00:13:00 -!- bfein_ has quit [Ping timeout: 252 seconds] 00:25:11 -!- jedahan has left ##crawl-dev 00:28:58 -!- bfein_ has joined ##crawl-dev 00:41:59 -!- dpeg has joined ##crawl-dev 00:42:07 mornings 00:48:59 !seen Napkin 00:48:59 I last saw Napkin at Mon May 24 12:35:40 2010 UTC (1d 17h 13m 19s ago) saying d6 cyclops - spend too much time cleaning a 8x5 kobold vault? :-P on ##crawl-dev. 00:49:17 Napkin: Können wir einen update für CDO kriegen? 01:18:01 -!- Zaba has quit [Ping timeout: 245 seconds] 01:27:09 -!- Zaba has joined ##crawl-dev 02:13:21 Moin 02:13:24 sicher das, dpeg 02:20:09 Yay! 02:20:17 Master branch on CDO updated to: 0.7.0-a0-1894-g0885fff (22.2) 02:20:34 finally the lost turn is gone \o/ 02:20:43 (when running & exploring) 02:21:48 Napkin: hot stuff 02:21:50 I wonder if that fixed the lost turn when passwalling 02:22:02 I've been playing my first MD in years and it's smooth sailing so far. 02:22:15 Okay, that's exaggerated, but shield+mail works well for them. 02:22:28 st_: could you research and file a bug report if not? 02:27:43 When you're on fire, should you do more damage do ice beasts? 02:30:41 Assert failure on load (https://crawl.develz.org/mantis/view.php?id=1619) by Kyrris 02:41:11 guys, I think there's a show-stopping error on that version I just installed 02:41:20 -!- Baraton has joined ##crawl-dev 02:41:20 seems like players can't load their savegames 02:41:37 greetings, kind developers and admins ) 02:41:45 seems so, yes, yes 02:41:54 ASSERT(cooky == QUIVER_COOKIE) in 'quiver.cc' at line 463 failed. 02:42:04 yes 02:42:04 -!- by has joined ##crawl-dev 02:42:06 (in crash file) 02:42:08 -!- jooosa has joined ##crawl-dev 02:42:09 somebody broke saves 02:42:17 Moin rob :) 02:42:23 without following procedure 02:42:31 http://crawl.develz.org/morgues/trunk/Baraton/crash-Baraton-20100526-072820.txt , for example (I generated tons of them already :) 02:42:33 why didn't this get fixed when it was discovered 02:42:45 by: It was discovered today 02:43:03 but it was discovered before cdo was updated 02:43:06 updated trunk 22 minutes ago 02:44:05 -!- joosa has quit [Ping timeout: 240 seconds] 02:44:05 -!- jooosa is now known as joosa 02:44:28 3f6687c 02:44:46 no time now, but you could just revert that commit 02:45:09 Baraton - want me to move your game back to the previous version? 02:45:23 yes, it would be nice 02:46:51 done, Baraton 02:48:17 * Baraton writhes, screaming 'The PPppppPOOOOoooWWWwwwErrRRR!' 02:48:25 thanks, Napkin 02:48:29 np :) 02:50:38 Hm, have there been recent changes to the quiver at all? 02:59:45 ok, prevented transfers of saves by raising version to 22.3 in versions.db 03:00:13 *further transfers 03:10:42 is someone fixing the save break? 03:10:46 or should I? 03:11:18 BTW, we need to either restore compatibility or increase major 03:11:46 this last statement addressed to me? 03:11:57 no fiddling with the minor will do any good; we can make CDO work but it will still screw non-server players 03:12:24 have any games started since the incompatible change? 03:12:32 yes, 1 03:13:19 hrm... :( 03:13:46 I can make it be compatible with pre-3f6687c4 or post-3f6687c4 but not both 03:16:12 someone broke something 03:16:20 yay 03:16:30 well, how about just raising savegame minor? 03:16:48 moin due ;) 03:17:23 what broke it? 03:17:50 ASSERT(cooky == QUIVER_COOKIE) in 'quiver.cc' at line 463 failed. 03:18:10 I meant which commit :) 03:18:24 by> 3f6687c 03:18:31 Napkin: raising minor is probably the way to go, but it will break that game in progress 03:18:56 oh... 03:19:19 raising major is the alternative -- you don't upgrade to any version but the most recent, right? 03:19:26 oh.. I forgot that raising minor is not helping 03:20:04 not preventing saves :-O 03:20:53 i should better raise major to 23 then? 03:21:15 what about raising major to 24 in master, KiloByte? 03:21:24 haxy - but should work? 03:21:37 and yes, I only upgrade to the most recent version usually 03:23:24 03kilobyte * rd9454ce9c160 10/crawl-ref/source/ (tag-version.h tags.cc): Restore save compat for 22 majors. BREAKS SAVES FROM 3f6687c4. 03:23:57 I've put in a fix for all older games for now. We may increase major instead, though. 03:24:39 22d564b|1274267096|22|2 03:24:39 0885fff|1274858404|23|0 03:24:41 who's the affected player? 03:25:03 ok, raised major 03:25:39 uhm... just on CDO? 03:26:16 just in the DB, which decides if a transfer of saves is possible 03:26:33 ah 03:26:50 so we need to go right to 24 then 03:27:00 exactly what I was about to suggest :) 03:27:25 by, due: do you have any changes you could use the save break for? 03:38:10 No, I haven't had time to do coding recently. 03:38:36 who broke save compat? Someone coding demonspawn changes? 03:40:32 due: tone! :) 03:41:32 Tone of? 03:42:18 due: no need to poke at Cryptic. 03:42:43 I guess I'm in a grumpyt mood today, sorry. 03:42:53 no sweat 03:42:58 there's a number of changes that can break with the increase of major, need to test them 03:43:36 by: you did check that #ifdefed code to handle Th/DK with tags != 22, right? 03:44:20 at least demonic guardians break, but it's a simple fix 03:45:32 uhm... DUR_DEMONIC_GUARDIAN is used in a number of places 03:45:49 including making sure only one is spawned at a given time 03:46:53 how is the guardian supposed to work now? Not timed at all? 03:47:12 -!- Spads has joined ##crawl-dev 03:51:19 KiloByte: yes 03:51:38 KiloByte: the idea is that the guardian is no longer an ally (who comes back after some time, if killed) 03:51:53 rather, the guardian will appear (with some chance) at high tension and disappear at low tension 03:52:22 Which is much less ally-like and much more guardian-like, and hence preferable gameplay and flavour wise, in my opinion. 03:52:34 * elly read that as "much less elly-like" 03:53:15 uhm no, it is timed now 03:53:54 g - a cursed -1 leather armour 41 gold 03:54:09 Need an Elly unique. 