00:00:03 I like suppression just affecting magical items 00:00:08 <|amethyst> yeah 00:00:09 it keeps the explanation nice and simple, easy to remember 00:00:19 <|amethyst> what about brands? 00:00:42 -!- Wensley_ has quit [Ping timeout: 246 seconds] 00:01:08 I remember some reason for weapon enchantment being kept. I dont remember if brands are 00:01:10 <|amethyst> and instantaneous item effects like wands etc 00:01:33 <|amethyst> brands are currently kept 00:01:36 evokable stuff is supposed to be blocked 00:01:47 <|amethyst> That is not the case yet 00:02:50 well, rings of levitation and such are supposed to be suppressed. I don't know if that extends to wands (I assume it's supposed to, and just hasn't been coded yet) 00:03:28 <|amethyst> rings of levitation generate a lasting effect, though 00:03:57 oh yeah, one thing to note about lowering MR... high level characters tend to rely on items for MR. Only a few races really have enough to allow you to get by just on racial MR 00:04:13 so, by suppressing items, its already going to take out a good chunk of MR for many characters 00:04:21 <|amethyst> I was thinking monster MR 00:04:26 ah 00:05:06 ??suppression[3] 00:05:06 suppression[3/3]: Evocable properties on equipped items have their success rate reduced. Rods and wands used within the field are still guaranteed to work, but have the power of their effects reduced. 00:05:15 I'll assume this is still the current plan 00:05:22 <|amethyst> oh, that works 00:06:02 <|amethyst> btw, was there consensus on reducing the aura radius? 00:06:14 <|amethyst> it seems far too big currently 00:06:45 Wensley wants it to be big, I think it should be a bit smaller. LOS-radius might be too small though, he pointed out that if it's too small you can just lure enemies outside of the zone 00:07:11 -!- Blade- has quit [] 00:07:20 evilmike: I think it should be at least 2xLOS 00:07:43 a radius of 16? 00:07:48 I'm not sure what 2x means here 00:07:52 sure 00:08:00 not even sure what the current radius is 00:08:03 I think it's 12 at the moment :P 00:08:14 just don't want people to be able to easily lure monsters outside of the aura without alerting the moth 00:08:22 well I'll have to increase the number then!! 00:08:49 what I want is to be able to use this monster without having it take over the entire map 00:09:05 for instance, I dont think a single moth should cover an entire branch ending 00:09:22 <|amethyst> remember that a radius of 12 is over twice the area of radius 8 00:09:42 area doesn't matter, it's all about length of travel 00:09:48 12 is huge, yeah. When he said 2x I dont know if he meant the radius, or the area 00:09:54 can't make it too easy to escape the aura on a beeline 00:09:57 I mean radius 00:10:06 area is irrelevant when it's already larger than los 00:10:37 <|amethyst> area is relevant when you're asking "what percentage of the level does it cover?" 00:10:53 "a whole damn lot, hopefull" 00:10:55 yyy 00:11:02 the way I see it, the bigger the area is, the less this enemy can really be used 00:11:10 to reiterate, this is not a monster, it is an environmental hazard 00:11:39 it is *supposed* to cover a huge area 00:11:48 <|amethyst> but if it has a radius of 12 you can't use it in a vault smaller than 24x24 without the effect leaking out 00:11:57 is that a bad thing? 00:11:58 Wensley: do you think it should spawn with a band of monsters? 00:12:01 <|amethyst> and 24x24 is huge for a vault 00:12:03 evilmike: perhaps 00:12:11 leaking is what this monster is all about 00:12:28 <|amethyst> that would make the vault obvious though 00:12:30 behavior of the aura propagation and size is identical to silent spectres 00:12:33 keeps things simple 00:12:36 <|amethyst> hm 00:12:46 |amethyst: not sure that vaults are ever supposed to be a surprise, that seems spoilery 00:13:08 knowledge that a certain vault exists on a level should not change your behavior 00:13:22 leaking outside tells you what is in the vault, though 00:13:34 and if the existence of the moth has effects outside of it, well to my eyes that's interesting 00:13:36 <|amethyst> I guess silent spectres are as good a comparison point as any, so I suppose the current behaviour is not entirely unreasonable 00:14:13 evilmike suggested making this a general effect and giving it to other monsters as a spell. we could also have a juvenile version just for vaults with a smaller aura, if you want 00:14:19 it's just a single number to change 00:14:33 <|amethyst> hm 00:14:39 I've said before, I think it would make a good lich spell. 00:14:48 <|amethyst> what if you make the aura scale based on something? HD? 00:15:03 Could replace banishment in one of their spellbooks, or something like that 00:15:03 <|amethyst> then it can be tuned by individual vaults 00:15:11 I am really pretty adamant that this monster must have a large aura to fulfill its intended purpose. it's up to the devs to decide whether that purpose is legitimate 00:15:27 if it's to scale by some stat, HD makes sense 00:15:33 @??moth of suppression 00:15:33 moth of suppression (09y) | 04UNFINISHED | Speed: 12 | HD: 9 | Health: 33-64 | AC/EV: 0/10 | Damage: 25 | Flags: fly | Res: 06magic(36) | Chunks: 07contaminated | XP: 504. 