl-dev 02:16:39 03evilmike * rbca38a398d71 10/crawl-ref/source/dat/des/branches/zot.des: A new Orb subvault. (nicolae) 02:16:39 03evilmike * rbe78005a71f8 10/crawl-ref/source/dat/des/branches/zot.des: Another new orb subvault. (nicolae) 02:16:39 03evilmike * rf50d9dcd8e90 10/crawl-ref/source/dat/des/branches/spider.des: Two new Spider vaults. (nicolae) 02:16:40 03evilmike * r56adde021308 10/crawl-ref/source/dat/des/branches/orc.des: Some more Orc accident vaults. (nicolae) 02:16:40 03evilmike * r31e1bf3c01c8 10/crawl-ref/source/dat/ (4 files in 3 dirs): Misc vaults. (nicolae) 02:35:17 -!- Pingas_ has joined ##crawl-dev 02:35:31 -!- Xiberia_ has joined ##crawl-dev 02:35:42 -!- Vandal has joined ##crawl-dev 02:36:31 -!- Ragdoll has joined ##crawl-dev 02:37:18 -!- Ganrao| has quit [Ping timeout: 252 seconds] 02:37:39 -!- dtsund has quit [Ping timeout: 252 seconds] 02:38:24 -!- Pingas has quit [Ping timeout: 252 seconds] 02:38:46 -!- Xiberia has quit [Ping timeout: 252 seconds] 02:38:57 -!- Xiberia_ is now known as Xiberia 02:44:35 -!- HousePet has joined ##crawl-dev 02:44:54 wow the tavern is funny and sad at the same time 03:07:38 -!- dtsund has joined ##crawl-dev 03:10:52 -!- ens has quit [Ping timeout: 245 seconds] 03:26:02 -!- dtsund has quit [Quit: dtsund] 03:36:49 -!- clouded_ has joined ##crawl-dev 03:39:37 -!- Mottie has quit [Ping timeout: 240 seconds] 04:08:16 !messages 04:08:17 (1/1) elliptic said (2h 48m 2s ago): Recent Yred army sighted: 6 bone dragons and 7 profane servitors following an XL19 char around. Didn't we try to make this sort of thing not happen? 04:08:37 this was talked about but with no actual action 04:09:43 dtsund requested a way to allow picking up certain monsters from other levels without having to do a level_excursion, this would allow implementing that off-level recall thingy 04:10:16 (for Crawl Light, it's needed to fix mummy stabbing) 04:12:22 I think it'd be easiest to abuse transit: remove relevant monsters (near stairs for Light, Yred gifts for us) from the map, recording their exact spot on the transit list, and if they won't get recalled in the meantime, put them back when you come back to the level 04:14:19 If we have that kind of inter-level transit, why not have reliable allies following instead of inter-level recall? 04:17:08 there are multiple proposals there, ranging from limited following to involuntary recall 04:18:23 I'd prefer finishing at least the mon-pick overhaul though -- I got missile combat and translations on my plate right now, got to finish _something_ 04:19:53 bad news for you: since you're caught online, I'm going to harass you for input for that mon-pick thing we discussed with elliptic yesterday :p 04:20:36 current best idea: http://sprunge.us/EdiN 04:21:06 the nasty-looking numbers in that wall of stuff are: min depth, max depth, rarity, flatness 04:21:58 I tried replacing flatness as a number by one of three enums for readability, not sure if that's a good idea 04:24:20 FLAT means full rarity on the whole range, PEAK means a linear progression from 0 at the ends to full rarity in the middle, SEMI is something in-between 04:27:47 galehar: it's damn likely we're trying to reproduce the old formula too much, thus a sanity check from you would be damn welcome 04:27:49 kilobyte: oh, one thing I realized is that those rarity numbers in your table should be replaced by something proportional to the chance that random2avg(100,2) <= rarity 04:28:47 since that's the actual peak weight for the monster 04:28:58 hmm right, yeah 04:29:08 but how to represent that in the distribution? 04:29:23 would it be sane to adjust just the peak? 04:30:01 when I suggested the linear progression from something at the ends to a peak at the middle, that was already taking this into account 04:30:16 ok, right 04:31:08 I think what alefury said the other day about it probably not mattering all that much what the distribution is so long as it is small at both ends and peaks in the middle is probably true, though 04:31:09 that "something" is actually 0 at mindepth-1 and maxdepth+1, as mindepth is defined as the shallowest depth with a non-zero chance 04:31:20 yeah 04:31:28 -!- alefury has joined ##crawl-dev 04:31:31 even these three distributions could be overkill 04:31:36 speak of the devil 04:32:16 kilobyte: finishing the mon-pick overhaul first is a good idea. The table looks pretty good. I understand it! :) 04:32:38 alefury: tail of http://s-z.org/crawl-dev/%23%23crawl-dev-20120531.log 04:32:55 even with the wall of numbers? 