03:54:47 KiloByte: could it be timed _and_ have a chance to disappear at low (or zero) tension? 03:55:03 Could also be that Cryptic only implemented spawning at high tension so far. This is new stuff. 03:56:30 the #ifdef logic was reversed in a couple of places 03:57:14 now ghosts. by: did you happen to test DK/Th ghosts? Do we even allow ghosts from different save majors? 03:58:47 we don't, ok 03:59:24 no, we shouldn't 03:59:36 By the way, can't reavers be removed with the same logic that removed death knights? 04:01:01 they make some sense, unlike necromancy which makes you weakened so you suck at melee 04:01:13 but that's only "some" sense, not "much" sense 04:02:01 03kilobyte * rd1767427b7af 10/crawl-ref/source/ (8 files): Increase major save version to 24. 04:02:38 ogaz: probably, but I'd rather like a total class reform, rather than many ad hoc removals 04:02:41 KiloByte: thank you 04:02:53 on the other hand, the conj books are much more skill-intensive than necromancy, and both the necromancy level 3 spells are good for hybridization 04:03:38 dpeg: makes sense, but I do think it makes sense to either remove or heavily modify reaver even before a total reform 04:04:16 that component proposal would fix all hybrids 04:04:50 KiloByte: yes, but I am not sure the proposal is good enough at the moment 04:07:17 yeah, definitely. That's why I think we should think more instead of doing random removals. 04:08:17 another issue: colorless scales. They are dubious for DS, and outright wrong for everyone else. Since it's quite rare to have two identical scale mutations, you end up with colorless scales three times. 04:08:30 KiloByte: yes, I did test the Th/DK stuff with bumped major, should work; thanks for handling this! 04:09:11 that looks silly, has no flavour, and is a nearly meaningless mutation 04:09:19 KiloByte: ah, we never discussed this. 04:09:34 I agree that colourless is not so hot for non-DS. For DS, it is interesting, though (I think). 04:09:36 by: I have it just a minor check 04:10:09 DS are guaranteed 3 levels, so the only effect is not letting them know what they got 04:10:15 KiloByte: exactly 04:10:54 We are already doing this with the Fire and Ice facets. 04:11:10 I think it's more fun to hand out the information in smaller doses. 04:17:51 -!- ogaz has quit [Remote host closed the connection] 04:25:55 small dwarves: the more I think, the more I see that having small dwarf-sized axes would be just wrong. They're supposed to use oversized ones. I'd just let them use short-hafted weapons (axes, maces) as if one size bigger. 04:26:19 two-handed swords being probably on the non-dwarf side of the aisle, since they're nearly 2m long 04:26:29 KiloByte: yes, that'd be okay, too 04:26:43 it would distinguish dwarves from the other small species 04:27:00 haha, one of the humans in the fake V:8 fault is a blowfly 04:27:49 for body armour, I'll go with that plan we discussed (dwarven armour being one size smaller, a guaranteed vault with a dwarf in plate for 0.7) 04:27:58 fake V:8 vault? 04:28:09 look at dpeque@CDO 04:28:23 That's one of my vaults, I still like it. =) 04:28:23 heh... and another is a dwarf 04:28:28 yes 04:29:13 just be ware, he's one of harder uniques now 04:29:21 I never know which of these uniques are the tougher ones these days. 04:29:29 Cannot blow all my resources on everyone of them. 04:29:31 KiloByte: thanks 04:29:36 Wayne or Norris? 04:30:24 Norris always used to be one of bigger ones, but Wayne is stronger damage-wise now 04:30:35 expect taking 60hp hits 04:30:40 oops 04:30:50 might call for berserk 04:31:07 I have AC 24, EV 16, SH 39, not sure if that helps 04:31:24 AC 24... ouchie 04:31:54 +2 plate, Armour 14 04:31:59 it's a slow process 04:32:33 and so I am dead 04:33:02 Wayne was almost dead, though 04:33:45 aren't you too deep for a xl14 char? 04:33:58 felirx: well, was running out of options... 04:34:01 !log dpeque 04:34:02 208. dpeque, XL14 MDFi, T:45424: http://crawl.develz.org/morgues/trunk/dpeque/morgue-dpeque-20100526-093223.txt 04:34:13 should have done Hive, but apart from that... 04:34:53 you've rpois and haven't entered swamp, is that because of heavy armor? 04:35:51 Hive doesn't give almost any xp, and has little loot other than food, so it wouldn't have changed anything. 04:36:54 !lm * cv=0.7 uniq=Wayne 04:36:55 No milestones for * (cv=0.7 uniq=Wayne). 04:37:06 ? 04:37:17 !lg * cv=0.7 killer=Wayne 04:37:17 No games for * (cv=0.7 killer=Wayne). 04:37:27 I killed Wayne 04:37:34 KiloByte: yes, I would have got to XL 15, that's all. I saved it because you can pump your Ely piety easily there. 04:37:36 !lm . char=hufi s=uniq 04:37:38 me too, and got killed by him as well 04:37:42 40 milestones for by (char=hufi): 1x the royal jelly, 1x Norris, 1x Jozef, 1x Mnoleg, 1x Norbert, 1x Lom Lobon, 1x Cerebov, 1x Louise, 1x Frances, 1x Aizul, 1x Frederick, 1x Kirke, 1x Maud, 1x Mara, 1x Dowan, 1x Psyche, 1x Blork the orc, 1x Xtahua, 1x Erica, 1x Donald, 1x Nergalle, 1x Duvessa, 1x Dispater, 1x Gloorx Vloq, 1x Saint Roka, 1x Duane, 1x Harold, 1x Margery, 1x Rupert, 1x Ilsuiw, 1x Uru... 04:38:11 !lm . char=hufi uniq=wayne 04:38:11 1. [2010-05-21] robx the Severer (L19 HuFi) killed Wayne on turn 101246. (Elf:3) 04:38:21 !lm * cv=0.7-a uniq=wayne 04:38:22 I probably killed wayne too, too much similar names to keep track 04:38:25 !lg * cv=0.7-a killer=Wayne 04:38:35 137. [2010-05-26] noom the Grappler (L19 TrTm) killed Wayne on turn 64858. (Snake:5) 04:38:35 10. dpeque the Severer (L14 MDFi), worshipper of Elyvilon, annihilated by Wayne (a +2,+3 dwarven broad axe of chopping) on D:17 on 2010-05-26, with 63355 points after 45424 turns and 3:02:38. 04:38:54 However, at least for MD (my first MD for some time, IIRC) heavy armour + shield is an option. 04:39:32 I switched to hand axes for bees, at least at some point. 04:39:48 niice... Wayne has 10 kills in a couple of months among trunk players 04:40:00 I was stupid... could have used Ely's Vigour before berserk. 