00:16:07 <|amethyst> or, for something unique, HP 00:16:30 hmm, that could leak information regarding how much hp the enemy has 00:16:37 unless we give it fixed hp, like uniques 00:17:04 (unless you mean this as a percentage of max hp) 00:17:13 <|amethyst> hm, good point; HD times percentage of max HP would be kind of what I was thinking 00:20:27 <|amethyst> but, ignoring that (an aura that scales with HP is an idea for another creature), what about radius2 = 16 * HD? 00:20:39 -!- nicolae-- has left ##crawl-dev 00:21:12 if HD is something that's defined in lua I'm not sure how to express that in the monster definitions 00:21:35 <|amethyst> that gives 144 (compared to 150; still radius 12, but pointier) for the current stats; LOS for a 4-HD creature; and 2*LOS for a 16-HD creature 00:21:48 moth of suppression (09y) | 04UNFINISHED | Speed: 12 | HD: 16 | Health: 67-108 | AC/EV: 0/10 | Damage: 25 | Flags: fly | Res: 06magic(64) | Chunks: 07contaminated | XP: 1714. 00:21:48 <|amethyst> %?? moth of suppression hd:16 00:21:57 Wensley: in mon-data.h there's a place where you will see 4 numbers. these determine HD 00:22:00 HP* 00:22:25 the first number is HD, the next two are HP stuff, and I forget what the fourth does 00:22:36 are we defining a general guideline for aura sizes here or what 00:22:50 -!- mikee_ has quit [Quit: HydraIRC -> http://www.hydrairc.com <- Wibbly Wobbly IRC] 00:22:59 not sure what HD has to do with it if we're not talking about varying the size of the aura according the hd of each moth in each vault 00:23:40 <|amethyst> I don't know about generalizing it, since the radius also depends on the aura type 00:23:50 <|amethyst> but I was thinking for suppression from a monster 00:24:17 <|amethyst> then, as you say, vault makers could set the HD with hd:N to get a different aura size 00:24:51 <|amethyst> though of course that has other effects that might not be desired 00:25:12 <|amethyst> I was just trying to think of a way to allow it to be tuned on a vault-by-vault basis 00:26:55 -!- nrook has quit [Remote host closed the connection] 00:27:01 okay well I am going to bed now, take the next day or so to think about this moth and come to the conclusion that you are wrong to think that the radius of this moth's aura should not asymptotically approach infinity :) 00:27:19 <|amethyst> Wensley: could your goals be achieved by having more moths with smaller (still > LOS) radius? 00:27:32 <|amethyst> :) 00:27:33 I said a few days ago, I think it will be easier to figure out a good radius after actual playtesting 00:27:38 <|amethyst> night 00:28:01 -!- ZChris13 has quit [Ping timeout: 265 seconds] 00:28:08 |amethyst: possibly, that would satisfy the goal of discouraging kiting, but another goal was to fuck with people on the other side of walls who have no way to do anything about the aura but man up and charge through 00:28:11 o/ 00:28:16 -!- Wensley has quit [Quit: Leaving] 00:32:43 abyssal knight trapped on d:1 (https://crawl.develz.org/mantis/view.php?id=5537) by rchandra 00:34:51 <|amethyst> evilmike: do you recognise the vault from that bug report? 00:35:22 |amethyst: Nope... it doesn't look like a vault at all, to me 00:35:29 but it could be one 00:36:04 would be helpful to see the .map file 00:37:47 <|amethyst> Requested feedback 00:37:52 <|amethyst> oh 00:37:54 <|amethyst> so did you 00:37:59 heh 00:38:38 <|amethyst> well, I changed the status too, so it was not for naught 00:38:40 <|amethyst> :) 00:41:40 Debian builds of master branch on crawl.develz.org updated to: 0.11-a0-959-g8726066 00:47:15 <|amethyst> evilmike: that's a tiny .map file 00:48:10 well, I dont recognise that layout at all. Highly doubt its a vault 00:48:57 <|amethyst> yeah 00:49:19 <|amethyst> wouldn't the .map include other bubbles if the level were disconnected? 00:49:37 I think it just shows areas you've explored, on the level you died on 00:49:43 <|amethyst> hm 00:50:16 this could be some disconnected zone on D:1, although I figured the game would check for that. Maybe there was a secret door 00:50:24 <|amethyst> so if there were a secret door he didn't find, it wouldn't appear in the .map? 00:50:34 <|amethyst> yeah, a disconnected level should be vetoed 00:51:35 |amethyst: it won't show an undetected secret door 00:51:40 it's basically just what you get on the X screen 00:52:26 it might be a nice feature if dead characters got the whole map, though 00:52:46 (same with how dead characters can get a list of vaults, although not by default) 00:52:53 <|amethyst> hm... I can't think of a way to get any more (useful) information without a save 00:53:17 <|amethyst> and the character's dead so I doubt the reporter has a save 01:02:20 03|amethyst * r557b6d736ba4 10/crawl-ref/source/makeitem.cc: Partially re-restrict Archmagi brand. 01:46:20 03evilmike * rf9501ecb9c6a 10/crawl-ref/source/dat/des/branches/vaults.des: Actually place the teleporter in the_teleporter. 02:26:24 -!- lord-naughty has joined ##crawl-dev 02:26:41 -!- lord-naughty has quit [Client Quit] 02:42:18 -!- dtsund has quit [Read error: Operation timed out] 03:01:58 -!