00:01:23 Shotax the Spry (L15 HaSk) (Lair:4) 00:02:04 03elliptic * r5831804c1085 10/crawl-ref/source/evoke.cc: Fix some other mistakes with skill_rdiv(). 00:02:19 !lm shotax crash -log 00:02:19 1. Shotax, XL15 HaSk, T:32263 (milestone): http://crawl.akrasiac.org/rawdata/Shotax/crash-Shotax-20110926-050123.txt 00:02:58 that looks strange 00:03:10 Unstable branch on crawl.develz.org updated to: 0.10-a0-919-g7dd7dc6 (32) 00:18:37 Windows builds of master branch on crawl.develz.org updated to: 0.10-a0-920-g5831804 00:22:24 -!- mumra has joined ##crawl-dev 00:27:37 -!- dtsund has quit [Ping timeout: 252 seconds] 01:10:00 -!- Keskitalo has joined ##crawl-dev 01:36:11 -!- ainsophyao has joined ##crawl-dev 01:37:20 -!- mikee_ has joined ##crawl-dev 01:40:08 -!- capablanca has joined ##crawl-dev 01:40:56 -!- capablanca has quit [Client Quit] 01:41:56 -!- mikee_ has quit [Ping timeout: 256 seconds] 01:43:08 -!- mikee_ has joined ##crawl-dev 01:45:48 -!- mikee_ has quit [Client Quit] 01:47:37 -!- ixtli has quit [Quit: Computer has gone to sleep.] 01:54:59 Unstable branch on crawl.akrasiac.org updated to: 0.10-a0-918-g4e0edf4 (32) 02:21:53 -!- galehar has joined ##crawl-dev 02:31:47 hi 02:37:34 -!- edlothiol has joined ##crawl-dev 02:42:04 03elliptic * r2108932a94f1 10/crawl-ref/source/player.cc: Make SH value non-random. 02:47:40 -!- Danei has joined ##crawl-dev 02:50:36 -!- Keskitalo has quit [Quit: Lost terminal] 03:02:44 -!- edlothiol has quit [Ping timeout: 276 seconds] 03:12:13 -!- kilobyte has quit [Ping timeout: 240 seconds] 03:13:10 -!- kilobyte has joined ##crawl-dev 03:21:10 -!- edlothiol has joined ##crawl-dev 03:25:34 -!- edlothiol has quit [Ping timeout: 255 seconds] 03:41:48 -!- monqy has quit [Quit: hello] 03:44:44 -!- dpeg has joined ##crawl-dev 03:53:46 !seen elliptic 03:53:46 dpeg: You have 4 messages. Use !messages to read them. 03:53:46 I last saw elliptic at Mon Sep 26 07:53:24 2011 UTC (1h 22s ago) saying with makhleb summons on ##crawl. 03:53:50 !messages 03:53:51 (1/4) mumra said (2d 18h 42m 29s ago): The piety gain/loss messages for all gods are hard-coded. If I'm going to move DG's ones to text files, it might be an idea to do the same for other gods whilst i'm at it. 03:54:00 !messages 03:54:00 (1/3) elliptic said (2d 5h 22m 38s ago): Disabling Hive is fine with me. :) 03:54:02 !messages 03:54:02 (1/2) evilmike said (2d 4h 10m 9s ago): I wrote some feedback on the zin wiki page about the proposed donation system, since I think it's still possible to game the system a bit. I also wrote some stuff about passive mutation resistance, but it looks like the wiki ate them (it was nothing important, though). 03:54:11 !messages 03:54:11 (1/1) evilmike said (2d 4h 5m 28s ago): re: elyvilon I have little to say other than the comment on the life-saving formula. All the tweaks suggested seem good. I'm a bit concerned about the weapon sacrificing being a bit complicated for the problem it's trying to solve (I don't think it would do that much to reduce tedium). 03:54:41 !tell mumra Sure, sourcing out more messages would be even better. 03:54:41 dpeg: OK, I'll let mumra know. 03:54:53 !seen kilobyte 03:54:53 I last saw kilobyte at Mon Sep 26 08:13:10 2011 UTC (41m 43s ago) joining the channel. 03:57:37 hi dpeg 03:57:37 mumra: You have 2 messages. Use !messages to read them. 03:57:51 !messages 03:57:51 (1/2) kilobyte said (2w 2d 22h 45m 7s ago): what's the point in duplicating the documentation for .des files on the wiki? These versions will end up obsolete and/or divergent. 03:57:57 !messages 03:57:57 (1/1) dpeg said (3m 16s ago): Sure, sourcing out more messages would be even better. 03:59:52 dpeg: i got the minion timer working on demigods. so mechanically, everything is there now. who's the best person on the dev team to help me regarding save data; i added a couple of properties to the player for tracking the minion timer, it works but it breaks save files ... 04:00:49 mumra: hm, good question. Ask at a time when both kilobyte and galehar around, they can either help you directly or blame someone else. 04:01:18 hey, that join was an automatic reconnect, it doesn't mean I'm listening :p 04:01:24 hehe 04:01:32 me neither! 04:01:39 kilobyte: I may have forgot your answer: you were fine with disabling Hive, too? 04:01:52 dpeg: for Hive, axing it for now would be generally ok, if we can add _a bit_ of guaranteed food. Tossing a huge pile in one go indeed makes any lack of food go away for most chars. 04:01:52 "experimental", as we put it 04:02:07 So let's do it. 04:02:14 I will add a note on the planning page. 04:02:23 I don't care about spriggans trying to cast level 7 spells in Lair 04:02:32 absolutely correct 04:02:42 Perhaps... more food shops? :) 04:03:23 when DEs get level 6 bolts or level 7 IOODs, spriggans get repeatable (ie, hungerless) Throw Frost and Mephitic Cloud, but unlike DEs, they can kite anything they want 04:04:01 maybe we'll have to do something about bee rooms. Getting several or none at all will have a huge effect on total available food. 04:04:24 what I do care about is that a single bad mutation can't screw you up, and also in many games you don't find a single piece of permafood before, say, D:8 04:04:33 galehar: yeah 04:04:44 galehar: but elsewhere we're fine with huge variance too (jewellery generation etc.). I'd treat bee rooms as nasty threat with (now more) relevant loot. 04:04:54 we could have one but guaranteed, in a mix with other food vaults for a variance 04:05:06 dpeg: nasty threat is good 04:05:11 you're too soft :) 04:05:16 have a unique bee that shows up in lair and drops jelly 04:05:23 dpeg: the thing is, you can survive without jewels. You can't survive without food. 04:05:48 pah, there is always Zin :) 04:06:03 maybe just reduce the amount of food in it. Keep the randomness, but reduce the variance. 04:06:05 if you want an extra need for permafood :p 04:06:27 seriously, I am fine with some tweaks but I don't think it is necessary to compensate immediately 04:06:41 It is only an issue for Sp, right? 04:06:50 Honestly I think hte biggest problem with hive is how annoying it is to play through,. 04:06:51 Or give Sp a second !porridge and call it a day. 04:06:54 Tr and Ce late game too 04:07:05 kilobyte: they should be able to care for themselves. 04:07:12 Bees running away, autoexplore trying to pick up a milion ghoneycombs, getting intrreptued every step etc. 04:07:13 sustenance exists for a reason 04:07:36 ghallberg: disabling will solve this nicely until racing hive is back as a portal vault 04:07:44 but I agree with dpeg. We can disable hive now, and adjust other sources later. 04:07:44 dpeg: Agreed 04:07:48 Disabling is trivial too. 04:08:07 for me, it is totally unacceptable to make the full game incompletable just because you did not get lucky. Not getting a spiked club of speed is fine, getting no food would be kind of the Horn of Geryon being a random drop. 04:08:15 * dpeg is already getting kicks from the changelog line "Removed Hive." and the ensuing reactions. 04:08:37 kilobyte: sustenance, kiku, acquirement? 04:08:39 I'd cheer it 04:08:41 dpeg: Zin can sustain _1/3_ of a Spriggan's metabolism, 1/27th of a Troll's 04:08:50 I switched god ^ :) 04:09:07 a troll with two rings of sustenance still consumes 5 food per turn 04:09:21 my reasosn for entering hive used to be things like "use bees for victory dancing dodging" and more recently, just to get the loot vault on hive:2. Sometimes for piety from bees 04:09:30 I can't remember the last time I actually did hive for the actual food :P 04:10:02 evilmike: 1000 Chei piety, or 100 for a blood god 04:10:10 mumra: did you look at docs/develop/save_compatibility.txt? 04:10:18 yeah, bee piety is pretty good 04:10:40 for most gods I tend to have 6* by hive though... going to hive for piety is usually if I blow most of it summoning demons with makhleb, or something 04:10:47 by the time you reach hive you have usually found a ton of permafood. if i ever have a problem with hunger it's in the early game. 04:11:02 galehar: no, didn't see that one; will have a read 04:11:11 kilobyte: if all game Sp and late game Tr/Ce nutrition turn out to be problematic, we can surely react. For a start, I'd just like to see how much the Hive removal affects everyone without compensation. 04:11:53 That's a reason to disable Hive early, btw. Perhaps today? =) 04:12:06 So we get more feedback and have more time to react, if needed. 04:12:25 dpeg: disabling it and watching games won't tell you anything, it affects one in 20 Sp games so the feedback from the remaining 19 will be "all is good" 04:12:42 do you think those spriggans ever use food shops? 04:12:45 kilobyte: we can call for special Sp games, if you think it is needed. 04:13:09 also: 04:13:10 !lg * win race=tr s=urune 04:13:10 161 games for * (win race=tr): 95x 3, 43x 4, 8x 15, 6x 5, 2x 6, 2x 7, 1x 2, 1x 14, 1x 0, 1x 8, 1x 9 04:13:42 -!- Xiberia has quit [Quit: ChatZilla 0.9.87 [Firefox 6.0.2/20110902133214]] 04:13:43 only 9 full or almost full troll games in the whole history of servers 04:13:52 kilobyte: from what I've heard, there are other things that can stop players from doing extended end games. I've read about lack of certain spells, lack of resistances and lack of randart gear. (Not sure how good those players were... our heroes might have been pulled those off without getting wet) 04:14:26 dpeg: if lack of certain spells can stop you, you're not supposed to count as a good player 04:14:33 you're supposed to know how to compensate 04:14:42 it's just food that cannot be compensated for 04:14:45 kilobyte: I don't think that is bad, to be honest. Troll is awesome early on, and more challenging later (this might be amplified with Trog even). That's good! We have way too many species going the other way around. 04:15:10 If Troll full game is a serious challenge, what's wrong with it? 04:15:27 it's not "challenging", it's "impossible unless you are lucky". There's no skill involved. 04:15:42 I mean, lack of food, not randarts 04:16:33 but, I fully agree with you that trivializing food issues for EVERYONE in early midgame just because a tiny percentage of builds need it in extended is not good 04:16:43 One long-standing idea is to give hells some natural monsters. I like this much better than giving more food to everyone (because it is targeted to those who actually need it, hell stormers). 04:17:08 I find it pretty easy to build up a stockpile of permafood with a troll, personally... 04:17:09 If you want, I can try to come up with a Marshall plan for how to add sultanas to the hells. 04:17:20 they get gourmand, so you don't really need to eat permafood much until later on 04:17:32 dpeg: how about "hell banquet" vaults 04:17:47 mumra: yes, can always add a few tortured humans in cages. 04:18:11 "Outrageous, those demons were about to dine on humans. That's my supper!" 04:18:15 so we might wish to dine in hell 04:18:22 :) 04:18:43 kilobyte: I'll add Hive-- to planning and add the concerns as well. 04:18:58 too random, I guess. For example, recently a draconian^Chei of mine needed 8 bread rations just to escape the Abyss 04:19:44 so permafood of some kind would be needed... but at the Hell/Pan stage, not at Lair:8 time 04:20:44 I'm afraid hell hounds in Pan break theme too :( 04:21:08 food basically isn't a problem once you hit lair. something could be done D:1-4 if anything (e.g. increase corpse drop likelihood for monsters originating on those levels; a semi-guaranteed early shop or food vault; etc.) 04:22:03 mumra: I quite fail to see how a chunk-eating race can have problems there 04:23:01 mumra: have you ever had food pressure that early? 04:23:03 kilobyte: i've had early food issues in an extremely low percentage of games. usually it's quite tense and fun when that does happen. but spriggans are a different matter. 04:23:08 trolls with their metabolism have gourmand which helps a lot, centaurs may be a bit pressed but speed means you can dive a lot more safely than most 04:23:08 I don't recall anything like that. 04:23:18 mumra: Sp have !porridge. 04:23:52 spriggans with food issues tend to be spriggan enchanters that are casting confuse way too much 04:23:56 dpeg: one... and no guarantee they'll find a single piece of food 04:24:02 dpeg: i don't play spriggans that often, but in a lot of my spriggan games that porridge has had to be used quite early. 04:24:03 kilobyte: from what I've read, Pan actually provides permafood. For Hells, I can make vaults (with good WEIGHTs, too). 04:24:05 evilmike: yeah, big spells too early 04:24:44 Just added Hive- to the planning page. Suggested a second !porridge there. It's easy to do and helps them when they need it most. 04:25:05 dpeg: hell food vaults would actually add an incentive for some characters to explore hells instead of just diving 04:25:11 mumra: I am aware of that. 04:25:37 I don't think we want that 04:25:40 perhaps we could have a late game Hive of some kind 04:25:43 (This implies they should be quite chunky so that you have an easier time finding them.) 04:25:49 galehar, mumra: right 04:25:51 kilobyte: ziggurat? :) 04:26:09 dpeg: heh 04:26:33 We could make the first few storeys of a ziggurat biased towards food. 04:26:44 not in the current incarnation probably, but yeah 04:26:47 people will be more confused by that than anything 04:26:55 "For some reasons, the inhabitants of this tower tend to store most of their food near the top." 04:27:18 there's a problem that Ziggurats are not guaranteed, and would be abused since you can find exceptional loot there 04:27:44 I think if you were at the extended endgame point and really desperate for food (for some reason), the best thing to do would be to explore pandemonium 04:27:52 kilobyte: someone who is considering extended end game can surely open a Zig from Pan? 04:28:16 pan should result in a net gain of food on most characters already 04:28:27 evilmike: kilobyte is concerned about Tr and Ce. 04:28:51 dpeg: you need to explore a whole lot of levels to find one 04:29:02 trolls definitely have an issue... they can build a stockpile throughout the game though. Centaur seems like the worst one 04:29:20 dpeg: the problem is, it's hard to keep a troll fed while not making everyone else swim in stockpiles 04:30:43 maybe sustenance could reduce your hunger rate by 2/3 rather than -2, or something 04:30:50 that way trolls could get something out of it 04:31:28 good point 04:31:34 how about if demons can drop corpses in hell (since the laws of reality work differently there) 04:31:46 Confusion trap! cool :D 04:31:51 basically lack of any other resource can be worked around: you might not get that vampiric demon whip but you'll always get a basic spiked flail from a random orc warrior. Spells can be nasty, but in the worst case you have to learn a new school. 04:32:39 like, my DrTm wanted to test dragon form or at least statue badly. Didn't find either for the whole game, had to make up for it with Shatter :p 04:33:03 kilobyte: The underlying question is: "Should every character be entitled to an extended game?" I wouldn't mind if the answer is "No, it depends on circumstances." but I don't mind helping Tr/Ce in a way that does not help the rest. evilmike's proposal is good for that. 04:34:16 dpeg: so this is where we disagree the most. For me, it's imperative that the game is always fully winnable with enough skill -- just some characters might require far more of skill than others. 04:35:16 kilobyte: I never was an endgame person. I understand the point of hell and Abyss. Pan not so much. 04:35:19 Hive is just placed badly 04:35:35 best branch in the game 04:35:50 (Pan not Hive of course) 04:36:48 I mean, scumming 54745693478 level of it is not the fun thing 04:37:16 what is fun is 1. open exploration without constraints, 2. restricted but random monster sets give different levels a different feel 04:38:51 galehar: food vaults in Hell could be helped with level sounds: "You hear the screams of the tortured." 04:39:47 dpeg: yeah, good idea 04:40:22 speaking of pan, i've started working on some new vaults (hopefully so every pan lord gets 3, instead of just lom lobon). Here's what I've done so far: http://pastebin.com/8DSHVfP3 04:40:38 we could stress 2 by having sets more themed rather than random, too 04:40:40 it's also kind of interesting that pan still uses the same 9 repeatable minivaults that it has had forever. would more of those be good? 04:41:14 kilobyte: yes, sounds good 04:41:21 evilmike: absolutely 04:41:26 dpeg: A vault filled with tortured corpses? 04:41:38 ghallberg: huh? 04:43:32 feel free to add points to the planning wiki 04:44:05 I support removing now and repairing later because everyone is annoyed about current Hive and the tweaks will be more subtle. 04:45:35 perhaps doing it will encourage someone to address the implementable 04:50:06 that is the plan! :) 04:51:07 I just realized that the concern about not finding a single piece of food by D:8 can be trivially fixed by a CHANCE:100 DEPTH:4 uniq vault with a ration or a couple. 