04:40:06 * dpeg is really stupid :( 04:40:20 slinkies had to not only stand next to him doing nothing but to actually strip down to let him kill him 04:40:30 KiloByte: in 0.5? 04:40:46 !lg mayhem killer=Wayne x=cv 04:40:46 1. [cv=0.4] Mayhem the Severer (L18 MDFi), worshipper of Okawaru, slain by Wayne (a +2,+1 scimitar of slicing) on D:22 on 2009-01-08, with 181844 points after 52742 turns and 8:56:51. 04:41:18 !lg * killer=Wayne s=name 04:41:18 13 games for * (killer=Wayne): 1x dpeque, 1x Eli, 1x heteroy, 1x Disco, 1x DashNine, 1x foolcircle, 1x Mayhem, 1x Nexos, 1x Zicher, 1x Lemuel, 1x Baraton, 1x noom, 1x syllogism 04:42:02 (ah, a stone giant simulacrum got the deathblow on me) 04:42:41 anyway, I don't think that melee is crippled to death 04:42:51 ...can never be sure with those screaming complaints 04:43:26 24 AC at the Vaults entrance... you used to have 50+ 04:43:52 KiloByte: we can (and should) change matters... the pendulum went a bit too far in the other direction. 04:44:08 but those AC were (and still are) ridiculous. 04:44:15 eww, I could do some 6+ exams during the summer if I wanted to 04:44:22 felirx: do it! 04:44:27 why 04:44:28 "wanted to"? 04:44:38 I'd improve AC _but_ increase penalties on EV 04:45:03 30+ EV in plate and large shield by early swamp is outright broken 04:45:17 parallel programming, calculus, analysis, requirements analysis, the principles of programming languages and so on 04:45:30 KiloByte: my large shield came with EV+7 and I had another +4 on my amulet. 04:45:35 yes; I've wondered if making the evasion penalty work as a multiplier might work 04:45:38 if you increase the -EV penalty, remember that it affects spellcasting too 04:45:47 i.e., plate -> base ev divided by 6 04:46:05 ah, KiloByte was not referring to the numbers of my guy 04:46:45 dpeg: my current DD had that. I got some EV from randarts I've since replaced, though. 04:47:04 also I have a feeling EV is doing too much for damage reduction 04:47:05 KiloByte: you shouldn't count those randarts, should you? 04:47:13 by: this I agree with. EV is the new AC. 04:47:32 felirx: no one says EV has to be hit by the same amount as spellcasting. In the old code, that wasn't the case. 04:47:33 Darshan always said that EV was already overpowered before we boosted it. 04:47:55 We boosted it? 04:48:03 we didn't, we simply nerfed it only the tiniest bit while nerfing AC into the ground 04:48:18 the Dex thing surely helped EV? 04:48:20 so it's a boost by faint nerfage 04:49:23 dex increases EV only for those who can afford getting it high. That is, Troggies and Cheibriadities. 04:49:47 anyone can afford getting dex high? 04:50:07 So we are now (where many more players rely on EV than ever) realising what Darshan said? 04:50:14 neglecting int -> crippled spellcasting 04:50:36 you can cast just fine with mediocre intelligence 04:50:44 it's a choice to make 04:51:18 you don't get much EV for pumping dex high 04:51:18 also, dex increases EV for those characters that naturally get EV higher than the 0.5 cut-off 04:51:57 so you choose that only if you either don't have better uses for stats or get that +15 for free 04:52:35 if people choose like that, it just means that EV is too good without high dexterity 04:52:49 The +15 of Che is a crutch. I just added that to make sure the god is viable, so players try to play with the conduct. We're free to come up with other, better bonuses for wearing ponderous items. 04:54:55 by: not really, it means that the gains from runaway dex are contained, which is not bad. We reward moderate dex but diminish rewards after that. 04:55:22 and those diminished rewards mean you keep stats balanced 04:55:22 I meant, if they never choose dexterity in favour of intelligence 04:57:04 what I mean is, the gains from dex above 16-18 are small. Better than str for a non-caster, but not good enough to warrant excluding everything else. 04:57:20 by: ideally, the weighting of Dex vs Int should depend on the amount of fighting (stabbing) vs casting. 05:03:06 If you only use Dex for defense (like a pure caster would do), then it's clear that Int will be more interesting. I don't think we can or should change this. 05:08:58 god forbid if I'd have to raise Dex too as a caster. That would be a total nightmare :) 05:11:33 felirx: because you already have to rise Str? :P 05:11:39 -!- ais523 has joined ##crawl-dev 05:12:20 yea 05:13:12 felirx: you could play more well rounded species than just Sp and DE! 05:13:30 I don't play DE at all anymore! 05:13:37 felirx: why not? 05:13:50 won the caster and DS is superior to De 05:14:02 well, DS is overpowered but we covered that yesterday :) 05:15:08 at least crawl doesn't have oblivion syndrome 05:15:32 I made a char there with 85% armor (capped) and total magic immunity, runspeed so high I could run the game map from one side to other in ~3 minutes 05:15:37 kinda lost interest after that :D 05:17:56 felirx: the invincibility syndrome? Yes, we explicitly claim to fight that. Although I am afraid we're not nearly there. 05:18:46 I did lose my interest on the DSWz after I cleared the pandazig 05:19:33 Hm, perhaps the ziggurats weren't such a good idea, after all. 05:20:22 felirx: was the ziggurat trivial to conquer? 05:20:32 pandemonium level at zig:27 05:20:45 firestorm did maybe 20-50 at most 05:21:09 felirx: were you ever in mortal danger? 05:21:11 I could run away with haste to wand of heal / evoke mana back to full 05:21:24 in true peril I doubt, I was at low health at points I think 05:21:37 can you get a replay of cdo games? 05:21:40 Well, this might be more a problem with the storms and mana regain. 05:21:43 felirx: yes 05:22:03 now the only question is how :D 05:22:59 felirx: what do you want replayed? 05:23:14 DSWz, second zig:27 05:23:53 !lm felirx char=dswiz zig=27 05:23:53 Unknown selector: zig 05:24:03 !lm felirx char=dswiz type=zig place=zig:27 05:24:03 No milestones for felirx (char=dswiz type=zig place=zig:27). 