- ToBeFree has joined ##crawl-dev 03:16:28 -!- lord-naughty has joined ##crawl-dev 03:20:15 03kilobyte * r3a8e79067add 10/crawl-ref/source/main.cc: Fix a lua stack leak. 03:20:25 03kilobyte * r0f6962bc2da8 10/crawl-ref/source/ouch.cc: When you die in wizmode to rotting, remove all rot. 03:20:25 03kilobyte * rdfe5f187e5bd 10/crawl-ref/source/main.cc: Assert if the Lua stack is non-empty. 03:33:27 -!- ToBeFree has quit [Read error: Connection reset by peer] 03:39:30 -!- jeanjacques has joined ##crawl-dev 04:19:41 -!- edlothiol has joined ##crawl-dev 04:42:11 -!- HousePet has joined ##crawl-dev 04:43:55 -!- FaMott has quit [Ping timeout: 252 seconds] 04:52:20 -!- ToBeFree has joined ##crawl-dev 05:00:07 OSX builds of master branch on crawl.develz.org updated to: 0.11-a0-964-gdfe5f18 05:07:29 -!- monqy has quit [Quit: hello] 05:08:43 -!- MarvinPA has joined ##crawl-dev 05:23:23 -!- vadatajs has quit [Ping timeout: 245 seconds] 05:32:20 -!- MarvinPA_ has joined ##crawl-dev 05:35:38 -!- MarvinPA has quit [Ping timeout: 240 seconds] 05:35:38 -!- MarvinPA_ is now known as MarvinPA 05:36:43 -!- st_ has quit [Ping timeout: 264 seconds] 05:37:54 -!- vadatajs has joined ##crawl-dev 05:47:40 -!- Pingas has joined ##crawl-dev 06:03:55 is that just me, or does Nemelex not grant decks of summoning anymore? 06:05:32 there was a change to corpse weights that makes decks of summoning less common 06:06:01 I mean, he just counts all corpses the same, now 06:07:37 -!- syllogism has joined ##crawl-dev 06:08:00 hrm, in 0.9 a yak corpse was worth 1900, in 0.10+ 50 06:08:20 you sure that made it into 0.10? 06:09:48 yeah, just checked 06:13:17 with the removal of mountain dwarves to mine for gold the dungeon economy has been experiencing deflation 06:14:56 bugger 06:21:44 -!- alefury has joined ##crawl-dev 06:27:15 idea: let's change 50 to max chunk count * 50 06:28:21 for a bunch of popular monsters, ~4 times smaller than 0.9 06:28:36 orc: 600 -> 4*50 = 200 06:28:48 yak: 1900 -> 8*50 = 400 06:29:33 could be reasonable, yeah 06:30:33 should I change just Nemelex, or Jiyva as well? 06:31:06 Jiyva seems to be more about destroying permanents, though, so I slightly prefer Nemelex only 06:31:13 but no opinion, really 06:31:46 yeah, could go either way probably 06:32:09 coinflip 06:32:40 jiyva's probably fine as-is, i guess 06:33:34 oh also, just remembered 06:34:02 HousePet: you're not in CREDITS.txt yet, right? 06:34:24 if not, what name should i add you under for the xom patch? 06:35:35 -!- petete has joined ##crawl-dev 06:35:40 ummmm no idea 06:36:01 i had no idea it existed 06:36:19 despite being blatently obvious... 06:37:04 kilobyte: why should nemelex and jiyva value some corpses more than others when oka, trog, etc do not? 06:37:21 I think best for all corpses to be the same 06:37:35 if you want to change the nemelex corpse value up a bit, that's fine 06:37:52 too much corpse sacrificing 06:38:22 also I'm pretty sure that change wasn't in 0.10... it definitely wasn't supposed to be 06:38:23 Nemelex values corpses as items, Oka or Trog as a token of a victory 06:38:35 what is nemelex doing with the corpses then 06:38:46 that makes him like big hunks of meat more than small ones 06:38:50 by the way, this reminds me that I was supposed to tie Okawaru's piety gains to the quality of kills/corpses 06:38:52 kleptomaniac god 06:39:17 or was that stealing? 06:40:08 elliptic: ah right, in 0.10 the corpse commit was already there, but it was then overwritten by corpse mass 06:42:43 kilobyte: Only for Oka? What about Trog and Makhleb? 06:42:57 I had the Okawaru formula mostly right (ok, for obviously thoroughly untested values of "right"), except for XL around 2-7 06:44:09 ghallberg: I don't remember the exact discussion, but it was agreed upon (by dpeg and some other folks) that they should be differentiated -- Oka wants hardest victories, Trog and Makhleb value bloodshed 06:44:35 you already get more piety for better kills for all gods 06:45:01 it's just that Oka would be better if you got almost nothing for popcorn 06:45:13 also, corpses. They currently are worth all the same. 06:45:39 kilobyte: I see. 06:45:47 Thanks. 06:46:21 I agree with bloodshed for Makhleb, but I could see the same argument about Trog. Also gameplay. 06:46:52 Even if we don't have Hive-vault anymore killing 15 bees and sacrificing their corpses is almost a guaranteed gift. 06:46:53 -!- gnu-dio has joined ##crawl-dev 06:47:02 (Or was last time I played Trogolodyte) 06:48:42 03MarvinPA * r5244e3efe8c3 10/crawl-ref/CREDITS.txt: Add HousePet to CREDITS.txt 06:49:39 So. I've been wondering about this for a couple years now, and never bothered to ask: Why did crawl go to openGL? I still play mostly .4 because after that it is a total pain to compile on a light distro, and near-impossible on a machine without a modern ati/nvidia/intel 3d chipset. It seems useless as well, since there's nothing 3d about crawl... 06:56:36 gnu-dio: there were some performance issues, but indeed, this seems to be a strange decision, and it indeed breaks a good portion of systems 06:58:19 for Oka, I could bias the formula at early XLs, but really, the core problem lies in threat tiers being wrong 06:58:21 I spent about a month of all my free time trying to hack the sdl code back into .5, but found it to be beyond my meager skill. 