00:03:18 Unstable branch on crawl.develz.org updated to: 0.9-a1-746-gf5743ff (32) 00:08:56 -!- ainsophyao has joined ##crawl-dev 00:18:29 Windows builds of master branch on crawl.develz.org updated to: 0.9-a1-746-gf5743ff 01:07:47 -!- neunon has quit [Ping timeout: 244 seconds] 01:23:52 -!- neunon has joined ##crawl-dev 01:33:12 -!- Twilight13 has quit [Read error: Connection reset by peer] 01:33:25 -!- Twilight13 has joined ##crawl-dev 01:58:42 -!- Wensley has quit [Ping timeout: 240 seconds] 02:13:21 -!- neunon has quit [Read error: Connection reset by peer] 02:17:01 -!- galehar has joined ##crawl-dev 02:44:23 03MarvinPA * re17dc81654bb 10/crawl-ref/source/ (showsymb.cc zap-data.h): Make Inner Flame harder to resist, highlight affected monsters to help prevent accidentally exploding them in melee 02:51:33 moin 02:52:09 -!- Twilight13 has quit [Read error: Connection reset by peer] 02:52:32 -!- Twilight13 has joined ##crawl-dev 02:57:50 hey 03:00:04 kilobyte: can we resume the tornado discussion? Yesterday we left at: you want to balance it against firestorm, we think it shouldn't be as strong as firestorm. 03:01:37 also it's currently stronger than firestorm so regardless of what it's supposed to be, it ought to be nerfed further 03:03:18 but yeah, i read the backlog and would also like to see it balanced as a single-school spell like we were in the middle of doing until it was inexplicably made conj/air again 03:04:48 marvinpa: what do you think about the ideas that were tossed around with making the tornado change size? 03:05:03 reduce size as it expires, right? 03:05:32 -!- monqy has quit [Quit: hello] 03:05:50 that'd be a more elegant way of giving it a timeout, if you couldn't recast it while it's active 03:06:12 reduce size at it expires and/or increase size as it starts up 03:06:15 I like the opposite. Increase size at the start. first turn, radius 2, then 4, then 6 03:06:28 or that, yeah 03:07:07 increase size at the start would work better with the current effect at the end with clouds dissipating harmlessly for a couple turns 03:07:31 i think the current damage level is pretty reasonable at least. pretty much however it's tweaked, with anything near the current damage output it will be appealing for lategame/extended endgame 03:08:10 so i don't think there's any fear of it becoming a terrible, niche spell like shatter (except that shatter isn't terrible or niche :P) 03:09:03 with my proposal, you get much less use for the first 2 turns. Monsters at the edge of the range will take 2/3 total damage for example. Or should we increase the duration by 2 to compensate? 03:09:21 also this automatic spell letter assignment is amazing, i keep learning spells and then instinctively hitting = to adjust them 03:09:25 then realising i don't need to! :D 03:10:12 MarvinPA: you said it's stronger than firestorm, then that the current damage level is pretty reasonable. ??? 03:10:35 sorry, yeah 03:10:53 i mean, the damage should probably still be reduced 03:11:06 but not by that much 03:11:45 up until that last damage reduction it was just leagues ahead of anything else, now it's at least in the right area :P 03:11:54 galehar: I'd say don't increase the duration, use this as a way of nerfing damage a bit 03:13:02 galehar: also, isn't it radius 5 currently? do we want it to get to radius 6? 03:13:12 that's what I was thinking. So don't touch damage, reduce the range of the first 2 turns, and add a cooldown timer. 03:13:25 oups, maybe :) 03:14:14 cooldown timer of just a couple of turns (approximately the length of time the clouds stay around at the end currently?) would still be nice, yeah 03:14:18 so range 1/3/5 03:14:44 how about 2/3/4/5 03:15:03 radius 1 tornado is a little sad :) 03:15:14 yes, good idea 03:15:41 I think it will feel more like a tornado this way 03:15:51 and of course, back to pure air :) 03:15:53 that sounds cool, yeah 03:17:51 hm, there was something else interesting in the backlog that i agreed with but i can't remember what it was 03:18:56 if 2/3/4/5/5/5 is too much of a nerf, increasing the duration by a turn or so could be reasonable too... or 2/3/4/5/6/6 03:19:05 lots of ways of tweaking 03:19:41 how are we handling the variable duration of the spell, by the way? just higher power = longer time at full size? 03:19:54 it might be better just to fix the duration, I'm not sure 03:21:57 marvinpa: red drac breath? something about bucklers/shields? 03:22:11 xom? 03:22:23 oh it was cbl i think 03:22:52 oh, I said something vague about wanting to lower the level 03:23:24 yeah, l6 air/conj might be too low though... it seems like it ought to be a pretty fun spell though, it's a pity it's underused 03:23:55 it is fun, yeah... L6 air/conj with the numbers tweaked a bit might be okay 03:25:06 but yeah, tornado was the main thing, i like the proposal 03:26:01 elliptic: the duration doesn't vary a lot with the power. It's 4 + power/60. Gives 6 at 120 and 6.5 at 150. 03:28:26 galehar: right, that's why I wonder whether that variation is really necessary... but I guess we might as well keep it unless we have some reason to want the duration to be fixed (say if we wanted to try a tornado that expanded to be really large) 03:31:24 it helps make power more relevant. At high power, since it boosts both damage per turn and duration, total damage is significantly higher. Could help with making the spell relevant both in 3 runes and endgame (which is what kilobyte is worried about). 03:31:55 Or not, since getting such a high power depends more on wielding a staff than maxing skills. 03:31:58 I think. 03:32:31 wielding a staff and finding a robe of the archmagi 03:32:42 and boosting int a lot 03:33:33 stuff you can do before zot 03:33:35 but yeah, keeping the duration variable is fine 03:36:31 -!- syllogism has joined ##crawl-dev 03:38:55 what is the cloud type left by tornado? Where is it handled? I can't find it. 03:43:46 looks like CLOUD_TORNADO 03:43:55 see line 310 of spl-tornado.cc 03:44:45 right, thanks 03:45:43 so, they have a 2 + random2(2) duration. Should we use that for the cooldown timer? Or just 3? 03:47:06 well, that means that 1/3 of the clouds will still be visible after 3 turns 03:47:09 er, 1/2 03:47:44 er, okay, I'm just wrong 03:47:58 sure, I'd say 3 turns if that's when all the clouds actually disappear 03:48:23 -!- herself has quit [Read error: Operation timed out] 03:48:30 fine 03:49:22 oh, I see, the clouds aren't actually all created anew each turn 03:49:37 so, looks like a plan: reduce range (2/3/4/5/5/5), add a cooldown timer (3 turns) and change school back to air. 03:49:49 now, all we have to do, is have kilobyte agree with that :P 03:50:04 do we really :P 03:51:45 He coded the spell, it would be better if we can have him agree with the design and the balance. 03:52:20 -!- herself has joined ##crawl-dev 03:53:34 it definitely seems far more interesting a spell concept to aim for single-school balance to me than to just add another storm-type spell 03:54:16 it seems pretty bad to me that the idea of making it level 10 was even brought up, for example :/ 03:56:48 yeah, I suggested it at some point, but then realised it's a terrible idea. 03:57:13 it just shows how totally ridiculously strong it really is/was, i guess :P 03:57:27 I could be very off here, but isn't part the problem of it being single-school is that it's trivial to get it relatively early using reskilling? 03:58:58 i don't think reskilling is particularly a problem any more than just focusing heavily on one skill with normal victory dancing 03:59:53 and i'd say shatter is still a very good comparison point, you can get it quite early if you go all-out and focus on just earth magic, and it's very strong if you do so 03:59:59 but it doesn't win the game for you 04:01:59 shatter is even a bit easier to get than tornado, since earth has very strong lower 04:02:11 lower-level pure spells (sandblast and LRD) 04:02:40 i mean, you can get firestorm by like the beginning of vaults if you're a deep elf of veh and you find some wizardry or something 04:02:47 turning off charms and victory casting swiftness doesn't take that much away 04:02:50 it's not like level 9 spells are *that* hard to cast if you really totally focus on it 04:02:53 i wouldn't say it's that much easier! 04:03:08 pacra: well, the point is that you don't need any other skills to kill everything with earth 04:03:21 true 04:03:35 with air, how are you killing things? lightning bolt probably, but you'll need to raise a decent amount of conj to get that castable at start 04:04:13 anyway, shatter is a cool spell that is very useful while at the same time not trivializing the game if I get it at L17 04:04:49 it should be possible to make tornado at that level also 04:16:16 elliptic: do you think there's any hope for a hex-themed AM book if they had inner flame in it somewhere? 04:16:44 like, that plus make cause fear a bunch less better, plus corona and some other stuff... 04:16:50 er, less bad* 04:17:11 well, I do like inner flame as an AM spell a lot 04:17:24 might well work better there than in FE even 04:17:26 i like it conceptually and keep trying to come up with ideas to bring AM back but it seems hard to make work :P 04:17:44 but corona + inner flame + cause fear still isn't very many spells, yeah 04:18:29 slow might conceivably be decent 04:18:38 it could even be buffed a little if necessary 04:18:57 yeah 04:18:58 and enslavement is cool combined with inner flame... 04:19:07 yesss that would be super fun 04:19:38 slow could be made harder to resist i guess, since it's not like it's even a very strong effect now 04:19:51 yeah 04:20:12 can we rename inner flame to soul fire? ;( 04:20:15 and that seems like a somewhat simpler method of making the underused hexes a little more usable without a larger-scale change to how MR works 04:20:28 or "enfire soul" which is also *cool* 04:20:46 corona, slow, inner flame, enslavement, cause fear, mass confusion? 04:20:59 that could be fun 04:21:00 the bigger problem with AM is that early-game bows are quite weak 04:21:12 ooh is tihs am without branding spells? 04:21:31 due: yeah, in theory 04:21:40 if we can come up with a good selection of spells :P 04:21:40 total thumbs up to that 04:21:58 I think the hexes would be fun and some of them have at least theoretical synergy with ranged combat, but the only thing going for old AM's ranged combat was poison weapon 04:22:09 can we have a "magnetism" spell that makes all metal-based projectiles have a higher to-hit against a specifc monster? 04:22:13 elliptic: in my local AM branch i also changed them to just get the same weapon selection as hunters do 04:22:13 a +0 bow with +0 arrows... does not do much damage 04:22:27 but yeah, that's still not great for the majority of races that get bows 04:22:56 (which is basically like a reflavoured corona) 04:24:07 with some sort of rebalancing of early ranged combat, I think hexy AM has definite potential... and maybe it is possible even without that 04:24:52 due: such a new spell could be useful if it also increased damage by a bit 04:25:06 elliptic: ooh yes "causes impact to be harder" or so? 04:25:16 yeah, it makes some sense I think 04:25:17 and it could have a bonus against metal or metal-wearing monsters. 04:25:37 wearing plate, chain, metal gargoyle, iron dragons, iron devils... 04:25:52 something like that instead of corona could really help to smooth out the early game for AM 04:26:19 corona is not bad but lack of damage is more of a problem than lack of accuracy 04:26:24 yes, exactly 04:26:41 am really sucked, to be honest; i didn't even bother using corona. 04:27:07 I mainly just used corona to conserve a bit of ammo on the first couple of levels 04:27:19 would it be so bad for AM just to get +1 ammo? 04:27:32 it seemed like +0 ammo caused a lot of its problems, to me 04:27:49 +1 vs +0 is less of a difference than it used to be 04:27:54 ??mulch 04:27:54 mulching[1/3]: The base chance of breakage is 1/X. X is: 3 for curare, darts. 4 for stones, arrows, bolts. 6 for non-curare needles. 10 for javelins. 25 for large rocks. Throwing nets only take damage when struggled against and are destroyed at -8. Other thrown things such as hand axes aren't technically missiles, and won't break. 04:27:58 ??mulching[2] 04:27:58 oh, hm 04:27:58 mulching[2/3]: Enchanted projectiles get an extra shot at avoiding destruction: ench / (3+ench) chance of survival. Flame- or frost-branded missiles have twice the chance of breakage. 04:28:05 re: damage or mulching? 04:28:13 mulching 04:28:33 +1 ammo would help a little with damage, but I'm not sure... we want hunter to have some advantages :) 04:28:48 yeah, true 04:29:21 well 04:29:51 how about boosting hunter's ammo to +2? 