04:33:30 the old formula had two numbers and a bunch of unobvious rules 04:33:55 yes, I find the wall of numbers easier to understand and tweak than the old thing 04:34:14 there is also a big bad problem that I think should be solved before implementing this: the OOD cap 04:34:15 replacing the 4th value by an enum helps too 04:34:19 i was just reading that :) 04:35:02 kilobyte: I don't understand what it is (the OOD cap). 04:35:28 btw, you say peak goes from 0 to max. Does that mean the ends of the depth range have 0 weight for the monster? because that would be weird. 04:35:58 alefury: I'd make it 0 at mindepth-1 to have the end be > 0 04:36:33 that sounds good, yes. 04:36:56 galehar: when a monster is requested from a depth deeper than the cap, a monster from the cap (D:31, Vault:15, Zot:4, etc) is picked instead 04:37:33 also, negative levels don't do anything with FLAT. Might be good to just leave them as they are though, in case the distribution gets changed. 04:37:35 the "9" symbol and "super OODs" go there in a vast majority of cases 04:38:36 this is what almost all of initial monster generation on Vaults:8 is about, for example 04:38:48 super OOD is called when you cross the OOD timer, right? So the OOD cap is not used in normal level generation? 04:39:08 except with "9" in vault 04:39:38 Zot and Hells exceed the cap during natural generation 04:39:53 re OOD cap: what happens if a 9 is requested on Orc:4? OOD cap is 23, 9 depth is 12 I think, and the deepest monster depth defined for Orc is 7. 04:40:18 even an 8 would give depth 9 04:41:40 i think having an OOD cap is good, otherwise monsters would have to be defined all the way down for ($+2)*2 04:41:44 '9' is Orc:13 on Orc:1, and Orc:19 on Orc:4 04:42:05 didnt you say it was (d+2)*2? 04:42:20 either way its deeper than the deepest monster definition 04:42:49 wouldn't it be easier to read with native_depth, range instead of min_depth, max_depth? 04:43:00 alefury: that's calculated in absdepth 04:43:00 (those negative numbers...) 04:43:55 galehar: i thought about that a bit, and im not sure. :( 04:44:18 galehar: the range makes it easier to read where the monster is allowed to spawn 04:44:43 right 04:44:52 an idea: perhaps that PEAK/SEMI/FLAT enum could be extended to something more than just flatness? 04:45:01 like, a triangle down 04:45:42 ? 04:45:51 what would you want that for? 04:45:55 also, vaults can be pruned of entries with max_depth <= 0, right? 04:46:32 galehar: the automated script marks these with // 04:46:53 oh right 04:47:28 same with monsters with min_depth above the OOD cap 04:47:49 using the enum for different distributions, sure. I just don't know what you would use them for. 04:47:56 mostly fiends and ancient liches in hells; they can still spawn through hell effects 04:48:11 about OOD cap, couldn't it be set to the native depth of the deeper branch monster? 04:48:35 alefury: PEAK with a negative min_depth 04:48:57 ah, thats what you meant 04:49:12 galehar: deepest, you mean? Lemme calculate that in the script. 04:49:27 i thought you meant a V-shaped distribution when you said "triangle down" 04:49:54 yeah, deepest 04:49:57 would be neat to get rid of negative depths 04:53:54 http://sprunge.us/geRj -- changes things quite a lot 04:54:59 -!- ens has joined ##crawl-dev 04:55:11 wait, no: except for Hells and Zot, capped OODs in most branches mean just "pick any monster" 04:55:26 one way would be to specify native depth. Even if it's (min+max)/2 most of the time. Set it to 0 for flat, but no pretty way of setting semi. 04:56:55 kilobyte: shouldn't maxlev be offset by the branch entry absdepth? 04:57:53 correction: the cap of max(native depth) would be a massive change in some cases; for example Elf would get a lot of blademasters/master archers 04:58:35 galehar: this file shows OOD cap and maxlev as relative depths 05:00:07 OSX builds of master branch on crawl.develz.org updated to: 0.11-a0-2079-g31e1bf3 05:03:13 let me try to understand this. Elf is absdepth 15, right? so when asking a "9" in Elf:3, (18+2) * 2 = 40. Capped at 16. 05:03:53 looking at branches, changing the OOD cap to max(native depth) would produce reasonable results for all branches except Elf, Tomb and those with wrong caps (Zot, hells) 05:04:01 galehar: 40-15 05:04:27 which is still bigger than 16 05:04:52 note that regular OODs and "8" will not hit the cap, though 05:06:07 absdepth+depth? That's 18 for Elf, not far from 16. it's unreasonable? 