06:58:25 take a look here: http://angband.pl/tmp/monster-tiers-by-xl.png 06:58:53 gnu-dio: I wonder, perhaps plain GTK or QT could do better? 06:59:14 GTK would be great for my personal situation ;) 06:59:37 kilobyte: perhaps increase the value of low amounts of damage 06:59:52 Just seems pointless to require all that 3d code-base for a 2d-tiles game... 07:00:06 especially for weapon-wielders 07:00:45 that'd boost goblins, hobgoblins, orcs, etc. 07:00:55 gnu-dio: yeah, I can't run GL tiles at all on 3 out of 5 platforms I have 07:02:00 And speaking of performance issues: openGL crawl in MESA on a Celeron (p3 class) with an intel 845g chipset..... 07:02:23 unplayable. but the sdl was just fine. 07:03:28 perhaps Keskitalo could enlighten us a bit, the SDL -> GL change was before my time 07:04:28 * kilobyte waves a dead kenku and chants the summoning ritual. 07:04:34 alternately: maybe we could add harder monsters in the first few dlvls :) 07:05:09 riiight... 07:05:42 Well, I'm going to assume there is something useful being provided by GL, (though I can't imagine it at this time), I wonder if one of the real programmers would consider figuring out a way to hack the old sdl code(or new gtk/qt/etc) back into the source as a compile time option.... 07:06:20 03kilobyte * r5a83ae6c9ea1 10/crawl-ref/source/test/stress/abyss.rc: Fix a recommended command in a comment. 07:06:20 03kilobyte * ra5c840eb7ce2 10/crawl-ref/source/godprayer.cc: Fix Nemelex giving nearly no decks of summonings. 07:07:22 it might be simpler to hack a local webtiles build instead :p 07:08:56 it has hefty dependencies (python, webtornado), and python is ridiculously non-modular (intentionally, too -- its authors say they want "batteries included" which makes embedding infeasible) 07:10:04 yet since you can hack up a webserver in a single page of bash, it would be doable 07:10:42 gnu-dio: if you want, you can try current trunk 07:10:43 That is an interesting idea. 07:10:56 the "local" webtiles. 07:11:16 I'll give it a try tomarrow. (I'm almost out of time today) 07:11:26 a bit tricky if you have to install things yourself -- but trivial on Debian unstable/testing 07:12:29 I'm going to be doing it on Puppy Linux (a VERY light distro.) As long as I don't have to do GL, I'm not afraid of python/webtornado/etc. (already have python installed and running well) 07:12:32 apt-get install python-tornado; make WEBTILES=y; webserver/server.py; $BROWSER http://localhost:8080 07:12:41 kilobyte: so why don't yak corpses give more piety for the blood gods? they presumably have more blood, after all... 07:12:57 you'd need a very recent version of tornado, though 07:13:03 they like variety? 07:13:12 and nemelex doesn't? 07:13:18 I have the current stable python installed atm 07:13:27 I just wondered about local webtiles recently too, though I cant remember why. Anyway, I can't remember the problems with SDL and reasons going to GL. 07:13:36 Should be no prob getting the current stable webtornado. 07:14:03 elliptic: inflicting suffering upon a cow doesn't seem more fun to me than doing that to a human 07:14:16 kilobyte: how is sacrificing a corpse inflicting suffering on anyone 07:14:33 elliptic: token of victory and such 07:14:49 kilobyte: does nemelex appreciate bigger corpses more because he is cooking and eating them? 07:15:38 Keskitalo: my box at work (some lousy radeon): hard crash. Same box, virtualbox + Windows: hard crash with GL passthrough, nothing without. 07:15:50 I just don't understand the reason why nemelex would like bigger corpses better 07:15:56 Keskitalo: my box at home: works ok; same box + virtualbox + Windows: all ok. 07:16:40 Keskitalo: n900 natively: doesn't even compile (GLES rather than GL). n900 + Debian chroot: works with visual corruption, ridiculously slow 07:17:37 the rationale for a god like Makhleb preferring bigger corpses would be clear (they have more blood, blood and souls for makhleb)... but what's the explanation for nemelex? 07:17:39 elliptic: more stuff to sacrifice; matters if you care about material stuff rather than kills 07:18:44 nem should just give out random decks and not accept corpses :) 07:18:54 really though, I think giving for the corpse the same piety (or a function of it) as for the kill would be better than what we have currently 07:19:19 I can sort of buy this, but I'd be a lot happier if things were consistent... at the very least, jiyva should clearly care about corpse size if you are using that explanation for nemelex 07:19:51 kilobyte: yes, basing piety on HD of corpse would be reasonable 07:20:31 HousePet: sounds like a good idea; you have little other stuff in the Lair though so they'd need to have similar value to orcs' gear, etc 07:20:38 it bothers me how a trogite (or okawaruite, or ...) can kill 20 green rats and get much more piety from the corpses than from the kills 07:20:55 i think that might be kinda good for nemelex too, yeah 07:21:05 the weighting stuff is all pretty complicated 07:21:49 if we remove weighting stuff and just give out decks randomly, then there's the question of whether item sacrificing is really the best piety gain mechanism 07:21:49 elliptic: per the discussion with dpeg, I think differentiating Oka from other blood gods may be good. Making Trog give _some_ heed to stronger corpses is indeed good, but let's make Oka care nearly exclusively about quality. 