04:29:56 orbetter yet 04:30:04 how about we boost +0 ammo to +1 and shift everything down one? 04:30:10 if +0 ammo is really that bad, maybe it needs a buff? 04:30:25 hunter is already a fairly good start really 04:30:46 at least with some races 04:30:57 speaking of that, yeah 04:31:05 i always wondered why MD get a crossbow *and* a hand axe 04:31:26 they get to be like MDFi except with an awesome ranged weapon too :P 04:31:43 the problem is really that +0 arrows with a plain bow are weak... crossbow + bolts is far stronger 04:32:04 so maybe tweak the mulch formula to make +1 ammo more durable. 04:32:12 then rebalance the formulae? 04:32:19 yeah. 04:32:36 bow catches up when you find a longbow (though then crossbow moves ahead again in yaktaur land, when you get many steel bolts) 04:32:55 magnetism might be eabit too much with crossbows and steel bolts, hm. 04:33:11 well, magnetism should presumably check MR 04:33:13 or perhaps it could have a cut-off point of damage bonus that makes ituseful early, but not totally game breaking later. 04:33:15 and have a low power cap 04:33:21 yeah 04:33:22 that! 04:33:28 so by late game, it won't be good for much 04:33:57 RC isn't that much use in the late game anyay, I've heard. 04:34:26 how big a factor is weapon speed early on, for crossbows vs bows? 04:34:38 actually ranged combat is very strong once you get a good launcher and ammo 04:35:05 elliptic: ah, good 04:35:09 marvinpa: crossbow is definitely slower but it also deals a lot more damage 04:35:11 !hs char=hahu 04:35:12 23. bookofjude the Halfling Catapult (L27 HaHu), worshipper of Trog, slain by a soul eater on Tar:7 on 2010-04-11, with 565571 points after 128411 turns and 20:28:52. 04:35:17 --^ point 04:35:59 the most successful non-cheating attempt at doing a ziggurat with trog was n78291 with a halfling slinger, I believe 04:36:16 i didn't cheat! 04:36:20 :P 04:36:22 you cast tornado :P 04:36:28 :( 04:36:53 speaking of which, you should probably get excommunicated if you get like 50 penance 04:37:02 it was sort of weird to be able to just carry on casting level 9 spells under trog 04:37:19 without it having that much of an impact :P 04:38:17 casting spells isn't excommunication-worthy for trog? 04:38:24 anyway, I think hexy AM certainly has potential... I really like the idea of using tricky hexes like inner flame and enslavement on a ranged char 04:38:32 no, you get some amount of penance depending on the level of the spell i think 04:38:38 * due obsessed with the idea of magnetism now 04:38:45 though whether to the point of actually writing code for it? who knows! 04:39:14 due: it is easy enough to cast a spell under trog accidentally, so excommunication would be a little harsh 04:39:31 prompting? 04:39:45 or keep a counter of spells cast? 04:40:11 you don't get a prompt at the moment :( 04:40:16 the book proposed before for hexy AM sounds great! 04:40:49 prompt and excommunication could work. 04:41:43 spell levels 123446 and almost entirely pure hexes except for inner flame, so i guess it wouldn't even be too hard for them to get their spells reliable 04:42:28 for mulching, how about making the chance to survive: ench / (2 + ench)? 04:43:29 spell levels. That reminds me: could shroud of golubria be L1? 04:43:37 marvinpa: hm, I see that inner flame turns allies hostile... not sure whether that makes sense given that it isn't actually hurting them... but maybe it would be too abusable otherwise 04:43:50 yeah, i wasn't sure 04:43:57 you can always inner flame then enslave second 04:44:16 I'm envisioning HO of beogh casting inner flame on all their orcs :P 04:44:20 haha 04:44:27 shroud of golubria? 04:44:28 for awesome chain reactions 04:44:47 presumably this would tend to get you beogh penance though 04:44:57 since you get credit for the explosion kills 04:45:00 yeah 04:45:05 shroud is cool! 04:46:01 elliptic: i think you can cast it on allied undead actually 04:46:20 could make for some fun kiku fire synergy! 04:46:35 without turning them hostile, that is 04:47:23 galehar: I don't really like the idea of making it L1... it is quite good (possibly I will decide to nerf it slightly) and L1 would make it too easy for anyone who found it to cast it reliably 04:48:02 ok 04:48:37 IMO skald has no need for a L1 spell... I know people have been clamouring for one, but crusader was fine for years without one 04:49:11 marvinpa: oh, it works on undead? I meant to ask about that actually 04:49:21 seems like possibly it shouldn't 04:49:33 ok, I'll test the current skald 04:49:53 possibly not, i'm not sure 04:50:05 so, about mulching. How about (ench+1)/(ench+4) (which makes new +1 like current +2) 04:50:09 marvinpa: I guess your description works fine for most monsters though 04:50:25 right, i was trying to think if there should be anything that it can't affect 04:50:47 but it seems to work fine for anything that's not magic immune 04:51:16 wraith (or other incorporeals), maybe 04:51:23 ah yeah 04:54:01 galehar: that sounds reasonable enough, or we could adjust the base numbers directly 04:54:14 ??mulching 04:54:14 mulching[1/3]: The base chance of breakage is 1/X. X is: 3 for curare, darts. 4 for stones, arrows, bolts. 6 for non-curare needles. 10 for javelins. 25 for large rocks. Throwing nets only take damage when struggled against and are destroyed at -8. Other thrown things such as hand axes aren't technically missiles, and won't break. 04:54:36 and change 4 to 5, say 05:03:05 hmm. That's actually better if we want to preserve AM ammo which isn't enchanted. 05:18:05 for octopode, how about we replace the beak attack (and maybe tone down tentacle slap) and give it reaching for all attacks. 05:18:36 Kilobyte already addressed the worst reaching abuse. 05:19:05 We make this change for 0.9, see how it's played out in the tournament, so that gives us a list of thing to fix with reaching for 0.10 05:19:09 the main problem with that is the fact that they're amphibious, i guess 05:19:24 oh, good point 05:19:29 yeah, water abuses are pretty bad 05:19:34 if you get an entry vault with water you'd have such an easy d:1 05:19:50 also, are octopodes going into 0.9? people have been saying conflicting things about this 05:20:24 I don't know. 05:20:26 marvinpa: and later on too... if there is water on a level, you can kill any monster who lacks a ranged attack without harm 05:21:01 this is already a problem with reaching on merfolk (or anyone with a good source of lev/flight), of course 05:21:04 just not on d:1 05:21:14 maybe it can be addressed with better AI. If the monster cannot attack you (no ranged and no pathfinding), and you attack him, then it flees. 05:21:27 could make it so you can't reach from one terrain type to another, or with a different terrain type in the way 05:21:33 but i have no idea how to justify that thematically :P 05:22:09 well, for merfolk and octopodes you are actually swimming in the water... it makes sense that it would be harder to reach up and out of the water 05:22:17 that doesn't handle lev/flight though 05:22:21 yeah 05:24:10 also i would be in favour of octopodes going into 0.9 as-is or nearly as-is, personally 05:24:31 they're a lot more interesting and fun and balanced than, say, felids :P 05:24:55 I like them better than the current status of felids also :( 05:25:06 I think the AI fix I proposed makes sense, and address a number of things. 05:26:51 I was one of the few real felid supporters when they were added, but the changes kilobyte made to them in 0.9 make me not want to have anything to do with them 05:27:11 which changes? level drain on dying? 05:27:16 galehar: it sounds a little tricky... what happens if the monster starts to flee and then you swim to the edge of the water 05:27:47 galehar: I'm not a huge fan of level drain on dying, but mainly I don't like the schedule for when felids get extra lives 05:28:19 levels 4, 7, 11, 15, 20, 25, and then long after reaching XL 27 you continue to get lives 05:29:15 Why not simply one every 3 level. Cats have 9 lives, don't they? 05:29:44 4, 7, 10, 13, 16, ... , 25 was how it was in 0.8, yes 05:29:58 though even then you continued to get extra lives after XL 27 05:30:11 completely undocumented and I didn't even know about it at the time 05:30:31 XL 28, XL 31, ... 05:30:55 I suggested the level drain to make death more significant 05:31:36 the thing is that level drain doesn't mean that much in crawl 05:32:19 difference between L15 and L16... a small amount of HP/MP, one spell level 05:32:38 well, it's better than nothing 05:32:44 for felids it puts them farther from getting their next extra life, which is something at least... but it still isn't a very exciting effect 05:33:26 I'm fine with that really, but getting an extra life at XL 31... just feels awful to me 05:33:33 I suggested no teleport to safety: but then it makes it looks like life saving instead of new life. 05:33:52 I suggested dropping inventory, but can be problematic in abyss and portal vaults 05:34:13 but yeah, I don't see what's wrong with capping extra-lives. 05:34:19 I suggested a few things here, and I had been testing some of them, but when kilobyte heard that I was working on it he quickly made his own changes instead 05:34:35 and he was very resistant to the idea of eliminating post-XL 27 extra lives 05:34:49 remember why? 05:35:07 because "you can recover from anything else with enough scumming" 05:35:26 so he thought you should be able to recover from losing your extra lives also 05:35:35 ... 05:35:52 can't you recover from losing your extra lives by maybe not dying again...? 05:35:56 If one ressource should be finite, it's felid's lives. 05:40:11 the commit for the change of schedule doesn't give any reason. 05:40:11 dpeg was around at the time and seemed to agree that there was no need for felids to have inexhaustible extra lives, yes 05:40:11 I mean, reducing the number of extra lives was a reasonable enough idea for nerfing (though I liked the "nine lives" flavour) 05:40:11 re 05:40:11 also I should say that the post-XL 27 extra lives aren't actually infinite because of the xp cap, as I understand things 05:40:25 (getting to the xp cap usually requires clearing quite a few ziggurats) 05:45:56 about the AI changed I talked earlier. I already though about it before we talked about reaching. If a monster can't attack nor pathfind to you (because of clouds or deep water or whatever), I hate it when it stays there stupidly while you kill it with ranged attacks 05:45:56 making it flee make sense. Or we can have a "retreat" status. It flees until outside of the range of your attacks. 05:45:56 galehar: what about the comment I made earlier? (which you might have missed in the felid discussion) 05:46:49 It probably shouldn't get right back at you, because then, you step back in the water and it's a stupid dance. 05:46:55 it would definitely be cool to improve this aspect of the AI, but it does sound tricky to me because of situations where you are moving in and out of regions where the monster can get to you 05:47:14 the AI already behaves strangely about this 05:47:50 often if you are swimming in a pond and a monster is outside it, you can cause it to walk all the way around the pond by making a single step 05:47:57 yes, it's not simple. Just making it flees is simple and better than current state. 05:48:26 if you make it flee and stay fleeing for a while, yeah 05:48:51 until ouside of LOS + some random duration. 05:49:14 galehar: elliptic: Normal fleeing behaviour? 05:49:32 -!- gustaf_arg is now known as ghallberg 05:49:52 I think normal fleeing behaviour suppose that the monster is heavily wounded. It will stop when sufficiently healed (or when it can't flee further). 05:51:00 there also can be situations where the current fleeing will just get the monster into a corner that is still in range of you 05:51:41 yes, it can happen. Well, in that case you've trapped the monster and outsmart it. 05:51:59 yeah, still good to fix the more typical cases 05:52:02 Hard enough to setup to not be abusable I think. 05:52:33 about octopode, it also has been suggested to give them an ink cloud ability, similar to ?fog. 05:52:49 inkilicious :D 05:53:11 I don't like that because fog is the sort of ability that becomes more powerful as the game progresses 05:53:20 we don't want that for Op 05:53:42 (reaching becomes less powerful in general, since there are more ranged attacks) 05:53:55 good point 05:54:00 ink only works in water, though 05:55:06 well, it has been suggested to make it work on dry land to. 05:55:18 a water-only fog ability sounds pretty narrow 05:56:20 it would be more thematic than gameplay, but still useful from time to time. 05:57:19 merfolk gets some nice bonus from water. octopode could have this little thing. 05:57:25 a water-only fog ability sounds all right to me, yeah 05:57:33 just not very exciting :) 05:57:52 ink cloud on dry land seems... pointles? 05:58:10 they squirt ink into the water, not actually creating a cloud. 