05:08:10 galehar: any value bigger than 8 means "pick any monster ignoring depth ranges" 05:09:43 -!- edlothiol has joined ##crawl-dev 05:14:39 if OOD level is capped at 7 there, it'd result in OODs meaning: Elf:1 (roughly) plus a blademasters/DEMAs 05:14:47 maybe what we need to do is change the OOD formula so it doesn't use absdepth. Maybe +min(x,depth*y). Depth being the branch length 05:15:32 +min(12,depth*2) 05:15:49 lunch time 05:26:47 -!- evilmike_ has quit [Quit: something happened] 05:28:56 -!- ainsophyao has quit [Remote host closed the connection] 05:54:57 -!- barbs has joined ##crawl-dev 05:55:31 lol. There's a Lantern of Shadows thread on the tavern and it got spammed by a decorative latern manufacturer 05:57:04 heh, nice 06:00:18 fr lantern that trails colored smoke everywhere you go 06:13:32 for that OOD cap problem: we could 1. add a branch flag "for out of bounds OODs, ignore depth ranges", 2. use that max(native depth) cap, and ignore Elf/Tomb issues, 3. think more (but thinking hurts!) 06:15:16 how about changing the OOD cap mechanic to actually make sense? As in make it an actual cap on the depth, instead of picking randomly when it's exceeded? 06:15:45 combined with using max(native depth) 06:17:07 OOD monsters are supposed to actually be OOD monsters, right? Why should that be circumvented? It seems much better to implement this in a way that actually makes sense, and then fix any problems this causes. 06:17:12 BlastHardcheese: "Crap lantern" 06:18:02 Either way any changes should get into trunk soon, because this will need testing, and 0.11 is not that far off. 06:28:14 -!- ussdefiant_ has joined ##crawl-dev 06:33:19 btw, one good test would be running those monster generation scripts and comparing to galehar's old results 06:34:03 i dont remember how detailed the results are, but at least you get xp per level 06:35:49 -!- Nomi has quit [Ping timeout: 248 seconds] 06:40:18 the results are very detailled and the script has a lot of configuration options 06:40:25 you can see per monster frequency 06:40:37 I didn't put all the results on the wiki 06:49:22 hm, that seems perfect 06:49:36 how long do they run? minutes? tens of minutes? 06:50:33 depends how many runs per level (default 150) and how many levels 06:50:34 also, can they test depths that only occur as OOD spawns? 06:51:12 probably not necessary, though 06:51:17 the script generate levels 06:51:29 so thats a no i guess. 06:52:35 could work around it with a testing vault, but it would only be helpful for seeing how super-oods got changed, so probably not important. 06:52:39 just tested, you can't generate invalid depths 06:54:30 running the script on V:8 can show how "9" ood are affected 06:54:47 maybe other branch ends use "9", don't know 06:55:18 it only matters how levels that exist are affected, and thats what the script tests 06:55:35 or, it's possible to create a vault with 9s, guarantee it at a certain depth, and run the script before/after the change to see the impact 06:55:41 but it doesnt do extra spawns, just initial generation 06:55:50 yes 06:56:14 galehar: it's even possible to create a vault with Lair:15 monsters to see how Lair:15 monster distribution is affected 06:58:37 -!- mikee_ has quit [Ping timeout: 240 seconds] 06:58:51 seeing how OOD timer spawns are affected might be useful if changes to how the OOD cap works are made 06:59:26 other than that just checking initial generation for all existing levels before/after changes should be good enough 06:59:46 for branch ends probably both including and not including vaults (which the script can do iirc) 07:00:40 early D levels should probably also be checked both with and without vaults 07:05:47 -!- mikee_ has joined ##crawl-dev 07:21:50 kilobyte: can you use rarity as a secondary key for ordering so abyss and pan monster lists are easier to read? 07:28:24 03thann * r7136bdd76718 10/crawl-ref/source/items.cc: Count items gained when merging a picked-up stack (Thann). 07:28:31 03|amethyst * r8c7ba4369b19 10/crawl-ref/source/ (itemprop-enum.h makeitem.cc mapdef.cc): Allow "any rod" in itemspecs. 07:28:32 03|amethyst * rfe6b02f91259 10/crawl-ref/source/mon-place.cc: Don't place band members across walls. 07:29:12 -!