04:53:44 true, but that's not the way how Crawl would usually do it -- at the least, I'd expect a depth range 04:53:44 evilmike: that seems to have some decent ninja protection 04:54:06 st_: heh yeah the pit fiend there might be a bit excessive... might open up the metal part with the rune a bit too 04:54:51 dpeg: nautrally, so you don't know which level to search 04:56:00 kilobyte: could give the vault a low chance of spawning a food shop instead of a ration? 04:56:34 Or both! 04:57:51 Free sample provided by the shop (leftovers that nobody would buy). 05:00:04 OSX builds of master branch on crawl.develz.org updated to: 0.10-a0-921-g2108932 05:00:48 dpeg: A food vault for hell, a bunch of mutilated corpses that you can butcher. (Won't help Ce I know) 05:01:17 ghallberg: wouldn't help -- it's the equivalent of a single ration 05:01:21 Ashan (L1 CeAK) ASSERT(num_visits == 0 && levels_seen == 0 || num_visits > 0 && levels_seen > 0) in 'place-info.cc' at line 43 failed. (D:1) 05:01:43 Ashan (L1 CeAK) ASSERT(num_visits == 0 && levels_seen == 0 || num_visits > 0 && levels_seen > 0) in 'place-info.cc' at line 43 failed. (D:1) 05:04:06 i like the idea of mutilated corpse vaults in hell just for the flavour 05:05:05 aren't they supposed to torture souls not bodies? 05:05:20 those would be good too 05:05:52 dpeg: I don't understand this line from the hive reform page - "The current wasp chamber should be randomised in layout and 'trick' (currently: secret door over deep water)." 05:05:53 I just like this mental picture I have of some slaughterfest with corpses and chunks everywhere, and lots of blood 05:07:42 mmm gore and carnage 05:08:38 mumra: sorry? 05:09:10 dpeg: s/wasp/bee/ ? 05:09:15 kilobyte: because nobody knows what a soul is, everyis falls back to torturing bodies. Which probably destroys the soul in the process. 05:09:32 dpeg: and i'm not sure what "secret door over deep water" is getting at exactly 05:09:35 mumra: the old Hive map (now just one of many) has a wasp room beyond a pond. 05:09:45 mumra: you're too young! :P 05:10:23 dpeg: right, i see; that's basically irrelevant to the new hive anyway ... 05:11:02 yes 05:11:13 dpeg: i've seen the wasp vault, of course, it just threw me having that as a bullet point at the end of the food face stuff :) 05:12:34 Ashan (L1 CeAK) ASSERT(num_visits == 0 && levels_seen == 0 || num_visits > 0 && levels_seen > 0) in 'place-info.cc' at line 43 failed. (D:1) 05:14:08 oh by the way, i found something really weird and interesting the other day. I was source diving crawl 3.3 to see what weird stuff it had, and there was a commented out vault in maps.cc with a "mollusc king" 05:14:48 mumra: improve the list :) 05:14:58 the mollusc king was only mentioned in two places in the source, and the vault itself wasn't just a bunch of islands with slugs + snails 05:15:35 evilmike: would it be possible to salvage it? 05:15:55 for fun, I re-edited it into a workable DCSS vault 05:16:19 didn't change anything excpet for the mollusc king, which I made into an elephant slug 05:16:48 evilmike: did the king have any special properties? 05:16:52 absolutely not usable in its current state: http://pastebin.com/Lg0aVamr 05:17:22 indeed :| 05:17:25 I was only able to find two mentions of the mollusc king in the source, one had its stats (but not its spells) 05:17:27 old != good 05:17:49 the other reference was that vault 05:19:51 evilmike: so we will never know what we're missing? Also, there must have been a reason why this was commented out.. 05:20:47 the reason is in there, vaguely 05:20:47 / For some reason this doesn't work 05:22:06 sorry that was supposed to be // 05:22:20 Do we know _what_ did not work? 05:23:23 nope :P 05:24:11 talk about useful comments 05:25:03 galehar: would you recommend storing in the props hash instead of adding new variables? the properties are relevant only for Demigods, although they're used throughout the game in that instance. it's tracking e.g. how long you've got left to kill the minion a god sent 05:25:24 there are some _really_ weird comments in the old source code I downloaded 05:25:34 evilmike: quote, quote! :) 05:25:52 for example some short comments (one word) make it pretty clear that the pan lords used to be gods 05:26:04 cerebov = okawaru, gloorx = kiku, lom = sif, mnoleg = nemelex 05:26:27 evilmike: I'd interpret this as the former being subjects of the latter. 05:26:43 oh yeah and their levels were named after the gods 05:26:59 so cerebovs castle was named "okawaru" 05:27:02 hm, my confidence is dwindling :) 05:27:22 the part that really gives it away though, is that in the file with all the monster descriptions, most lines have the normal monster name as a comment... except the pan lords 05:27:55 So you'd kill the gods? What if you follow one of those gods at the time?? 05:28:02 no idea 05:28:22 I can't find anywhere to download anything older than crawl 3.3, which was the last version by linley henzell 05:29:32 This interests me a lot. I should write him an email. 05:29:38 this version does have guardian robots though... and a great swords skill :P 05:30:15 the robots became orbs or electric golems later, I think 05:30:37 orbs of fire were "swords"... same description (almost), same abilities, same glyph even 05:31:44 I was actually curious about *really* old versions, so I looked through some old usenet posts. Apparently crawl 1.0 had square-los and non persistent levels (but you could only travel downward until you got the orb, like in Rogue) 05:32:22 I wonder why he dropped squarelos 05:32:35 st_: no other roguelike used it? 05:33:23 kilobyte: leaving for food in a moment. Could you live with Hive-, another !porridge for Sp, and tweaking =sust to make sense for Ce/Tr as a start? 05:36:17 dpeg: too late, I've another solution for Sp coded and being tested. And no Hive. 05:36:17 a guaranteed single bread ration on D:2-5, a guaranteed food vault with 4 breads or 6 honeycombs/jellies on D:7-14. 05:36:20 doesn't solve Ce/Tr yet 05:39:54 not sure how to avoid letting them a single ring trivialize their biggest concern before the endgame 05:40:52 oh speaking of weird leftover comments, there's a bunch of stuff that shouldn't be there in abyss.cc... including something part of a list of gods: 05:40:53 case 10: return "Lugafu the Hairy"; 05:41:26 evilmike: er, what? 05:41:34 sounds familiar 05:41:51 evilmike: are you reading the same abyss.cc as me? 05:41:52 moin guys 05:41:56 kilobyte: nope 05:41:58 mine is from 1999 05:42:14 question: how is dokuwiki in general? should we stick with it? 05:42:20 * kilobyte sniffs Napking. 05:42:28 linley henzell left a ton of commented out stuff in it, and a lot of the comments are downright bizarre 05:43:25 CIA is sleeping on the job, I've dropped dpeg's bomb. 05:44:22 1R88dO-0001mf-RI mail.cia.vc [204.152.223.100] Connection timed out 05:44:24 1R88dO-0001mf-RI == cia@cia.vc R=dnslookup T=remote_smtp defer (110): Connection timed out 05:44:28 --- 05:50:23 -!- galefury has joined ##crawl-dev 05:50:48 could someone please confirm that wiki is read-only now? 05:51:58 Napkin: i appear not to be able to log in 05:52:17 that's not related 05:52:32 *shouldn't be related 05:53:00 when i click login it redirects me to mantis; but i'm logged in on mantis 05:53:16 i tried logging out on mantis, then going thru that again, but it still doesn't log me in on wiki 05:54:15 cache, mumra? 05:54:27 maybe delete your cookies for crawl.develz.org? 05:54:48 i can't reproduce that 05:55:05 do you have html code in your password? 05:55:49 napkin: cleared cache, cookies, etc. - restarted browser for good measure - still the same 05:55:56 napkin: my password is just text 05:56:20 napkin: was logging in fine 20 mins or so ago before whatever you changed ;) 05:56:59 that's not related 05:57:40 can anyone else confirm whether they can or can't log in to wiki? 05:57:51 I don't see the edit button on pages, Napkin 05:58:12 st_: does it show your user name, or do you still see Login button at the bottom of the page? 05:58:37 I'm logged in but I also see a login button 05:58:51 if you're logged in, you don't see a login button 05:59:06 well, I do 05:59:11 so do I 05:59:17 even though I'm logged in 05:59:26 that's a different bug, I believe 05:59:32 hmm ... interesting! 05:59:40 what does it tell you at the bottom left, mumra? 05:59:50 oh yeah lol 05:59:52 "Logged in as: abc (REPORTER)"? 05:59:56 it does say logged in 05:59:58 in tiny tiny text 06:00:10 i was looking in the top-right for some reason (isn't it usually up there?) 06:00:57 apologies; i'll keep quiet now 06:01:09 well, let's try to update, then we can have a look at that 06:01:25 no problem, mumra :) 06:01:38 there should definitely be a Logout button; so it *is* a bug ;) 06:02:01 actually, it's a missing feature rather ;) 06:02:56 or the hack to show the logout/login button is ugly hardcoded on permissions, rather 06:03:16 there is no such thing as a "missing" feature. there are only "features" ... which may be defined as "this has a feature of not having a logout button" ;) 06:03:53 non-sense 06:03:54 or "this level has a feature of crashing Crawl if you approach the bottom left corner" 06:05:39 Ashan (L1 CeAK) ASSERT(num_visits == 0 && levels_seen == 0 || num_visits > 0 && levels_seen > 0) in 'place-info.cc' at line 43 failed. (D:1) 06:05:55 -!- Fail_Bot has joined ##crawl-dev 06:06:02 Heh - unchivalric_attack_type is_unchivalric_attack 06:06:18 -!- syllogism has joined ##crawl-dev 06:10:17 !learn swap hive[1] hive[3] 06:10:17 Swapped hive[1] with hive[3]. 06:10:37 !learn swap hive[2] hive[3] 06:10:37 Swapped hive[2] with hive[3]. 06:11:04 !learn edit hive[2] s/^Accessed/In 0.9-, accessed/ 06:11:05 hive[2/4]: In 0.9-, accessed from the main dungeon somewhere between levels 11 and 16. Two levels deep. Tons of honeycombs and royal jellies can be found in the bee hive on the bottom level. The bottom level has several possible layouts, but each contains a small amount of treasure (in addition to random floor objects), usually in some form of hidden vault. 06:15:34 -!- evilmike has quit [Quit: something happened] 06:19:26 back 06:20:05 Lugafu the Hairy makes Lugonu a downright sensible name choice. 06:21:56 well, it took me two years to learn "Kikubaaqudgha" 06:22:00 Kikubaaqudgha? 06:22:05 Yredelemnul! 06:22:13 Cheibrados. Elyvilon (I'm never sure of that). 06:22:20 yrdy is quite easy to remember 06:22:24 Lugonu, Zin, Jiyva... 06:22:35 Cheibr_i_ados 06:22:36 due: s/Cheibrados/Cheibriados/ ... 06:22:36 I suppose Kiku's the hardest :) 06:22:46 mumra: Irony: I came up with that name. 06:22:52 lol :) 06:23:10 (That was actually a "I didn't press i hard enough the second time" typo) 06:24:21 -!- greensna1k has joined ##crawl-dev 06:25:27 -!- greensnark has quit [Read error: Operation timed out] 06:27:04 -!- Mu_ has joined ##crawl-dev 06:29:17 greensna1k? 06:29:17 Iiiinteresting. 06:29:18 mu! <3 06:29:28 Mu_, how is your sprint map? 06:29:54 (I should sacrifice mumra to Trog for ruining tab complete and making me rewrite that line six times) 06:31:10 -!- mikee_ has joined ##crawl-dev 06:31:27 Home Del Del Del Del Del Mu_ End? 06:32:22 kilobyte: No, it's faster to just down-arrow and retype. 06:32:35 hang on - you're pressing tab to save you typing the letter 'u' ? 06:32:59 mumra: sure. Everyone did until you were around :O 06:33:06 mumra: so do I 06:33:18 * dpeg is a confessing single letter completionist. 06:33:50 I tend to forget about Home though, and almost always hold Alt-B, then Alt-F Ctrl-W 06:34:43 -!- Zaba has quit [Read error: Operation timed out] 06:34:55 mumra: No, I typed Mu and it got you instead of Mu. 06:35:11 I'm a two-letter tabbler but IT IS NOT NOT ENOUGH YOU SUCK. 06:35:38 :P 06:35:39 -!- Zaba has joined ##crawl-dev 06:35:46 after tab completing to mumra, add "can you please tell mu_..." 06:37:14 mikee_: that works, yes 06:41:03 -!- Keskitalo has joined ##crawl-dev 06:43:12 sprint map on hold atm 06:43:28 i'll get to it eventually 06:44:00 dpeg: More food shops in return for axed Hive sounds good to me! :) 06:44:17 my snake end got 2 kills so far though :P 06:44:21 I also think it's ok to remove Hive now and see what problems come up. 06:44:44 Mu_: i died to it offline on an MfIE 06:44:49 hooray 06:44:50 i went there way too early though 06:44:52 Keskitalo: too late, a commit to remove the Hive would cause a conflict 06:45:02 with no rp and dmsl 06:45:15 so it was more my stupidity than your snake end that killed me ;) 06:45:17 unless you do it exactly the same way so git realizes it's already there 06:46:06 i usually only do hive if i'm missing some jewellery, and pretty late 06:46:17 so i'd miss the two floors worth of trash loot more than the food :P 06:46:39 the loot is usually good, even just floor drops 06:46:44 ye 06:47:22 and you had several | there 06:47:40 -!- Torokasi has joined ##crawl-dev 06:48:24 hey it's not trash 06:48:24 you can get good things in hive 06:49:16 conflict? 06:50:50 pop in the ring so bees kill each other 06:51:20 I mean, it's already done, just CIA being dead kept the tides of player comments off so far 06:51:38 oh :) 06:52:41 -!- ortoslon has joined ##crawl-dev 06:53:26 kilobyte: I like guaranteed food shops instead of guaranteed food on the floor; Sp might be motivated to do Orc earlier. Anyway, nice to see this move forward! 06:54:36 it's just a single bread ration, then one of old hive entries (or an equivalent bakery) 06:54:48 the bakery is nice :) 06:55:14 could adapt other entries too 06:55:53 and doing something with bee room variance is badly overdue. If you get three you get the Hive's worth already. 06:56:01 03kilobyte * r1153542442c1 10/crawl-ref/source/ (branch.cc dat/des/builder/food.des): Disable the Hive, guarantee some very basic food supply. 07:01:37 -!- Xiberia has joined ##crawl-dev 07:04:55 -!- ainsophyao has quit [Remote host closed the connection] 07:10:53 -!- dtsund has joined ##crawl-dev 07:16:15 bakery! :) 07:16:18 -!- Pacra has quit [Quit: This computer has gone to sleep] 07:17:11 kilobyte: one good first step would be limiting the size of bee rooms... if you get a 14x14 one or something, then either you can't kill bees and already had to leave the level, or you can and it's a staggering amount of food 07:17:29 limiting where it can show up would be good too 07:18:59 perhaps make them spawn only in lair, with a weight such that there's an expected value of 10 honeycombs + 2 RJ? that seems like a fair amount of food from bees in an average game 07:19:56 alternately, axe bee rooms in their current form and use stuff more like the hive entrance vaults 07:20:07 which are presumably now unused? 07:21:27 it's just a single bread ration, then one of old hive entries (or an equivalent bakery) 07:21:41 bee/undead/orc/whatever rooms are sort of weird anyway 07:21:44 -!- MarvinPA_ is now known as MarvinPA 07:21:52 yes 07:21:57 i don't think they would be much of a loss to get rid of 07:22:09 especially when they show up in tartarus or etc. 07:22:13 haha yeah 07:22:16 they don't anymore :P 07:22:17 I thought they didn't anymore 07:22:17 i think that was finally fixed 07:22:25 but it was pretty funny while it lasted 07:22:37 * dtsund has been sifting through the commits 07:22:48 if we want ADOM-style tension rooms, that would be cool 07:22:55 but they should be filled with something more interesting than trash monsters 07:23:05 they are more often a source of XP and loot than any serious threat. Tm loves finding a kobold room. 07:23:11 like the door vault, i guess 07:23:23 galehar: kobold rooms are actually good because of how early they can show up 07:23:31 and because they use equipment 07:23:34 anyway, special rooms are just vaults 07:23:56 i think we should keep around early kobold and orc rooms, but as more cave-y vaults or something 07:25:06 Anyway, I can't support ADOM-style tension rooms because that'd make autoexplore worse than regular travel. 07:26:19 mumra: regarding save compat, props is limited to 256 entries so shouldn't be abused. Temporary data should definitely go there. Permanent one limited to one race should probably not. 07:27:46 galehar: another thing, the string speed-up patch on Mantis does a nasty hack to reduce the amount of mallocs -- with indeed quite a good speedup, but what struck me the most is that it is improving just props 07:27:56 kobold room is sort of ok regarding threat level 07:28:00 this suggests props cost us a lot of CPU 07:28:09 it's one of the tactical problems new players ask for help with the most 07:29:11 kilobyte: huh. I haven't looked into it. I thought it was about pathfinding, which doesn't use props right? 07:29:21 dtsund, they would need a warning system (as ADOM has) 07:30:47 mikee_: Yes, and once you get the warning you'll need to not autoexplore until you find it. Players will LRD doors from afar. 07:31:34 Doesn't Crawl already have vaults with doors marked as "do not touch"? 