05:58:41 could create slippery ink trails, localized liquefaction-like effect? :P 05:59:24 "The orc slip on the ink and falls down on his butt." 06:01:10 well, having a different effect on dry land would be fine 06:37:07 elliptic: regarding skill cost being dependent on skill training order. All the hacks removed it's now more consistent. There's still a case where order matter: antitraining. If you want 2 opposite schools, you should train the worst aptitude first. Relevant? 06:37:28 -!- greensnark has quit [Ping timeout: 240 seconds] 06:38:14 -!- greensnark has joined ##crawl-dev 06:40:13 -!- edlothiol has joined ##crawl-dev 06:57:53 -!- Napkin_ has joined ##crawl-dev 06:58:29 -!- Napkin has quit [Quit: leaving] 06:58:44 -!- Napkin_ is now known as Napkin 07:13:37 -!- b0rsuk has joined ##crawl-dev 07:13:42 Hi 07:14:14 moin b0rsuk 07:15:04 I retreated into a U-shaped wall section. Nagas followed me, blocking black mamba from reaching me. Then it occured, an amazing example of teamplay: 07:16:12 The rock wall was actually a secret door! 07:16:15 _A naga opens the door. 07:16:31 nice 07:17:03 I don't know how it happened, but naga opened the door so a black mamba could reach me from the other side. Was that a fluke, or are monsters smart enough to open doors for their teammates who can't ? 07:17:23 I don't think anu such smartness is actually implemented 07:17:32 any* 07:17:49 must've been coincidence 07:20:55 there's a certain word 07:21:06 that game designers use when developing AI 07:21:21 to make it/them do something out of the ordinary and seemingly at random 07:21:37 but that the player can concievably look at and think "whoa, it knew how to do THAT?!?" 07:21:46 but the term escapes me 07:22:04 emergence? 07:22:30 emergent behaviour meybe 07:22:36 Pacra: Do you like cellular automata ? 07:22:47 there's some code to make monster in melee range move sideway to let other monsters get to you. Maybe the naga opened the door because it wanted to sidestep there to let the black mamba get to you. But the mamba moved there before the naga. 07:23:32 No, mamba was on the other side of the wall. There was a naga behind the naga, though. That must be it. 07:23:37 It was still nice. 07:39:54 b0rsuk: mmm, conway's game of life 07:40:54 in college I made a program in Max/MSP that would generate an electronic music ditty with a few homemade digital synths using a step-through of a single game of life that the user could define on a grid 07:41:10 always had a soft spot in my heart for glider guns and r-pentominos. 07:41:27 and pulsars! 08:02:06 so, how about swapping throw flame and throw frost ranges? 08:12:56 Pacra: You might want to try Conway's Inferno too - http://www.newgrounds.com/portal/view/560921 08:13:29 Conway's Inferno has procedurally generated music too. 08:14:15 And every pulsar fan should see this: http://www.youtube.com/watch?v=QH2-TGUlwu4&feature=youtu.be 08:16:14 wrt octopodes: removing beak is fine by me 08:16:22 since, after all, humans have a mouth but don't have a bite 08:16:51 b0rsuk: this is glorious 08:17:01 i think i would like to get at least some attempt at constriction in at the same time as reaching 08:18:02 say: if you attack something, you have a small chance to have them be en-netted if they aren't something that hurts you on attacking them, except it's a virtual net and if it 'breaks' they simply come unconstricted 08:18:24 attacking anything else de-nets the first creature 08:18:54 i'd like to keep the AC 1 for now, i think the apts are okay if a bit boring, and i'm not in love with the idea of regen 1 but it could be an option 08:19:43 the neutral apt combined with low AC push them a lot toward magic 08:29:43 b0rsuk: is the procedurally generated music generated from each level? 08:29:49 pretty neat if so 08:30:48 galehar: i'm not sure this is avoidable, since giving them good dodging or AC doesn't make sense, and i don't think they should get great HP (since that benefits casters for no investment) 08:30:53 -!- edlothiol has quit [Quit: edlothiol] 08:31:11 -!- edlothiol has joined ##crawl-dev 08:32:07 i think they could have a good niche as favoring hybrids, since that lets either melee or magic rings be useful to them 08:47:06 Pacra: Yes, it is. 08:47:55 love the idea and execution 08:48:00 love that low-fi feeling 08:48:16 only thing is, not enough levels; you could do some mind-bending stuff here 08:48:34 Pacra: Can you find an elegant solution to every level ? 08:48:39 and there should be an option implemented to speed up the turn-per-second 08:49:03 it looked to me like there was a specific solution to every level 08:49:28 Pacra: You've finished it all ? I don't want to spoil. 08:49:35 but there could be multiple solutions! idk 08:49:44 no, taking a break to finish my crawl-ogre. 08:50:25 was reading the comments on its page, stopped before getting spoiled i.e. last level 08:59:44 -!- petete has joined ##crawl-dev 09:13:14 -!- b0rsuk has quit [Quit: Lost terminal] 09:20:30 -!- petete has quit [Remote host closed the connection] 09:27:20 Eronarn: what about allowing 2 handers with a shield? That's a nice buff to melee. 09:30:19 that'd be a huge lategame buff though 09:30:27 and it's not like they're bad at melee later on 09:38:16 yeah, i'm more concerned with balance up till lair, initially 09:38:36 right now, they do get the ability to use offhand punches when using 2 handers 09:38:42 (or shields, i guess) 09:39:24 i think i might prefer: same combat rules as humans, except also a constriction aux, which is either 'squeeze' (just damage) or 'constrict' (they get caught) 09:39:46 i think the way i want to do this is by splitting tentacles into two mutations 09:39:56 MUT_TENTACLE_LEGS and MUT_TENTACLE_ARMS 09:40:46 legs 1: cosmetic. legs 2: no boots, kick replaced by squeeze. legs 3: better damage, constriction 09:41:18 arms 1: cosmetic. arms 2: no gloves, reaching. arms 3: many tentacles, rather than two big ones, so gain a tentacle-slap aux 09:41:36 this sound reasonable for something that could be a) generic b) DS facet? 09:42:13 that sounds pretty good to me 09:44:30 -!- ortoslon has joined ##crawl-dev 09:47:54 -!- edlothiol has quit [Ping timeout: 255 seconds] 10:17:04 Eronarn: That constriction idea is exactly the same as proposed for nagas before, and it's good 10:19:30 -!- Wensley has joined ##crawl-dev 10:52:01 -!- edlothiol has joined ##crawl-dev 10:54:25 Keskitalo: it's been proposed since several years ago :( 10:55:21 those muts are clearly stronger than current DS bodyslot facets, no? 10:55:57 elliptic: not sure about that... they'd be less damage 10:56:29 and you can't realistically constrict stuff late game because it's all gigantic flying demon-y things 10:56:56 possibly could make them monstrous DS only, though 10:57:31 that'd be one way to balance them, if they not onyl got a lot of body slot muts but also better ones 10:57:48 well I don't know how constriction would work... but bodyslot facets currently don't give great damage 10:58:15 probably use the same rules as nets, except fix those if they appear broken (i think they're basically ok but might have some quirks) 10:58:32 and have the net strength be based on your strength and size 10:58:58 net-constriction sounds strong 11:01:27 -!- galehar has quit [Quit: Page closed] 11:07:36 perhaps it could be tiered - net level if you get high enough success, otherwise, just prevent movement, or cause floundering like water 11:11:12 wrt op in .9: it does have the comedic value that i announced they'd be in .9 several versions ago, but we only have about a week before we freeze, right? seems like last-minute changes might be bad for that 11:17:31 I always figured that the sorta-crappy DS body slot muts were a balancing factor for how awesome the other DS mutations were :P 11:51:14 Nate (L24 KoBe) ERROR: range check error (-1 / 80) (Shoals:2) 11:51:29 !lm Nate crash -log 11:51:30 1. Nate, XL24 KoBe, T:114206 (milestone): http://crawl.develz.org/morgues/0.8/Nate/crash-Nate-20110627-165110.txt 12:01:11 -!- evilmike has joined ##crawl-dev 12:01:11 -!- evilmike has quit [Client Quit] 12:03:23 -!- evilmike has joined ##crawl-dev 12:10:48 -!- galehar has joined ##crawl-dev 12:13:04 Eronarn: yes, if they make it in, it's probably in their current state (or maybe with minor changes). But we can talk about futur changes :) 12:14:22 -!- ortoslon has quit [Ping timeout: 246 seconds] 12:19:01 -!- gnsh has quit [] 12:19:31 i think for .9 i would try: current, remove beak & 'claws', add regen 1, maybe some apt or HP tweaks 12:20:09 regen? Didn't you said you don't like it? Or was that someone else? 12:20:29 i don't like it, but it's probably better than their survivability coming from UC 12:20:37 early on 12:21:17 and the other methods of giving them more survivability feel wrong 12:21:56 reaching is a good idea but i'm not sure we'd want to put it in for .9 12:22:43 koboldlord suggested on the forum to give them -2 to conj and some positive aptitudes in M&F or polearms to move them a bit away from their magic bias. 12:22:55 that's what they originally had 12:23:16 it was pretty annoying 12:23:28 people didn't like them being that bad at conj, or favoring a specific weapon type 12:23:33 yeah, we'll try to add reaching for 0.10 12:23:48 well, actually it was originally -3 not -2. but yeah 12:23:48 since they fit casters nicely and was weird for them just to be arbitrarily terrible at conj 12:23:55 it was weird* 12:24:18 yeah, i'm not sure it is bad if they strongly favor casters 12:24:21 personally i like the flat apts 12:24:32 what is bad is if people can't start Op because bats kill them 12:24:41 Eronarn: so, what did you have in mind for apt tweaks? 12:24:56 nothing in mind yet, but the current apts are not set in stone, by any means 12:25:38 for example, people suggested giving them higher tmut 12:26:39 though, btw, octopode transmuters are probably kind of broken right now, we might want to think about how rings work when merged (or maybe we just let it work that way, anyways) 12:27:39 -!- gnsh has joined ##crawl-dev 12:36:39 Eronarn: no ink clouds yet? 12:42:43 Wensley: i don't think ink clouds are a good idea 12:42:55 hiding in water is scummy enough already 12:55:05 -!- petete has joined ##crawl-dev 13:02:19 -!- Textmode has quit [Ping timeout: 250 seconds] 13:06:40 -!- monqy has joined ##crawl-dev 13:15:42 -!- Zaba has quit [Ping timeout: 240 seconds] 13:22:16 -!- Zaba has joined ##crawl-dev 13:35:50 -!- evilmike has quit [Quit: Leaving] 13:42:45 -!- ainsophyao has quit [Remote host closed the connection] 14:01:11 -!- b0rsuk has joined ##crawl-dev 14:20:18 -!- b0rsuk has quit [Quit: leaving] 14:20:51 -!- b0rsuk has joined ##crawl-dev 14:36:56 -!- Hehfiel has quit [Ping timeout: 240 seconds] 14:41:41 -!- Hehfiel has joined ##crawl-dev 14:50:51 Don't you think that Stasis is perhaps too hard of a counter to some enemies ? It reduces sphinxes to smiting. Actually Stasis blocks most of MR effects. It won't help against Confusion (not so bad) and Banish (dangerous, but rare). 14:52:07 -!- Twilight13 has quit [Read error: Connection reset by peer] 14:52:28 -!- Twilight13 has joined ##crawl-dev 14:53:29 stasis also has drawbacks, you know... I barely ever use it because there are usually better counters (like getting some MR) 15:02:15 -!- dpeg has joined ##crawl-dev 15:02:25 hi dpeg! 15:02:29 Hi! 15:03:38 The xp nerfs go in the correct direction, so we don't worry, right? 15:04:00 yes, that's what I'm thinking. 15:04:12 and from the simulations I ran, the change isn't that big. 15:04:15 ??xp nerfs 15:04:15 I don't have a page labeled xp_nerfs in my learndb. 15:04:15 I am still not sure about the +3 HP... seems so much :) You'll have to take the responsibility if now everyone and their grandma streak! 15:04:26 skill cost change 15:04:34 b0rsuk: c-r-d mail 15:05:06 Oh, you're so mean :(. I'll have to switch to browser, type an adress, click a couple of times... 15:05:32 Well, I think it's more important how streaks are done and undone, than how long and frequent they are. 15:05:47 HPhas 15:05:50 been boosted starting at XL13. The growth used to slow down, now it's 15:05:52 linear all the way to XL27. HP from fighting reduced and scaled to species 15:05:55 HP. Big nerf to frail species with high fighting like SpBe. 15:05:57 * tornado's damage has been 15:06:00 reduced 15:06:03 by 15:06:05 about 40% (and recently changed to conj/air). 15:06:08 * skill costs (how many XP for a skill point) are no more dependent on skill 15:06:10 level. I ran a few skill training 15:06:13 simulations 9wP>, 15:06:18 and it doesn't seem to break the balance. It mostly changes the mid-game 15:06:21 where low level skills are a bit more expensive and high level ones a bit 15:06:23 cheaper. So it's a slight nerf to generalists and a slight boost to 15:06:24 What's the title of the mail ? 