- MarvinPA has joined ##crawl-dev 07:30:41 <|amethyst> !tell HangedMan You've got your "any rod" now 07:30:42 |amethyst: OK, I'll let HangedMan know. 07:32:36 oh wow, you fixed the band members bug! 07:32:53 -!- ussdefiant_ has quit [Ping timeout: 248 seconds] 07:34:53 <|amethyst> it could end up making some bands small, if placed in tight spots 07:35:51 btw, are you sure you fixed it? iirc it was problematic due to the order in which stuff in vaults gets placed (monsters before walls, or something like that) 07:35:59 I HAD NOTHING TO DO WITH oh never mind 07:36:07 there are some comments in the duplicates 07:36:54 i think i didnt remember correctly 07:37:43 <|amethyst> it seemed okay, let me do a little more testing 07:37:53 <|amethyst> I was using &L to test but that might be a bit funny 07:37:54 check the related report too 07:38:06 and its duplicates 07:38:23 <|amethyst> I didn't address that one 07:38:26 <|amethyst> hm 07:39:16 -!- Mottie has joined ##crawl-dev 07:44:14 <|amethyst> yeah, it seems to still have a problem with vaults 07:45:02 <|amethyst> that said, it looks like only one existing vault would be affected 07:52:22 iirc there was some discussion about how to fix it, soon after i cleaned up that duplicate mess a little 07:52:30 you could check the irc logs for it if you feel like it :P 07:52:53 <|amethyst> hm 07:53:10 <|amethyst> I was thinking about doing two passes, one for features and items and another for monsters 07:53:44 yeah, that came up 07:55:37 -!- Wensley has joined ##crawl-dev 07:58:29 Napkin: since it is almost world ipv6 day, how hard is it to configure a server to be ipv6 ready? (I figure that develz has been ready for about a decade) 07:59:11 -!- xnmojo_ has quit [Ping timeout: 245 seconds] 08:01:40 well, only webtiles isn't ipv6 yet - the rest is :) 08:01:56 Wensley: you need a ring of ipv6 and network skill at level =~ 15 08:02:33 galehar: but I am a troll, my networking aptitude is -3 08:02:53 <|amethyst> alefury: this was shortly after you marked the duplicates etc you say? 08:03:11 <|amethyst> do you recall if anyone had objections to doing two passes? 08:06:46 <|amethyst> kilobytes said "it's a bit more complex if there's more than one monster" 08:11:10 <|amethyst> seems to work... 08:11:15 Wensley: best is to get DNS done properly. your FQDN should resolve into ipv4 and ipv6. if the service is properly written, it'll listen on both ips if you tell it to listen on the FQDN. 08:15:41 03|amethyst * r2d761a2458cf 10/crawl-ref/source/dungeon.cc: Don't place band followers in vault walls. 08:17:16 -!- ark____ has joined ##crawl-dev 08:19:09 -!- Mottie has quit [Ping timeout: 250 seconds] 08:22:46 <|amethyst> alefury: thanks 08:24:38 -!- HousePet has quit [Quit: Divide by cucumber error] 08:24:49 Napkin: cool, I'll get started on that 08:27:27 was afk 08:30:39 <|amethyst> minmay says the effect on minmay_rotating_lava sounds fine--that was the intended behaviour anyway 08:30:58 Wensley: unless you're implementing this on a low level, you're not supposed to even care if it's ipv4 or ipv6 08:32:14 Wensley: webtiles tell the browser "please connect to tiles.crawl.develz.org", and it'd resolve it, try connecting via ipv6, immediately get "network unreachable" from your router, retry over ipv4 08:34:03 -!- mikee_ has quit [Read error: Connection reset by peer] 08:35:32 |amethyst: the problem with walls/monsters is: all kinds of symbols get resolved top-to-bottom, left-to-right (or a mirror of that :p), then if a band is placed, monsters that spill downwards don't yet know whether that place will be wall or floor. And at that moment it's likely to be (still) floor 08:35:43 <|amethyst> kilobyte: yes, I fixed that 08:36:04 <|amethyst> kilobyte: apply_grid() now makes two passes 08:36:12 the monster/monster problem is: the monster up has a band, which gets placed one down. Then, the next monster tries to be placed... 08:36:56 so it would have to place leaders first, band members later? 08:37:00 <|amethyst> hmm 08:37:01 <|amethyst> I see 08:37:49 i dont really see how the current behavior is worse than the old one regarding multiple monsters 08:38:23 but seeing as that is a bug that will likely never be reported, it might be good to fix it before it gets forgotten :P 08:38:30 <|amethyst> hm, and the subsequent monsters won't be retried because they had a specified place 08:39:15 <|amethyst> I don't know, someone might notice if prince ribbit's blink frog band gets placed (outside lair) but not ribbit himself 08:39:50 <|amethyst> well, doing ordinary monsters in pass 1 (or 1.5) and bands in pass 2 would work 08:40:31 |amethyst: doesnt the band always get placed after the leader? otherwise checking los to leader wouldnt even work because null pointer. 