07:31:55 Seems like the same idea could be extended to tension rooms. 07:31:55 Yeah, but if you used one of those here it'd defeat the entire point. 07:32:01 galehar: performance-critical paths do query a lot of props, including pathfinding and what not 07:32:24 The point being to force players to use alternative door opening methods on that level? 07:33:45 The point being to ambush the player. If you use one of those special doors, the player can just leave them all closed and forget the rooms exist until he's overpowered for the monsters inside and come back for the experience later. 07:34:03 galehar: ok i'll keep the new fields on the player class; save_compatibility.txt clears things up (although it might be slightly out of date, it doesn't mention anything about also using "#if TAG_MAJOR_VERSION == 32") 07:34:20 Right. So the idea was not to announce these ambush rooms then? 07:34:34 (Unlike how ADOM does it) 07:34:44 Then the player should *never* autoexplore. 07:35:03 The idea just doesn't work. 07:35:50 But can't you arrange the vaults to be a bit bigger? Like, have a big room in front, a big bunch of monsters in the next room, and a bunch of doors that lock up or whatever behind you? 07:36:47 Then the player should *never* autoexplore. He should read the des files so he can recognize these vaults from a distance and plant exlcusions. 07:37:15 mumra, are you still in? 07:37:18 Sure, but if you pre-exclude the doors, he'll get a warning and then decide whether he wants to go in anyway. 07:37:23 could you check the wiki again now? 07:37:58 Napkin: i can edit now 07:37:59 * dtsund points to what he said earlier about players uniformly leaving them alone 07:38:16 Not every idea has to work. 07:38:30 dtsund, yes i anticipated that problem too 07:38:40 the main difference here is...... 07:38:48 ADOM has locked doors 07:39:16 Yeah. That gives you a hint as to whether it might be a tension room. 07:39:38 -!- Xiberia has quit [Ping timeout: 276 seconds] 07:39:38 it has enough locked doors that if you want to fully explore you have to risk opening the tension room 07:39:47 -!- dtsund has quit [Quit: Coursework] 07:39:49 I guess. If there was guaranteed loot there I would be more willing to chance it, though. 07:39:50 and sometimes (pretty uncommon), you need to go through the tension room to reach the stairs 07:40:15 well it is not uncommon to have to go through a locked door to get past a level 07:40:17 kilobyte: then optimizing props access is a very good move (and the hack doesn't look that nasty to me). Maybe there's more to do there, but I'm not really skilled at low level optimization, so I can't tell. 07:42:18 uhm no... it is a slight optimization to a place that could be easily optimized a good deal 07:42:31 like, we don't need std::string there at all 07:43:01 for the key, right? 07:44:05 it might be doable to do it with C strings, but ints are 4-5 orders of magnitude faster 07:44:46 oh, you mean using enums instead of strings? 07:45:38 yeah, we use #defines for their names anyway 07:46:02 I can check which props are the most problematic, too 07:46:14 I guess it's just a few that are queried over and over 07:46:27 it'd be better to use enums (#defines aren't used consistently anyway, i've seen plain string keys all over the place) 07:46:42 kilobyte, maybe they could be made proper class members? 07:46:53 hm, I have been doing too much music when I read "C strings" as anything other than ... c strings. 07:47:40 due: we'll call them null terminated strings from now on, to avoid any confusion ;) 07:50:59 replacing only performance critical string keys by enums is less work, but left the door open for future new critical string keys. Replacing them all is safer in the long term, but reduce coding flexibility and defeat somewhat the hashtable principle. 07:52:45 I suggest we do the former, but add some comments about the potential performance problems of using strings as keys. 07:54:45 galehar: it's okay, I'm having trouble when people talk about C# too :0 07:55:21 galehar: I'd benchmark them first though. Props have one advantage: they take up space only when actually used. 07:57:04 -!- Fail_Bot has quit [Ping timeout: 256 seconds] 07:59:56 kilobyte: oh, I think I misunderstood what you meant when you talked about using integers. If you're making them static, then we can just make them normal class members (as Zaba suggested) and move them out of props then. 08:01:01 -!- mikee_ has quit [Ping timeout: 240 seconds] 08:01:14 yeah, but if one in hundred thousand monsters has a property, all of them will need the memory for that class member 08:05:46 -!- capablanca has joined ##crawl-dev 08:05:48 -!- capablanca is now known as mikee_ 08:06:43 -!- valrus has joined ##crawl-dev 08:08:08 -!- valrus has quit [Remote host closed the connection] 08:12:12 my slime pit has no route to the bottom 08:12:32 i'll have to dig around 08:12:35 is that supposed to happen 08:12:42 any hatches? 08:12:45 any stair mimics? 08:12:51 they got removed a while ago though 08:12:53 from slime 08:13:02 * due turns into a stair mimic; chases galehar. 08:13:12 there's a hatch down, yeah 08:13:27 i wouldn't be able to get back up though 08:13:34 you probably would 08:13:47 but id rather dig... 08:14:06 size 1 corridors are horrible in slime though 08:14:20 -!- ainsophyao has joined ##crawl-dev 08:14:23 might have to learn the spell to dig in deep levels 08:14:47 * galehar was a mimic! 08:14:49 -!- ainsophyao has quit [Remote host closed the connection] 08:15:24 well this is annoying 08:15:54 oh, new mimics are nice btw 08:16:26 and stair mimics seem to cause surprisingly little trouble 08:16:35 regarding connectivity 08:16:52 -!- ais523 has joined ##crawl-dev 08:21:11 i do sometimes get disconnected bubbles in orc though 08:21:20 even after using all hatches 08:21:45 no idea if mimics are to blame for that 08:22:01 Mu_, magic map and ctele 08:22:18 this is the best solution 08:22:23 i just spent a billion turns digging 08:22:26 might as well use the hatch then 08:22:45 using a hatch in slime is bad 08:23:00 mhh, i guess its a bit easier to get back with ctele 08:23:14 specifically i mean it doesn't check for walls 08:24:02 it does 08:24:16 kilobyte: I was looking into database.cc to see how it easy it was to add some new functionality to text parsing, and I noticed some interesting things. 08:24:52 -!- Xiberia has joined ##crawl-dev 08:24:58 kilobyte: are you aware that it is basically using a complete SQLite instance solely to manage a list of key/value strings (i.e. a hashtable?) 08:25:35 yeah, it does emulate dbm using sqlite 08:25:48 Zaba, was that about hatches in slime? 08:26:14 mikee_, yeah, they actually do try to shift the target position a bit away from walls if that's possible 08:26:20 -!- st_ has quit [] 08:26:44 i've had them plop me right into a mess of walls 08:26:44 Zaba: it just seems a bit excessive for a simple string dictionary 08:27:01 probably understandably i've stopped using them after that so i could've mixed a fix to them =P 08:27:03 mikee_, it didn't use to do that, though 08:27:15 not sure when it's been added 08:27:48 mumra: we're talking about speeding things up! 08:28:11 mumra: just think: if a hash reference is excessively slow, what would a SQL query take? 08:29:19 mikee_, Zaba: I changed hatches destination just after 0.9 release. They used to to put you on same the coordinates, now it's randomized. Maybe I missed the part about slimy wall safety. 08:29:22 kilobyte: well it's hard to know, depends how good the SQL implementation is at indexing 08:29:52 kilobyte: although since SQLite is only used for text lookups, i'd say it's not critical for optimisation anyway 08:38:41 mikee_, Zaba: just checked. It's still working, hatches won't put you in the middle of slimy walls. 08:38:55 nice 08:39:12 i will modify my future advice about slime 08:40:05 i assume everyone is aware that there are still some problems with new abyss? 08:40:19 mostly morphing disregarding enemies 08:40:31 and putting them in walls, water or lava 08:42:24 also exits seem to be morphable, which makes them very hard to find once morphing speed becomes fast 08:43:08 and i havent noticed any abyss vaults during my search for the demonic rune, maybe weights could be turned up a bit 08:43:17 especially for rune and exit vaults 08:45:15 I updated the pagemove plugin, which should move history too now. please verify everything worked after using it 08:49:07 -!- Zaba has quit [Ping timeout: 255 seconds] 08:49:12 galefury: monsters are supposed to be displaced, but I wasn't aware it doesn't work. Exits aren't morphable anymore. Abyss vaults have always been rare, it's not related to new abyss (although rune and exit vaults could be a bit more common, yeah). 08:53:01 -!- Zaba has joined ##crawl-dev 08:53:44 dpeg? which documents would you like to check with the wiki-text output plugin? 08:54:09 just add a "&do=export_text" to the url you'd like to see in text format 08:55:48 galehar: ive seen an exit be morphed last week 08:56:00 im pretty sure it was after the commit that said it prevents this 08:56:14 but it might not have been in the windows build at that point, im not 100% sure 08:57:03 might have been on friday 09:00:46 Napkin: it was for the vault design reference, e.g. https://crawl.develz.org/wiki/doku.php?id=dcss:help:maps:syntax:headers&do=export_text 09:01:03 and? how does it look? 09:01:44 It's stripped out all the HTML formatting - including headers and bullets. 09:03:02 galefury: I don't understand how it would be possible. Just tested in wizmode. Create an exit, set abyss to full speed, disable monster generation, hold on 5. Hundreds of turns later, everything is morphing around but the exit it still there. 09:03:39 hm, maybe something just morphed to block los to it 09:03:43 Napkin: I'm thinking it's basically doing too much; most wiki formatting is fine as-is. All that really needs doing is taking the wiki source text and converting headings to a numbered hierachical list 09:03:51 and maybe it doesnt work with wizard placed exits 09:03:53 i dont know 09:04:04 Napkin: So I'm just gonna have a quick look at the code for that plugin and see how easy it might be 09:04:11 i wanted to use that exit, so im pretty sure it was gone 09:04:39 but i could be wrong 09:07:25 the items change was in, and i think the change to exits was before that 09:07:25 ok, cool 09:07:34 would you like me to link you, mumra? 09:08:58 Napkin: is it this one? http://www.dokuwiki.org/plugin:text 09:09:07 Napkin: I gave dpeg the original link 09:09:15 yes 09:09:23 are you familiar with git, mumra? 09:09:38 Napkin: I had a quick look at the plugin and it's pretty easy to customise 09:09:50 I've disabled shifting for easier testing, and after 5000 turns holding s, the exit is still there. Most likely, walls morphed around it. 09:10:00 Napkin: yes, well, I'm starting to be familiar with git 09:10:06 ok, then wait 09:10:25 ill keep an eye out for exit weirdness 09:12:21 -!- ainsophyao has joined ##crawl-dev 09:12:33 speaking of the dev wiki: is there any decent way of subscribing for updates? I've tried RSS, but formatting is terrible and it only sends some updates. 09:13:18 galehar: Napkin just updated the dokuwiki version i think, you could try it again and see if it's better 09:14:55 mumra: http://git.develz.org/?p=personal/dokuwiki-text.git;a=summary 09:15:25 kilobyte: re: #4471, did you mention a while ago that it might be okay to go ahead and push it as-is for now, and resolve the existing ugliness later? would be nice to try out the functionality in trunk, at least 09:15:30 please base on your changes on that 09:16:56 -!- dtsund has joined ##crawl-dev 09:17:11 (the returning to pan from abyss patch, that is) 09:17:57 galehar: what about subscriptions? 09:18:38 Napkin: what subscriptions? email? 09:18:57 yes 09:21:16 Napkin: I'll give it a try 09:29:11 back 09:32:47 Napkin: which version of Dokuwiki is the site using? latest stable - 2011-05-25a? 09:33:25 yes, mumra 09:33:29 wb, david 09:34:41 -!- ussdefiant has quit [Ping timeout: 276 seconds] 09:34:54 wordpress next, but tomorrow 09:37:48 -!- Torokasi has quit [Ping timeout: 260 seconds] 09:43:06 galehar: uhm, why would you use RSS for the wiki? mail >>>>> RSS for it. 09:44:23 kilobyte: seems so. Will use mail instead. 09:44:35 RSS can't be marked or saved, it times out, requires a WWW browser, etc 09:44:41 there is one bug i still can't fix.. 09:44:59 it's that sometimes people manage to subscribe to pages, even though they are not logged in 09:45:04 galefury: a quick test doesn't show any problem with monsters in walls and such. Can you be more specific? 09:45:10 which leads to root receiving their emails 09:45:23 MarvinPA: sure, why not. I'll try to untangle the branch mess but it's too many things to do too little time. 09:45:25 if anyone manages to do that, please let me know 09:46:18 Napkin: best guess would be a session timeout 09:46:27 kilobyte: okay, cool :) 09:47:13 MarvinPA, kilobyte: what to do about Zigs? 09:47:41 if you return to the same Zig level after banishment, you could in theory use that to clear it in several go. 09:47:44 hmm... personally i like the idea that you wouldn't be able to escape easily by banishing yourself 09:47:49 but yeah, that's potentially a problem 09:48:34 i could see that being a good thing but i think some levels could use some rebalancing in that case 09:49:13 or just prevent banishment in Zigs. Doesn't work down there. 09:52:52 Are there zig monster sets using banishment a lot? Shouldn't be, unless someone added something. 09:53:22 In other words, no banishment in zigs hurts the player more than the monsters, so is good. 09:55:36 yeah 09:55:51 the only thing likely to banish you is maybe a pan lord's branded melee or something 09:58:14 galehar: not sure how i can be more specific, monsters ended up in walls and over lava a lot 09:58:18 also in water 09:58:31 all kinds 09:58:52 as i said, item change was in, so should have all the abyss commits 10:00:04 It'd allow the player to kill a few things, banish, rest, and go back for another go 10:00:35 draconian levels have banishers with decent HD 10:01:22 and of course some other monster sets have banishers but that is the most likely place to be banished from, probably 10:01:52 imo prevent banishment but provide some way of easy exit 10:02:27 galefury: there is already one. A portal conveniently placed on every level... 10:02:40 right, at the wrong end of the room 10:03:18 im not saying the easy way out should be free, just that one should exist 10:03:28 The portal at the wrong end of the room on the previous level is the easy way out. ^_^ 10:03:35 I quite fail to see how banishment would be a problem bigger than DDoor or Sanctuary 10:03:58 Banishment is available to all. 10:04:02 its very tedious 10:04:14 kilobyte: Sanctuary is expensive 10:04:23 Death's Door is risky. 10:04:47 The change would weaken Lugonites in ziggurats but I am not sure that is a bad thing. 10:04:55 banishment is risky too, actually 10:05:11 well... 10:05:20 If you're strong enough to consider a zig run, the Abyss holds little fear for you. 10:05:24 well, it depends, like all things 10:05:25 either very costly or very unreliable 10:05:30 no that's not what i mean =P 10:05:31 or Borgnjor's, for a more direct comparison 10:05:39 Ah, yeah. 10:06:19 galefury: are you really asking for an easy exit on a new Zig storey? That would defeat the whole point. 10:06:27 a good penalty for easily leaving a zig would be no more zigs for that char imo, and possibly loss of all loot from the zig 10:06:44 What about putting an exit portal on the entry side on the other end of something the player really wouldn't want to walk through normally? 10:06:45 banishing yourself once the monsters have spotted you can be pretty risky, yeah, i'm not sure it'd even be that feasible a strategy 10:06:53 I'd suggest mutafog, but some chars can ignore that 10:07:13 you don't particularly want to be repeatedly hurting/confusing/teleporting yourself while surrounded by the very things you're desperate to escape from 10:07:28 galefury: there's one exit. To the Happy Hunting Grounds. 10:07:39 if you use it as a strategy you would use it before things get too dangerous 10:07:53 and wear stasis i guess 10:08:30 its not really that risky i think, with proper planning 10:08:53 im sorry for breaking autocomplete btw 10:09:02 you can banish yourself with Lugonu (costs as much as Borgnjor's + Sanctuary), or distortion unwield (1/5? 