15:06:26 specialists. There was also a special case to make magic schools cheaper in 15:06:29 the early game. This has been removed, slightly nerfing early game casters. 15:06:31 sorry 15:06:41 last changes for 0.9 15:06:53 WOW ! EPic spam :D 15:07:13 galehar: yes, not really serious there. Of course it is not at all good if all streak breaking comes from randomness on D:1. 15:07:34 b0rsuk: yes, I had a /msg b0rsuk before that, but failed. 15:08:01 bug: that lists tornado as being conj/air, which would clearly make no sense at all 15:08:25 !seen kilobyte 15:08:25 dpeg: yes, it's quite bad - no time to get invested in the character to make the players suffer more 15:08:25 I last saw kilobyte at Sun Jun 26 23:11:57 2011 UTC (20h 56m 28s ago) saying Wensley: I havent seen him recently, it would be good to ask if he thinks its good enough. on ##crawl-dev. 15:08:43 Speaking of victory dancing... if you want to get rid of it, it may be not possible with a consistent approach. Maybe a hybrid approach would be better. For example it would be fairly easy to come up with formulas for melee combat. 15:09:02 b0rsuk: galehar has a good plan 15:09:16 But awfully hard for formulas for skills like Enchantments. How do you tell if player has a valid reason to cast an enchantment, and when he's just victory dancing ? 15:09:26 b0rsuk: I've got all the design in my head (and on the wiki). I just couldn't find enough time to code it. 15:09:27 It is very explicit (and I'd like some other solution, but cannot come up with one), but it will work,. 15:09:51 ADOM sort of has hybrid skill system. Weapon skills are treated differently than the rest of skills. It works reasonabl well. 15:10:05 dpeg: I also think the automatic mode can be made quite natural and effective. 15:10:19 galehar: I'd like to nerf Tornardo in the way you described. Didn't want to put that on c-r-d, lest everyone sends in their opinion on the spell. 15:10:21 have we settled on how pool removal affects draining? i'd like to reiterate my proposal for that: temporary skill loss, which you regain at the same time as gaining normal XP (so 20k XP debt = lose 20k worth of X skill = for each 1 XP you gain, add 1 to XP, and remove 1 from debt penalty) 15:10:28 galehar: yes, there is hope 15:11:27 dpeg: what do you think about felids? I'd really like to remove the still-getting-more-extra-lives-after-XL27 business 15:11:28 it'd be pretty different from the current draining effect but a lot less scummable which is good 15:11:35 galehar: regarding energy randomisation abuse: doesn't it seem better to just remove the feature if it causes so much pain? 15:11:48 Eronarn: how about simply draining skills? Maybe for a lower cost than XP. 15:12:26 elliptic: yes, nine should be all in my opinion. What is your opinion on Fe? Are they good (design-wise, not power-wise). 15:12:31 galehar: if you make it temporarily drain skills, but requiring progress (XP gain) to recover, you can make those skill penalties much larger than if it permanently erased your skills 15:12:32 dpeg: it was added for a reason. No randomisation means long pillar dance to regenerate. 15:12:46 galehar: I know that! 15:12:56 but it seems to have caused more problems than it solves 15:13:01 Eronarn: I don't see the point. Just drain the skills so you have to train them again. 15:13:21 dpeg: I honestly liked the design for felids better in 0.8 than the current status :/ 15:13:36 galehar: draining would have to drain *tons* more XP than it does now to meaningfully affect skills 15:13:52 dpeg: I wouldn't said that. I think it's better, the advanced pillar techniques are used only early by good players trying streaks. 15:13:53 dpeg: currently you get extra lives at levels 4,7,11,15,20,25, and then more substantially after XL 27 15:14:10 it's more interesting if getting drained is actually tactically relevant (e.g., lose enough skill to not be able to cast your top level spells) 15:14:38 elliptic: would be easy to sneak nine into 27? (I don't know what a good number of lives is.) 15:15:07 Eronarn: but like summoners: teleport, wait it out, come back. 15:15:21 dpeg: in 0.8 you got extra lives at 4,7,10,13,16,19,22,25 (and more after XL 27) 15:15:27 Eronarn: draining could be much more severe, but you could regain the lost xp over time (as opposed to gaining new xp from killdudes) 15:15:45 discounting the "more after XL 27", that made 1 life to start + 8 extras = 9... which I liked 15:15:57 dpeg: what i'm saying is that draining is 'temporary', but rather than recovering by resting, it recovers by gaining XP (for every 1 XP you gain - which works normally otherwise - erase some number of draining points) 15:16:06 elliptic: yes 15:16:14 so you cannot un-drain except by progressing (either in the same branch, or elsewhere) 15:16:15 (I didn't know about getting more lives after XL 27 because it isn't documented anywhere afaik) 15:16:54 but draining becomes not actually lost XP, but just 'locked up' XP which you have to win back 15:16:59 Eronarn: ah, I see. Interesting, but could still have undesired backlashs. 15:18:26 well, it would remove players killing themselves at turn 0 due to xom abandonment draining - unless we made the amount of debt you have also lower your character level the same way actually-lost XP does now 15:18:58 anyways, i just mention it because i've seen a lot of talk about the new pool stuff but none about draining (which has been in need of a fix for a long while, anyways) 15:20:18 (as a recap - the very different effect if you get hit with full pool / with no pool, and the scumminess of using it to do perma-damage to monsters) 15:21:21 I hope the cure to victory dancing turns out better than Zin, Ashanzari and Petrify. 15:21:34 what is wrong with draining monsters? 15:21:52 b0rsuk: say something useful or stay silent 15:22:08 Monsters which drain or monsters which (haha) are drained ? 15:22:56 dpeg: if you drain a monster, you do maxHP damage to it, which means you can pillar dance to weaken it (permanently) before fighting it 15:23:07 b0rsuk: have specific problems with zin? 15:23:09 elliptic: I agree, going back to 0.8 live schedule and capping it at 27 (for a total of 9) sounds great. 15:23:51 do felids get 9 lives, or 9 extra lives (9 free deaths + 1 final one)? 15:24:08 currently they can scum to get even more. 15:24:08 Eronarn: yes, but you reduce xp in the meantime and it takes a while (so is risky if you do it with something dangerous) 15:24:13 in 0.8 it was 8 extra lives (not counting XL 28 shenanigans) 15:24:36 I think it's easier to fix felids by making the Crawl around them more interesting and varied, more possibilities. 15:24:45 galehar: well, kilobyte also buffed felid apts substantially (undoing most of the nerfs that had been made to them) 15:25:01 dpeg: it's not a huge deal, but i do consider it a problem with the brand that its most useful feature is being able to deal perma-damage 15:25:19 since otherwise it's just a bad damage brand that reduces your XP when you don't want it to 15:25:20 b0rsuk: stop trolling 15:25:21 dodging and stealth +2 -> +3, UC and fighting -2 -> -1, hexes 3 -> 4, tloc 2 -> 4 15:25:33 galehar: Stop saying stupid things. 15:26:17 so felids have been buffed substantially aside from getting 2 fewer extra lives at XL 27 and losing a level on each death 15:26:18 There seems to be a trend where new races are distinguished more by slots than function. 15:26:25 I barely even said anything... 15:26:38 b0rsuk: ...what 15:26:42 deep dwarves 15:26:43 vampires 15:26:44 felids 15:26:51 octopodes, lava orcs, nomes 15:26:56 b0rsuk: coming in here and calling things stupid isn't very friendly 15:27:00 The issue I have with that is that messing with slots boils down to "can't use X, Y, and Z" 15:27:06 literally every new race either in or being seriously worked on is the opposite of what you said 15:27:41 as octopode slot differences are so dramatic that they're really like playing something totally different rather than 'lol cannot wear hat' 15:27:42 And besides, I thought aptitudes were supposed to do that. Could it mean that aptitudes fail to represent that ? 15:28:29 aptitudes are not great for differentiating races because someone can always fight against aptitudes if they want to 15:28:37 they're a tool but not the best or most powerful one 15:29:02 I think they are a pretty good tool, especially since the aptitude reform 15:29:49 Players can largely work around aptitudes because for most part they can make the most important strategical decisions at character generation. 15:30:04 this is a pointless discussion 15:30:15 this is ##crawl-dev, not ##b0rsuk-musings 15:30:34 galehar: here's a thought about felids: go back to 0.8 life schedule, don't keep on giving new extra lives after XL 27, and keep kilobyte's aptitude buffs... but make it drain two levels instead of one on each death 15:30:42 'crawl is going in the wrong direction' doesn't help anyone make decisions about what to do for .9 15:31:22 I see that as a balance between "adaptation" and "planning". Low "adaptation" means that crawl doesn't reward versatile races all that much, there aren't many items you find that make you want shift gears. 15:31:44 elliptic: hmm... haven't thought about that. 2 levels? seems harsh 15:31:51 but maybe balanced :) 15:32:06 what is the level loss supposed to actually be doing? 15:32:26 punishing people for dying and making it harder for them to get more lives, I think 15:32:32 even if you're stocked up on lives, you'll still be careful not to die. 15:32:42 -!- b0rsuk has quit [Quit: leaving] 15:33:50 galehar: do people really just not care when they die on felids, though? specifically, ones before endgame 15:34:16 on the orb run or whatever, people will still use dying tactically even if it costs 2 XLs, just because it's an available resource so you might as well use it 15:35:13 I don't care much in midgame usually if I'm careless and lose a life or two... because I probably had more than two extra lives anyway 15:35:26 well, I'd like if you drop your stuff too when you die. At least the orb. But it can be problematic in abyss or portal vaults. 15:36:00 convert to lugonu to get your stuff back :) 15:36:12 About D:1 balance. Here's from the philosophy section of the manual: 15:36:14 giving people fewer extra lives is one way to deal with this, and it is what kilobyte decided... but making the level loss harsher does something similar 15:36:15 deally, a game like this would be really challenging and have both random layout and random course of action, yet still be winnable with perfect play. 15:36:16 there may be situations where you are doomed - no action could have saved your life. But then, from the midgame on, most deaths are not of this type 15:36:35 *Ideally 15:36:48 galehar: I vaguely remember having written this :) 15:37:02 why does it have to wait until the midgame to be fair? 15:37:17 all situations are winnable, depending on luck 15:37:20 harsher level loss definitely makes more sense than post-27 lives 15:37:21 well, maybe I'll see whether kilobyte is at all open to making any felid changes... I'm not optimistic though 15:37:29 though i still like molting felids more :P 15:37:41 if we guarantee early survivability, we overshoot by far (this is my estimate, although everyone back then agreed) 15:38:21 dpeg: the early game is possibly my favorite part of the game, because I have to be so careful and run from things and am afraid of most everything. I'd hate to see that gone. 15:38:37 galehar: it is completely okay when players die randomly, in my opinion. If we make sure players don't have to win, we end up like Nethack (I don' 15:38:40 well, it's never guaranteed. But by now, even the best players say that D:1 is too hard without a lot of pillar dancing. 15:38:43 mid/endgame should be less forgiving :) 15:38:45 t say this to flame, I am serious) 15:39:04 What means "too hard"? Too hard to streak? 15:39:12 Streaking should not be our concern. 15:39:30 streaking is certainly possible with the current setup 15:39:33 player memory: the more often you win, the harder the game gets! 15:39:34 too hard to avoid otherwise unavoidable death? 15:39:43 It is our responsibility to reduce ways how players bore themselves (all dancings). 15:39:57 !tell kilobyte The current felid life scheduling is pretty ugly... no longer having "nine lives" makes the flavour weaker and the post-XL 27 lives are undocumented and unnecessary (and "unlimited extra lives" makes the race sound worse than it is). 15:39:57 elliptic: OK, I'll let kilobyte know. 15:39:58 press Tab to auto-pillar-dance 15:40:00 is kiting a dance I hope? 15:40:31 dpeg: i'd say that d1 is too hard with the specific reason that you cannot engage stuff in straight-up fights due to the variance 15:40:34 Wensley: Tab is already occupied by toggling autofight, isn't it? 15:40:52 (well, you can, but you are very likely to die that way in a way you aren't later on) 15:40:59 "too hard" is subjective... I enjoyed 4.1. 