05:24:09 -i 05:24:10 !lm felirx char=dswiz place=zig:27 05:24:10 No milestones for felirx (char=dswiz place=zig:27). 05:24:15 !lm felirx char=dswz place=zig:27 05:24:16 4. [2010-05-17] felirx the Farming Genius of the Arcane (L27 DSWz) left a Ziggurat at level 27 on turn 417340. (Zig:27) 05:24:23 -!- Spads has left ##crawl-dev 05:24:24 !lm felirx char=dswz type=zig place=zig:27 05:24:24 2. [2010-05-17] felirx the Farming Genius of the Arcane (L27 DSWz) reached level 27 of a Ziggurat on turn 416199. (Zig:27) 05:24:41 !lm felirx char=dswz type=zig place=zig:27 -tv 05:24:41 2. felirx, XL27 DSWz, T:416199 (milestone) requested for FooTV. 05:25:02 or maybe -tv:>$ if it ends too early 05:27:06 that was reach 05:27:21 reach? 05:27:35 enter 27, and the tv broadcast ended 05:29:07 !lm felirx char=dswz type=zig place=zig:27 -tv:>$ 05:29:07 2. felirx, XL27 DSWz, T:416199 (milestone) requested for FooTV. 05:29:26 that should make it broadcast until cancelled? 05:32:58 yes 05:40:46 !footv -cancel 05:40:55 wrong :9 05:44:37 now how to cancel it mmm 05:45:20 !lm felirx char=dswz type=zig place=zig:27 -tv:cancel 05:45:20 2. felirx, XL27 DSWz, T:416199 (milestone) cancel requested for FooTV. 05:45:50 cheers 05:45:51 http://sprunge.us/DIjQ -- defense stats vs. orb guardian with different armours and different levels of dodging and armour skill 05:46:10 Seems like I dipped at 66hp lowest 05:48:27 plate mail GDR seems to lower max damage from 45 to 36 reliably, does this sound correct? 05:48:41 wow, that's a lot of scrollback 05:49:11 * Zaba goes to read git log instead 05:49:32 -!- ais523 has quit [Read error: Connection reset by peer] 05:50:20 effective damage starts at around 20 for a nude character; mastering dodging brings it down to something like 10.5 05:51:09 a dodging character in full kit with no armour skill gets around 7 05:51:13 -!- ais523 has joined ##crawl-dev 05:51:45 a plate mail character doesn't get below 8 05:52:00 (unless they also learn dodging a lot) 05:52:12 in summary, completely broken? 05:52:57 -!- Luca__ has quit [Remote host closed the connection] 05:53:18 -!- Luca__ has joined ##crawl-dev 06:00:41 -!- syllogism has joined ##crawl-dev 06:02:06 <+by> http://sprunge.us/DIjQ -- defense stats vs. orb guardian with different armours and different levels of dodging and armour skill 06:02:16 syllogism: might be interesting? 06:02:58 it looks to me like heavy armour is worse than dodging in terms of average damage reduction 06:03:28 (comparing a player in plate mail with high armour, low dodging and a player in a robe with high dodging, low armour) 06:03:32 -!- Mu_ has joined ##crawl-dev 06:04:25 do we have 0.5 stats for a comparison? 06:04:37 I don't 06:05:57 http://sprunge.us/SAQE 06:06:19 that's a diff against wiz-fsim.cc to raise both dodging and armour skill simultaneously 06:07:23 it might be better to analyze the function (ac,ev,gdr)->damage and (race,skills,eq)->(ac,ev,gdr) separately 06:07:44 stats also go into the latter 06:08:07 looking now 06:10:35 I guess we could consider it ok that a heavy armour character should invest both into armour and dodging skill 06:11:09 26% accuracy seems lower than empirical tests indicate 06:11:34 the accuracy stat may consider 0 damage hits as misses 06:11:36 those numbers claim it only has 61% accuracy against a 2ev character, that also seems off 06:12:14 by: just back, trying to understand the numbers 06:12:47 syllogism: it only counts a hit that does damage as a hit 06:13:01 that may explain it then 06:13:04 dpeg: fsim could do this after all 06:13:19 by: oh. I should have called for help before 0.6, eh? 06:13:44 don't know, who to call? 06:13:55 you :) 06:14:12 syllogism provided the excel tool, though 06:14:16 and we looked into this 06:14:22 I did not want to be involved in the AC nerf 06:14:37 I can understand that. 06:15:02 But a nerf was needed nonetheless. Something we have definitely gained by now is GDR which is noticeable. 06:15:05 syllogism: right? 06:15:54 not to try to complain or stir an upset or anything, but is heavy armour useful now in trunk? 06:16:05 after the nerf it felt like it wasn't worth the hassle :) 06:16:14 and I haven't played in quite a while 06:16:15 cbus: it is useful, yes 06:16:19 oki 06:16:23 I'll try it out later then :) 06:16:25 I haven't played any heavy armor characters since 0.5 06:16:30 syllogism, same here 06:16:33 My MDFi died to Wayne, and I wouldn't blame the plate mail for that. 06:16:50 the strength required for each armour is a pretty big turnoff too 06:16:51 wayne is really tough 06:16:55 ?? wayne 06:16:55 wayne[1/1]: A fat, evil dwarf in a stupid looking hat. Same spells as an orc priest: smite, pain, cantrip, heal. Your goal is to be the first player on CAO to die to him. (Mayhem wins!!!!!!!!) Reflavored as a priest of okawaru in 0.7 and capable of dishing out some major damage. 06:17:13 well, the AC you had was ok... for a weak EV char 06:18:31 cbus: it is playable, but seems to be clearly the inferior option to going with dodging 06:18:46 KiloByte: assuming that most pure melee fighters would have used Okawaru or Makhleb, I don't really see much of a problem. I could have won, too, if I had used Elyvilon more intelligently (Vigour or Major Healing). 06:19:16 I do find these numbers fairly surprising 06:19:45 it may well be that the fsim code is completely broken 06:19:47 you were a MD. The race that's supposed to be insane in armour. No god should be needed for that. 06:20:20 but I suppose since the ac difference is only 25 at most, it makes sense 06:20:24 and you fought Wayne in a corridor 06:20:28 KiloByte: no god needed until Wayne. 06:21:09 okay, I pacified hydras, but that was because of being an axe dwarf. 