08:40:49 unless checking for the actual location 08:40:54 <|amethyst> the leader isn't the issue, it's other monsters in the same vault 08:40:56 strictly speaking, still incorrect as one band would block another, but I somehow don't exactly see vaults with multiple bands next to each other 08:41:15 |amethyst: yes i know, but that was the example you used 08:41:18 alefury: if placing leader fails, band is not even tried 08:41:21 <|amethyst> the prince ribbit vault uses MONS: Prince Ribbit, blink frog band 08:41:29 <|amethyst> so Ribbit isn't the band leader 08:41:40 oh 08:41:53 <|amethyst> if the "2" is placed before the "1" there might not be room for him 08:42:07 and vaults get rotated... 08:42:12 <|amethyst> right 08:42:26 yeah i get it. and with your "no bands through walls" change this might actually happen a lot in cramped spaces 08:43:11 <|amethyst> the ilsuiw band doesn't have quite the same problem... the band monsters might block some of the vault monsters, but they won't block ilsuiw herself 08:43:35 also, they will no longer spawn in walls! :D 08:43:58 <|amethyst> kilobyte: band monsters shouldn't block subsequent band monsters (if the leader isn't blocked), because _place_monster_aux will retry at a different position 08:44:12 <|amethyst> kilobyte: at least, until the room is full, and at that point they should be blocked 08:44:20 so doing monsters in pass 1, bands in pass 2 should work 08:44:30 <|amethyst> problem is that it really breaks layers of abstraction to do that 08:44:49 Why is it that certain deaths like 'splashed by acid' don't indicate the monster they came from? Is there a technical reason? 08:44:57 yeah, it sounds hacky without even looking at the code :( 08:45:20 -!- Fa has joined ##crawl-dev 08:50:21 Patashu: they're implemented as different death types, with hardly any code sharing between them 08:51:02 yuck 08:51:27 -!- mikee_ has joined ##crawl-dev 08:53:15 -!- ToBeFree has joined ##crawl-dev 08:56:27 Vault monster bands can block other monsters. (https://crawl.develz.org/mantis/view.php?id=5710) by neil 08:57:07 -!- Fa has quit [Ping timeout: 245 seconds] 08:59:20 -!- Grunt has joined ##crawl-dev 09:00:59 -!- MarvinPA_ has joined ##crawl-dev 09:03:56 <|amethyst> maybe in the first pass set a flag saying "a monster will be placed here" 09:04:09 <|amethyst> a mask that is 09:04:12 -!- MarvinPA has quit [Ping timeout: 245 seconds] 09:04:12 -!- MarvinPA_ is now known as MarvinPA 09:04:46 <|amethyst> then avoid cells with that mask in _valid_monster_generation_location 09:04:56 <|amethyst> err, well, not quite that 09:06:36 Exiting the abyss can place the player in a wall. (https://crawl.develz.org/mantis/view.php?id=5711) by st 09:09:13 -!- Fa has joined ##crawl-dev 09:12:30 Unstable branch on crawl.akrasiac.org updated to: 0.11-a0-2083-g2d761a2 (33) 09:34:34 kilobyte: re your tornado vault, some dude on the tavern says he played around with it in wizmode cant fly accross and back before dying even with a berserk l27 fighting 27 ogre 09:34:45 argh typos 09:35:50 use blink, apport, tele, cblink 09:37:28 alefury: hmm, weird, after all tornado nerfs it did like 15 damage during the way in the last time I tested it 09:38:00 it does more damage to flying things, yeah? 09:38:14 no 09:38:20 no, thats airstrike. tornado makes things fly. 09:38:47 even Leda doesn't help against it 09:39:09 being immovable does 09:39:35 (but I really doubt I can get tree form for 0.11 :p) 09:45:26 maybe tornado got buffed again since you tested it 09:45:35 its quite strong again ime 09:45:45 when we add treefolk, let's add a unique one named Git 09:46:13 since playing diablo 3 i hate trees. trees suck. 09:46:15 "He has many branches." 09:47:43 they are kind of like crawl catlobes, except instead of petrifying you they kill you 09:48:48 and their gas is short range instead of line shaped, so you have to step back instead of aside, which means melee chars have to go out of attack range to not get killed. 09:57:48 alefury: I just tried the obelisk, doing a full turn deals about 50 damage 10:01:09 -!- ivan`` has quit [Ping timeout: 248 seconds] 10:02:42 okay, so the dude just wasnt doing it right 10:03:45 -!