15:41:02 !tell kilobyte What do you think of something like the following: go back to the 0.8 life scheduling, cap lives at 9 (1 + 8 extra), lose two levels on death rather than just one? 15:41:02 elliptic: OK, I'll let kilobyte know. 15:41:10 deaths that let you go alive -> dead very quickly without being very predictable are bad 15:41:22 bhaak: right, pillar dancing and autofight can be combined. wouldn't make autofight any more lethal :P 15:41:28 -!- syllogism- has joined ##crawl-dev 15:41:39 :-) 15:41:42 -!- syllogism has quit [Ping timeout: 240 seconds] 15:41:43 Eronarn: but early on, alive -> dead is so short... which is why I don't worry not so much aboud early game unfairness 15:42:00 dpeg: well, it's short, but it's still annoying 15:42:13 by increasing HP, it wouldn't be so short :) 15:42:44 galehar: I am fine with making the game easier, but we should really know for what cause. 15:43:02 If players die with weird combos, or on streaks, who cares? 15:43:08 Also, on D:1, you have very few options. Especially non-casters. If you try to survive, all that's left are abusing AI, energy system. Meta-gaming. 15:43:11 If a particular background is too weak, we can improve that. 15:43:53 !lg * cv=0.8 ckiller!=quitting|leaving x=sum(dur) 15:43:53 galehar: I once suggested a single use, small healing item (think a fruit) for everyone. That was rejected by Darshan. +3 HP seems stronger. 15:43:54 78700 games for * (cv=0.8 ckiller!=quitting|leaving): sum(dur)=161716728 15:43:59 !lg * cv=0.8 ckiller!=quitting|leaving place=d:1 x=sum(dur) 15:44:00 23191 games for * (cv=0.8 ckiller!=quitting|leaving place=d:1): sum(dur)=2741092 15:44:55 I'd mostly like to know the context(s) of the "too hard". 15:45:10 so 1.7% of playtime is spent on games that end on d:1... not bad 15:45:35 what are those ratios for previous versions? 15:45:59 is that 30% death rate on D:1? 15:46:25 yes 15:46:46 wensley: 2.8% for cv=0.7, surprisingly 15:46:48 !lg * cv=0.7 ckiller!=quitting|leaving x=sum(dur) / place=d:1 15:46:50 155018 games for * (cv=0.7 ckiller!=quitting|leaving): sum(dur)=0.7077923E7/0.254935274E9 (2.78%) 15:47:32 that probably actually is a matter of the game getting easier later on 15:47:38 so people's games tend to last longer 15:47:49 !lg * cv=0.6 ckiller!=quitting|leaving x=sum(dur) / place=d:1 15:47:51 77019 games for * (cv=0.6 ckiller!=quitting|leaving): sum(dur)=0.3170396E7/0.125513825E9 (2.53%) 15:47:52 so these statistics aren't good for much 15:47:58 still interesting, though 15:48:13 !lg * cv=0.5 ckiller!=quitting|leaving x=sum(dur) / place=d:1 15:48:14 135781 games for * (cv=0.5 ckiller!=quitting|leaving): sum(dur)=0.5614578E7/0.256551108E9 (2.19%) 15:48:45 could also be a matter of having a higher proportion of good players for whatever reason 15:49:44 0.7 had the most total games and the highest percentage of time spent on d:1, which seems like it could indicate an influx of new players (rampant speculation) 15:49:45 dpeg: I understand what you mean. We don't want to overdo it and make D:1 trivial. Maybe +2 is enough. 15:49:48 is pan scumming significant enough to give it a noticeable bias? 15:50:01 dpeg: I think the main reason to increase starting HP is that it would reduce how much people pillar dance substantially... the great majority of pillar dancing happens on d:1 15:50:07 only if enough players pan scum 15:50:07 galehar: even +3 is okay, if we know why :) 15:50:35 elliptic: but that seems like a lukewarm solution. Can't we address pillar dancing in a better way? 15:50:59 does nobody else just take it like a champ when hobgoblins wreck them on d:1, rather than dancing? I will probably never streak :P 15:51:03 For example, even with +3 HP we're just pleading "players, please don't dance with the pillars" -- it'd be just as optimal to keep doing it. 15:51:06 well we can change the energy system. But people will just die instead. 15:51:23 galehar: if they die without boring themselves, that is progress 15:51:42 dpeg: there were suggestions the other day here, but nothing anyone was that happy with I think 15:51:50 Wensley: I also never do that (and only have two puny 2-streaks :) 15:52:09 elliptic: we realised that there are two types of pillar dancing 15:52:30 well, I had this idea for changing energy randomisation (not for 0.9 though): each turn, every monster has one chance in 10 to have its speed change. Either by +1 or -1, capped at +2/-2 range. 15:52:33 running away at 2/11 HP and running away at 8/11 HP? 15:53:06 elliptic: to wait for the energy randomisation to kick; or to regain HP/MP 15:53:12 FR: slime walls everywhere. pillar dancing solved! 15:53:24 Wensley: remove the walls. 15:53:26 dpeg: vast majority of the time people are doing the latter 15:53:28 Wensley: I like that, but I also supported slime walls from day one 15:53:47 elliptic: yes, I wasn't aware of the first type until you told me the other day 15:53:57 dpeg: you'll love my alchemist class proposal! 15:54:14 I think I more frequently dance for the randomisation to kick in, actually; maybe about the same 15:54:15 FR: adrenaline: when monsters get more wounded, they move more quickly 15:54:17 I try not to dance, though 15:54:38 can't we stop pillar dancing in some other ways: like growing toadstools from time to time? 15:55:01 FR: player trips occasionally 15:56:05 dpeg: my suggestion from earlier: if you go over the same tile too many times in a row it wears out and becomes a shaft! 15:56:14 dpeg: you can't remove fleeing. It's the basic crawl's strategy :) 15:56:18 Eronarn: similar to my one 15:56:22 I remember one idea about the persuant eventually calling for assistance, which would ostensibly prevent or at least hinder further dancing 15:56:31 monqy: not even sure it needs to be eventually 15:56:36 monsters could stand to be a lot noisier, now that noise is sane 15:56:58 I think there are some ways to attack the problem's root. 15:58:46 we should collect them on a wiki page 15:58:50 let me check if there is one 16:00:55 dpeg: yeah, attacking the problem directly might be possible... I don't think a small increase in HP would necessarily be a bad thing though anyway 16:01:17 dpeg: by the way, what do you think about quad damage being in the main game now? 16:01:30 only the speed randomisation page 16:01:36 https://crawl.develz.org/wiki/doku.php?id=graveyard:speed_randomization&s[]=pillar 16:01:45 (with a funny comment by b0rsuk near the top) 16:01:52 elliptic: I don't understand the point of the thing. 16:02:07 it fails the 200-year rule... 16:02:17 elliptic: what is the gameplay purpose? 16:03:07 quad-damage in the real game seems like a bad idea :P 16:03:13 some vague incentive to use misc acquirement (this is the only way you can get it) 16:03:28 st__: not good enough, imo 16:03:34 dpeg: I think kilobyte just randomly decided to add it to the main game while trying to buff misc acquirement 16:03:42 what, it's already in?! 16:03:49 has been in for weeks 16:03:54 that's absurdly broken 16:04:27 I got it from a lair:8 vault that gives an acquirement item my current game 16:04:29 the real solution is to make better misc items 16:04:53 there are tons of proposals on the misc items page 16:05:05 at least, when I checked last year there were 16:05:05 I'm not opposed to the idea of a misc item that provides a temporary boost to your attacks, but does it have to be a direct reference to a FPS video game? :/ 16:05:33 it's okay where it is in sprint. not in the main game 16:07:48 I initially found it quite funny, but yeah, it's unappropriate. 16:07:59 I am fine to forget the 300 year rule, but I don't see the point of that particular item. 16:08:29 There are some interesting proposals on the misc item page (teleport anchor, and more) 16:08:36 can we at least have a 20-year rule or something :P 16:09:02 with a 20-year rule, we could add quad damage in 2016 16:09:27 I'll get started on implementing lightsabers :P 16:10:02 Wensley: <3 16:10:17 300 is not so bad, it seems :) 16:11:23 problem is we're stocking up on "have to revert this thing done by kilobyte": tornado air/conj, felid lives, quad damage. 16:11:32 I'm not sure how he will take it. 16:12:02 people have suggested reflavouring quad damage as vehumet-related 16:12:05 and he's MiA since yesterday's discussion about tornado 16:12:09 since vehumet likes explosions and all 16:12:14 uh, was there tension? 16:12:27 I came here to talk about Tornado rather than reply on the c-r-d... for kilobyte 16:12:28 xom would have been amused 16:12:33 no, just disagrement up until the point where he stopped responding. 16:12:57 I really liked his idea of removing Vehumet (rather, merging S and V) 16:13:03 ugh no 16:13:09 sure I do :) 16:13:15 and I said so in public 16:13:19 maybe if we get another caster god or two 16:13:20 the idea is good, but it'd leave us short a god, for sure 16:13:26 we need to differentiate them more, merging them would only be a short-term solution 16:13:31 summoner pact god! 16:13:42 elliptic: we removed one dwarf, to dish out another one some versions later 16:13:55 dpeg: 'some versions' can be a while though, see gnomes 16:14:01 do we really want to do that with a god? 16:14:01 sure 16:14:05 yes 16:14:07 dpeg: but currently casters don't have many god choices already! 16:14:32 I just got a mail by b0rsuk where he presumably asks me to remove him from the credits. Sigh. 16:14:41 elliptic: really? 16:14:49 a lot of chars, people are mainly choosing between sif and veh (who play quite differently)... combining them would mean no choice 16:15:04 you're the best player, but I didn't expect you to cling so much to Vehumet-baby :) 16:15:10 they are the go-to gods for conjurer-type casters 16:15:12 dpeg: i had a good thought on the pact god, btw... summons pretty naturally fit into a few types (nature, elements, demons, undead, otherworldly would be good) 16:15:17 dpeg: it's true, just today I taught somone how to !rng to decide between sif and veh 16:15:29 those could be the basis of the pacts 16:15:45 dpeg: I don't particularly like veh design, but this merge would decrease god variety significantly for casters 16:16:04 elliptic: perhaps more casters would consider Makhleb or Fedhas? 16:16:15 do we really want casters considering makhleb/ 16:16:22 sure 16:16:38 giving fedhas more spotlight would be nice, but I feel like they would only turn to fedhas after becoming bored with sifumet 16:16:43 that would basically mean there isn't a caster god more fun for them than 'passively regain HP/MP on kills' 16:16:49 gods should work across playing styles 16:16:54 dpeg: not more... people who want a god who doesn't help with casting or finding books already choose makhleb or nemelex or fedhas 16:17:18 elliptic: ah, so that is covered, good 16:17:30 we could make new vehumet a priority for version 0.9+ 16:17:31 but people mainly like to choose a god on their casters who actually helps with casting 16:17:41 rather than being a good generalist god like those 16:17:53 and this means sif or veh or kiku 16:17:55 they should just go all Kikubaaqudgha, like every self-respecting Crawler 16:18:02 true :) 16:18:45 ash is also useful for everyone in different degrees 16:18:57 right, ash also on the list of generalist gods 16:19:06 You know that you've devoted your little black heart to Kikubaaqudgha, if you can spell his name without thinking: Kikubaaqudgha! Kikubaaqudgha!! Kikubaaqudgha!!! 16:19:46 elliptic: my comment on S/V merge was off-topic, just another bit of dislike against quad damage 16:20:49 well, certainly if vehumet is subsumed by sif then flavouring anything with vehumet is hard :) 16:21:19 yes! 16:21:26 speaking of gods, can we make beogh non-evil already 16:21:44 I am never really sure on these questions, but I just trust dolorous. 16:21:51 ok well, I think I'll go with the +3 HP and see how it plays out. 16:22:07 * dpeg suspects that dolorous has a little miniature Crawl pantheon in his home. 16:22:09 yay 16:22:17 dpeg: any other ideas of something to do about quad damage? it could just be removed but I fear kilobyte will be resistant... 16:22:28 galehar: fine. Would you still tell me who said that it is currently too hard? :) 16:22:50 elliptic: perhaps he should deal with it, people get overruled all the time 16:22:57 elliptic, casmith and eronarn supported the HP boost. 