1/6? chance of working, otherwise you most often take a serious hp hit) 10:09:14 1/5 10:09:26 and its only something like 30 HP usually 10:09:37 The only issue I have with ziggurats at the moment is that a player can potentially get rather far in them before learning what the worst they have to offer is. 10:09:48 galefury: I won't support that. 10:09:57 personally I'm in favour of no banishment in zigs 10:10:02 dtsund: yes, I can see that 10:10:05 support what? 10:10:25 galefury: you want to remove the central challenge by providing a free exit on the new level 10:10:31 Lair-Snake-Lair-Dungeon-Dungeon-Snake-WTFMUMMIES 10:10:34 banishment from miscast can turn into malign gateway 10:10:37 galefury, like i said, it depends 10:10:37 -!- edlothiol has joined ##crawl-dev 10:10:39 certainly not free, no, that would be silly 10:10:40 dtsund: I think the storeys should stay random but use a queue 10:10:50 there is no guarantee of "oh i will do xyz and it will be perfectly safe" 10:10:50 actually, there's a more complex formula than 1/5 -- you are eligible for miscasts of any severity, it's just strongly biased towards severity 3 10:11:05 dpeg: hell yeah 10:11:15 kilobyte: I thought distortion unwield was always severity 3, hm 10:11:23 it is also good for fun -- who wants to play the same storey three times? :) 10:11:47 dpeg: So, like, the first floor would draw from {Snake, Dungeon, Shoals}, future floors would draw from their own sets? 10:11:49 Oh 10:12:13 kilobyte: should I come up with a formula? It would be about weights, of course... only because you've seen all but one level sets, the remaining one is not guaranteed, just a lot more probable. 10:12:14 dpeg: most Ziggurats are easy so people can claim that "technique X pwns all", then you get a Pan / mummy Zig:27 and you're pink mist 10:12:31 * dpeg <3 pinky misty players. 10:12:48 imo angels are the worst... 10:12:59 see the chaosforge wiki for guides about technique A-Z pwning all 10:13:00 proper holy zig levels would be nice though 10:13:01 galefury: uhm right, that's the worst level 10:13:13 Also, I need Zig feedbak, on the monster sets. I invented the things but cannot beat them. So too strong/too weak ==> please tell me. 10:13:13 it just has a 0.00000000000001 rarity 10:13:31 The lack of holy monsters :| 10:13:35 Hrm. I've only legitimately cleared one once, and it was offline. 10:13:40 And I never saw angels. 10:13:42 same here 10:13:45 i died in two 10:13:48 finished one 10:13:49 lair and snake are obviously still too weak 10:13:51 and did a bunch halfway 10:14:05 angel is too strong and possibly mummy as well 10:14:10 mikee_: I think they're not too weak, they just need to not show up so early. 10:14:15 deaths were to angels and pan lords, never had a really deep mummy level 10:14:17 s/early/late 10:14:22 mikee_: any monster that could potentially help lair/snake? 10:14:23 lair/snake levels? you can autofight them 10:14:27 dtsund: it's basically: Zig:8 "4 smites a turn", Zig:27: "20 smites a turn" 10:14:28 dpeg, catoblepas 10:14:44 it could be given a level-dependent weighting like ancient liches in vaults levels 10:14:54 yes, we can do that 10:15:00 i'm not sure about snake 10:15:03 go lair --> dragon depending on depths? 10:15:05 dtsund: ah, and constant cleansing flame 10:15:07 there's such a limited monster set 10:15:15 IT BURNS 10:15:18 could work for snake too 10:15:22 both are reptiles 10:15:24 :) 10:15:33 dpeg: hydras with exceptionally high numbers of heads 10:15:35 although dragon levels are not that hard either currently 10:15:43 could combine lair/snake/shoals into one set 10:15:59 it doesn't auto-kill Vp/Mu anymore but you don't need autokill to go in cinders in a single turn 10:16:12 elliptic: could work 10:16:12 What Zig could use is easy levels for the first few, then a consistent ratcheting up of difficulty. 10:16:14 dpeg: how about god-themed floors generated from the demigod minion code? 10:16:44 mumra: sure! 10:16:51 should be rare, to spice things up 10:16:51 sounds great, but should be pantheon floors imo 10:16:53 not single god flavored 10:16:55 is there a plan to dilute angel levels with the new holy monsters? 10:16:58 So that you know that the next level will be a bit harder than the last one... but you successfully cleared the last one, and there's loot to be had, how can going on be a bad idea? 10:17:04 And then splats. 10:17:08 mikee_: sure, with pearl dragons! 10:17:12 ??pearl dragon[2] 10:17:12 haha 10:17:12 pearl dragon[2/2]: Up to 162 per breath plus 48 per turn in clouds. Cerebov's firestorm is up to 136 (76 with rF+++) + resistable clouds. "Caution advised" is an understatement. 10:17:18 mikee_: only a vague plan. It was supposed to be one application of the holy monsters (the other was a holy pan lord) 10:17:26 possibly a slight improvement over another daeva 10:18:03 dpeg: you're thinking small. A "Heavens" rune branch! 10:18:05 the hell and crypt floors are incredibly boring and quite common, too 10:18:22 crypt is pretty boring but i guess it can have silent spectres 10:18:24 crypt is okay imo, due to silent spectres 10:18:27 since the actual monster lists for the hells are just junk mostly, it's the hell effects that make it interesting 10:18:49 depths dependent weighting for fiends? 10:18:57 galefury: pantheon floors would be slightly crazy; i think it'd work better with a smaller number of god-specific minions, remembering they are tougher than normal dungeon monsters! 10:18:58 I don't have time to follow everythign in detail. Please add ideas to https://crawl.develz.org/wiki/doku.php?id=dcss:brainstorm:ziggurats 10:18:59 MarvinPA: basically, any floor other than Pan lords / mummies / holy is boring for any character who can except to survive those three 10:19:02 can be rough and short 10:19:05 that would be like 30 fiends 10:19:11 s/except/expect/ 10:19:11 the new fire and ice and air are good 10:19:19 * dtsund wonders whether his neo-spectres might be liked in Stone Soup 10:19:20 air is far from boring, yeah 10:19:25 as are late dracs 10:19:40 and pan lords 10:19:58 kilobyte: I've seen players die to Orc (the moths) and to Slime (the eyes) 10:19:59 they are a bit too harsh imo, due to lack of information 10:20:10 the pan lord floors i mean 10:20:21 orc is really underrated i think 10:20:36 do they get crossbows often? 10:20:40 and how much the # of pan lords can vary 10:20:42 i dont remember 10:20:46 they don't need crossbows 10:20:55 I will move the old ziggurat content to the new one (linked ^). Please add ideas there, I will take care of them. 10:20:56 also they'll be berserk anyway, probably 10:21:05 orc and slime are a bit more interesting, if not as outright challenging as the really nasty ones, yeah 10:21:24 MarvinPA: "a bit more interesting" == good enough for now? 10:21:34 i'd say so 10:21:37 cool 10:21:50 oh, late vaults is good of course 10:21:53 ancient liches galore 10:21:55 dpeg: mostly a matter of unpreparedness, I'd say. You can do the former easily with melee on a mostly-caster char, the latter is a matter "why didn't you wear stasis?" 10:22:05 oh yes, <3 late vaults 10:22:10 kilobyte: there are several types of eyes 10:22:10 what about spider? 10:22:17 magic draining can also hurt, I think 10:22:19 is it horrible enough with ghost moths? 10:22:23 most of the levels pose some challenge unless you've build a character for which there are no challenges 10:22:29 built* 10:22:30 slime isn't a huge challenge but it at least requires you to slow down and think a bit, right 10:22:41 maybe some hellspider code could be used for spider levels once hellspider is done 10:22:47 so for that i'd say it's pretty okay 10:22:50 several hellspiders with random upgrades 10:22:58 mikee_: exactly.... and if you've built a character for which there _are_ challenges, you'll splat on Pan/mummies 10:23:13 spider is a huge pain, not only can you kill it all with autofight, you *have* to kill it all with autofight because of so much mp draining 10:23:44 oh, theres also eye of draining levels 10:23:47 so there are basically two Zigs: one for most chars, one for über ones only. And you don't know which one you got in advance. 10:23:57 those are either nearly impossible, or you can smite them 10:25:05 well, you don't need to be uber to survive late pan/tomb levels 10:25:12 you just need to be uber to clear them 10:25:16 heh 10:25:25 that depends on the pan level... 10:25:36 and once again, you dont know in advance 10:26:37 galefury: as the number of Pan lords approaches 100, the chances of the level having all of: silence, torment, smite, summon eyes, multiple firestormers approaches 1 10:26:49 Originally, the idea was that ziggurats are unclearable... they'd kill everyone eventually. But greensna1k and I lost it. So we had to restrict levels to 27, otherwise some players would still be clearing levels. 10:27:16 i remember the original concept from rgrm 10:27:25 dpeg: so what you want is unclearable Zigs? 10:27:27 mikee_: is it _that_ old? 10:27:32 kilobyte: well, open-ended. 10:27:46 dpeg, i think so =P 10:28:04 this means, a guarantee that late levels feature the hardest ones 10:28:54 kilobyte: you're thinking about current zigs (and I agree that challenges should come up -- we have a number of hard monster sets, right?). The original version had no Zig level limit. 10:29:18 making zig:27 guaranteed hard doesn't sound too bad 10:29:49 hard = pan/mummy/holy, with those levels possibly nerfed 10:29:51 MarvinPA, do you have any suggestions for increasing crypt/hell level difficulty? 10:30:02 fiends fiends fiends! 10:30:07 not really... just throwing fiends into hell levels would work but be sort of boring 10:30:08 i'm editing the wiki page now and i could just write 'increase difficulty' if you don't 10:30:13 dpeg: do you mean an infinite amount of levels? 10:30:31 yes 10:30:32 i considered adding hell effects to hell levels =P 10:30:40 but i dont want people to think i'm insane 10:30:47 that would be better than just adding fiends! 10:30:52 I'd still have it keep a ramp so it's not a matter of luck 10:30:53 and thematic :D 10:30:58 imo it wouldnt be better 10:31:08 haha 10:31:10 hell effects in their current form are pretty meh 10:31:21 galefury: yeah. And are time-based. 10:31:21 i am going to write it in and someone can remove it if they wish 10:31:27 kilobyte: no level limit, and each harder than the one before (in the same sense as now, the new level could be easier because of monster set) 10:31:53 Ziggurats: the IVAN of Crawl 10:32:38 something could be done with the spriggans from the forest vault 10:33:13 wizard level or something, with all wizardly casters in the game on the list 10:33:16 spriggan level is an interesting idea 10:33:32 dpeg: can't be inifinite. At some point, you'll have a 80x70 zig level. 10:33:48 mikee_: not very varied currently, with mostly just berserkers + airstrikers 10:34:07 @??spriggan air mage 10:34:07 spriggan air mage (11i) | Speed: 16 | HD: 14 | Health: 33-54 | AC/EV: 1/25 | Damage: 16 | Flags: spellcaster, see invisible, fly, DMsl | Res: 06magic(130), 10elec++ | Chunks: 07contaminated | XP: 1466 | Sp: airstrike (0-38), b.lightning (3d19), swiftness, shock (d16), haste. 10:34:16 good, they do have airstrike :) 10:34:16 it would also have druids and defenders 10:34:25 are those relevant? 10:34:34 druids summon 10:34:34 without trees i mean 10:34:43 yeah, but they summon bears and crap 10:34:51 also they have a spell that hurts undead quite a bit 10:34:52 Sunray ignores DMsl and such 10:35:00 oh, i forgot about that one 10:35:07 @??spriggan druid 10:35:08 spriggan druid (03i) | Speed: 16 | HD: 12 | Health: 28-45 | AC/EV: 1/25 | Damage: 16 | Flags: spellcaster, see invisible | Res: 06magic(112) | Chunks: 07contaminated | XP: 1095 | Sp: sunray (3d19), awaken forest, caniforms, minor healing. 10:35:19 @??spriggan defender 10:35:20 spriggan defender (08i) | Speed: 16 | HD: 15 | Health: 36-54 | AC/EV: 3/30 | Damage: 30 | Flags: see invisible | Res: 06magic(140) | Chunks: 07contaminated | XP: 1578. 10:35:21 the defenders are at least relevant in the way all of these monsters are 10:35:28 (they are fast enough to reach the player) 10:35:48 maybe do spriggans/something else? 10:35:56 in their natural habitat it's defenders who are the hardest 10:36:12 they dont have a lot of good weak monsters 10:36:17 Perhaps spriggan-heavy levels could be forested 10:36:21 so spriggans + bears or something? 10:36:46 @??spriggan rider 10:36:46 spriggan rider (12i) | Speed: 16 | HD: 11 | Health: 35-52 | AC/EV: 1/16 | Damage: 21 | Flags: see invisible, fly | Res: 06magic(102) | Vul: 09poison | Chunks: 07contaminated | XP: 1077. 10:36:47 just say: place:Forest:$ like most levels do 10:37:09 with perhaps an additional weight to heavy hitters 10:37:16 theres definitely a lot of nice spriggans 10:37:33 do the riders use polearms? :D 10:37:50 galehar: the hope was that everyone is dead by then :) 10:38:24 oh, and more furniture is always an option 10:38:34 galefury: for a wimpy speary value of "polearms" but yeah 10:38:38 dpeg: i'm thinking the stairs to the next floor should be locked until everything is dead 10:38:57 reaching will be nice either way 10:38:59 mumra: that won't go well with zig regulars, I think :) 10:39:14 Is avoiding combat too easy? 10:39:20 umm, no 10:39:29 on late levels you usually cant cblink 10:39:30 that seems somewhat unnecessary, making it to the exit without killing everything is pretty challenging on the nasty floors 10:39:43 and theres mosnters all around 10:40:01 and in the cases where you're just trying to get to the exit you're probably also skipping most of the loot, that's the penalty there :P 10:40:12 * dpeg really, really wants that @-vs-panlords zig level with audience 10:40:19 MarvinPA: If the floor's too challenging to kill everything, the player should leave the zig, not go to the next level :) 10:40:31 meh, "Forest:$" is too wimpy, even for the regular branch end 10:40:33 MarvinPA: I wasn't suggesting locking the exits; just the stairs to the next floor 10:40:53 mumra: here I disagree. Getting in over your head and then spending resources or even prayers in order to make it out alive is good stuff! 10:40:56 but it's better to encourage them to go deeper! "the next floor might be easier and i can get more loot!" -> deathhhhhh 10:41:04 mumra: ah, I see 10:41:42 anyone editing the ziggurat wiki entry? 10:41:44 Hm, could be done. I'd leave that to devs who actually frequent the ziggurats, though. 10:42:49 -!- ussdefiant has joined ##crawl-dev 10:43:13 mikee_: what about silent spectres for crypt? 10:43:41 dpeg: silent spectres are in ziggurats 10:43:43 -!- mikee_ has quit [Read error: Connection reset by peer] 10:44:05 dpeg: in fact it's the one thing that nearly killed my character in a whole zig! 10:44:25 -!- mikee_ has joined ##crawl-dev 10:45:02 -!- galehar has quit [] 10:46:01 * dpeg scolds the silent spectre department at Crawl HQ for incompetence. 10:47:05 elf and shoals are good, right? 10:47:07 the combination of silent spectre and buffed reaper were one of the few things that nearly killed my chei character, too 10:47:20 adding a list of good levels and problematic levels to the wiki 10:47:29 since with chei you can't really safely approach things with reaching 10:47:36 yeah i noticed that too with my chei char 10:47:44 and with the spectre you can't blow them up, either! 10:47:55 required a bunch of random teleporting to get around 10:48:10 something maybe should also be done for some things that depend on the number of turns you get 10:48:13 Chei Zig or regular game? 10:48:15 like cloud damage and blocking 10:48:17 regular game for me 10:48:22 chei zig 10:49:30 but yeah, it's something to worry about in regular games too, outside zig there's probably more ways to get around or avoid them though 10:50:28 added some stuff to wiki, check if i missed anything? 10:50:36 i definitely missed some problematic levels 10:50:42 i remember dragon not being good 10:50:47 -!- mikee_ has quit [Read error: Connection reset by peer] 10:50:51 and some levels i probably completely missed 10:51:05 dragon might be okay with quicksilver dispel 10:52:09 03MarvinPA * rb464ac6246c1 10/crawl-ref/source/ (directn.cc effects.cc stairs.cc travel.cc): When you escape from the Abyss, return to where you came from. (GreatZebu) 10:52:44 earth theme is pretty boring too 10:54:10 ranged theme isn't that hard but they can do a lot of damage if you're not careful, i guess 10:55:21 dwarf? 10:55:29 i think its annoying, but somewhat interesting 10:55:30 ughh dwarf 10:55:36 yeah... 10:55:51 the only real threat is mirror damage 10:55:55 theres also giants 10:55:57 but what an annoying threat it is 10:55:59 the level i mean 10:56:42 kind of meh imo 10:57:14 I recall ice level being kind of hard with ice storm only. 10:57:19 yeah, probably another one that could just be biased earlier 10:57:20 So many ice fiends. 10:57:52 and dragon needs some bias, badly 10:57:58 although i guess even the really hard levels are okay when you get them on early-to-mid floors, they're pretty good at those points even 10:57:59 drakes, ugh 10:58:04 so it'd be sad to have them only show up really late 10:58:05 Does the giant level get more titans as you go deeper? 10:58:06 -!