- ivan`` has joined ##crawl-dev 10:06:37 -!- bhaak is now known as kerio____ 10:06:57 -!- kerio____ is now known as bhaak 10:08:22 -!- ainsophyao has joined ##crawl-dev 10:08:35 -!- ussdefiant_ has joined ##crawl-dev 10:28:41 -!- Sequell has quit [Quit: Sequell] 10:28:53 -!- Sequell has joined ##crawl-dev 10:34:12 -!- Fa has quit [Read error: Operation timed out] 10:38:04 -!- ussdefiant_ has quit [Ping timeout: 265 seconds] 10:40:28 -!- ens has quit [Ping timeout: 260 seconds] 10:40:37 -!- Patashu has quit [Ping timeout: 240 seconds] 10:46:03 -!- Mottie has joined ##crawl-dev 11:16:47 -!- ussdefiant_ has joined ##crawl-dev 11:18:02 -!- blabber has joined ##crawl-dev 11:38:14 -!- barbs has quit [Ping timeout: 246 seconds] 11:55:50 -!- ussdefiant_ has quit [Ping timeout: 244 seconds] 11:55:58 -!- ussdefiant_ has joined ##crawl-dev 12:04:07 -!- ainsophyao has quit [Remote host closed the connection] 12:05:04 -!- Wensley has quit [Quit: leaving] 12:19:03 -!- ussdefiant_ has quit [Ping timeout: 260 seconds] 12:19:45 -!- ussdefiant_ has joined ##crawl-dev 13:09:37 -!- monqy has joined ##crawl-dev 13:10:40 guys, we need to find an alternative to hosting webtiles on CDO 13:11:17 didnt someone recently buy a new server? in the us? 13:11:56 i'm starting to get annoyed that even though there are 4 cores with 2.8ghz the server has a permanent load between 2 and 5 and is very laggy/delayed for every kind of action.. even something simple as writing a single sentence here with irssi 13:12:17 i forgot who, though. might be greensnark. 13:12:21 anyway, bye for now 13:12:25 -!- alefury has quit [Quit: ChatZilla 0.9.85 [Iceape 2.0.11/20120421074340]] 13:13:01 alternative: don't host webtiles :I 13:13:31 optimize! 13:15:16 -!- ToBeFree has quit [Read error: Connection reset by peer] 13:16:22 -!- dtsund has joined ##crawl-dev 13:17:36 -!- elliptic has quit [Quit: Leaving] 13:26:22 -!- mikee_ has quit [Quit: HydraIRC -> http://www.hydrairc.com <- The professional IRC Client :D] 13:59:36 Napkin: once I have new hardware it might be able to handle it 13:59:50 Although maybe not even then if it's that bad for you, wow 13:59:54 How many games at a time is that? 13:59:55 rax :)) 14:00:05 ~30, i think 14:00:12 but haven't checked for a while 14:00:28 Wow that is significantly more resource-intensive than the non-webtiles version 14:01:05 yes 14:02:04 Yeah I dunno if I want to dedicate four cores on the machine just to crawl 14:02:29 I mean I can wall it off and make just crawl slow at least because I do virtualization, but 14:02:57 -!- |amethyst has quit [Ping timeout: 252 seconds] 14:03:28 -!- ussdefiant_ has quit [Ping timeout: 245 seconds] 14:03:56 -!- |amethyst has joined ##crawl-dev 14:05:24 -!- tswett has quit [Ping timeout: 244 seconds] 14:07:03 -!- alefury has joined ##crawl-dev 14:07:37 -!- tswett has joined ##crawl-dev 14:15:04 Napkin: do you have data about how much the usage (as in: avg concurrent players) have risen? Is it a matter of Crawl going slower or of just more players? 14:15:15 -!- |amethyst has quit [Read error: Connection reset by peer] 14:16:15 -!- Mottie has quit [Ping timeout: 244 seconds] 14:16:42 -!- |amethyst has joined ##crawl-dev 14:17:04 -!- ussdefiant_ has joined ##crawl-dev 14:17:15 Napkin: also, what is the ratio of tourney to regular periods? 14:17:36 no idea 14:17:59 doh, Sequell knows the latter 14:18:40 doesn't cao gather this info? 14:19:01 no clue about cpu usage stats 14:19:10 i just see the lag.. maybe sar could help? 14:19:18 !lg * end>=20120501 end<=20120307 14:19:20 No games for * (end>=20120501 end<=20120307). 14:19:23 !lg * end>=20120301 end<=20120307 14:19:24 13765. polystyrus the Grave Robber (L7 OpNe), worshipper of Kikubaaqudgha, blasted by an orc wizard (puff of flame) on D:5 on 2012-03-06, with 778 points after 5147 turns and 0:15:38. 14:19:30 !lg * end>=20120501 end<=20120507 14:19:32 10453. Zurktot the Sneak (L1 NaFE), slain by a goblin on D:1 on 2012-05-06, with 34 points after 182 turns and 0:01:12. 14:19:59 beh, that small a difference? 14:20:50 CAO suggests crawl has gotten more resource intensive over 5 years but I don't keep good stats :( 14:20:52 !lg * end>=20100501 end<=20100507 14:20:53 5901. Cerrus the Poker (L3 HODK), succumbed to poison on D:1 on 2010-05-06, with 57 points after 1359 turns and 0:02:54. 