16:23:14 kilobyte did also 16:23:20 he even suggested 4-5 HP 16:23:32 yeah, my proposal was going to be higher - based on my experiences in other games 16:23:32 I forgot to ask MarvinPA about it. 16:23:59 well, I told you the story about the healing fruit in everyone 16:24:05 s starting inventory 16:25:13 My opinion is that the game should be harder, not softer. The mid game much more so, in particular. But I'm not at the helm, thank Trog. 16:25:38 wait who is at the helm 16:25:41 dpeg: I'd agree that the midgame should be much harder, yes 16:26:21 syllogism-: Hi! galehar is in the driver's seat, I'd say. 16:26:34 oh 16:26:48 am I? 16:26:50 :) 16:27:33 It is really good to have someone at the helm, for otherwise we cannot resolve some conflicts (like quads)... and then Crawl becomes NEthack. 16:27:40 it's like a small boat - you can have someone shouting orders but in the end it's all about the oars 16:27:49 there's a difficulty gap in the mid-game, because you can post-pone branch end until you've cleared all the rest. 16:28:24 galehar: there is a proposal to force players to finish some branches by midgame. Not many liked it, but I think it has a lot of appeal. 16:29:02 yeah, I've read that. Not too fond either. 16:29:35 dpeg: lair could still use buffing 16:29:45 moving loot into branch ends, instead of lying around on the floor, could fill a similar role 16:29:46 Making the lower levels of branches such as snake and swamp scale more in difficulty could help. But more difficult monsters also means more XP... 16:30:33 Eronarn: yeah, loot in rune vaults could work. Players are greedy :) 16:30:36 shorter but harder/more branches? 16:30:43 i've long been in favor of making most of the snake loot be in snake 5 in a large guardian naga vault 16:31:05 monqy: we have been cutting levels since quite a while 16:31:19 a good trend 16:31:24 dpeg: tome has cut more than us! we're losing! 16:31:47 Eronarn: also relatively? 16:31:52 elliptic: what do you think about swapping the ranges of throw flame and throw frost? (and bolt spells too). It doesn't make sense that basic ice spells are better than fire ones. 16:32:22 dpeg: actually, it's a much longer game. i'm not sure about who has cut more %-wise 16:32:58 galehar: I'm fine with that, though I'll note that fire is already much less resisted than cold early on, so throw flame has that advantage 16:33:19 You want to make it attractive for players to do branch ends... the simplest solution really is to ask for a rune by D:13. 16:33:55 elliptic: yeah, but ice has throw icicle, refrigeration, freezing cloud... 16:34:21 asking for a rune by d:13 sounds good since it would prevent me from getting shafted past *both* lair and vaults entrances :) 16:34:30 :) 16:35:07 galehar: yeah, I'd certainly agree that ice is better than fire on the whole currently and this should change 16:35:17 dpeg: asking for a single rune at D:13? You know what, I don't remember why I didn't like the idea... 16:35:19 nerfing freeze would help 16:35:43 speaking of things tangentially related to things we're not speaking about anymore, here's a wiki entry for banner suggestions for the upcoming tourney: https://crawl.develz.org/wiki/doku.php?id=dcss:tournament:start#august_tourney_banner_ideas 16:35:48 galehar: fun or serious? 16:35:58 dpeg: I'm serious :) 16:36:01 hey! 16:36:19 elliptic: you can get shafted past D:13 just fine. No problem, only need to bring a rune if you want to go to D:12 :P 16:36:33 dpeg: haha 16:36:33 dpeg: make d:14 a special level that requires a rune to enter. 13 dungeon levels before, 13 dungeon levels after. and make d:14 itself especially challenging 16:36:49 it'd also be nice to add more non-damaging fire spells; the schools should feel weighted but right now fire's useful spells are either conj-ish damage or not actually useful 16:36:58 Wensley: yes, something like this (another awesome possibility for story-building-without-words) 16:37:05 make it a big castle full of monsters! 16:37:08 no wait... 16:37:12 haha 16:37:16 troll bridge 16:37:21 filled with rune trolls 16:37:22 galehar: level entirely full of bees! bees as far as the eye can see 16:37:24 the River Styx. 16:37:25 they're made of runes 16:37:35 Eronarn: I like the approach (spearheaded by galehar?) that Fire _is_ about damage spells 16:37:54 dpeg: have you seen the inner flame spell? 16:37:58 Eronarn: I'm against that. I think fire should stay the "damaging school", have the least utility spells and the best damage. 16:38:00 yes, awesome! 16:38:01 dpeg: well, damaging is fine, but see conjure flame or ring of flames - not solely about doing damage at range 16:38:15 yes, but no utility spells 16:38:25 -!- bmh has joined ##crawl-dev 16:38:25 -!- bmh has quit [Changing host] 16:38:25 -!- bmh has joined ##crawl-dev 16:38:28 galehar: i agree it should have the least utility spells and the best damage, i just think it should have *some* more utility-ish spells (though preferably ones that cause damage indirectly, such as a damage shield effect) 16:38:33 hi 16:38:33 bmh: You have 1 message. Use !messages to read it. 16:38:35 Fire should be the caster's Trog. Or so. 16:38:36 !messages 16:38:41 (1/1) Wensley said (22h 20m 3s ago): Do you think that the new abyss is good enough to be a candidate for merging this week? 0.9 cutoff is rapidly approaching and I'd love to see it get in :D 16:38:51 when is the 0.9 cutoff? 16:39:06 early july, vaguely? 16:39:12 Eronarn: well, as you said, there's conjure flame, ring of flame and there's now inner fire. 16:39:21 who of us saw the new Abyss in action (I didn't)? 16:39:28 I did, and I approve heartily 16:39:35 it is the best new feature since nomes! 16:39:36 galehar: yes, those are good - i think we could use 2 or so more like that 16:39:40 elliptic: another fire balance. What about fireball? It has been suggested to reduce it to L5. 16:39:45 galehar is totally right about modifying rot 16:39:55 Wensley: I still don't understand that vault behavior you saw 16:40:10 bmh: you mean colors changing or vaults decaying over time? 16:40:17 either 16:40:22 I haven't been able to reproduce them 16:40:28 bmh: we don't have a defined date yet, but beginning of july. 16:40:39 well, neither was a deal-breaker :) as long as the AK starting abyss can be made usable, I see no problems 16:40:55 in fact, with regard to vault decay, that was exactly what I wanted to happen :) 16:40:56 hm... what about changing AK spawning 16:40:56 galehar: I tried fireball out yesterday for the first time in quite a while and it seemed pretty good as is... and I think marvinpa agrees with that 16:41:04 Start the AK in a real entry vault and then Abyss them 16:41:12 that way there would be an abyssal entry vault :-) 16:41:18 elliptic: one thing i'd be in favor of for fireball is making the center tile do more damage 16:41:21 it would definitely be too strong at L5 without a nerf 16:41:24 I could replace the upstairs with gateways out of the abyss. 16:41:34 bmh: sounds good 16:41:38 elliptic: yes he does. So we'll leave it at that. 16:41:40 thanks 16:41:44 bmh: that sounds cool 16:41:51 so: fireball aimed directly at one person = more damage than bolt of fire even though same level; corridor = bolt of fire is better; crowd = fireball is better again 16:41:55 not all entry vaults may qualify, but we could solve that by TAG 16:42:01 Eronarn: yes, some more variety in dealing damage with fire can be nice, just like marvin did with inner fire. 16:42:04 galehar: it could maybe use a small buff to damage 16:42:09 but not much 16:42:17 hmm 16:42:18 ok 16:42:18 that segment of code is just weird though. I-started-in-the-Abyss is passed by some weird GDT flag 16:43:23 galehar: I don't know why you saw tons of items in that game. I haven't touched the item code in the slightest 16:43:33 And the rate of entombment is very low 16:43:40 bmh: probably just luck :) 16:44:47 https://crawl.develz.org/wiki/doku.php?id=dcss:brainstorm:misc:pillar_dancing 16:44:58 eronarn: remember that fireball is undodgeable and bolt of fire isn't... that means it shouldn't be more damage 16:45:12 it could be better at center, but I'm not sure 16:45:53 elliptic: could you comment on the wiki page at some point (not urgent)? 16:46:53 sure 16:47:18 thanks 16:48:34 I don't think I can hit a July 1 drop dead date on the abyss. 16:48:52 wensley: thanks for putting the banner ideas on the wiki... I'll certainly look back there for suggestions at some point 16:48:53 If I can get a little guidance, I think I can hit July 3 16:49:24 bmh: what are the problem left? AK? 16:49:35 galehar: AK, maprot, sanity checking 16:49:38 elliptic: really just condensed a little bit of old banner discussion from the wiki with the banner ideas from the learndb and the banners from the previous tournament, it is open for new additiions 16:49:40 and state persistence 16:49:48 wensley: a few of the banners listed in the learndb have been changed now, by the way... sorry I didn't do that earlier, but I was distracted by discussion here 16:50:11 ah, well, the point of all this was to have somewhere centralized to keep everyone up-to-date :) 16:50:59 bmh: seems too short to me. Let's take our time and aim at 0.10. We need some time to test it. 16:51:23 galehar: The only code path I've hit is apply_terrain 16:51:46 galehar: now that the tournament has been pushed from the presumed august 1 to a likely august 13, could we also push the cutoff date back by a few days? it also depends on the status of the fight rewrite and the skill system changes 16:51:47 still, I'd rather have it in and be awesome, than have it be half-assed 16:52:20 Wensley: oh, I didn't no about the tournament date shift 16:52:35 so yeah, it makes sense to push the freeze date too. 16:53:14 bmh: I'll try to have a look at your code, but I'm not making any promise. What was the problem with AK exactly? 16:53:16 yeah, elliptic suggested august 13 - 28 for a three-weekend tournament. 16:53:25 galehar: AKs get spawned in a box :-) 16:53:36 but the box isn't a box for long :P 16:53:36 ok, seems easy to reproduce :) 16:53:45 blech. my local branch just got screwed up 16:54:01 really the only problem is that it encourages spoilers to tell all new AK to "press 5 once to escape this box" 16:54:05 I ended up with modifications to crawl-ref/source/contrib/sdl that I could reset and now I'm not rebasing cleanly 16:54:11 Wensley: 2 weeks and 3 weekends is a great idea. 16:54:34 it worked well for the may tournament :) 16:55:02 ah-ha, fixed 16:55:06 will the new skill system make it in time, or do you think it will have to wait? 16:55:12 bmh: phew, you had me worried :P 16:55:13 bmh: kilobyte made an update in the sdl contrib. try git submodule sync (and then update) 16:55:29 galehar: done :-) 16:56:03 Wensley: you mean manually selecting skills to train? 16:56:17 galehar: yes 16:56:35 oh, by the way, is there any hope of getting CAO trunk updated sometime? since it being 2 months old is sort of bad 16:57:16 Wensley: 2 weeks... I'll give it a try. 16:57:28 worse than not having CAO trunk at all, really 16:57:28 also remove erocrawl 16:57:31 galehar: :D I'll test it for you nonstop if you push it out 16:57:37 because people keep playing it for some reason 16:57:37 but now that 16:57:57 now that I'm in the driver's seat, I have less time to code :) 16:58:01 isn't CDO's trunk version updated via cron jobs? would it require greensnark to get the same thing for CAO? 16:59:19 ugh I just saw kilobyte's CRD e-mail about tornado 16:59:55 all full of "bad idea" and "terrible idea" 17:00:00 :( 17:02:09 does Labs timed rot? 17:03:01 elliptic: because pure Air _is_ terrible idea? The only other thing I named a bad one can be done in a different way that doesn't hurt theme. 17:03:11 kilobyte: [citation needed] 17:03:38 elliptic: for Quad Damage: I guess this is mostly a megabat-type issue, but I won't fight hard 17:04:14 elliptic: also, you seem to be too keen about adjusting the numbers without fixing code 17:04:39 this includes outright bugs like keeping tornado up while translocating 17:05:04 which you listed among "typical spells being cast while it's on" 17:05:35 outright bug? we've discussed it before and decided it wasn't clearly a bug 17:05:43 kilobyte: shatter is stronger than any L7, L8 spell 17:05:44 it is a grey area, sure 17:06:10 I don't mind blinking canceling tornado but don't blame me for not fixing it 17:07:05 if the winds were in the original place, how come they instantly appear in the new one? 17:07:07 ok, the first thing to sort out is the schools. What's wrong with tornado being pure air and balanced like shatter? 17:07:30 as elliptic said yesterday, not every L9 should be like fire storm. 17:07:35 that you can get it in mid-D 17:07:51 especially with Ashenzari 17:08:20 hey, I think tornado's a really cool spell no matter what and wish that people would relax and be jovial :) library and hamburgers time o/ 17:08:20 this does not mean it is a no-brainer: you could link duration to spell power -- even if you can cast it at mid-D, it might not be good enough 17:08:37 I'm not sure ashenzari helps more than wizardry of Veh for casting L9 17:08:44 skill boost is low at high level 17:09:05 galehar: where should I look for fixing up abyssal maprot? 