- Torokasi has joined ##crawl-dev 10:58:15 yeah, i had fun on early angel levels even 10:58:36 jle: no 10:58:44 http://gitorious.org/crawl/crawl/blobs/master/crawl-ref/source/dat/dlua/ziggurat.lua 10:58:55 levels start at line 328 11:03:12 -!- ais523_ has joined ##crawl-dev 11:03:51 -!- monqy has joined ##crawl-dev 11:04:38 Are some places supposed to just be easy, like Lair? 11:05:02 s/places/zig levels 11:05:38 i dont know what the original intent was, but nobody seems to like that they are easy 11:06:15 because blasting hordes of yaks with firestorm is just not very interesting 11:06:34 and levels where you can just hold tab suck 11:07:33 Clearly we need catlobes whose number increases with level :) 11:07:39 i had lair on zig:27 ... it was a huge anticlimax :) 11:08:10 Though that would probably not be much of an inconvenience since iirc petrification chance at xl 27 is quite low. 11:09:09 lol yeah, they would just petrify the yak horde 11:09:13 mostly anyway 11:09:23 Even dire elephants seem a little weak at that point. Maybe you could spice it up with hellephants, or is that too theme-breaking? 11:09:31 how about mixing giants and dwarves? 11:09:38 dire death yaks 11:09:51 it would vary the giants level a bit, and dwarf sucks anyway 11:10:04 also the giants level could benefit from some animate dead :) 11:10:39 It would certainly make the challenge less about managing damage mirror. 11:11:15 ill just add it i guess, dpeg can sort it out 11:12:27 what about dragons (not draconians) 11:12:32 since those show up in lair 11:13:23 i suggested that a while ago, but nobody reacted 11:14:00 the dragon level is also in need of attention, so merging those might be good 11:14:06 snake/lair/dragon 11:15:29 Pretty big merge. 11:16:10 Kill them all and let dpeg sort it out. 11:16:12 back later 11:16:50 If only hellephants had hellfire, I'd suggest an elephant level. :) 11:17:12 i think i saw one hellephant in all of crawl :( 11:17:15 @??hellephant 11:17:16 hellephant (04Y) | Speed: 10 | HD: 20 | Health: 138-203 | AC/EV: 13/10 | Damage: 45, 20, 15 | Flags: 05demonic, !sil | Res: 06magic(133), 12drown, 04rot, 13neg+++, 13torm | Vul: 08holy++ | XP: 4411 | Sp: fire breath (3d40), blink. 11:17:18 Or maybe it would be challenging without, they do quite a lot of fire damage. 11:17:31 And blink around. 11:17:40 its basically xtuaha with an Y glyph and blink, right? 11:17:43 @??xtuaha 11:17:43 unknown monster: "xtuaha" 11:17:48 argh 11:17:55 @??xtahua 11:17:55 Xtahua (05D) | Speed: 10 | HD: 19 | Health: 133 | AC/EV: 15/7 | Damage: 35, 17, 20 | Flags: see invisible, fly, !sil | Res: 06magic(177), 05fire++, 03poison, 12drown | Vul: 12cold | Chunks: 07contaminated | XP: 5449 | Sp: flame blast (3d38). 11:17:59 Except no vulnerabilities or resistances apart from demonness. 11:18:06 03kilobyte * red3764ac24f7 10/crawl-ref/source/religion.cc: For now, fix incorrect rod description for Trog. 11:18:07 03kilobyte * r3f757922d4e3 10/crawl-ref/source/evoke.cc: A whitespace fix. 11:18:07 03kilobyte * rebb05b803bc5 10/crawl-ref/source/ (store.cc store.h stuff.cc): A way to see which props slow us down the most. 11:18:17 03kilobyte * rc1447d13131c 10/crawl-ref/source/mon-pick.cc: Adjust the Forest monster set a bit. 11:18:17 03kilobyte * r24f71c93c636 10/crawl-ref/source/dat/dlua/ziggurat.lua: Spriggan levels for Ziggurats. 11:18:18 03kilobyte * r53212c704b9b 10/crawl-ref/source/dat/dlua/ziggurat.lua: More catlobes in Lair Zig levels. 11:18:36 Haha, what. More catlobes, yesssss. 11:19:26 hmm... need moar kittehs 11:19:58 galefury: Anyway, I think something with increasing numbers of hellephants might be challenging since they blink around and are, in theory, thus harder to firestorm all at once. 11:21:20 might be a good way to deal with the abyss monster set 11:21:29 idea: dwarf ; executioner's axe ego:pain . plate mail / merfolk ; demon trident ego:flame / ghoul / human spells:magic_dart ; nothing / felid spells:firestorm,airstrike,haste / etc 11:21:31 good stuff can appear there, its just too rare 11:21:48 ie, simulated player characters 11:21:55 all hd:27 of course 11:22:11 that human seems a trifle out of place 11:22:11 ugh, magic dart at hd:27? 11:22:18 heh 11:22:23 @??rat hd:27 spells:magic dart 11:22:24 Unknown spell name: 'magic' in 'magic' 11:22:29 galefury: humans suck :p 11:22:30 @??rat hd:27 spells:magic_dart 11:22:31 could be a fun idea though, yeah 11:22:31 rat (07r) | Speed: 10 | HD: 27 | Health: 48-86 | AC/EV: 1/10 | Damage: 3 | Flags: !sil | Res: 06magic(1) | XP: 479 | Sp: magic dart (3d7). 11:22:42 galefury: just like in RL. Dumbest species around. 11:22:57 well, it always hits 11:23:01 but usually wont get through ac 11:23:02 They get 3d7 magic dart? Monsters are such cheaters. :p 11:23:16 i was worried it might be something like 3d20 that always hits 11:23:17 @??rat spells:magic_dart 11:23:17 rat (07r) | Speed: 10 | HD: 1 | Health: 1-4 | AC/EV: 1/10 | Damage: 3 | Flags: !sil | Res: 06magic(1) | XP: 1 | Sp: magic dart (3d4). 11:23:29 -!- jeanjacques has joined ##crawl-dev 11:24:19 sounds interesting (and is easy to make interesting if it isnt) 11:24:33 problems: lack of information, using known monsters would usually be better 11:24:52 this is also a problem with the pan levels and the suggested god levels 11:25:00 Any love for an elephant level? Elephants, dire elephants, hellephants, weighted to have 10 or so hellephants by zig:27. Some herders that haste them if there exists a monster for that. 11:25:19 Or just moths of wrath. 11:25:23 :) 11:25:38 kind of limited threats 11:25:39 the above is what the demigod minion system is doing; rolling randomised and/or themed player character-alikes 11:25:59 so it could be leveraged for non-god-aligned characters as well 11:28:13 kilobyte: the problem is that reflection is the only threat in dwarf levels 11:28:20 for a char with good defenses 11:28:32 you can just stand in a corner, tank them, and take them down one by one 11:28:55 its incredibly boring, and not that hard 11:29:28 the dwarf monster set isn't really finished tbh 11:29:36 well, if you are immune to torment anyway 11:29:42 if you arent there might be a problem 11:29:55 galefury: if you are undead then you have to worry a little bit about dispel undead 11:30:09 its not smite targeted, so it doesnt matter much imo 11:30:13 i never had problems with it 11:30:13 also most chars aren't immune to torment :P 11:30:19 in zigs? 11:30:22 yes? 11:30:31 huh, how do they beat mummy levels? 11:31:05 makhleb healing can outpace the damage you take if you kill everything quickly enough 11:31:08 i certainly would go deeper than 15 or so without necromut or at least statue form 11:31:11 On my one time meeting mummies, I lured them off and ran for it. 11:31:13 *wouldn't 11:31:14 or you can use ddoor 11:31:23 or you can run past them, yeah... none of them are fast 11:31:43 I've also cleared tomb zig:27 levels with crusader-types with TSO 11:32:04 How many mummies at a time did you fight like that? 11:32:14 since TSO healing + angels + silence is pretty good 11:32:22 mhh, silence helps i guess 11:32:25 you have to try to avoid fighting the whole level at once, of course 11:32:56 Has anybody fought a deep silence pan lord level successfully yet? 11:33:01 or I've cleared late tomb levels with dispel undead and haunt on kiku chars 11:33:08 -!- Pacra has joined ##crawl-dev 11:33:20 jle: sure, with storms and/or tornado 11:34:09 Oh, cool. They always seemed the scariest thing to me. 11:34:40 pan lord levels are certainly the worst 11:35:00 Now they have torment too, don't they? 11:35:03 personally i hate angels more 11:35:19 Well, some rare ones of them. 11:35:26 but im sure its possible to make a case for pan lords too 11:35:45 angels are bad if you don't have a healing source and can't storm them at edge of line of sight, but otherwise they aren't a big deal 11:36:02 also easy to run/teleport past because there aren't that many of them 11:36:41 yeah i think im a bit biased because i used imbatornado against my only really deep angel level 11:36:45 I've cleared angel levels with 0 problems with refrigeration + makhleb, for instance 11:36:55 so all they do is stand at the edge and smite 11:37:00 because they dont want to move in 11:37:13 Code question. If I want to test out a simple design of my own, do I need to do something beside add to the list that begins in ziggurat.lua:328? 11:37:45 kilobyte: I know elliptic's opposition counts a lot more than minmay's, but I really not happy about Trog accepting rods. 11:37:58 (With weight:100 or whatever to get it to spawn more often) 11:38:34 dpeg: my reasons for opposing it are completely different from minmay's as far as I can tell 11:39:19 elliptic: did not mean to accuse you 11:39:37 I just don't find rods very interesting on chars who can cast spells, and I don't really see the flavour objection given that wands are apparently okay... 11:40:14 elliptic: Trog does not oppose magic, he opposes spells. 11:40:25 if rods are supposed to be like casting spells, perhaps they should actually use a skill other than evocations 11:40:30 are rods casting spells, though? 11:40:43 since it really isn't clear to me how using a rod is worse than zapping a wand 11:40:53 I mean, most wands duplicate spells, but they aren't casting spells 11:40:54 i think of them as the same sort of thing as wands, yeah 11:41:18 elliptic: not that I am happy with rods as such... 11:41:51 rods could maybe use some rethinking, yeah 11:42:08 I'd be against Trog banning rods because rods are kind of how some Trog followers get their tactical diversity. 11:42:57 What annoys me a little is that an oversight is turned into a feature, in a blink. Trog did not accept rods for the longest time, if I understood correctly. 11:43:03 dpeg: huh? 11:43:07 umm, no 11:43:11 the description is the bug 11:43:14 not him accepting them 11:43:18 trog has been fine with rods for ages at least 11:43:24 for as long as I've been playing 11:43:25 * dpeg is out of the loop 11:43:36 Bevoker was considered a very strong build in pre-SS days, IIRC. 11:44:50 That word rings an almost forgotten synapse in my brain. 11:44:55 Okay, I withdraw. 11:51:22 if we had a better assortment of misc evokables, though, I'd be in favor of removing rods completely 11:53:29 it doesn't involve any learnin', anyhow 11:57:17 Issue I have with rods: The fact that you can get them from staff acquirement is somewhat spoilery. 11:57:39 that *is* a bit odd 11:57:40 the manual describes them as staves 11:57:41 -!- ais523_ has quit [Quit: Page closed] 11:57:43 staff acquirement is somewhat spoilery 11:57:44 Unless that changed at some point and I forgot. 11:57:55 staff acquirement is sort of bad, yes 11:58:24 * dtsund ponders giving rod its own acquirement section 11:58:24 given that if you are a caster it is nearly guaranteed to be a specific item and if you aren't so much a caster then you will get a rod 11:59:39 maybe there should be some more staves, too, so you might get something other than what you wanted... 12:00:37 I've considered trying to tweak the staff acq code to make it less deterministic, probably by making the non-enhancer staves a possibility even if spellcasting isn't your top skill 12:00:40 Well, the one I typically *want* is energy, which I usually can't get. :P 12:01:14 Heh. Increase chance for energy if lots of your spells use a lot of food, maybe? 12:01:30 That's abusable 12:01:40 yeah, early on i want energy and wizardry, later on some element and maybe channeling 12:01:50 Not *very* abusable, mind, but the potential is there. 12:01:54 so usually i really cant get the one i want 12:02:07 By learning Tornado at almost-useless just before acquirement? 12:02:11 I was thinking something like 50% chance for the enhancer staff for your top skill, 50% chance for one of the four non-enhancers 12:02:14 Something like that, yeah. 12:02:17 okay, s/want/expect/ 12:02:35 elliptic: I think that'd be an improvement. 12:02:40 so you would no longer be guaranteed to get that staff of earth or whatever, but you would have a pretty good chance of getting something useful from it 12:02:46 though power and channeling are pretty bad usually 12:02:47 Less chance for the others if it already exists? 12:02:48 what about olgreb and wucad-mu? should there be a chance to get those? 12:02:55 jle: yeah 12:03:04 How about no chance for channeling if you follow Sif Muna 12:03:08 Or reduced chance 12:03:44 bye for now 12:03:48 -!- galefury has quit [Quit: ChatZilla 0.9.85 [Iceape 2.0.11/20110701115916]] 12:04:54 -!- mikee_ has joined ##crawl-dev 12:05:12 -!- ais523 has quit [Remote host closed the connection] 12:08:20 maybe trog should excommunicate for staff acquirement, too... 12:09:01 elliptic: would you like to brainstorm what we want of the future misc items? They shouldn't just be an assortment of cool tricks, but actually allow a playing style? 12:10:08 not sure I really have the ideas... to some extent I think they should just be an assortment of cool tricks (I'm thinking of stuff like disc of storms and tome of destruction), but they should be useful 12:11:14 elliptic: it was a question... We could say that for an Evoc-based playing style, go Nemelex. Fine by me. But I think that miscies should do things unavailable anywhere else, at least in part. 12:11:47 making staff acquirement less deterministic sounds good to me, yeah 12:12:54 same here 12:13:13 and breaking rods out into their own category? 12:13:15 dpeg: I think maybe we should just keep our eyes open for good ideas for interesting misc items and add them to the game 12:13:33 and once we have a few more, we'll have more of a basis for evaluating things 12:14:55 elliptic: hey, I made a wiki page for that :) 12:15:21 It has some really good ideas, in my opinion. (Not my ideas, I compiled this from elsewhere, mostly SF.) 12:15:46 yeah, there might be some good stuff already there waiting to be implemented :) I'm going to look now 12:17:00 make comments what you think would help the game most, please 12:17:08 Currently, it is just a mass of ideas 12:19:30 -!- mumra has quit [Quit: offski] 12:21:18 some good stuff there, yes. I'll try to make a list of the ones I think would be best sometime soon 12:24:57 cool 12:26:21 -!- Galefury has joined ##crawl-dev 12:30:07 -!- MarvinPA_ has joined ##crawl-dev 12:33:31 -!- MarvinPA has quit [Ping timeout: 245 seconds] 12:34:43 -!- ZChris13 has joined ##crawl-dev 12:38:56 -!- MarvinPA_ has quit [Ping timeout: 245 seconds] 12:42:23 How are zig ghosts treated currently? 12:43:10 they don't remember which zig level they died on 12:43:18 this means they usually generate early in zigs 12:43:43 it would be good to fix this of course... not sure how hard it would be but the ghost code is sort of scary to me 12:44:25 thanks 12:45:04 -!- MarvinPA has joined ##crawl-dev 12:46:06 -!- ais523 has joined ##crawl-dev 12:50:18 -!- ortoslon has quit [Ping timeout: 260 seconds] 12:55:13 sorted out ziggurat page 12:59:46 wow, it's so much more detailed now 12:59:51 note that the problem in angel levels are mostly the daevas, not the angels 12:59:56 because those are the ones that smite 13:00:58 i assumed angel was referring to genus but that is good to point out 13:01:17 adding pearl dragons isn't likely to make angel levels easier :P 13:01:49 pearl was nerfed a bit, right? 13:02:17 it was nerfed to a quokka in that vault if that's what you mean :P 13:02:27 oh no wonder it was easier =P 13:02:39 no i think it's speed 10 now instead of 13? 13:02:47 but in any case i agree 13:10:25 Unknown crash on an open level (https://crawl.develz.org/mantis/view.php?id=4619) by thevogonpoet 13:11:55 please improve what you can 13:12:36 Holy levels could simply have fewer monsters than other sets at the same depth. 13:12:55 they already do 13:13:06 could be even less 13:20:07 03galehar * red77eddab217 10/crawl-ref/source/wiz-you.cc: Don't crash when setting a skill above 27 in wizmode. 13:20:17 03galehar * racaf588c0d41 10/crawl-ref/source/skill_menu.cc: Don't show decimal for mastered skills. 13:25:38 -!- mikee_ has quit [Quit: HydraIRC -> http://www.hydrairc.com <- IRC with a difference] 13:28:07 03galehar * r6a0c9c1be515 10/crawl-ref/source/abyss.cc: Prevent abyss morphing from putting monsters on unhabitable terrain. 13:29:15 -!- ainsophyao has quit [Remote host closed the connection] 13:41:36 -!- ussdefiant has quit [Ping timeout: 256 seconds] 13:44:33 oh, I bet CIA's hooks only get run when ref is *changed*, not when it's first created... 13:48:33 Would it make sense for Lugonu to reward you for banishing dudes instead of killing them? It feels a bit odd that he doesn't give out piety when you effectively kill enemies with his chosen weapon type. 13:50:37 -!