06:21:44 fsim doesn't do blocking, does it 06:21:52 at least it doesnt print stats 06:21:54 damage aside 06:22:12 I removed the shield 06:22:44 but no, I don't think it would give explicitly mention it 06:22:56 yes, I'm just interested in seeing the difference it would make 06:23:14 something better than the current fsim would definitely be nice to have 06:23:45 I remember someone from something awful was interested in improving it... 06:23:50 !seen gunofdis 06:23:51 I last saw gunofdis at Wed Jan 27 16:16:45 2010 UTC (16w 6d 19h 7m 6s ago) quitting with message Read error: 110 (Connection timed out). 06:24:03 that one maybe? 06:24:04 haven't seen him on SA either 06:24:08 in a while 06:26:42 did 4.0 change the basics of combat (comparing tohit vs evasion to determine hits, reducing random2(maxdam) by random2(ac))? 06:28:05 greensnark may know, doubt anyone else around here does 06:28:27 sorear would also know, I think 06:31:03 if we were sure those are good, a next step would be to figure out sensible values of (ac,ev,sh) 06:31:54 then adapt the way (race,skills,stats,equipment) are mapped to (ac,ev,sh) 06:32:06 -!- st_ has quit [] 06:33:45 by: as always, there are so many good approaches. 06:33:56 I don't think the basic setup is flawed. 06:34:26 what approach were you thinking of? 06:34:30 btw, my MDFi (XL 3) could take on a gnoll band on D:2, without any tools. 06:34:48 by: tohit vs EV, and d(maxdam)-d(AC) 06:35:13 together with the new approach to GDR 06:35:59 did the way GDR percentage went into the calculations change? 06:36:14 (that value belongs with ac,ev,sh above) 06:36:14 -!- st_ has joined ##crawl-dev 06:37:32 dpeg, scalemail? 06:37:44 This I don't know. Sorry. I meant the (your?) change that GDR does not need Armour skill anymore. 06:37:51 cbus: yes 06:37:55 I'm not convinced the GDR makes a big difference 06:37:56 6/6 AC/EV 06:38:02 probably an extra point of EV does much more 06:38:06 by: 20% GDR in plate? 06:38:31 by: that has been my position 06:38:52 btw 06:38:57 did the no death by low stats happen? 06:38:57 I just don't know 06:40:02 what about having a MDFi start with 6 AC, 0 EV, 50% GDR 06:40:17 50? really+ 06:40:36 it's 50% of max damage 06:41:28 and in <0.6 gdr was never higher than your ac 06:41:33 at least that's how I understood it 06:41:58 is it not even AC/2? 06:43:36 gdr is min(maxdam * gdr_perc/100, ac/2) 06:44:05 isn't that even worse 06:44:11 syllogism: than what? 06:45:01 but yes, I misremembered it 06:46:46 I remember runnning some 40ac tests with 0 gdr and 50% gdr 06:47:04 and the difference rather unspectacular 06:47:16 being 06:47:38 I like the idea of GDR, and also that it scales with AC. Of course, the factor (0.5 above) could be changed. 06:48:47 I definitely dislike the fact you can get fairly high AC in a robe and barely being any worse than a 27 armour skill character in a crystal plate 06:49:02 yes 06:49:04 I was at 74 ac in a robe... 06:49:41 but how to fix this? 06:49:57 limit light armour maximal enchantment? 06:50:10 capping it at +2 would help somewhat 06:51:00 reduce auxiliary armour base AC to 0 and add it to the heavy armours? 06:51:01 that would somewhat obsolete armor enchant scrolls for light types though 06:51:01 it doesn't help gdr is only applied in melee 06:51:33 good point, gdr could affect any damage 06:51:43 or at least any physical damage 06:52:07 it's the main reason why I underplayed the importance of GDR 06:52:14 any physical damage sounds reasonable 06:52:21 SA still keeps going on and on about how that's why armor sucks 06:52:25 because "gdr was nerfed" 06:52:49 we could rule that body armour enchantment is capped at base AC 06:53:39 that might actuallly be kind of interesting 06:53:48 also makes some sense, I guess 06:54:05 you could at most double the ac gain from a given armor piece 06:54:05 but I don't think that armour enchantment is responsible for the runaway AC in robes 06:54:08 instead of going from 2->10 06:54:44 felirx had +15 ac from jewellery plus +13 from ozocubu 06:54:52 46 ac is still high though 06:55:10 +5 from weapon too 06:55:17 but yeah 06:55:23 not in the morgue 06:55:33 ozocubu was somewhat nerfed at least 06:55:34 but that has ac 68, right 06:55:41 since it now melts 06:56:11 otoh note that without ozocubu's ice magic is probably totally inferior to fire magic 06:56:35 refrigde is totally cool, but in essence yes 06:57:03 although perhaps nerfing both ozocubu and fire storm could be the right solution 06:57:19 ice storm only marginally worse than fire storm 06:57:21 is 06:57:34 fire storm blocks LOS and is smite target 06:57:40 other than those factors, they're equal 06:57:47 ice storm is fireball target which is almost as good 06:57:49 ice storm delas a bit more damage to compensate iirc 06:57:58 felirx: and lets your enemies something to do 06:58:23 The storms being underpowered is a different issue than AC. 06:58:30 they aren't underpowered :P 06:58:34 underpowered?! 06:58:48 I'd say they are rather overpowered 06:58:50 it's a competition of two overpowered spells 06:58:55 most players I know seem to think they're overpowered 06:59:11 and if something's actually bad but players think it's good, why not let them use it? 06:59:13 ais523, and the other don't live long to get them ;) 06:59:14 in ziggurats fire storm is significantly better because enemies are slowed by fighting your vortices and you can shoot while cornered 06:59:34 and you can butterflie yourself and firestorm 06:59:35 luca: i had that backfire on me once as some ancient lich keps spamming Shadow creatures outside my LOS 06:59:37 syllogism: erm, of course 06:59:53 typo 06:59:54 flooding the level with more ancient liches and 1s 06:59:56 outside zigs, yes, ice storm is probably good enough 07:00:01 dpeg: that's a great typo 07:01:19 -!- Textmode has quit [Quit: Over the ages, man has been certain of a great many things. some of these things were even true.] 07:15:09 03dolorous * r3faeec86e23e 10/crawl-ref/source/dat/descript/spells.txt: Fix punctuation and wording. 07:15:10 03dolorous * rab4db4c9ea2a 10/crawl-ref/source/spells2.cc: Comment fixes. 07:15:20 03dolorous * re55b135c3b5b 10/crawl-ref/source/ (dat/descript/spells.txt makeitem.cc spells2.cc): Remove the restriction on poisoning blunt weapons. 