14:21:28 double the number of games played per week 14:21:39 !lg * end>=20120301 end<=20120307 xl<10 14:21:41 12061. polystyrus the Grave Robber (L7 OpNe), worshipper of Kikubaaqudgha, blasted by an orc wizard (puff of flame) on D:5 on 2012-03-06, with 778 points after 5147 turns and 0:15:38. 14:21:49 !lg * end>=20120501 end<=20120507 xl<10 14:21:50 9193. Zurktot the Sneak (L1 NaFE), slain by a goblin on D:1 on 2012-05-06, with 34 points after 182 turns and 0:01:12. 14:22:13 !lg * end>=20070501 end<=20070507 14:22:14 983. mr0t the Martial Artist (L2 SETm), slain by a snake on D:2 on 2007-05-06, with 118 points after 1766 turns and 4:32:23. 14:22:50 rax: with over ten times as many games, that's no surprise 14:23:37 I have no data how CPU/bandwidth costly webtiles vs console is 14:23:43 !lg * end>=20120301 end<=20120307 won 14:23:44 Bad timestamp value: 20120307 won 14:23:55 !lg * won end>=20120301 end<=20120307 14:23:56 146. Paroid the Eclecticist (L25 DgFE), escaped with the Orb and 4 runes on 2012-03-06, with 1730232 points after 113475 turns and 20:21:42. 14:24:03 !lg * won end>=20120501 end<=20120507 14:24:04 82. nmf the Acrobat (L27 DEFE), worshipper of Ashenzari, escaped with the Orb and 5 runes on 2012-05-06, with 2064192 points after 141083 turns and 9:38:26. 14:25:39 kilobyte: But even across tournaments, which have decently similar load I think 14:25:52 Some of it is that there are more secondary things on the machine as well 14:26:50 rax: do you remember the dates of an ancient tourney? 14:28:49 it's been August pretty much every year I think 14:29:35 Aug 2008, Aug 2009, Aug 2010, Aug 2011, with others in May 2011 and Feb-Mar 2012 14:29:54 aug 2011 was 2 weeks, i think. the others were the full month 14:29:59 Planning on one for Aug 2012 once the hardware and such was updated 14:30:05 That sounds right 14:30:06 -!- ens has joined ##crawl-dev 14:30:06 the others as in the previous augusts 14:30:14 All of them are archived on the front page actually 14:36:53 Lack of Monsters on Vaults:8 (https://crawl.develz.org/mantis/view.php?id=5712) by pivotal 14:37:05 -!- Nomi has joined ##crawl-dev 14:37:44 !lg * end>=20080810 end<=20080817 14:37:46 3225. Anym the Cutter (L2 VpWr), slain by a kobold (a +0,+0 club) on D:2 on 2008-08-16, with 123 points after 1396 turns and 0:10:01. 14:38:04 !lg * end>=20050810 end<=20080517 14:38:06 138992. Grimm the Grave Robber (L1 SENe), slain by a giant gecko on D:1 on 2008-05-16, with 19 points after 32 turns and 0:00:14. 14:38:13 !lg * end>=20080510 end<=20080517 14:38:14 1673. Grimm the Grave Robber (L1 SENe), slain by a giant gecko on D:1 on 2008-05-16, with 19 points after 32 turns and 0:00:14. 14:46:33 -!- HangedMan_ has joined ##crawl-dev 14:48:22 -!- Fa has joined ##crawl-dev 14:49:11 -!- HangedMan_ is now known as HangedMan 14:50:52 -!- dtsund has quit [Ping timeout: 244 seconds] 14:51:00 -!- dtsund has joined ##crawl-dev 14:56:02 -!- HangedMan has quit [Quit: Page closed] 14:56:38 -!- HangedMan has joined ##crawl-dev 14:59:39 -!- HangedMan has quit [Client Quit] 15:00:26 -!- Jason has joined ##crawl-dev 15:00:51 -!- Jason is now known as Guest54947 15:01:02 -!- Guest54947 is now known as HangedMan 15:47:51 -!- blabber has quit [Quit: leaving] 15:58:00 -!- headzone has quit [Ping timeout: 276 seconds] 16:01:30 -!- headzone has joined ##crawl-dev 16:04:05 -!- ussdefiant_ has quit [Ping timeout: 250 seconds] 16:07:11 -!- ussdefiant_ has joined ##crawl-dev 16:10:43 -!- ussdefiant_ has quit [Read error: Connection reset by peer] 16:32:11 -!- Fa has quit [Ping timeout: 272 seconds] 16:40:37 -!- Mottie has joined ##crawl-dev 16:51:45 -!- blabber has joined ##crawl-dev 16:58:25 -!- Ragdoll has quit [Read error: Connection reset by peer] 17:00:32 -!- blabber has quit [Quit: leaving] 17:01:48 -!- Wensley has joined ##crawl-dev 17:03:47 -!- dtsund has quit [Quit: dtsund] 17:14:21 -!- dtsund has joined ##crawl-dev 17:15:59 minmay notes that the scroll of summoning is requested for in lugonu_bribe.des or whatever 17:21:43 -!- ens_ has joined ##crawl-dev 17:22:26 -!- ens has quit [Read error: Connection reset by peer] 17:35:49 -!- Wensley has left ##crawl-dev 17:46:06 -!- Pingas_ has quit [Read error: Connection reset by peer] 18:09:01 -!- LiteralRickshaw is now known as NotEvenARickshaw 18:09:47 03kilobyte * rce0c410e38de 10/crawl-ref/source/dat/des/altar/lugonu_bribe.des: Don't try to place a renamed scroll. 18:11:29 -!- evilmike has joined ##crawl-dev 18:18:53 evilmike, some vault wip feedback? 