- Torokasi has quit [Ping timeout: 252 seconds] 13:56:05 that does sound rather odd 13:56:28 piety for banishing stuff would be cool, yeah 13:57:04 should it be more, less, or the same as for killing it? 13:57:24 I would favor the same, since you take them out just the same. 13:57:32 I mean, on the one hand Lugonu should be thrilled 13:57:42 potential new worshippers 13:57:48 But on the other hand you banish like every fifth foe. :) 13:57:52 on the other hand, aren't they in Abyss and still killable? 13:59:28 They are, though mechanics-wise most don't survive of course. 13:59:29 read the Lugonu wiki page, folks 13:59:50 Btw, have there been any reactions to the Hive cut? 13:59:56 piety for banishment is part of the big lugonu overhaul proposal but it'd be a nice improvement regardless of all the bigger stuff, imo 14:00:24 absolutely 14:02:08 sure 14:02:30 I have a possible "leading idea" for evocation: all items that use evocation should have limited charges. 14:03:04 You get a wide selection of (hopefully varied and wacky) effects, but they all run out and you need to move on. 14:03:12 is there a list of what items use evo somewhere? 14:03:26 Problematic: rods, evocable rings/artefacts, ... 14:03:33 I was going to say 14:03:36 Keskitalo: interesting. That's a departure of the one-off vs permanent use idea, not that we've done anything with that. 14:04:09 I don't think that idea makes much sense for rings, and certainly not for artefacts 14:04:13 What about disability to re-use, i.e. timers? 14:04:18 dpeg: You could work that in, so that you either could or couldn't use the one-off power with an evocable without charges. 14:04:35 artefacts are supposed to be essentially immutable 14:05:02 SamB: it is about not being able to blink twice in a row from the same source 14:05:11 not about sterilising the artefact 14:05:41 oh, if it would work like rod charges that would be fine then 14:05:51 Rods have timers, yeah. 14:05:54 yes 14:06:08 just not like wands ;-) 14:07:05 so, are you suggesting each power get its own timer? 14:07:34 dpeg: Well, timers make the items not run out, which would mean the "misc item rotation" wouldn't happen. 14:07:57 (You should also be able to use recharging on all misc items (well not decks)) 14:08:51 hm, needs more thought :) 14:08:54 So that would be a contrary idea. 14:10:21 All activatable items suddenly work like rods, with charges that can be expanded upon by reading recharging at them? 14:10:49 jle: does not match Keskitalo's original concept 14:11:10 Oh right, limited charges, sorry. 14:11:14 jle: Like wands 14:11:34 add lantern of shadows to the problematic list I guess 14:11:55 not that many misc items were unproblematic now :) 14:11:57 Keskitalo: it would fit the theme if you could combine misc items 14:12:39 (I like the single-use/permanent-use idea, just "multiple-use" instead of "permanent-use" :)) 14:12:52 Heh. I use LoS all the time when ninjaing runes. It's a fun item though, the tradeoff of limited attack options for less visibility is interesting. 14:13:08 (any overhaul would be great, I'm not adamant that this "all have charges" thing would be be-all end-all) 14:13:40 overhaul is just a question of someone having a concept plus effort :) 14:13:45 jle: It's one of the better misc items in my mind to be truthful 14:14:07 yes 14:14:11 Keskitalo: Yes. I think this is because a lot of the misc items are not very good. :) 14:14:17 It could well have a limited charges, and spend one per wielding 14:14:21 I don' 14:14:28 t think it uses Evoc skill right now? 14:14:39 No. It's wield to activate. 14:14:51 Unless it produces more shadows at high evocations I guess. 14:14:59 I think infinitely reusable misc items are okay so long as there are tradeoffs for using them, really 14:15:08 I don't really pay attention to that part. 14:15:30 dpeg: I haven't tought about this idea much more than that, apart from vague ideas that there should be more benefits from high Evocations, with wands as well 14:15:55 elliptic: you recall the time when rods could be spammed? There was a reason why Erik and Darshan (over)nerfed rods for 0.1. :) 14:15:56 crystal ball of energy punishes you with MP drain and stat drain and confusion, for instance 14:16:13 dpeg: I wasn't around then, no 14:16:20 elliptic: do you think it is a well-designed item? (I have never used it.) 14:16:38 Lantern of shadows has two negative enhancers for spells, it ruins your stealth and of course you can't hit things. 14:16:58 jle: yeah, I think lantern is all right also 14:17:34 dpeg: crystal ball of energy is reasonably designed with creating a risk/reward tradeoff, at least 14:17:46 yeah, you always need a backup plan if you use it 14:17:50 since the negatives of it can be quite serious when you are using it 14:17:56 it is a difficult item to balance though 14:18:08 and the int drain can drain resources over time 14:18:22 if you use cboe a lot 14:18:25 I think the cloud traps are a definite improvement over most traps.. often you at least have a situation where you have to get out of the clouds to minimize the damage, i.e. you need to make some decisions after walking into them, even if there aren't any monsters around. 14:19:01 they suffer from the same problem all clouds do, they cant overwrite existing clouds 14:19:06 stuff like rods and the elemental summoning devices are not good though as is... infinitely reusable for no particular price 14:19:07 also they are very good for luring enemies into 14:19:15 but both only works with T&D to detect them first 14:19:20 so this is good imo 14:19:21 Galefury: how? They're single use. 14:19:24 keskitalo: my main concern about them is that they make blink much better 14:19:26 so? 14:19:26 yes, that is no problem 14:19:35 a single use of miasma can be pretty nice 14:20:05 Are the elemental summons infinitely reusable? I thought the risk for a hostile elemental went up if one already exists. Or is the risk capped at some (low) number? 14:20:10 i wouldnt usually lure monsters toward them or anything, but they can be useful 14:20:21 jle: well, I didn't mean infinitely reusable in a row 14:20:28 elliptic: Right, right. 14:20:32 elliptic: Didn't think that was possible! 14:20:39 elliptic: Do you have a problem with magic spells? 14:20:42 but like rods, they are infinitely reusable over the course of the game 14:20:46 keskitalo: what? 14:21:01 For Blink to be any better than it already was. ;) 14:21:02 dtsund: I have a problem with making an already great spell even better 14:21:12 keskitalo: heh 14:21:13 No, you misinterpret my meaning. 14:21:20 Spells are also infinitely reusable. 14:21:22 elliptic: I don't disagree with your point about infitely usable misc items btw 14:22:14 dtsund: this is true, yes... but spells require lots of different skills to use 14:22:22 evokables only require one, and it is a cheap one 14:22:47 Yeah, I've (vaguely) tought about the skill requirements too 14:22:50 hmm, is there someplace we could take the addresses of some of the operators on coord_def ? 14:22:51 There are also armour tradeoffs to spells. 14:22:56 Yes, but you get many different spells from one skill too 14:24:43 dtsund: you are right that rods aren't that different from spells... this is part of what I don't like about them :) 14:24:59 There are evocables that use multiple skills already 14:25:08 The elemental items 14:25:39 Add tukima's weapon rack that creates dancing weapons based on your weapon skills and evoc! 14:26:40 As I see it, you trade "magical skills + variety + armour restrictions + spell slots restrictions" (spells) against "Evocations skill + harder access + no further restrictions" (rods), I don't think that's worth it. 14:27:41 dpeg: Heavier misc. objects would create another restriction (inventory) 14:28:04 yes 14:28:32 It would help if rods have different spells, but that is a very long way. And probably not worth it :) 14:28:42 Some do already. 14:28:47 I love the rod of inaccuracy. 14:28:56 I'd like to see misc. evokables be either (a) limited use (wands, tome of destruction) or (b) dangerous use (disc of storms, crystal ball, lantern of shadows) 14:29:49 or both of course :) 14:31:51 -!- einokesk has joined ##crawl-dev 14:33:07 -!- Keskitalo has quit [Quit: Lost terminal] 14:33:55 -!- einokesk is now known as Keskitalo 14:34:52 There's striking, smiting, summon swarm and bolt of inaccuracy that are only in rods. Is that all? 14:36:11 -!- syllogism has quit [] 14:37:51 sounds about right 14:38:42 I had the vault with the shedus and the ophan, that was nice. 14:39:19 it is guaranteed currently on d:17 in trunk 14:40:57 That's where I had it. 14:46:42 by the way, spears of dragon slaying are getting identified upon hitting my demonspawn 14:46:50 no idea if they actually did extra damage, i hope not 14:46:54 also no idea if it happens for other species 14:48:18 galefury: hm, I thought that bug was gone for some reason 14:48:29 maybe I just haven't run into dragon slaying weapons :) 14:51:09 -!- ussdefiant has joined ##crawl-dev 14:52:15 Oh, I noticed that when my hoak died yesterday, the reaper's weapon was ID'd as dragon slaying. 14:53:03 Not sure if it was post- or pre-hit (not the killing blow either way, so I didn't get it from end-of-game notes). 15:10:17 -!- Zaba has quit [Ping timeout: 255 seconds] 15:16:05 New Grinder Tile (https://crawl.develz.org/mantis/view.php?id=4620) by Curio 15:18:10 so ... when compiling with DEBUG but not NO_OPTIMIZE, "inline" methods don't generally get out-of-line versions generated, right? 15:18:58 this makes it tricky (i.e. impossible) to call them from GDB 15:18:59 -!- Zaba has joined ##crawl-dev 15:20:20 this can be a pain for stuff like coord_def's operator ==... 15:27:56 -!- Torokasi has joined ##crawl-dev 15:29:52 dpeg: for Trog and rods, I don't have an opinion. I disabled the highlight only because 1. it was misleading as for the current state, 2. the commit caused a bunch of other problems (Poison Arrow having the prompt even if you use it against undead, etc) 15:30:17 hmm, I've an idea ;-) 15:30:52 dpeg: so don't assume I'm against you there... but I'm neither against elliptic. It's one of the cases where I'm completely without a preference to either side. 15:32:48 dpeg: as for Ziggurat ghosts... the whole Ziggurat depth code is a pile of crap. For example, you can't ~Zig:27, the kill stats will claim everything you killed in a Zig was on the last level, save file is peppered with invalid names, etc, etc. 15:33:42 * SamB makes a note not to add zig*.cc to his shopping list, ever 15:35:12 -!- evilmike has joined ##crawl-dev 15:36:08 -!- Pingas_ has joined ##crawl-dev 15:36:13 -!- Pingas has quit [Quit: Spottet Sigmund] 15:36:23 -!- Pingas_ is now known as Pingas 15:38:07 kilobyte: not sure if you saw it but I withdrew my opinion on Trog vs rods. It seems I misunderstood something... I thought that Trog used to dislike rods and it was by accident that he does not anymore. 15:38:38 So let's leave everything there and hope this adds to the motivation of doing something about miscellaneous items 15:38:55 Regarding Ziggurats: it is greensnark's code, it can't be bad :) 15:48:29 -!- Keskitalo has quit [Quit: leaving] 15:49:34 is it intentional that all hell upstairs are listed as portals to hell in ctrl-O? 15:49:48 i mean, sure, they are portals to the vestibule, but its pretty spammy 15:52:48 Galefury: not needed in any case. 15:54:47 so i'm reading the "hopping" (rune lock page on the wiki) 15:54:55 this idea got my attention: "a large, very nasty vault (think Swamp branch end); in front a portal with rune requirement, at the end a portal without rune requirement (effectively, this would mean that players have a third battleground beyond the two easy Lair subbranches for going deeper)" 15:55:53 that sounds nice 15:55:54 this seems like a fairly good way to go, if a rune lock is put in, at least of all the ideas on that page. 15:56:24 it lets the (hard part) of the d:14 vault function as a sort of gauntlet or something, with the rune lock being a way to skip it 15:56:29 evilmike: it is the softest version, yes. 15:56:58 I prefer the simplest one: small vault (no monsters) with the portal to show your rune. Plus another vault with shafts. 15:58:43 being able to intentionally shaft past sounds softer to me actually 15:59:13 I just really like the idea of being able to "prove your worth" - fight past a horde of ood stuff, and show that you don't need a rune 15:59:15 elliptic: well, you don't get up without a rune. Forced to get silver, so it's nasty. 15:59:45 dpeg: that could be quite a newbie trap 15:59:49 evilmike: my beef with that is the content. We have to make vaults and stuff, just to test the rune lock idea. I'd rather test it on its own. 15:59:51 Yeah, I was about to say that. 16:00:05 "Crap, I left my stash in Lair!" 16:00:07 ais523: pre-identified shafts, as it says on the wiki. 16:00:12 dpeg: sure, it is a tradeoff... I meant soft in the sense of it being easiest to get to d:14 :) 16:00:21 dpeg: /still/ a newbie trap unless there's a huge warning not to use them without a rune unless you're crazy 16:00:26 elliptic: yes. Different measures for soft :) 16:00:29 and I agree that something simple like what you just said is what we should test now 16:00:35 if you want I could put together a vault for testing that. If the rune valt didn't go in it could easily be changed into a loot vault 16:00:44 I have a feeling the level might need to be -cTele though 16:00:51 using the shaft is going to be nearly always the Wrong Decision unless you want to bump up the difficulty level 16:01:00 evilmike: KFEAT: no_tele_into for the vault itself would do 16:01:00 more exciting things like vaults can be added later 16:01:01 the idea evilmike mentioned is my favourite implementation of the concept too 16:01:08 ais523: that's not at all clear to me 16:01:14 so if he's volunteering to make a vault for it... :) 16:01:43 well, it depends on how much that particular version of the idea is supported, whether it has any chance of happening soon, etc 16:01:45 if you take your stash with you, shafting beyond seems clearly better than dying in a lair branch if you don't feel like your character is well-suited for that 16:01:55 evilmike: sure, go ahead. _That_ work is never wasted. Even if we scrap the rune lock idea because it is impractical or player send letter bombs or whatever, we will keep the vault. 16:02:04 -!- mumra has joined ##crawl-dev 16:02:22 alright 16:02:58 There was no outrage because of Hive cut, it seems. So it was long overdue or the players are not aware :) 16:03:13 I'm not surprised that there's no outrage 16:03:18 most players really hate the hive 16:03:36 There's always someone who goes "Bbbbut my spriggans..." 16:03:40 usually people who enjoy it are people with ignite poison 16:03:41 also we've been carefully spreading rumours about hive being removed for a version or two 16:03:48 so people are prepared! 16:03:53 evilmike: It used to be four levels and I did get complaints when suggesting to cut it in half. 16:04:04 elliptic: yes, that might have helped :) 16:04:05 haha yeah hive:4 was torture 16:04:18 dpeg: i was watching ##crawl earlier when the CIA notice went up ... literally nobody even noticed, there were certainly no comments 16:04:24 evilmike: and players were campaigning for loot compensation... funny times 16:04:34 of course a lot of players don't know about it yet 16:04:51 hive:2 does have ok loot, but many people don't know about all the loot vaults 16:04:56 i, for one, will not mourn hive 16:05:02 in one of them it's extremely well hidden 16:05:02 the best loot in hive didn't come from the loot vaults 16:05:08 and i have yet to find anything good in the H: 2 loot vaults 16:05:11 just random stuff on the ground... two levels worth of it 16:05:20 elliptic: yes. Plus piety. 16:06:01 If we decide that no replacement is needed, I have to find another use for my fine-tuned racing portal vault concept =) 16:06:01 Oklobs vault tended to have some pretty good stuff 16:06:05 dpeg: removal of hive is much more of an issue with elyvilonites than with spriggans, I think 16:06:15 given how food-intensive pacification is 16:06:27 but I still am not worried about it 16:06:35 elliptic: E users fare well enough, in my opinion. You don't have be pure. 16:06:50 right 16:06:51 And if you're pure, we can expect you to stretch a little more. 16:07:48 btw, are volcanoes still generated? I haven't seen one in ages. 16:07:55 Yes 16:08:02 They're just rarer than ice caves 16:08:10 And only appear in Orc? 16:08:11 they've always been fairly rare i think, fewer places they can spawn 16:08:18 RichardHawk: also Lair 16:08:24 Never seen a Lair one. 16:08:40 default-depth: Lair:1-8, Orc:1-4, Hive:1 16:09:00 so even rarer now! 16:09:23 default-depth: Lair:1-8, Orc:1-4, Hive:1 16:09:25 damn :) 16:09:56 I have never seen one in lair! 16:10:05 they are pretty rare in lair for some reason 16:10:06 all my volcanoes have been in orc, I think 16:10:10 ice caves are far more common 16:10:25 !lm * br.enter=volcano cv>=0.9 s=oplace 16:10:30 429 milestones for * (br.enter=volcano cv>=0.9): 93x Orc:1, 85x Orc:3, 84x Orc:2, 79x Orc:4, 42x Hive:1, 10x Lair:2, 8x Lair:4, 7x Lair:5, 6x Lair:3, 6x Lair:7, 5x Lair:1, 4x Lair:6 16:10:31 they all CHANCE:5% 16:10:34 that's really weird 16:10:47 perhaps the Lair ruination? 