07:16:17 nice :) 07:16:26 but is poison weapon worth using if you already have sting? 07:18:52 cbus: not everyone has sting. 07:19:07 dpeg, nah but was thinking on VM 07:21:06 the very little tests I did as SpAM, poison weapon was really cool 07:23:36 felirx: it is. I really VM: a great early game, and you've got to be flexible with the books you'll find. 07:26:39 Hey cool, bike-shed painting re: DK. 07:28:01 due: yes :( 07:28:31 the best thing about the name is DK=decay 07:38:42 True. 07:38:48 Decay Knight? 07:39:20 * due sleep. 07:40:03 Decaf Knight 07:40:45 yay... we have two implementations of weapon_str_weight() (with the same name, diff arguments), and they give different results 07:42:29 -!- Luca__ has quit [Remote host closed the connection] 07:42:49 -!- Luca__ has joined ##crawl-dev 07:42:51 KiloByte: ouch! For how long, can you see that? 07:46:39 both predate stone soup 07:47:40 differences are small. For example quick blades are 1 in fight.cc, 0 in itemprop.cc. gsc 9 / 10. 07:48:00 KiloByte: it would still make sense to consolidate them, I guess? 07:48:18 of course 07:49:20 for some reason 0.6 tiles built from the tarball and prebuilt tiles have very different floor colors. 07:49:25 itemprop.cc has the data in the table of weapons, fight.cc does that with code, using weapon skill, handedness and special cases 07:49:45 KiloByte: hm, what's better? 07:51:54 hard to tell, it's about what set of numbers you prefer 07:52:14 KiloByte: which code is cleaner? 07:52:52 by: should anything come out of the AC discussions today? 07:53:30 Improve GDR (higher GDR_rate, or better AC factor)? Link total AC to body armour base AC? Limit body armour enchantment to base AC? 07:53:33 Something else? 07:54:35 KiloByte: afaict, one is only used for launchers 07:54:41 fight.cc has commented code, itemprop.cc has no code at all (the data is just next to to-hit/dam/mass/etc) 07:54:51 Luca__: yeah 07:55:37 fight.cc assigns weights based on your handedness, itemprop.cc doesn't 07:55:58 but fight.cc has no data for launchers 08:10:34 even when I try to be constructive, it feels more like I'm being disruptive. https://crawl.develz.org/mantis/view.php?id=974 08:10:43 it's a gift and a curse :) 08:11:32 bhaak: heh, same here 08:11:36 that's why I haven't said much for weeks 08:12:27 bhaak: haha, yes, your thread 08:12:44 bhaak: you are like CSU boss Seehofer :P 08:13:43 now that's something I don't hear often 8-O 08:25:15 hmm - if a unique in a level kills itself somehow, will it create a milestone entry that I killed it? 08:25:49 just entered a new level and Gretell said "Chojin (L7 TrDK) killed Menkaure. (D:6)" 08:26:06 Napkin: yes 08:26:20 and I just killed an Ogre and an orc in the small room I see so far 08:26:30 oh, ok 08:27:53 dpeg: the understanding that current state is broken; personally I don't know what to do about it, but recording the suggested changes would be a good idea 08:29:19 Napkin: this is the cause for Lernaean Hydra spam in Sprint, for example 08:29:50 oh, ok 08:30:21 but outside of that special case, even if they die to a trap or monster infighting, it's safe to assume the player had his hands in it 08:31:07 by: okay, I'll add them to the wiki. You (and anyone else who's been around today) might have a look, okay? 08:40:08 -!- casmith789 has joined ##crawl-dev 08:52:16 -!- eith has joined ##crawl-dev 09:01:51 -!- by has quit [Disconnected by services] 09:01:52 -!- by_ has joined ##crawl-dev 09:03:30 -!- by_ is now known as by 09:03:45 -!- Enne has joined ##crawl-dev 09:26:48 -!- Luca__ has quit [Remote host closed the connection] 09:27:08 -!- Luca__ has joined ##crawl-dev 09:38:12 -!- Cryp71c has joined ##crawl-dev 09:40:43 Cryp71c: Hi! 09:41:10 There is a pretty interesting comparison of Fire and Ice Storm on http://forums.somethingawful.com/showthread.php?threadid=3285755&userid=0&perpage=40&pagenumber=57 (by German Joey). Is it valid? 09:41:24 Morning 09:42:29 dpeg: some of it is, some isnt 09:44:20 syllogism: I would really like it if good content moved from SA to Mantis. It'll just rot there. 09:45:18 the storms are too alike, but other than that, I don't think it's urgence balance-wise 09:45:38 it would be nice to replace ice storm by something completely different 09:46:10 by: sure, Haste and Mephitic Cloud are much more urgent. 09:46:27 existing vortexes are immune to subsequent fire storm casts. <- isn't this false 09:47:22 I looked at one of my firestormer's logs and saw I'd killed ~3000 fire vortices so I would guess yes 09:47:36 I had the same on my defe 09:49:03 I don't think ice storm misses either but he's right about the vortices distracting everything 09:49:13 they seem to be immune 09:51:00 does it count the vortexes killed by monsters then? 09:51:09 probably 09:51:13 since fire type casters end up with thousands of vortexes killed 09:52:04 Ice Radiance or something? 09:52:29 like ozo's refridge but in a 4 square radius centred on the caster 09:52:34 and lasts multiple turns 09:52:56 ring of fire err ice 09:52:56 -!- Cryp71c has quit [Read error: Connection reset by peer] 09:53:07 -!- Cryp71c has joined ##crawl-dev 09:53:19 s/fire/flame 09:53:35 something that freezes monsters in actual blocks of ice, so resistance doesn't help? 09:53:52 maybe making them more susceptible to physical damage 10:00:13 a LOS-wide paralysis-like effect of putting monsters into ice blocks? 10:00:38 probably overpowered, and yet another los-wide ice spell is boring 10:00:49 needn't be los-wide, could be targeted 10:01:37 I don't think it would fit in conjurations, though. 10:01:56 ice's level 9 spell doesn't have to be conj 10:03:35 and the difference from paralysis would be, well, checking different resistance, perhaps, being partly irresistible depending on power, I guess, but that's still not a lot. 10:07:32 there could be particular synergy with IOOD, which would allow blowing up frozen enemies 10:08:42 ice/enchants would be an interesting combo for it 10:09:36 by, it could allow certain attacks to shatter the ice block together with the monster inside, I guess. 