07:22:28 kilobyte: I'm pretty dubious that players will be able to tell the difference 07:22:36 they just go and sacrifice everything they don't want to eat 07:22:43 yep 07:22:48 the #ifdefed out Oka code cares exactly as much for corpses as for kills, Trog could do the same -- just with a different formula what piety that is 07:23:15 elliptic: with the green rats issue, probably yeah 07:23:21 the primary offender, Hive, is gone 07:23:45 kilobyte: something like that sounds fine, but it won't do much to differentiate them I think 07:24:05 too much kills and corpses for piety 07:25:06 -!- evilmike has quit [Read error: Connection reset by peer] 07:25:53 vehumet should accept armour sacrifice :) 07:26:47 what about changing nemelex piety to piety for exploration? 07:27:31 (and also piety for using decks still) 07:28:00 piety for using unided stuff 07:28:07 that is a big of a gamble 07:30:51 nemelex isnt much of a gambler though, hes more about tarot 07:31:12 at least thats what i think 07:32:02 hmmm 07:33:36 oh, by the way, the nemelex description claims that cards in decks are "painted in the ichor of demons"... clearly this means nemelex should give piety for killing demons :P 07:34:41 lol 07:36:11 -!- Ragdoll has joined ##crawl-dev 07:38:00 by the way, my Okawaru formula is, after simplifying out all unnecessary squaring, square roots, etc, etc, turns out to be simply piety of XP (divided by something depending on nothing but your XL) 07:38:14 s/of/for/ 07:38:49 i think i would divide by piety 07:38:50 threat tier is ~ square root of monster's XP 07:39:14 there's a further slowdown in gain_piety(), but that's common to nearly all gods 07:39:27 ok 07:44:52 -!- gnu-dio has quit [Quit: Page closed] 07:44:54 -!- ToBeFree has quit [Read error: Connection reset by peer] 07:45:05 -!- Pacra has quit [Quit: This computer has gone to sleep] 07:50:00 -!- Xiberia has joined ##crawl-dev 07:52:45 -!- Pacra has joined ##crawl-dev 07:55:26 hrm, monster generation is indeed bad. For example, let's look at typical D:3 07:55:29 Monsters: 2 kobolds, adder, bat, hobgoblin, 2 jackals, 4 giant geckos, 5 rats, and 3 giant cockroaches. 07:55:32 19 monsters, 42 total exp value (42 non-uniq) 07:57:07 and no wonders -- no monsters are native for D:1; D:2 has goblin, giant newt; D:3 giant cockroach, ooze and giant python, ... 07:58:07 first real monsters start at D:6 (gnoll shaman) and D:7 (orc priest); their rarity allows them to spawn earlier 07:58:07 is that really a typical D:3? 07:58:21 Monsters: kobold, 2 bats, 2 goblins, giant newt, giant gecko, 2 rats, quokka, and 2 giant cockroaches. 07:58:24 12 monsters, 20 total exp value (20 non-uniq) 07:58:41 ugh 07:58:42 Monsters: 6 kobolds, 3 adders, 2 bats, 3 goblins, 3 hobgoblins, 4 jackals, 3 giant newts, and 3 rats. 07:58:45 27 monsters, 75 total exp value (75 non-uniq) 07:58:52 Monsters: ooze, 3 kobolds, 4 bats, goblin, giant newt, giant gecko, 5 rats, 2 giant cockroaches, and giant mite. 07:58:55 19 monsters, 39 total exp value (39 non-uniq) 07:59:02 (keep pressing &^R &") 07:59:33 hm, that will never regenerate uniques, though 07:59:39 right? 08:00:12 (also, other unique vaults) 08:00:36 generate yes, regenerate not sure 08:01:11 vaults are returned to the pool, subvaults are not 08:01:45 i dont remember D:3 always being so boring 08:01:53 but looks like it very often is 08:02:16 heh, and since D:3 has a Kiku overflow in this game, I keep getting a human a good part of the time 08:02:53 and whoops, D:3 gnoll castle: 08:02:54 Monsters: 2 kobolds, ogre, 9 plants, ball python, adder, bat, 2 goblins, 3 hobgoblins, 44 gnolls, 14 jackals, 3 giant newts, 08:02:58 33 rats, giant cockroach, and 2 worms. 08:03:00 117 monsters, 853 total exp value (853 non-uniq) 08:04:45 now thats more like it! 08:06:24 usual exp spikes go to early 90s 08:06:38 most levels have 20ish 08:07:32 a single ogre is enough to bring it well into 150ish 08:07:34 @??ogre 08:07:34 ogre (07O) | Speed: 10 | HD: 5 | Health: 16-40 | AC/EV: 1/6 | Damage: 17 | Res: 06magic(20) | Chunks: 07contaminated | XP: 119. 08:08:22 also, why centaurs give less XP than ogres? 08:08:32 @??centaur 08:08:33 centaur (07c) | Speed: 15 | HD: 4 | Health: 12-32 | AC/EV: 3/7 | Damage: 10 | Res: 06magic(16) | XP: 112. 08:08:55 arena results: 6:93 08:18:15 well, the arena is an open level 08:18:40 if you encounter an ogre when turning a corner its more dangerous than a centaur 08:19:28 9 monsters, 14 total exp value (14 non-uniq) -- great D:3 08:19:48 usually it's at least a corridor 08:20:51 well, roguelikes are about being random, and crawl is a very long roguelike, so huge exp fluctuations are fine. some games will be harder than others, but they all have some hard and some easy bits. 08:21:40 i dont think somehow clamping the exp is necessary, but it probably wouldnt hurt either 08:22:13 kilobyte: place-poulation.lua is very handy to get monster type ratio and average XP value of a level 08:22:55 -!