16:11:14 * dpeg really thinks that races make for interesting minigames. If the Nethackers won't listen, I'll get a race in Crawl. 16:11:16 !lm * br.enter=ice cave cv>=0.9 s=oplace 16:11:16 No milestones for * (br.enter=ice cave cv>=0.9) 16:11:18 probably worth bug reporting if no one is going to fix it right now... 16:11:22 !lm * br.enter=icecv cv>=0.9 s=oplace 16:11:29 2099 milestones for * (br.enter=icecv cv>=0.9): 202x Lair:1, 176x Lair:2, 171x Lair:3, 168x D:11, 122x Lair:4, 118x D:12, 111x Lair:5, 101x Lair:6, 89x Lair:7, 88x D:13, 82x Lair:8, 79x Orc:2, 64x Orc:4, 64x Orc:1, 57x Orc:3, 51x D:14, 43x D:15, 32x D:16, 26x D:19, 26x D:17, 23x Elf:3, 20x Vault:7, 19x Elf:2, 19x D:18, 19x Vault:3, 18x Vault:1, 17x D:20, 16x Vault:5, 16x Elf:4, 15x Elf:1, 15x Vaul... 16:11:42 ice caves don't seem particularly rare in lair though 16:12:11 I don't understand dungeon generation/vault placement enough to have any idea what is going on here but it does seem strange 16:13:15 not sure anyone does, perhaps Zaba 16:13:29 evilmike: yes 16:14:38 Some of the volcano maps look like a lot of fun. 16:14:49 they often time out in orc 16:14:51 due to bubbles 16:14:54 ice cave entries don't have any real difference from volcano entries, that I can see 16:14:57 and they seem very rare in lair 16:14:59 Galefury: this is true 16:16:18 i thought volcanoes weren't in any longer 16:16:27 i haven't come across one in half a year or so 16:16:50 both Icecave and Volcano entry vaults have CHANCE: 5% 16:17:15 Icecave: default-depth: D:11-19, Lair:1-8, Orc:1-4, Elf:1-5, Vault:1-7 16:17:25 Volcano: default-depth: Lair:1-8, Orc:1-4, Hive:1 16:17:35 That is the reason, 16:17:37 I think 16:18:10 hmm, odd that i haven't come across any in lair at all 16:18:16 i'm probably just unlucky 16:18:22 2100 : 430 = 33 : 13 16:18:35 not quite, but at least a bit 16:18:41 -!- galehar has joined ##crawl-dev 16:18:43 does that explain why ice caves are so much more common in lair...? 16:18:47 Xiberia: me neither, for a long time, that is why I asked 16:18:52 evilmike: no :) 16:19:06 re 16:19:13 !seen galefury 16:19:29 I last saw Galefury at Mon Sep 26 21:14:57 2011 UTC (4m 31s ago) saying and they seem very rare in lair on ##crawl-dev. 16:20:18 Galefury: thanks for pointing me to the abyss monster bug. I introduced it when I fixed items :S 16:26:13 could it be because map names are sorted alphabetically and volcano comes last and there's a lot of concurrency in Lair for vault placement? 16:26:35 wow, Crawl's makefile is *really* finicky 16:26:35 galehar: that is an ingenious idea. 16:26:52 galehar: Competition not cuncurrency I guess? 16:27:16 I add a single normal target to the make command line, and it decides it has to start from scratch 16:27:19 ghallberg: yeah right. Sorry, translation mistake. 16:27:52 No probes, I only mentioned it because it's almost the same in swedish :D 16:28:13 SamB: yeah, that happens a lot. It's worse under windows. Use ccache if you can. 16:28:23 Competition = konkurrens 16:28:25 rename volcano to ivolcano and icecv to vicecv and see whether it makes the frequency go the other way? :P 16:28:50 Can anyone operate greensnark's vault generation test file? 16:29:02 galehar: no worries, kilobyte already told me that 16:29:05 so I am 16:29:24 dpeg: genlevel.lua? 16:29:46 I forgot. 16:29:57 But it produces a loooong list of map statistics. 16:30:17 galehar: so, is ccache available on Windows? 16:30:49 SamB: I don't know 16:31:55 dpeg: test/vault_generation.lua? 16:32:24 SamB: yes it is; i installed it but didn't get it working yet 16:32:25 let me have a look 16:34:07 crawl --mapstat 16:34:10 mumra: oh, what's the trouble? couldn't figure out how to get it invoked? 16:34:21 if section P of advanced.txt is still valid 16:36:16 only one dash 16:37:07 both - and -- work 16:37:12 does anyone here know of using "9" (+5 depth monster) is the same as placing a monster on d:14 with "place:D:19"? 16:37:24 they seem to give slightly different monster sets but maybe that's just me imagining it 16:37:52 evilmike: no idea 16:37:54 dpeg: it seems like it only reports the veto %. And it crashes in the abyss :( 16:38:24 galehar: does it create a huge file full of vault names? 16:39:28 dpeg: no because it crashes before 16:39:33 I'll see if I can fix it 16:39:36 what a pity, so it is dated 16:39:57 no, it seems to be working alright. Just the new abyss make it crash 16:40:02 ah 16:42:44 oh, evidently it gets confused by explicit targets that contain words like "debug" and "profile" as substrings 16:44:46 SamB: yeah i installed it but it didn't seem to make much difference, so i assume i have to enable it somehow, and/or tweak the settings 16:45:09 mumra: yeah, it probably needs enabling 16:46:37 here on Debian, I had the choice of installing symlinks to the "ccache" executable or making the build system prepend "ccache " before the compiler name in compilation commands 16:47:08 the former seems likely to be out of reach on Windows ;-) 16:47:35 SamB: i don't see why; Windows has symlinks 16:47:35 (well, by "installing" I mean "putting somewhere along the PATH") 16:47:46 mumra: well, some versions do 16:47:47 and it has a PATH variable 16:47:59 certainly PATH is not the obstacle 16:48:17 unless you want to chain several gcc wrappers, just having it in PATH is enough 16:48:47 on Debian, export PATH="/usr/lib/ccache:$PATH" 16:48:59 for some reason I went the other route and added GCC= and GXX= lines to my make invocation 16:49:42 er. not lines... 16:49:45 it's when you want colorgcc or distcc when mess starts. But I like highlighted errors and warnings so much, especially in parallel builds! 16:50:15 oh, well, I use compilation-mode anyway so I don't really need colorgcc 16:50:55 SamB: ok - just looked it up again 16:51:03 SamB: msysgit includes /src/ccache-win32/release.sh? 16:51:17 (ignore the question mark, paste error) 16:51:48 so, i ran that, and it downloaded and compiled ccache-win32, including applying a patch 16:52:46 SamB: it also ran an msysGit post-install.sh script; so initially i assumed that had done everything needed 16:53:50 yes, well, it doesn't come with the mind reading device needed to tell which compiles you want it to apply to, so ... 16:56:29 SamB: hilariously, i never added the msysgit/bin directory to my windows path, which could explain everything 16:58:39 03galehar * r11430422ed69 10/crawl-ref/source/dungeon.cc: Fix a crash with crawl --mapstat. 16:59:03 yay 17:03:00 the stats are quite interesting. Lair:8 has 56% level vetoed 17:04:00 because of the large endgame vaults perhaps 17:04:19 galehar: can you figure out the Volcano/Icecave anomaly? 17:05:06 it's confirmed. With 30 generations, volcano has only been generated in orc 17:05:09 dpeg: did you see my conv with napkin earlier about the wiki text generator? 17:06:27 mumra: a bit. Had no time, I hope you're doing alright. I think you know our intentions well enough :) 17:06:41 galehar: and if you swap the names? 17:08:03 dpeg: No problem. Haven't done any real work on it - but basically i can modify that text plugin to emit something very similar to the original docs (but with e.g. chapter references properly worked out from wiki links) 17:08:54 kilobyte: would it bother you if the vault docs were all in one file instead of four? 17:13:35 -!- Zaba has quit [Ping timeout: 255 seconds] 17:23:41 -!- mumra has quit [Quit: offski] 17:24:09 I wish CIA could announce new git refs 17:26:46 Basic Crystal Plate Armour Tile (https://crawl.develz.org/mantis/view.php?id=4621) by Curio 17:26:55 -!- Zaba has joined ##crawl-dev 17:28:37 Zaba: hi! 17:28:45 We need you, portal vault generation is borked. 17:31:45 -!- Zaba has quit [Ping timeout: 260 seconds] 17:32:10 hm, didn't work 17:32:10 03naesten 07refs/merge-requests/9 * r194cea5d4f2c 10/crawl-ref/source/ (Makefile Makefile.obj dbg-inlines.cc): Allow some "inline" methods/functions to be called from GDB. 17:33:23 (I guess those of you who use "make debug" can probably already call those methods/functions, but those of us who debug optimized builds have a harder time.) 17:34:58 SamB: why don't you use "make debug" for debugging? 17:35:08 galehar: it's so spammy 17:35:23 also, it's huge 17:36:45 if we had a more flexible option system, we can use options to control debug message spam. For now, just use show_more = false 17:36:59 *we could 17:37:18 and I like to use a binary I can actually play with, so I can sit there and play with the debugger attached and investigate crashes and so forth 17:37:31 -!- mumra has joined ##crawl-dev 17:37:40 also, since the build system isn't really set up for multiple build trees ... 17:38:46 -!- Letchik has joined ##crawl-dev 17:39:03 anyone trying to debug the profiling build would also benefit from this, FWIW 17:39:32 SamB: that's counter to the very purpose of optimized builds 17:39:44 kilobyte: what is? 17:39:57 this doesn't actually prevent the inline versions being used or anything 17:39:59 and debugging these is a pain... 17:40:20 it just generates out-of-line copies that GDB can use 17:40:30 yeah but it bloats the binary when you specifically requested it to be meant for play rather than debugging 17:40:54 it only activates when DEBUG is set, though 17:41:07 what's the point in debugging an optimized build unless you're specifically looking for a bug in optimization? 17:41:50 Butterfly mesmerize is not doing anything (https://crawl.develz.org/mantis/view.php?id=4622) by Kyrris 17:41:55 also I'm running on a PII 450 MHz and I'm afraid the unoptimized build would be too slow 17:42:10 you can do that by -O0 -- and the binary will be actually debuggable. With -O2 most variables will be hidden away. 17:42:18 -!- Zaba has joined ##crawl-dev 17:42:31 i've run crawl debug on a PII 17:42:41 huh, I haven't had too much trouble... 17:42:53 I mean, sometimes they aren't available as soon as you expect 17:42:57 do you mean, you compile it on a PII 450 with optimization? 17:43:20 I for one would prefer hour-long compiles rather than two hours... 17:43:29 hmm 17:43:39 (ok, I actually prefer fourty seconds ones :p) 17:43:56 I have trouble telling the difference between one- and two-hour-long compiles 17:44:20 I did nothing to the bubbles. 17:44:53 anyway, it's the play speed/executable size I'm concerned with 17:45:16 plus, don't optimized builds give better warnings? 17:46:17 -Os is typically best in terms of play speed nowadays, because smaller is often faster due to cache effects 17:46:26 -O0 is basically only useful for debugging 17:46:31 but can be very useful there 17:47:04 next question: why do we use -O2, not -Os, anyway? 17:47:54 better speed > better space 17:48:03 but ais523 just said ... 17:48:07 nah, better space is typically faster nowadays because of cache 17:48:10 just people stick to the old ways 17:48:16 -O2 is a good compromise in terms of compilation speed, btw 17:48:26 -O3 versus -Os is the decision if you go beyond there 17:48:36 -O2 is often a good compromise between fast compile and fast play 17:49:06 better binary size doesn't mean anything about the memory needed to be used to play the game 17:49:34 just compared it, -O2 compilation takes user:5m38.961s, -O0 user:4m14.208s 17:49:51 that's on a relatively fast box 17:50:34 oh hah, a P2 350Mhz box at work is still connected, I can benchmark it there :p 17:55:14 I wouldn't say which optimization level is the best until I benchmarked perfomance with each level given a specific application 17:56:49 indeed, I'm just going general principles here 17:56:50 well, -O0 and -O1 can be ruled out 17:58:41 good night 17:58:56 -!- Textmode has joined ##crawl-dev 17:59:03 -!- Galefury has quit [Quit: Stay sane inside insanity!] 18:03:54 anyway, I don't think people who compile with DEBUG but not NO_OPTIMIZE are too likely to be complaining about the slight files size increase, <2 cache lines per level, and <2 pages of RAM that this would cost ... 18:04:14 03MarvinPA * rb64770d53168 10/crawl-ref/source/dat/database/monspeak.txt: Remove a misleading speech line 18:05:18 ... and those who do compile with NO_OPTIMIZE, well, I'm not sure what exactly it might cost them, but well nobody forced them to use NO_OPTIMIZE 18:06:47 -!- Letchik has left ##crawl-dev 18:08:01 -!- dpeg has quit [Quit: zzz] 18:09:17 kilobyte: I was about to say "five freaking minutes?" but then I noticed the fast box bit. 18:10:43 the ratio was still nowhere near the 2:1 he had supposed earlier, though 18:13:34 dtsund: s/fast/a year old cheap one/ 18:15:32 kilobyte: what difference does that make? 18:15:34 * kilobyte wishes he had a 2*8-core*HT modern box. Or better. 18:15:36 Still seems fast to me. 18:15:52 I have to do all my compiling on a dual-core 2Ghz laptop. 18:16:35 2.8Ghz is not really "fast". Not a laptop or a phone, though. 18:16:47 are eldritch tentacles ok to place in vaults now? They seem to work fine but I want to know if there are any weird/unforseen issues 18:16:56 but hey, they're going to release a 5-core phone now... 18:17:13 * dtsund mutters about phones more powerful than his laptop 18:17:13 I thought it would be.. amusing to have one by the rune in a mnoleg vault (would be easier than giving mnoleg the spell, like marvinpa wanted) 18:17:15 evilmike: if they were unforeseen, how would we know about them? 18:17:25 because calling your mom is much more efficient with 5 cores 18:17:39 dtsund: nobody said they would be very powerful cores... 18:17:42 yet 18:17:42 well, they're a monster meant for a weird summon spell 18:18:09 evilmike: I'd expect problems without the malign portal 18:18:14 evilmike: or giving mnoleg tentacles like I wanted 18:18:22 though I suppose it's best to ask about anything with "eldritch" in the name 18:19:04 -!- galehar has quit [Remote host closed the connection] 18:20:08 MarvinPA: <3 <3 <3 <3 <3 for the butterfly change 18:20:32 heh, i think it just slipped through when the huge amounts of new speech got added a while ago 18:26:09 03kilobyte * r41f9acc5e781 10/crawl-ref/source/rltiles/ (7 files in 3 dirs): Move "deformed foo" tiles to vault/ 18:26:10 03kilobyte * rf1f5d41cf228 10/crawl-ref/source/rltiles/ (5 files in 4 dirs): Curio's new Grinder, consistent with new shadow imps. 18:26:10 03kilobyte * r03af59c2f47f 10/crawl-ref/source/ (19 files in 7 dirs): Organize tiles a bit. 18:26:11 03kilobyte * redd277bc17c8 10/crawl-ref/source/rltiles/dc-mon/undead/wight.png: Fix wight transparency. 18:26:11 03kilobyte * r485a86c7bb14 10/crawl-ref/source/rltiles/UNUSED/monsters/shedu.png: Put co's alternate shedu tile in UNUSED/. 18:26:11 03kilobyte * r4e0d75b8c5af 10/crawl-ref/source/dat/dlua/ziggurat.lua: Use all currently available holies on holy Zig levels. 18:26:11 03kilobyte * rec5e27b6c1e0 10/crawl-ref/source/ (mon-data.h mon-gear.cc): Give cherubs their gear, differentiate stats from angels. 18:26:11 03kilobyte * r4ab68a9f040f 10/crawl-ref/source/ (8 files in 4 dirs): Decapitalize "Angel", "Cherub" and "Daeva". 18:26:12 03kilobyte * reb58c105a575 10/crawl-ref/source/dat/dlua/ziggurat.lua: Add cherubs to holy Zig levels. 18:27:11 have cherubs actually been tested at all? I have no idea what they are 18:27:27 @??cherub 18:27:27 Cherub (16A) | 04UNFINISHED | Speed: 15 | HD: 12 | Health: 84-119 | AC/EV: 10/20 | Damage: 25, 10 | Flags: 08holy, see invisible, fly, !sil | Res: 06magic(128), 10elec++, 03poison, 04rot, 13neg+++, 08holy | XP: 2087 | Sp: minor healing. 18:29:05 @?? angel 18:29:06 Angel (16A) | Speed: 15 | HD: 12 | Health: 84-119 | AC/EV: 10/20 | Damage: 25, 10 | Flags: 08holy, see invisible, fly, !sil | Res: 06magic(128), 10elec++, 03poison, 04rot, 13neg+++, 08holy | XP: 2087 | Sp: minor healing. 18:29:24 ??cherub 18:29:24 I don't have a page labeled cherub in my learndb. 18:29:33 i guess they're unfinished. as in not yet different from angel 18:30:33 pearl dragons in zig levels?? 18:31:08 -!- Textmode has quit [Ping timeout: 258 seconds] 18:31:08 cherubs actually work since they're just angels with different name/tile 18:31:35 evilmike: 01:26 < CIA-34> kilobyte * rec5e27b6c1e0 /crawl-ref/source/ (mon-data.h mon-gear.cc): Give cherubs their gear, differentiate stats from angels. 18:31:50 oh, heh 18:32:42 just using what due wrote in his holyhaul document, not sure if he still means that 18:33:23 oh, gretell doesn't have the latest build yet, of course 18:33:33 @??cherub 18:33:34 cherub (07A) | Speed: 10 | HD: 14 | Health: 96-137 | AC/EV: 10/20 | Damage: 25, 10 | Flags: 08holy, see invisible, fly, !sil | Res: 06magic(149), 05fire, 10elec, 03poison, 04rot, 13neg+++, 08holy | XP: 1965 | Sp: minor healing. 18:33:46 @??daeva 18:33:46 daeva (08A) | Speed: 10 | HD: 14 | Health: 100-137 | AC/EV: 10/13 | Damage: 25, 10 | Flags: 08holy, see invisible, lev, !sil | Res: 06magic(149), 03poison, 04rot, 13neg+++, 08holy | XP: 2887 | Sp: smiting (7-17). 