10:10:38 fire storm already one-shots and definitely two-shots most things 10:10:46 syllogism: in any case, do SA users know that their comments (and some are really good) will not be seen much by the devteam, and that posting them on Mantis would improve matters? 10:11:05 dpeg: probably 10:11:07 -!- ais523 has quit [Ping timeout: 248 seconds] 10:11:11 -!- ais523 has joined ##crawl-dev 10:11:47 posting on mantins takes a lot more effort so they may not be inclined to do that even if someone suggests 10:11:53 casmith789, enchantments already have too much stuff in them 10:11:55 mantis 10:12:04 I was just thinking that 10:12:11 what about ice/summon 10:12:20 no-one uses demonology at all 10:12:28 -!- ais523_ has joined ##crawl-dev 10:12:33 although it might be better after enchantments are split up, which I believe it going to happen, eventually. 10:16:18 -!- ais523 has quit [Ping timeout: 258 seconds] 10:17:41 -!- ais523_ is now known as ais523 10:24:46 Zaba: YES 10:24:49 oops 10:26:38 is there some wiki page or any other resource that discusses the enchantments split up in detail? 10:27:30 Zaba: there is a SF entry 10:27:45 would be useful to move that to the wiki, but it's work :) 10:28:12 I remember something on the wiki, but it might have been swallowed by Eronarn's scary spell rework page 10:30:50 http://sourceforge.net/tracker/?func=detail&aid=2748451&group_id=143991&atid=757516 10:31:29 let me first add the AC stuff, lest I forget 10:32:14 by: is there a newer AC page than https://crawl.develz.org/wiki/doku.php?id=dcss:brainstorm:misc:ac&s[]=ac ? 10:34:22 I don't know 10:34:25 :) 10:34:37 our wiki is truly labyrinthine 10:34:40 there's ac stuff everywhere, e.g. on the 0.7 page 10:34:47 but still miles better than SF, of course 10:35:04 Yes, if both Eino and I neglect the wiki for a bit, it shows. 10:40:24 #-randoming seems to be broken in trunk currently 10:44:05 by: added some AC stuff to the wiki 10:44:20 I may not have captured your intentions correctly. Same for syllogism. 10:44:30 thanks 10:51:28 * dpeg ponders whether now is a good time to move the enchantments FR 10:53:21 -!- Enne has quit [Quit: Enne] 10:56:23 stupid question: is anybody actually working on an online tiles or is it a faraway project? 10:56:44 someone is 11:00:45 xale: there is an approximate solution at http://repo.or.cz/w/crawl/crawl-nettiles.git 11:01:43 I'm also trying to refactor the code so that the client and server part can eventually be split 11:02:02 this is non-trivial and very intrusive though 11:02:52 by: did you ever get around to push your double timer for portal vaults? 11:03:03 It can go in as is, in my opinion. 11:03:16 it needs wrapping up with messaging etc 11:03:37 I asked due a week or so ago, but he didn't seem to want to do it 11:06:27 -!- Luca__ has quit [Remote host closed the connection] 11:06:53 -!- Luca__ has joined ##crawl-dev 11:10:27 03dolorous * ra58dbceb447f 10/crawl-ref/source/ (dat/descript/spells.txt makeitem.cc spells2.cc): Revert "Remove the restriction on poisoning blunt weapons." 11:10:37 03dolorous * r162ab44c42fc 10/crawl-ref/source/makeitem.cc: Disallow the venom brand on staves again. Olgreb's is a special case. 11:14:04 -!- Enne has joined ##crawl-dev 11:16:46 Luca__: you mean, without tiles trying to parse the console output? 11:16:55 yes 11:17:34 my idea is to start by refactoring code so that structures with player knowledge exist in memory 11:18:06 once that is done, it is "only" a matter of (de)serializing them and creating a client/server mode 11:19:57 -!- ogaz has joined ##crawl-dev 11:22:30 has anyone considered adopting boost in crawl? 11:23:07 it was discussed a while ago, and decided against 11:23:23 why? I don't think it would be a portability problem, and would simplify the code 11:23:54 e.g. I added my homebrew signal/slot implementation in nettiles 11:24:19 also things like CrawlStoreValue and FixedVector could be replaced with boost data structures 11:24:30 (= smaller code, and someone else maintains them) 11:24:50 also things like graph algorithms could perhaps be useful 11:28:53 I don't remember the details, but remember something about boost's size 11:29:02 New Jiyva mutations and tweaks (https://crawl.develz.org/mantis/view.php?id=1620) by 78291 11:30:15 how about reconsidering that? 11:30:26 we already pull in lots of stuff, especially in tiles build 11:31:04 it's also a template library, so in theory you get only what you use 11:31:19 "There are already a few broken windows." is not really a good argument. 11:31:47 there is also the argument that size is much less important than development convenience 11:32:54 even things lie cellphones now have gigabyte drives (and anyway we would want to run tiles there, hence the above argument) 11:33:54 NIHness of a few data structures is the least of the problems in crawl's code. 11:34:33 yes, but it's a part too 11:41:19 but anyway, the code is usually good/very good locally 11:41:28 the global organization and design is a bit messy though 11:43:26 it's usually good when you view individual functions, yes. But that makes as much sense as judging how well a building is built by inspecting individual bricks. 11:44:29 Luca__: exist in a memory? 11:44:58 I mean, having code that builds such structures 11:45:25 well, for example, there's "you". 11:45:27 as opposed to just having accessors on internal structures 11:45:39 "you" contains lots of stuff that shouldn't be told to the player 11:45:53 for instance, exact expiration turn counts for effects 11:46:05 but haven't looked at that yet 11:48:28 though it might be better in that case to keep the structure and insert dummy values, not sure 11:50:43 what about the dungeon map? 11:51:10 there is show_def/map_cell, which presumably should be expanded to include info on everything visible in a cell 11:51:24 and tile_* built from that instead of in a parallel way 11:58:25 the map code already seems to have some notion of player's knowledge though? 