- HousePet has quit [Quit: Divide by cucumber error] 08:23:41 or a range of levels 08:23:45 or a whole branch 08:24:08 does it also give min max std? or just the average exp and distributions? 08:24:09 it has options for uniques and vaults, but I haven't tested them 08:26:45 Count | XPAv | XPMin | XPMax | XPSigma | CountMin | CountMax | CountSigma | % | Monster 08:27:59 crawl -script place-population D:3 08:28:03 http://pastebin.com/3Wpq5BgL 08:31:37 whats count? because the numbers dont make any sense 08:32:14 assuming count is actually how many monsters of that type there were on average (or min, max, std) 08:35:21 how doesn't it make sense? Looks fine to me. 08:36:19 minCount > Count, for example 08:36:40 which is impossible if count is the average monster count and mincount the minimum monster count 08:37:47 i guess minCount could be the minimum nonzero count 08:38:21 s/minCount/CountMin 08:40:54 yeah, looks like minCount returns 1 instead of 0 08:41:35 otherwise it would probably be almost always 0 08:41:53 this way it can be used to recognize band monsters and vault-only monsters 08:48:50 there is an information leak with autotravel and steam 08:49:04 and smoke too 08:51:01 ah, it happens in shoals too with items and tides 08:52:58 -!- syllogism has quit [] 08:53:36 !tell nfogravity what's the point in disabling the asserts in skills.cc? 08:53:37 galehar: OK, I'll let nfogravity know. 08:53:40 should minotaurs be able to retaliate while confused? 08:53:40 the clouds one is already reported 08:54:09 i would say no but maybe it's so instinctive that they can do it anyway, or something 08:56:55 -!- syllogism has joined ##crawl-dev 08:57:09 attacking still works, if you randomly try to move in the right direction. and attacking with a weapon is a pretty complex movement that requires a lot of skill. 08:58:11 not sure if/how those attacks are penalized, but even with penalties the assumption that retaliation would still work is reasonable (only about just as reasonable as assuming it wont work, though, which is probably the reason why you are asking) 08:58:51 flip a coin? :P 08:59:41 the last time we flipped a coin, elliptic started demanding Jiyva gets the same corpse treatment, so no way :p 09:00:22 heh. then fall back to the first commandment of crawl development: when in doubt, nerf! 09:00:24 (no opinion about minotaurs here) 09:00:28 03kilobyte * ref90fb232866 10/crawl-ref/source/mon-stuff.cc: Re-enable the monster tier counters spam. 09:00:39 03kilobyte * re9e3817707dc 10/crawl-ref/source/godprayer.cc: Give the Nemelex formula for corpses to Jiyva as well. 09:00:39 03kilobyte * ra4f20a1fb183 10/crawl-ref/source/wiz-dgn.cc: When recreating a level with &^R, make uniques re-eligible for spawning. 09:00:39 03kilobyte * r3510bdbd85da 10/crawl-ref/source/mon-pick.cc: Don't attempt to generate killer bee larvas in D/Vaults (it'd fail anyway). 09:00:45 I demand that Jiyva gets the same retaliation treatment 09:00:48 also it looks like monsters get a to-hit penalty while confused, but players don't 09:01:45 MarvinPA: I'd say no retaliation while confused 09:01:47 ever been around a drunken guy who starts raging? You don't know where he'll hit, but that makes it that much dangerous (a sane person is predictable). 09:02:56 and this to-hit code makes me sad 09:03:35 if (attacker->is_player()) { loads and loads of stuff! } 09:03:37 :P 09:03:40 just write a comment "here be dragons" 09:03:45 then nuke it from the orbit 09:03:46 heh 09:04:11 it'd upset balance, but hey 09:05:03 its a long time to the next release 09:12:14 have you noticed that we have about no monsters with xp around 20? 09:13:30 early levels get a big pile of xp 1-3 stuff, with 100ish xp monsters and orc priest bands becoming more and more frequent as you go down 09:14:46 // If you can't see yourself, you're a little less accurate. Maybe monsters should get this penalty too? (would nerf orc wizards a bit) 09:15:14 worth unifying for both players and monsters while i'm doing melee_attack.cc stuff? or is that even a penalty worth keeping? 09:16:13 dunno if it would affect much other than orc wizards on the monster side 09:16:22 maybe a few uniques that cast invis 09:16:46 sigmund! 09:17:02 oh no :( 09:17:16 would make the difference between a visible orc wizard and an invisible one a bit smaller, that's good 09:18:42 MarvinPA: remove that penalty imo 09:19:46 it's spoilery that invis makes you worse at attacking unless you have see invis 09:20:46 it makes logical sense 09:20:50 kilobyte: there was the idea to increase normal orcs HD a bit. Maybe it would fill the XP gap (and make Orc worth more than just loot) 09:21:07 if you can't see your sword / arrow then it's hard to judge where it is 09:21:12 yeah, but still spoilery. if its kept there should probably messages like for armor (your plate armor prevents you from hitting the foo) 09:22:17 yay, harder orc. maybe finally adjust depth to usually be below lair? 09:23:07 galehar: I recently gave them 3 xp rather than 2 09:23:29 galehar: I think the idea was to make orc HD be higher in the mines, especially in the lower levels 09:23:30 alefury: early orc is good if you're looking for a weapon or armour 09:23:31 could go for 20ish 09:24:04 elliptic: do we even need small-HD orcs? There's too many goblins, hobgoblins, kobolds and other chaff. 