18:33:52 and now he does :P 18:34:04 so... smiteless daevas with better defenses? 18:34:09 @?? angel 18:34:09 angel (16A) | Speed: 15 | HD: 12 | Health: 84-119 | AC/EV: 10/20 | Damage: 25, 10 | Flags: 08holy, see invisible, fly, !sil | Res: 06magic(128), 10elec++, 03poison, 04rot, 13neg+++, 08holy | XP: 2087 | Sp: minor healing. 18:34:31 I guess its more like an angel-daeva cross actually 18:34:32 evilmike: or rather, holy yaktaur captains 18:34:38 more like slightly weaker rf angels 18:34:43 weaker, slower 18:34:48 oh nice, they have a ranged attack 18:35:16 er, not weaker, i guess. just less xp 18:37:35 brown A is pretty ugly 18:38:07 but hey, now that we have 'an angel', we can add a unique named Angel 18:39:10 that might be a bad idea 18:39:20 wouldn't they get the same description? 18:39:22 having capitalized "Angel" but lowercase "devil" was religious discrimination 18:39:32 yeah, the capitalizations were bad 18:39:42 SamB: i have no idea if descriptions are case sensitive 18:39:42 we need to have "red angel" and "blue angel", clearly 18:39:53 @??imp 18:39:53 imp (055) | Speed: 10 | HD: 3 | Health: 9-18 | AC/EV: 3/14 | Damage: 4 | Flags: 05demonic, fly, regen, !sil | Res: 06magic(36), 05hellfire, 03poison, 04rot, 13neg+++, 13torm | Vul: 12cold, 08holy++ | XP: 48 | Sp: blink. 18:39:54 are tier 1 demons still capsed? 18:39:59 this change still didn't happen? :( 18:40:02 @??executioner 18:40:02 Executioner (151) | Speed: 20 | HD: 12 | Health: 49-86 | AC/EV: 10/15 | Damage: 30, 10, 10 | Flags: 05demonic, evil, see invisible, !sil | Res: 06magic(144), 05fire, 02cold, 10elec, 03poison, 04rot, 13neg+++, 13torm | Vul: 08holy++ | XP: 2373 | Sp: pain (d14), haste. 18:40:06 so I see they are 18:40:43 kilobyte: having 4 unholy branches and 0 holy branches (soon to be 1) is religious discrimination! 18:40:46 :P 18:40:56 or 5, if you count pan 18:41:22 abyss? anyway, crawl is a bad place for Good 18:41:30 Where's the wiki page for the FSM branch? 18:41:31 .!arena angel col:red name:red n_adj v angel col:blue name:blue n_adj 18:41:31 yes 18:41:41 Fyren: <3 18:41:55 Fyren: oh hrm, dpeg won't allow pirates :( 18:41:56 Fyren: :D 18:42:04 kilobyte: ah, but what about corsairs? 18:42:45 fsm, unfortunately, doesn't meet the old criteria, which is 300 years i think 18:43:01 hmm right... we can have corsairs, buccaneers, parrots and ninja corpses 18:43:14 spriggan ninjas 18:43:20 what about surfers? 18:43:41 surfing spriggachu! 18:43:53 Suddenly, Missingno 18:44:02 (is that one guy still an ex-surfer?) 18:44:14 yes 18:44:16 what about giant goldfish 18:44:20 in that he's no longer a surfer 18:44:23 ??anachronism 18:44:23 I don't have a page labeled anachronism in my learndb. 18:44:28 ??timecop 18:44:28 I don't have a page labeled timecop in my learndb. 18:44:29 sigh 18:44:36 it's in there somewhere i swear 18:44:59 https://gitorious.org/crawl/crawl/blobs/raw/master/crawl-ref/source/rltiles/UNUSED/monsters/norris_with_board.png 18:45:11 those refer to children's book with a roughly-current-day setting 18:45:20 ??norris[2] 18:45:21 norris[2/2]: < petato> I see Norris lost his board. poor guy :( <+sorear> anachro police got it < petato> timecops go home 18:45:21 haha great 18:48:53 -!- Textmode has joined ##crawl-dev 18:49:27 random: if we have octopodes, we should have antipodes 18:51:52 How about invisible pink unicorn instead? 18:51:56 Unicorns are over 300 years old. 18:52:09 but invisible pink ones are not 18:52:27 Invisibility is, pink is, it's just a simple composition of ideas. 18:52:58 composition of ideas is not over 300 years old either :P 18:53:09 (jk) 18:54:46 -!- ussdefiant has quit [Ping timeout: 245 seconds] 18:55:39 -!- ussdefiant has joined ##crawl-dev 19:04:23 Eronarn: octopodes as 'X' look really out of place, especially if krakens are nearby 19:04:43 perhaps, like most human-sized and smaller creatures, 'x'? 19:05:43 Props to whoever's idea it was to color code constituents of the visible monsters list by relative difficulty. That's a really nice feature for new players. 19:08:48 kilobyte: oops, i'm not sure why i made them X instead of x 19:19:02 -!- ussdefiant has quit [Ping timeout: 276 seconds] 19:19:15 -!- dtsund has quit [Quit: dtsund] 19:19:35 dammit, Ziggurats don't allow customizing monsters :( 19:20:21 so my plans for "octopode spells:bolt_of_fluffiness ; shit-covered stick" have been thwarted 19:20:52 haha 19:21:32 -!- ussdefiant has joined ##crawl-dev 19:26:08 03kilobyte * rfb7d0f73061c 10/crawl-ref/source/ (mon-gear.cc monster.cc tilemcache.cc): Give monster octopodes some gear. 19:26:08 03kilobyte * rac48a2b3b89c 10/crawl-ref/source/mon-pick.cc: Banish some octopodes. 19:26:09 03kilobyte * rb2ad95ca2cc0 10/crawl-ref/source/mon-data.h: We know octopodes are abominations, make them not as big. 19:28:32 -!- dtsund has joined ##crawl-dev 19:29:06 haha 19:33:36 should TSO worshippers be able to get cherubs as well as angels/daevas? 19:34:21 just to make the ability a bit more varied, if they wouldn't be too powerful 19:37:42 I wouldn't go too hasty here 19:38:29 the ability works well as-is, we can rethink it once more of the holyhaul is in 19:38:51 especially not half an hour after a dubious first stab gets in 19:39:50 as evilmike noticed, they're pretty much daevas, when they're supposed to be more angelic 19:40:10 daevas with no smite but bows/slings, that is 19:44:54 kilobyte: are you sure you can't customise monsters? all the definitions at line 328 and beyond look like normal MONS definitions ... 19:46:12 i think they're just X because everything with tentacles is on X? :P 19:46:17 mumra: they're later converted to bare monster enums with place: coded otherwise 19:47:21 Mu_: X and x (although abominations are described as mostly random piles of body parts) 19:47:38 -!- edlothiol has quit [Ping timeout: 276 seconds] 19:47:41 okay :P 19:48:12 i thought X was "things that are supposed to be scary looking", personally 19:50:09 03dolorous * r5f4262013a1c 10/crawl-ref/source/dat/descript/monsters.txt: Decapitalize "Chaos spawn". 19:51:07 mumra: the vault_mon_list code seems to be an unholy hack involving props and lists of ints 19:51:35 they save type, "base" and weight, or packed_place 19:52:07 marked by a magic value of "type" of -1 19:53:14 oh, "base" is used by draconians only, too 19:53:23 so you can't even choose a zombie type 19:53:41 huh, I guess it's not exactly easy to grab the MONS specs, though, is it? 19:53:57 kilobyte: the whole thing could probably be massively simplified by just using normal monster_list (like MONS does) 19:54:17 mumra: agree 19:54:23 SamB: it's trivial to build mons specs from strings, and the Zig monsters are already formatted appropriately 19:55:03 oh, in that case, why the heck doesn't it already do that? 19:55:36 I guess because it's old portal vault code 19:55:37 and yes, I know it's trivial 19:55:38 SamB: for the Demigods stuff, i'm building mons specs from the TextDB system used for speeh etc. (which means you can randomise everything with recursive substitutions) 19:55:47 I've seen the code for &M 19:57:18 you could also pull ziggurat monsters from a textdb file ... it'd add a lot of flexibility 19:57:38 I still don't get what sqlite buys textdb 19:57:46 nothing 19:58:00 makes it slow and quirky 19:58:17 -!- ussdefiant has quit [Ping timeout: 255 seconds] 19:58:46 SQL is great if you use it, but here it's just a key-value mapping 19:58:53 I noticed 19:58:53 -!- Wensley has joined ##crawl-dev 19:59:14 it *could* be used for better things, since it's in their already 19:59:40 e.g. adding a "language" field to the table, to start supporting translations 19:59:50 heh 20:00:24 "Fix disc of storms producing 49 times as many effects" <-- not certain this wasn't the correct behavior 20:00:32 or other things involving addition parsing / filtering of the text files to be more useful (for instance, i'd like to be able to pass parameters into them) 20:01:28 Wensley: hilarious deaths[62/62]: !lg turnerjer place=slime:1 1 -tv 20:01:34 -!- Twinge_ has joined ##crawl-dev 20:01:58 mumra: still a key:value mapping 20:02:56 kilobyte: yeah but you could have extra fields to filter on (which combined with appropriate indexing could be way faster than implementing the same processing in code) 20:03:20 kilobyte: and it'd be easy to scale up to full mysql on cdo for instance 20:03:46 which would be far faster than the current system (which involves a lot of string parsing whenever the db is accessed) 20:03:59 -!- dtsund has quit [Quit: dtsund] 20:04:29 -!- dtsund has joined ##crawl-dev 20:04:35 -!- Twinge has quit [Ping timeout: 255 seconds] 20:04:37 sure, if you have something that could use it 20:04:54 current db lookups do not and thus are a waste 20:04:55 -!- ussdefiant has joined ##crawl-dev 20:06:27 kilobyte: agreed, there most be a ton more efficient ways to provide a key/value lookup! 20:07:03 kilobyte: in C#, i'd just go: new Dictionary() 20:10:09 03kilobyte * r851cb3178368 10/crawl-ref/source/dat/dlua/ziggurat.lua: Ettins on giant Zig levels. 20:12:14 mumra: I don't think TextDb can get much more out of a relational DBMS than it already does, really 20:12:37 and I don't see any reason to implement languages with another column, either 20:12:50 oh, are people talking about the translation project? 20:12:59 just stick the .txt files in different directories per language 20:13:09 Wensley: someone started a project? 20:13:27 SamB: I thought someone was planning to, eventually 20:13:28 translations can't be done with a database anyway 20:14:04 kilobyte: the descriptions databases would need translating, though, certainly 20:14:04 there is a lot of string->string lookups, yeah, but a lot of custom grammar code as well 20:15:04 SamB: of course, but that's a trivial part (code wise) 20:15:15 a cheap half-baked translation might be doable by replacing the descriptions with text in some other language and just getting rid of the grammar stuff 20:15:30 leaving all the normal text in english 20:15:37 kilobyte: i was thinking a lot could be done with substitutions and lua, if you could pass parameters into text lookups 20:15:43 textdb already supports lua chunks 20:15:46 SamB: and what's the point of that? 20:16:14 might help people with partial english skills 20:16:18 mumra: not really, you can't do grammar in a context-less way 20:16:44 kilobyte: exactly, that's why you need to be able to pass in parameters (to provide context!) 20:17:00 but yeah, i know there are huge complications with grammar 20:17:22 but, anyway, english-only speakers are probably not the best people to start any actual effort 20:17:32 you'd need to provide the whole parse tree 20:18:14 at least crawl doesn't support even pseudo- natural language input ;-) 20:20:02 most European languages for example have grammatical gender. In Polish for example the word "cat" is masculine and "mace" feminine, no matter what's the actual cat gender. If you do know the cat in question, there's a separate word for "female cat". 20:20:48 and verbs do change according to the noun's gender 20:20:52 yeah... 20:21:50 it gets even more confusing in some languages, like the word for "girl" in German being neuter 20:21:57 even though girls are always female by definition 20:22:04 then, there's German or Latin which often place the verb at the end of the sentence so you might have "N1 blah blah N2 V2 V1" -- you can't even reference just the previous noun 20:22:25 IF and MUD systems tend to have a lot of stuff to help deal with such complications... 20:22:28 -!- ais523 has quit [Remote host closed the connection] 20:23:41 hmm, postscript tends to be more readable... 20:37:49 -!- ZChris13 has quit [Ping timeout: 240 seconds] 20:49:27 -!- valrus has joined ##crawl-dev 20:52:30 -!- Twinge_ has quit [] 20:56:22 -!- st_ has joined ##crawl-dev 21:07:27 -!- mikee_ has joined ##crawl-dev 21:07:33 join #rrogue 21:09:15 -!- Pingas has quit [Ping timeout: 256 seconds] 21:21:27 -!- Pingas has joined ##crawl-dev 21:31:43 UseBees the Cutter (L2 NaGl) (D:2) 21:40:07 -!- ixtli has joined ##crawl-dev 21:41:44 hrmm 21:42:11 simply putting zotdef abilities on capital F and up does not have good finger-memory compat :-( 21:47:25 -!- ainsophyao has joined ##crawl-dev 21:47:29 did you F something up 21:50:11 *groan* 21:50:30 -!- ainsophyao has quit [Remote host closed the connection] 21:56:24 kilobyte: so, you think we should either just avoid duplicate altars in zotdef or let the user pick (for a higher price)? 21:58:58 SamB: The current system is pretty ugly, having to go to some corner to spit out a random number of altars 21:59:11 Picking at a higher price would be great 21:59:33 RichardHawk: I know, I reported that bug 22:01:32 but I wasn't sure what the best solution would be 22:01:36 <|amethyst> would it make sense to avoid duplicate altars but increase the cost every N uses? 22:01:55 <|amethyst> to kind of simulate to current cost without using the floor space 22:02:09 <|amethyst> s/ to / the / 22:02:18 I was thinking maybe do it like the magic beans in Ocarina of Time 22:02:32 <|amethyst> and that is? 22:02:47 wait seven years for the power to kick in 22:02:50 increase the cost by the initial cost each time 22:03:15 so like 2,4,6,8,.. 22:03:45 <|amethyst> unless you let the user choose the god that's a nerf 22:12:18 <|amethyst> currently the expected cost of getting the altar you want, assuming the probabilties are equal (not sure if this is true or not), is 36 22:12:20 -!- jeanjacques has quit [Ping timeout: 260 seconds] 22:12:39 <|amethyst> if you eliminate duplicates but increase the cost by 2 each time, the expected cost is 126 22:12:47 hmm 22:13:59 well, if you want to work out the formula for the expected cost of getting 1st, 2nd, etc. distinct altar, be my guest ;-) 22:14:50 or I guess we could give the user choice of altar for 36 ZP 22:17:38 <|amethyst> if you make it 2, 2, 3, 3, 4, 4, ... that would have an expected cost of 41.7 22:18:56 anyway, the issue is https://crawl.develz.org/mantis/view.php?id=4582 22:19:17 as I see it the alternatives are to let the player pick the altar or to make the cost actually significant 22:19:39 elliptic: er, wait 22:19:59 shouldn't the cost *be* significant if the player gets to pick? 22:21:18 <|amethyst> letting the player pick does reduce a perhaps-useful tension of "do I go with this second-best god, or do I keep spending ZP in hopes of the one I really want?" 22:21:51 <|amethyst> s/useful/interesting/ 22:21:56 samb: well, it could be... but I also don't see anything particularly wrong with giving people free god selection once they reach XL whatever 22:22:15 the bad thing is if they have essentially free selection but it is annoying to use 22:22:18 as it is currently 22:22:26 yeah 22:23:18 2,4,6,8,... with no duplicates doesn't sound terrible to me 22:23:25 could even be harsher about the increase 22:23:49 -!- Torokasi has quit [Ping timeout: 258 seconds] 22:23:55 -!- Mu_ has quit [Quit: Defecator, may everything turn out okay so that you can leave this place.] 22:24:02 you don't want to encourage startscumming for a cheap altar though 22:24:32 just let players pick for 100 22:24:53 hmm, that's pretty damn steep 22:25:20 <|amethyst> 2,4,6,8 costs 126 to get what you want if you won't settle for second-best 22:25:26 <|amethyst> so it's cheaper if your picky 22:25:32 <|amethyst> s/your/you're/ 22:26:04 true 22:26:24 <|amethyst> hm 22:26:24 <|amethyst> wait 22:26:29 elliptic: how about "cinder dragon" 22:26:30 <|amethyst> recalculating 22:27:11 <|amethyst> aha, 108 if the cost is 2,4,6,8... 22:27:11 wensley: hm, I like that maybe 22:27:22 <|amethyst> so pretty close actually 22:28:09 -!- evilmike has quit [Quit: something happened] 22:28:56 -!- valrus has quit [Remote host closed the connection] 22:33:23 * SamB sees a dragon covered in soot with a wisp of smoke trailing from between its jaws, with a fairly sad expression on its face 22:33:23 cinder dragon? to replace normals? 22:33:50 * SamB notes that the art is rather stylistic 22:35:27 Eronarn: yeah 22:35:45 sounds fine to me 22:36:13 -!- ZorbaGama has quit [Read error: Connection reset by peer] 22:36:27 -!- ZorbaGama has joined ##crawl-dev 22:41:27 -!- Torokasi has joined ##crawl-dev 22:45:42 -!- Pingas has quit [Read error: Connection reset by peer] 22:50:11 03elliptic * rec7076438f62 10/crawl-ref/source/chardump.cc: Tweak formatting of the spell usage list. 22:50:21 03elliptic * rf0278bbff5fe 10/crawl-ref/source/chardump.cc: Change sorting for spell usage list. 22:57:06 -!- Pacra has quit [Quit: This computer has gone to sleep] 23:11:16 -!- ussdefiant has quit [Read error: Connection reset by peer] 23:25:25 -!- mikee_ has quit [Ping timeout: 240 seconds] 23:35:29 -!- Textmode has quit [Ping timeout: 258 seconds] 23:40:48 -!- Keskitalo has joined ##crawl-dev 23:45:29 -!- Wensley has quit [Quit: ChatZilla 0.9.87-rdmsoft [XULRunner 1.9.0.17/2009122204]]