17:09:25 bmh: sorry, no idea 17:09:52 dpeg: unless there's massive anti-stepdown, it is good enough for one-shotting anything you can find in lair branches or most of V, tough monsters start appearing only a good deal later 17:10:37 but how can Shatter be balanced when Tornado is always overpowered as a single school spell? 17:10:38 kilobyte: why couldn't it be balanced like shatter is? I keep hearing that shatter is well balanced and useful for the whole game; 17:10:44 is it because of the large area? 17:10:49 tornado isn't known for one-shotting things 17:10:57 it takes a few turns at least to kill monsters 17:11:16 that's what make it different and interesting 17:11:27 yeah, I meant killing in one cast, obviously 17:11:46 galehar: that's why your idea of having it wind up is a good one 17:12:34 -!- Wensley has quit [Ping timeout: 246 seconds] 17:12:38 if it is too good, we can change it... saying that it can't possibly ever be balanced at L9 air is rather much unless you have a logical proof of this somehow 17:13:03 Any hope for nerfing Mephitic Cloud for 0.9? 17:14:04 bmh: no! 17:14:04 kilobyte: good. can you do the wind up code? I'll try to finish new skill training in time. 17:14:18 dpeg: as in it won't be nerfed, or it won't be nerfed in time? 17:14:36 galehar: of course, this is far easier than other changes I have in mind 17:15:45 kilobyte: what's wrong with having tornado weaker than firestorm? 17:16:43 ok, given that I'm not going to hit 0.9, I'll try to concentrate on improving the code quality rather than tacking on more features 17:16:48 elliptic: it's mostly about casters dedicated to one school vs balanced chars. The latter are far more fun to play. Trying to balance for both Swamp and Ziggurats means there'd have to be a superlinear dependency on power, which reduces most of the incentive to use it by non-one-school conjurers. 17:17:12 so now your personal notion of "fun to play" determines all crawl design decisions 17:17:23 bmh: if the tourney would start in mid-Aug, there's still some time 17:17:53 also, since when do we worry as much about ziggurat balance as swamp balance 17:18:29 elliptic: there's so many other spells. What's the problem with being useful for more than one type of build at the cost of not being available in midgame? 17:19:07 Everyone seems to agree that Aug 13-28 is great. Tournament starts Aug 13th. Code freeze is 13th july. 17:19:11 I do care more about improving full games rather than 3-runers; there are mid-level spells which can do 3-runers better. 17:19:19 bmh: not for 0.9 17:19:24 needs a nerf, of course 17:21:29 kilobyte: most people don't do "full games" 17:21:41 kilobyte: I believe that it can still be useful in Zig even if pure air and nerfed. Doing Zig by spamming the same overpowered spell over and over doesn't sound like the most exciting thing to do. 17:22:01 galehar: this can be addressed by stretching the duration 17:22:05 marvinpa also is on record for saying that it could still be useful in zig at a much lower power level 17:22:16 and he has completed more zigs than anyone 17:22:25 uhm, it _already_ has half the power it had in 0.8 17:22:29 ... 17:22:31 so? 17:22:36 completing zig... darn freaks 17:22:40 if people did zig by spamming fire stom before tornado arrived, the fix is not to replace it by another OP spell. 17:23:18 kilobyte: yes, it has been nerfed. yes, it needs to be nerfed more. 17:23:27 so we can make it pure air again. 17:24:00 the initial aim for Tornado was to have a roughly equal alternative to storms that has similar power but a different mechanics 17:24:20 can we please settle on it being pure air since nobody but kilo disagrees with that yet it's somehow been a recurring issue for literally months 17:24:21 pure air doesn't allow similar power, unless you care only about one character build 17:24:40 Eronarn: that is not really helpful 17:24:57 I used Tornado but never really the Storms... wouldn' 17:25:05 t it be better to think about nerfing the latter? 17:25:06 kilobyte: but why do you insist on having it on par with storms? 17:25:06 dpeg: sure it is, if it kills the discussion on the topic 17:25:41 If the storms are overpowered, then we shouldn't worry about how to get Tornado up to their level, but rather how to get them down. 17:25:46 we don't need a third big gun. 17:26:08 galehar: I don't insist on having it be equal in power -- it might instead capitalize on synergy with other spells in that school -- but I do insist on using that casting difficulty niche 17:26:08 dpeg: maybe. But we don't have to aim for the same power level 17:26:14 also it's interesting to have spell combos that don't already exist 17:26:14 kilobyte: why does the initial aim matter so much? you made the spell, great, it is very cool... but it is months later now, and I haven't heard a single person other than you who wants it to be conj/air 17:26:19 there are already two high level air/conj spells 17:26:26 especially since storms are dual school and we'd rather have tornado single school 17:26:38 if tornado is a third, you'll just pick them up in a row, whereas if tornado is pure air, different characters will pick it up at different times 17:26:51 Eronarn: these spells are bad jokes currently 17:27:13 kilobyte: the best spell in the game for doing damage to cerebov quickly is a joke? 17:27:27 kilobyte means badly overpowered, I think 17:27:28 Eronarn: having Chain Lightning be _harder_ to cast than Tornado means no one will ever bother 17:27:38 ah, the other Air/Con spells, sorry 17:27:57 we cannot solve everything at once, let us leave the other Air spells for later 17:28:09 chain lightning is harder to cast than tornado? tell that to conjurations-heavy chars 17:28:36 elliptic: it is rarely ever useful, about everything late game resists elec 17:28:59 make less stuff resist elec? 17:29:05 kilobyte: I've used chain lightning in plenty of places 17:29:07 kilobyte: if those two spells are so broken that we should ignore them in balancing, why are you arguing that we can't have pure air tornado because it will step on their niche (which doesn't even make sense as an argument anyways since they have different skill profiles) 17:29:08 elf, pan, zig 17:29:21 galehar: I could try to make maprot rot only those tiles that have changed 17:29:26 I don't know if that leaks too much 17:30:36 bmh: you can start by using the same code as labs maybe. It needs to be improved, but it would be better than the current no map at all. 17:30:58 and I think it needs to rot more than that. 17:31:34 also reminder that conj/air tornado means you can go freezing cloud -> ice storm or tornado -> the other one and use both, if you actually care about thinking about zig balance 17:31:42 ok, I can rot 2 or 3 squares for every square that's changed. That should contribute to a very lively rot 17:33:23 kilobyte: even pure air, if you want to cast it mid-game, it needs dedication and sacrifices. Currently it works because the spell is still OP. But balanced properly, it won't be enough on its own. 17:33:51 What happens if you cast tornado and there are other clouds around? 17:34:11 it twirls them around 17:34:14 bmh: can't you just rot with time? 17:34:25 that's cute 17:34:46 galehar: don't know. I've never touched the maprot code. 17:35:08 heavy pure air on a hybrid sounds fun 17:35:28 kenku or something 17:35:34 galehar: I'll dig through the code when I have a chance. night 17:35:36 -!- bmh has quit [Quit: bmh] 17:35:44 (does tornado deflect ranged stuff yet?) 17:36:14 well yeah, AE are usually hybrids. So focusing only on air to cast it midgame seems pretty hard to do. 17:36:34 luckily, kenku have high air apt and high weapon apt 17:36:54 and you know what? i am Totally 100% Okay with kenku AE hybrids being good at tornado 17:37:34 !apt air 17:37:34 Air: Ce: -1, DD: -3, DE: 1, DG: -1, Dr: 0, Dr[black]: 2, Dr[green]: 0, Dr[grey]: 0, Dr[mottled]: 0, Dr[pale]: 1, Dr[purple]: 0, Dr[red]: 0, Dr[white]: 0, Dr[yellow]: 0, DS: -1, Fe: -1, Gh: -2, Ha: 1, HE: 2, HO: -2, Hu: 0, Ke: 3!, Ko: 0, LO: , MD: -2, Mf: -2, Mi: -3, Mu: -2, Na: 0, Og: -3, OP: , SE: 1, Sp: -1, Tr: -4*, Vp: 0 17:37:57 so the only ones with even a shot of doing this are HE, Ke, one kind of drac 17:38:02 very OP races 17:40:39 Eronarn: so you want only specific races to do this, while my aim is to make it non-niche 17:41:38 it wouldn't be niche pure air 17:41:43 what 17:41:48 ugh, stop being terrible 17:42:00 kenku would be able to get it earlier, but everyone can learn and cast tornado 17:42:01 Eronarn: but you too! Let's calm down... 17:43:13 Proposal: the wind up change. We try 0.9 with pure Air tornado -- the tournament will give it a lot of feedback. Afterwards we evaluate and decide whether to make Tornado Air/Con or to nerf the Storm spells. And whether the other high Air spells need a change. Okay? 17:43:38 'decide whether to make Tornado Air/Con or to nerf the Storm spells' - since when was this ever what we were trying to decide 17:43:59 dpeg: ok, but when it sucks like it will do, it's all your fault! 17:44:12 dpeg: sounds good to me at least :) 17:44:33 yay 17:44:37 pardon me if I'm missing something, but what does nerfing the storms have to do with balancing tornado? 17:44:44 nothing 17:44:49 ("suck" not as in "be underpowered" but "either too strong or too niche") 17:45:11 too niche: L9 air spell 17:45:16 not too niche: L9 air/conj spell 17:45:17 monqy: balance, they're supposed to have around the same level of utility or people will use only a particular school 17:45:18 kilobyte: yes, I will take the blame. And I will be fair when evaluating the spell. (I have been able to pull off Tornado in the last tournament, I'll do it again in the next.) 17:46:08 monqy: it should be obvious: if the storms outshine tornado, the latter sees little use 17:46:27 kilobyte: I completely disagree. The elemental schools are different. They don't need to have the same spells with different colours. 17:46:31 kilobyte, dpeg: so it's been decided that it should be around storm-level rather than shatter-level? 17:46:38 no 17:46:51 galehar: the more spells differ, the better -- but the utility level should be the same 17:46:53 I don't think so, but the power of one spell can clearly affect other spells 17:47:03 dpeg: currently the storms are clearly stronger than shatter but they both see decent use (because shatter is easier to cast and different schools) 17:47:21 also because it's not like a MDEE is ever going to cast ice storm 17:47:34 galehar: like, if CBL was actually useful, having the two work well together is interesting 17:47:38 elliptic: yes, and that is fine. It is still a valid question whether the storms are too strong. But let's leave that for later. 17:47:46 dpeg: yes 17:47:54 especially fire storm 17:48:31 kilobyte: I think fire (and earth) should be the more damaging schools. Air has lot of utility spells. 17:48:51 and it relevant to think at least a bit about it right now because Tornado is a competitor of Fire Storm, and if the latter is a bad target to aim at (no matter whether we're talking utility or power), then we should be aware. 17:49:22 tornado is only a competitor of firestorm because it's been pushed that way 17:49:25 there is no reason why it has to be that 17:49:29 it's the only spell that is a real Zig winner. You can claim that "but I did a Zig with spell X!", but currently only three spells can give you a real chance of finishing a Zig:27 pan level. 17:49:51 Of course: having four level 9 spells does not mean that they should even out -- the Fire can deal most damage, and the Air one do considerably less, given what Air can do in principle for you. 17:49:59 s/the only/obvious what I meant 17:50:19 * dpeg is not happy that people storm ziggurats. That was explicitly not the plan. 17:50:36 dpeg: I'd _guarantee_ a Zig:27 pan level 17:50:46 kilobyte: would that help enough? :) 17:50:54 if storming ziggurats isn't the plan perhaps there shouldn't be panlord zig:27s 17:51:01 just sprinkle moths on Zig:27 17:51:04 or at least that either Zig:25, Zig:26 or Zig:27 is a pan level, so you don't know what to expect 17:51:13 Eronarn: the idea what that it is endless 17:51:21 huh? 17:51:22 *was 17:51:28 for the record, you can ninja past pan lord zig:27 levels without a storm or tornado in a few ways :) 17:51:48 elliptic: "finishing" 17:52:13 not kill everything, sure... but since when is killing everything the main point of crawl? 17:52:19 elliptic: Death's Door can get you a long way, but won't actually kill anyone 17:52:25 kilobyte: escaping and trying another day with the hope of not getting a pan lord zig:27 is also possible. 17:52:52 i think it's pretty reasonable to say that zig levels that are unkillable without storms (even if beatable other ways) are kind of lame 17:53:32 * dpeg puts on the broken record: no channelling in ziggurats! 