09:24:09 also, amulet of inaccuracy is only -5 acc, huh 09:24:40 kilobyte: this seems pretty cool: http://nowoczesnapolska.org.pl/2012/04/03/free-textbooks-are-part-of-digital-school-programme/ 09:24:50 kilobyte: increasing orc HD across the board would make early orc bands a lot more dangerous, and they really don't need that 09:25:15 unless you are also moving them deeper... but they are a pretty good early threat 09:25:45 could replace them with gnoll bands (+leader) 09:26:23 gnoll bands don't have anything comparable to wizards/priests currently 09:26:45 elliptic: there's one issue: in a band, you never care about basic orcs unless one got a branded weapon; you walk around them to get to priests/wizards 09:27:00 kilobyte: depends on the character and how early they are 09:27:11 on the other hand, they indeed have bands which make them more interesting than the interchangeable chaff 09:27:19 I often care about basic orcs even without brands... they hit hard for the depth with a flail or whatever 09:28:11 03MarvinPA * r165db27b5e54 10/crawl-ref/source/melee_attack.cc: Don't constrict things you can't see, don't constrict while confused 09:28:11 03MarvinPA * r8ee7becbe696 10/crawl-ref/source/melee_attack.cc: Don't let minotaurs retaliate while confused 09:28:12 03MarvinPA * rc173530e01f4 10/crawl-ref/source/melee_attack.cc: Unify a couple of to-hit penalties 09:28:18 walking around orcs to get to priests/wizards is a pretty terrible idea at the depths I'm thinking of... instead, you run away and try to pick off the basic orcs first 09:28:36 and catch the priest/wizard around a corner 09:29:52 new orc --> orc, orc --> piglet 09:31:00 but we probably dont want even more orc types? 09:32:57 -!- edlothiol has quit [Ping timeout: 252 seconds] 09:40:27 elliptic: a flail makes the orc hit for 14, without it's irrelevant 5 09:42:59 -!- absolutego has joined ##crawl-dev 09:56:15 03MarvinPA * rbec61c68e3af 10/crawl-ref/source/show.cc: Show the location of invisible monsters you're constricting 10:30:31 -!- edlothiol has joined ##crawl-dev 10:36:07 Unstable branch on crawl.akrasiac.org updated to: 0.11-a0-975-gbec61c6 (32) 10:38:33 kilobyte: yeah... maybe make orcs generate with weapons more often? 10:39:05 -!- edlothiol has quit [Ping timeout: 260 seconds] 10:39:29 I guess currently they actually get weapons 80% of the time? 10:39:37 could make that 100% 10:40:13 -!- alefury has quit [Quit: ChatZilla 0.9.85 [Iceape 2.0.11/20111108220737]] 10:40:14 also I find it strange that case MONS_ORC falls through to case MONS_DRACONIAN... 10:42:00 -!- edlothiol has joined ##crawl-dev 10:46:22 03MarvinPA * r79e0d0b45446 10/crawl-ref/source/spl-damage.cc: Don't heal players when casting vampiric draining under DDoor 10:46:32 03MarvinPA * rd7628e0bfe42 10/crawl-ref/source/mon-cast.cc: Fix monsters not being healed when casting vampiric draining on players 10:46:33 03MarvinPA * r77f76ec28bdb 10/crawl-ref/source/melee_attack.cc: Don't let monsters heal from vampiric weapons while under DDoor 10:49:06 -!- edlothiol has quit [Ping timeout: 265 seconds] 10:49:38 that last one was sort of wrong, they didn't heal anyway but it stops the related messages from showing up at least 10:51:03 (for the incredibly important case when nergalle picks up a vampiric weapon and casts deaths door and hits you with it and it would've looked like she's getting healed but actually isn't) 10:55:58 an important fix for sure 10:56:36 can a player illusion cast ddoor? 11:02:04 -!- edlothiol has joined ##crawl-dev 11:04:04 -!- alefury has joined ##crawl-dev 11:05:32 -!- blabber has joined ##crawl-dev 11:06:49 -!- absolutego has quit [Quit: leaving] 11:12:02 -!- elliptic has quit [Quit: Leaving] 11:14:15 -!- elliptic has joined ##crawl-dev 11:22:24 -!- syllogism- has joined ##crawl-dev 11:23:04 -!- syllogism has quit [Ping timeout: 245 seconds] 11:24:24 -!- FaMott has joined ##crawl-dev 12:16:24 -!- absolutego has joined ##crawl-dev 12:21:58 -!- ToBeFree has joined ##crawl-dev 12:23:11 -!- qqryq has quit [Quit: Page closed] 12:24:13 -!- qqryq has joined ##crawl-dev 12:34:32 -!- blabber has quit [Quit: leaving] 12:37:57 -!- ToBeFree is now known as SphenisciTuxBot 12:38:58 -!- ZChris13 has joined ##crawl-dev 12:42:01 -!- Blade- has joined ##crawl-dev 12:43:00 It seems like autopickup should go back on once you leave a level with an invisible monster on it 12:43:31 because otherwise the player almost always will have to manually reactivate autopickup and it's an unnecessary hassle 12:46:15 -!- Pacra has quit [Quit: This computer has gone to sleep] 12:46:44 you don't always know if it might have followed you up the stairs ... 