18:19:22 alright 18:19:57 http://pastebin.com/puY8kXCC 18:23:21 first one: pretty sure I've given you feedback on this already 18:24:28 "guardian serpent perm_ench:shapeshifter" does this actually make it any different from a regular shapeshifter? 18:24:38 the teleport traps are truncated! the equipment is re-balanced! 18:24:47 it creates a shapeshifter that always starts as a guardian serpent 18:24:49 yes, thats bascially what i suggested last time 18:24:57 will it be a guardian serpent when you come across the vault? 18:25:00 yes 18:27:17 hangedman_cross_veins: wow you actually made a normal vault, congrats 18:27:31 hahaha 18:27:35 also, what's with the #########|#########|#########|#########|#########|#########|#########|#########| dividers 18:27:46 self-reference with how wide a vault is 18:27:53 ah 18:28:10 good text editors will tell you how many columns something is 18:28:23 you're a windows user, right? try notepad++, that one is reasonably popular 18:28:28 -!- Patashu has joined ##crawl-dev 18:28:42 hmm, will try 18:31:10 hangedman_surveillance_ring: probably want to place an entrance somewhere 18:31:33 'k 18:32:19 ...would mini_float work? 18:32:24 yeah 18:32:40 "needle q:3 ego:sickness" you don't really need q:3 here, if you don't specify a quantity you'll get a random amount of needles 18:32:57 i'm saying this because i find it weird when a vault will always produce a stack of exactly 3 needles, or whatever 18:33:14 hmm, fair 18:33:21 q:1 is ok though 18:33:40 then that it shall be 18:35:41 the whole "four visible abyssal runes" is fine? 18:44:25 -!- PatashuXantheres has joined ##crawl-dev 18:44:25 -!- Patashu has quit [Disconnected by services] 18:45:02 -!- PatashuXantheres is now known as Patashu 18:46:07 -!- alefury has quit [] 18:46:07 -!- Fa has joined ##crawl-dev 18:50:59 mimics are a bit overplayed but there's no rule against using them 18:51:12 if i were you i'd make the chance of mimics lower 18:51:14 they probably just need to generate a lot lot less 18:51:20 so they're more of a special occurence when they appear 18:52:09 I was trying to work on the principle of "you know these are mimics, but which ones aren't", which I think works a little better then the monster equivalent of traps 18:52:32 also I agree with lowering the chance of mimics appearing in general 18:52:48 as do i 18:53:02 i just want them to potentially appear in more places :3 18:57:13 -!- edlothiol has quit [Ping timeout: 272 seconds] 19:03:29 The wall is a mimic! 19:03:34 The floor is a mimic! 19:03:37 The mimic is a mimic! 19:04:21 The dungeon is a mimic! 19:04:30 The world is a mimic! 19:05:55 (All the world's a mimic, And all the adventurers merely victims...) 19:05:57 The computer you are using is a mimic! 19:07:43 computer hits you with crawl, causing you massive damage to your free time! 19:08:38 The computer mimic addicts you. 19:14:02 -!- elliptic has joined ##crawl-dev 19:15:03 -!- ainsophyao has joined ##crawl-dev 19:18:30 -!- headzone has quit [Remote host closed the connection] 19:24:05 -!- headzone has joined ##crawl-dev 19:52:09 -!- ainsophyao has quit [Remote host closed the connection] 20:03:57 -!- MarvinPA has quit [Ping timeout: 260 seconds] 20:29:52 the cloudy nemelex altar vault having depth:* ended up placing the vault in dis:7 during wizmode testing, is this fine 20:33:23 -!- clouded_ has quit [Read error: Connection reset by peer] 20:33:41 -!- st_ has joined ##crawl-dev 20:40:27 -!- MarvinPA has joined ##crawl-dev 20:45:14 please say the name of the vault 20:46:13 nemelex_altar_shiori 20:59:41 well, it's not the only vault to have that, and it's forbidden from some branches. So it does appear to be intentional 21:10:28 -!- headz0ne has joined ##crawl-dev 21:11:51 -!- headzone has quit [Remote host closed the connection] 21:16:48 -!- ainsophyao has joined ##crawl-dev 21:27:58 -!- HangedMan has quit [Quit: leaving] 21:36:15 -!- Nomi has left ##crawl-dev 21:53:27 -!- ark____ has quit [Quit: Leaving] 22:00:41 -!- dtsund has quit [Ping timeout: 250 seconds] 22:01:30 -!- mikee_ has joined ##crawl-dev 22:05:22 -!- st_ has quit [Ping timeout: 265 seconds] 22:08:11 -!- ussdefiant_ has joined ##crawl-dev 22:16:15 -!- ussdefiant has quit [Remote host closed the connection] 22:20:53 -!- Mottie has quit [Ping timeout: 248 seconds] 22:51:05 -!- ussdefiant_ has quit [Ping timeout: 252 seconds] 22:58:40 -!- gnsh has quit [] 23:07:17 -!- dtsund has joined ##crawl-dev 23:56:09 -!- Grunt has quit [Quit: leaving]