17:53:44 Eronarn: +1. If zigging is about spamming the same spell over and over, the problem is with zigs. We don't need to balance all the high level spells because of this. 17:54:00 i'd rather see zigs be trivial for ubercharacters and not that hard overall than see them made uberhard for ubercharacters and undoable for anyone else without ninjaing 17:54:08 galehar: no, the problem could be elsewhere 17:55:13 dpeg: no, the problem is definitely with zigs, panlord zig 27s are nuts 17:55:36 I'll leave judgement to someone else. 17:56:18 Just a remark: Eronarn, your comments are usually "this is like this, _definitely_" -- no conjunctives, no "in my opinion", no "perhaps" -- how can you be so sure about all these things? 17:56:18 Well it does sound insane 17:56:32 how many pan lords are there in a Zig:27? 17:56:37 dpeg: because i have opinions? obviously anything i'm saying is my own view 17:57:00 galehar: 30-40ish, I forget exactly 17:57:48 (and they can have storms, of course) 17:58:07 !lm * place=zig:27 x=cdist(name) 17:58:07 666 milestones for * (place=zig:27): cdist(name)=59 17:58:14 another Zig oddity: you know how much loot you'll get beforehand: a Zig that costs 9k to enter has several times as much as a 1.5 one 17:58:38 only 59 people (with some duplicates) have even seen a zig:27 online... let's not make it become a major balancing point for every high-level spell :) 17:58:56 elliptic: of course not, but it could still indicate that something is off 17:58:59 kilobyte: yeah, that's pretty strange 17:59:43 huh, zig loot scales with entrance fee? 17:59:45 -!- edlothiol has quit [Ping timeout: 255 seconds] 18:02:08 is there any reason to keep this? 18:02:17 not unless bazaars/troves do it too 18:02:31 bazaars don't have a fee 18:02:52 people can scum Pan longer for a cheap Zig, but then, if they want multiples, amount of loot doesn't really matter 18:02:59 oh, right, oops 18:03:06 playing mtg on my other monitor :) 18:03:44 kilobyte: I'd be fine with scrapping it, yeah... I don't think anyone expects it and it doesn't really serve much purpose 18:03:48 kilobyte: you can change it, but I did it on purpose (thinking about D ziggurats, of course) 18:04:31 hi deep leg 18:04:37 hey, a due 18:04:44 not an urgent thing, though 18:04:51 due: about to leave though 18:05:36 before you do: are there any plans for CAO? 18:05:53 having three months old trunk there does more harm than good 18:06:37 kilobyte: I have nothing to do (and no plans) for CAO -- I am a European :) 18:06:51 what prevents us from upgrading CAO? 18:07:30 I doubt it would be a sane thing for rax to give full access to random devs 18:07:41 and without maintenance, it goes to rot 18:08:01 just look how much effort Napking puts into CDO 18:09:58 -!- syllogism- has quit [] 18:11:10 What do you need on CAO? 18:12:24 * kilobyte claps wildly for rax. 18:12:59 current problem: the trunk build there comes from some time before the 4th Crusade 18:13:08 Hi rax! Applause!! 18:18:47 what is needed for trunk: regular updates (and someone who can fix problems if it gets wedged), and a way to trigger a rebuild if an urgent bug gets in 18:22:31 Okay. Uh, if someone can write me documentation that tells me what to do, I can do that 18:22:56 But I don't know how to build crawl anymore and I am terrified of screwing it up. 18:23:33 the person who knows the new CAO setup is greensnark... 18:24:18 Napkin knows CDO well, and I guess they're similar now, but I don't know the details. My testing DGL rig did not use the new goodies. 18:24:37 it would be possible to ask greensnark for help 18:25:57 If greensnark could write something up, that would be amazing. In general I'd like to get more stuff documented so that I can hand instructions to a couple of other people who I trust. 18:27:49 email to Darshan sent 18:38:35 03galehar * r9d1860e8cd42 10/crawl-ref/source/scripts/training_simu.lua: Update the training simulator with Cr -> Sk. 18:38:45 03galehar * r6826cf6aacca 10/crawl-ref/source/initfile.cc: Allow redefining crawl_dir for builds with SAVE_DIR_PATH (#4137). 18:38:45 03galehar * r8ee289483b89 10/crawl-ref/source/ng-setup.cc: Increase base HP by +3. 18:39:59 cTele nerfs (https://crawl.develz.org/mantis/view.php?id=4182) by dpeg 18:42:28 rax: not sure whether you've already heard this, but the plan for the tourney is to have it run for just Aug 13-28 (same length that the May tourney was)... how does this sound to you? 18:43:40 -!- ainsophyao has joined ##crawl-dev 18:45:01 D command for dropping item last picked up (https://crawl.develz.org/mantis/view.php?id=4183) by dpeg 18:45:49 -!- ainsophyao has quit [Remote host closed the connection] 18:46:25 galehar: you could have changed just this line: you.hp_max_perm = hp - 5; 18:47:40 lol 18:48:03 elliptic: If that's what people prefer, that's fine with me. I think it changes some of the "most of X" trophies around a little bit but that may not be a bad thing. 18:48:37 kilobyte: good point. I'll fix it. or do it if you want. Bed time for me. 18:48:58 rax: "most of X" meaning stuff like "most combo high scores" and "most uniques killed"? 18:49:04 -!- galehar has quit [Remote host closed the connection] 18:49:23 folks, I added some Implementables -- I hope they are uncontroversial 18:49:54 if not, please get in touch with me, don't discuss in the Mantis item (that should be reserved for volunteers speaking up) 18:49:58 can I start complaining before I read them? (jk) 18:50:03 Shift-O command for autoloot (greedy, non-exploration) (https://crawl.develz.org/mantis/view.php?id=4184) by dpeg 18:50:04 :) 18:50:18 If you have other ideas for Implementables, tell me, or write them yourself :) 18:51:07 still no mail about them 18:52:56 elliptic: Yes. 18:53:57 oh 18:54:07 i have parts of a working rst2txt converter 18:54:14 kilobyte: (re: your mail to the list) 18:55:11 due: <3 18:55:50 due: any big dependencies? 18:56:32 kilobyte: just docutils so far 18:59:14 due: ie, python and tex ): 19:00:06 Autoswitch launcher/weapon (https://crawl.develz.org/mantis/view.php?id=4185) by dpeg 19:00:48 kilobyte: well, yes 19:01:56 kilobyte: that is the native platform for rst 19:02:53 it's quite a problem since you can't do that during regular builds 19:03:07 what, python? 19:03:20 and prebuilds suck... in some places they are forbidden, too 19:03:54 no python on msys, default BSD, Mac 19:04:16 Debian unless you manually install it, too 19:04:27 and tex is even bigger 19:05:04 IIRC it's a matter of several lines of sed, too 19:05:08 Dangerous monster warning (https://crawl.develz.org/mantis/view.php?id=4186) by dpeg 19:05:38 debian doesn't default with python? 19:06:49 dpeg: the first two were good, in this one being true by default would be really spammy 19:09:06 due: only Ubuntu does 19:09:07 of course, you already need a number of dependencies so on Debian there'd be no extra burden 19:09:07 but making msys and Mac builds harder is not nice 19:09:50 -!- blackpenguin has quit [*.net *.split] 19:09:50 -!- Adeon has quit [*.net *.split] 19:09:50 -!- elliptic has quit [*.net *.split] 19:10:09 Announce (https://crawl.develz.org/mantis/view.php?id=4187) by dpeg 19:10:31 kilobyte: no, I don't think so. Did you look into it? First, it is about another line in the _description_. Then there is an option to enhance the see-monster message by a sentence. 19:10:31 but I am done for today, gotta sleep 19:10:31 what is wrong with python and tex: you only need it on one dev machine, right? 19:11:05 kilobyte: but we really only need to run the script once 19:11:24 couldn't it put the txt version into trunk? 19:12:21 dpeg: having generated data in version control is not nice, yaccage already causes problems at release time 19:12:26 due: also yay! 19:13:07 kilobyte: could the build scripts download the txt file from the internet, if ordered to do so? 19:13:30 -!- blackpenguin has joined ##crawl-dev 19:13:31 -!- Adeon has joined ##crawl-dev 19:13:31 -!- elliptic has joined ##crawl-dev 19:13:34 Debian explicitely forbids shipping things not built from source, too 19:14:28 okay, won't solve this tonight :) 19:14:30 -!- dpeg has quit [Quit: zzz] 19:15:21 kilobyte: then just store the pre-built manual as text as well as rst? 19:16:37 same as yaccage, and it's updated a lot more often 19:17:13 but lemme see if we're not overengineering things 19:17:19 -!- Textmode has joined ##crawl-dev 19:30:29 -!- MarvinPA_ has joined ##crawl-dev 19:34:12 -!- MarvinPA has quit [Ping timeout: 276 seconds] 19:34:29 -!- Pacra has quit [Ping timeout: 250 seconds] 19:40:18 Disappearing Snake Rune of Zot (https://crawl.develz.org/mantis/view.php?id=4188) by cybersaint2k 19:44:34 we really do need to improve the rune display somehow 19:45:08 was pressing } going to display them? 19:47:09 that'd be cool 19:47:20 i still favor ascii art for displaying which ones you have though :( 19:47:40 currently } displays something else, I forget what except that it is one of those keys I never use 19:48:25 expanded quiver, maybe...? 19:50:29 MarvinPA_: think we can get cause fear as a duration buff into .9? it's not like anyone currently uses it 19:51:37 (it might also be cool to add warding style flinching to fear, if you attack the thing you're afraid of? or maybe even make warding cause things it wards to be afraid of you) 19:56:24 oh another thing we should get into .9: fixing spriggan speed 19:58:47 we'd have to break it first 19:59:16 kilobyte: congratulations, you just did... months ago 19:59:24 @??spriggan berserker 19:59:24 spriggan berserker (04i) | Speed: 16 | HD: 12 | Health: 27-45 | AC/EV: 2/25 | Damage: 21 | Flags: see invisible | Res: 06magic(112) | Chunks: 07contaminated | XP: 1094 | Sp: brothers in arms, hand, berserker rage. 19:59:37 ^^ note the broken speed ^^ 20:00:29 and...? 20:01:19 I think what eronarn means is that player spriggans don't act 1.6 times more quickly than player humans 20:01:31 but monster spriggans do 20:01:55 because monster weapon delay is broken in general 20:02:48 and as i said earlier, fix weapon delay in general rather than trying to fix it in this specific place in a way that shoves the problem somewhere else (now spriggan mages are 1.6x at casting spells) 20:03:32 couldn't you make spriggans be 1.6x at moving and melee but not spellcasting? I seem to remember this is possible 20:04:01 so are all other monsters with speed != 10 20:04:12 elliptic: if monsters not getting faster than 1 attack speed with high HD and a shortblade is a problem, then we should fix that problem 20:04:21 rather than make spriggans attack faster just because shortblades are fast 20:05:22 kilobyte: yes and that certainly creates problems too (double breaths each multi-bouncing when going around a corner into a storm dragon), but this one is particularly offensive because the player can play as a spriggan 20:05:32 then what, executioners are broken because they can cast 6 times while your CB char gets to move once? 20:06:45 executioners can't cast anything useful, and their gimmick is being globally much faster than you 20:06:46 I guess we should revert making a number of monsters speed 10 then 20:07:10 how does that follow at all 20:07:19 (it doesn't) 20:07:55 because you're complaining that monsters may have a speed other than 10 20:08:16 or that non-10 aut turns exist at all 20:15:35 no 20:15:39 here are the things i am saying: 20:15:55 1) it is bad for player spriggans and monster spriggans to have wildly different speed capabilities 20:16:39 2) making a monster 1.6x faster doesn't make it 1.6x harder, but rather more than that, and it makes it less predictable, particularly if the monster has spells or can hit for very high damage in melee in a single round 20:17:15 3) if spriggans are supposed to get multiple quick attacks because of using shortblades, any other monster using shortblades should do that, and that's something fixable 20:17:49 -!- evilmike has joined ##crawl-dev 20:20:44 -!- ainsophyao has joined ##crawl-dev 21:27:30 -!- MarvinPA_ has quit [Ping timeout: 276 seconds] 21:39:04 -!- Pacra has joined ##crawl-dev 21:55:56 -!- Sequell has quit [Remote host closed the connection] 21:56:40 ryw (L8 SETm) ERROR: range check error (52 / 52) (D:6) 21:56:50 -!- Sequell has joined ##crawl-dev 22:08:27 03elliptic * rbec177b56670 10/crawl-ref/source/ (6 files): Add a milestone for reaching ****** piety. 22:10:36 -!- ainsophyao has quit [Remote host closed the connection] 22:33:48 Unstable branch on crawl.akrasiac.org updated to: 0.9-a1-751-gbec177b (32) 22:48:00 -!- Pacra has quit [Read error: Connection reset by peer] 22:48:25 -!- Pacra has joined ##crawl-dev 22:58:58 -!- petete has quit [Remote host closed the connection] 23:28:01 -!- ghallberg has quit [Ping timeout: 244 seconds] 23:28:19 -!- ghallberg has joined ##crawl-dev 23:51:20 -!- hoody has joined ##crawl-dev 23:56:36 -!- st__ has quit [Ping timeout: 255 seconds]