12:47:23 yes, but it is unlikely that you will autoexplore in that case 12:49:18 it would also probably be switched off again right away because you got attacked. depending on what direction you move (3/8 chance to move away from the monster) 12:49:24 -!- SphenisciTuxBot is now known as ToBeFree 12:50:11 that too. 13:16:31 -!- monqy has joined ##crawl-dev 13:28:15 -!- FaMott has quit [Read error: Connection reset by peer] 13:29:29 -!- casmith789 has joined ##crawl-dev 13:42:19 -!- absolutego has quit [Quit: Lost terminal] 13:45:34 elliptic: hmm, i just went to try and test that and it crashes whenever an illusion gets created 13:45:58 in dbg-scan.cc, so i guess it's some weird debug mode thing 13:46:51 not that i have any clue what dbg-scan.cc does 13:54:04 -!- casmith789 has quit [Ping timeout: 272 seconds] 14:06:57 -!- dtsund has joined ##crawl-dev 14:10:42 -!- st_ has joined ##crawl-dev 14:13:48 -!- ZChris13 has quit [Ping timeout: 245 seconds] 14:19:07 -!- jeanjacques has quit [] 14:32:22 -!- ZChris13 has joined ##crawl-dev 14:38:10 -!- Pingas has quit [Read error: Connection reset by peer] 14:38:24 -!- Pingas has joined ##crawl-dev 14:41:50 -!- ZChris13 has quit [Quit: ChatZilla 0.9.88.1 [Firefox 11.0/20120312181643]] 14:53:08 -!- Blade- has quit [Quit: Page closed] 14:55:01 -!- ais523 has joined ##crawl-dev 15:13:23 how much mp is enough 15:17:01 -!- heteroy has joined ##crawl-dev 15:36:40 !tell kilobyte was trying to figure out why detailed monster descriptions don't show up in debug mode anymore and it seems like b42c4f46f08a9cd9 is the cause? reverting it made them appear again and i couldn't repro any crash at least, quite possibly i'm missing something though 15:37:22 MarvinPA: OK, I'll let kilobyte know. 15:39:04 -!- Ragdoll has quit [Read error: Connection reset by peer] 15:45:33 -!- lord-naughty has left ##crawl-dev 16:03:35 MarvinPA: edlothiol quite recently changed mon_info to store absolute coordinates instead of player-relative ones (it's an info leak, but it was too unmanageable), and it looks like this place was indeed missed 16:03:35 kilobyte: You have 2 messages. Use !messages to read them. 16:04:00 except for && -> || which seems to be correct 16:07:55 aah right 16:08:13 !tell minqmay octopodes walk on land, so "The dead are walking" is correct 16:08:13 kilobyte: OK, I'll let minqmay know. 16:09:17 03kilobyte * r274fa8d69408 10/crawl-ref/source/ (viewchar.cc viewchar.h): Use ucs_t rather than "unsigned" in a few places. 16:09:17 03kilobyte * rd34caa7dfc99 10/crawl-ref/source/ (makeitem.cc player.cc): Don't unnecessarily check for SPARM_DARKNESS on non-cloaks. 16:09:17 03kilobyte * r26f1bf4fa12f 10/crawl-ref/source/ (decks.cc decks.h): Remove an unused deck property (from Mark Four times). 16:09:18 03kilobyte * r457f92c1a75f 10/crawl-ref/source/dat/database/pl/godspeak.txt: A typo in .pl. 16:09:18 03kilobyte * r8ae89f4b34d9 10/crawl-ref/source/ (21 files): Make another bunch of functions static or gone. 16:09:18 03kilobyte * r0d2ca58aa737 10/crawl-ref/source/ (goditem.cc goditem.h): Cut some massive code duplication. 16:09:19 03kilobyte * rdf412964a868 10/crawl-ref/source/describe.cc: Fix monster info (except for desc) not showing up in wizmode. 16:12:07 sorry about that, I guess I missed it because I renamed pos to find all its occurences, but didn't compile with DEBUG_DIAGNOSTICS 16:13:21 <|amethyst> kilobyte: yay @ 0d2ca58 16:14:17 03MarvinPA * rd1d064f94b6a 10/crawl-ref/source/describe.cc: Make monster spells take up less space in debug mode descriptions 16:16:47 edlothiol: I keep doing the very same mistake the other way :p 16:17:25 -!- lord-naughty has joined ##crawl-dev 16:26:17 erdraug (L2 TeNe) (D:2) 16:26:53 -!- elliptic has quit [Quit: ChatZilla 0.9.88.1 [Firefox 3.6.27/20120216013254]] 16:28:52 -!- elliptic has joined ##crawl-dev 16:29:24 !lm erdraug type=crash -log 16:29:24 kilobyte: You have 1 message. Use !messages to read it. 16:29:29 1. erdraug, XL2 TeNe, T:1555 (milestone): http://crawl.develz.org/morgues/0.10/erdraug/crash-erdraug-20120405-212604.txt 16:30:21 03kilobyte * rd9f2200eb16f 10/crawl-ref/source/goditem.cc: Don't look up spells on a non-spellbook non-rod. 16:30:39 some webtiley stuff in this crash 16:43:41 yeah, looks like it 16:48:00 MarvinPA: d7628e0b will make you take damage without any message if the monster for some reason couldn't heal (DDoor, etc) 16:52:49 oh hm yeah 16:52:56 well there's still the "infused with holy energy" message 16:53:38 but maybe "draws life force from you" should always show, and then "and is healed" if it actually heals 16:54:02 actually, there's apparently no case when it can fail 16:56:20 03elliptic 07stone_soup-0.10 * r5996d127b127 10/crawl-ref/source/dat/lua/autofight.lua: Don't try to hit orbs of destruction with autofight. 16:56:20 03elliptic 07stone_soup-0.10 * rd7d43a48f8fa 10/crawl-ref/source/evoke.cc: Fix outdated comment. 16:56:21 03kilobyte 07stone_soup-0.10 * r4632b7e0c090 10/crawl-ref/source/Makefile: Force timestamps on .d to be same or older as those of .o 16:56:21 03elliptic 07stone_soup-0.10 * r2d52b66999ac 10/crawl-ref/source/melee_attack.cc: Don't waste turns trying to hit rock worms when it will do nothing. 16:56:25 if the monster has maxhp - 1, maybe? 16:56:31