12:03:23 yes, but it seems rather limited to me 12:03:30 i.e. not enough for a tiles client 12:04:10 (because it only shows whatever is on top) 12:07:30 interesting! 12:14:44 -!- Zaba has quit [Ping timeout: 240 seconds] 12:15:40 what part of all this you're working on now, if you don't mind me asking? 12:17:08 I'm testing monsters 12:17:24 I also did items, but that's less testable right now 12:20:27 the idea for items is to keep item_def but erase all non-known parts 12:20:54 for monsters, make monster_info contain all visible monster properties, and use that internally for descriptions, name generation, etc. 12:21:18 (including erased item_defs for visible equipment) 12:22:17 ah, s/colour/color, s/armour/armor/, s/sation/zation/ would be nice too :) 12:23:37 -!- Zaba has joined ##crawl-dev 12:33:12 ~. 12:33:14 . 12:33:44 Oops. 12:37:18 by: I put a preliminary (but apparently basically working) version of the item info erasure + monster_info changes at http://repo.or.cz/w/crawl/crawl-nettiles.git/shortlog/refs/heads/clientserver 12:39:48 did you consider filling, instead of erasing? 12:41:26 -!- morik has quit [Disconnected by services] 12:41:31 -!- bfein_ is now known as morik 12:41:53 -!- morik_ has joined ##crawl-dev 12:42:17 that might indeed be cleaner 12:43:22 it also probably would be apparent to the client if you fill too little. 12:46:48 03N78291 * r6d268f1676e8 10/crawl-ref/source/ (15 files): Add several Jiyva specific mutations, Jiyva tweaks 12:47:08 03by * re031ec17b196 10/crawl-ref/source/ (fight.cc itemprop-enum.h makeitem.cc): Make acid bite use new SPWPN_ACID. 12:47:11 03by * r743c67268d10 10/crawl-ref/source/fight.cc: Make fangs and acidic bite give extra aux bite attacks. 12:47:11 03by * r9e9dfb190fd1 10/crawl-ref/source/wiz-you.cc: Allow giving slime mutations through wizard mode. 12:49:15 by, thanks! 12:50:02 are you ok with these mutations? I figured they can't do much harm, if only jiyvaites get them 12:51:59 fun mutations 12:54:23 Luca__: just reading the commit messages, that looks wonderful 12:54:38 note however that it changes some things and also surely breaks something 12:54:49 s/also/almost/ 12:54:57 by: cool, thanks 12:55:04 zipcode, too, of course 12:56:46 btw, rewriting the item patch to whitelist instead of blacklist, and use a single switch(object-type) for all fields 12:58:35 by, definitely. I think its great that Jiyva has some mutations (especially slime or ooze themed ones) that you can only get as a Jiyvaite. 12:59:39 Would be cool to also give Jiyvaites the acidify wall abiltiy. 13:00:10 dpeg: to stop monsters camping in corridors against them? 13:00:20 My sister likes IF (interactive fiction), she could start writing a novel on "the life of Dissolution" :) 13:00:26 ais523: well, they would take damage, yes. 13:00:46 I suppose it could have a few niches 13:00:49 It would be local (say LOS) and could be only done once per level, I guess. 13:00:49 depending on how powerful it was 13:01:08 are jiyvites immune to acid wall damage? 13:01:09 main problem is how monsters should react 13:01:15 ais523: I hope so. 13:01:18 daftfad (in ##crawl) raises a good point (which I have shared, but never voiced). The prompt on troves seems backwards. Logically, the Y/N prompt should ask "Do you want to enter the trove" and not "Do you want to lock the trove" since if you pressed '>' the implicit understanding is that you want to enter the trove. 13:01:42 Thus, ">Y" should enter the trove, not lock it. 13:01:56 hmm, that would go quite a way towards explaining how the mechanic worked, too 13:01:57 Has anyone else shared these thoughts, by chance? 13:02:10 you try to enter, can't afford it, then the imp turns up and offers to lock it 13:02:34 speaking of >... I have recently wondered why > and < are separate commands at all. Why not make them both follow the staircase/enter the portal/whatever the player is on, regardless of direction? 13:02:49 Zaba: we wondered that too during the usability project 13:02:56 :P its a good point 13:02:58 it most likely stems from older requelikes 13:03:04 keeping > and < separate is good imo 13:03:15 you can merge them on portals and the like if you want 13:03:19 casmith789, any objective reason? 13:03:30 X > 13:03:31 it's good when travelling to stairs in NetHack 13:03:31 X < 13:03:32 but sometimes when I'm using G> and I don't move I don't want to go back up 13:03:50 the other reason would be helping to catch thinkos 13:03:58 well, G>> etc or X 13:03:58 G[<>] and X[<>] were not covered by the proposal. 13:03:58 where you think you're on a downstairs and you're actually on an upstairs 13:03:58 zaba: If I am on a staircase and wasn't paying attention to whether its up or down well enough, I'd rather not go the wrong way 13:04:01 casmith789, thats an entirely different prompt, which should continue to operate as it presently does. 13:04:15 well sometimes I'm standing on a > and I press < 13:04:20 morik, there are plenty of other things that result in worse things if you're not paying attention. 13:04:24 (because I think I'm on a <) 13:04:32 not something I want 13:04:39 cryp71c: is the fact that there are worse things a good reason to make a design decision? 13:04:43 perhaps the correct approach is to just add a "use stairs/portal" button in the tiles version 13:04:49 sometimes I go up into a separated region in, say, orcish mines, then forget I didn't come from the above, try to <, fail, think that there's no way to escape and try something else, and most likely die. 13:05:03 casmith789: press G? sometime 13:05:12 hm? 13:05:12 (I think there is already some way to do that with a modified click) 13:05:40 casmith789: you said you sometimes don't want to take the stairs, just move towards them. Press G? to find out how. 13:05:53 No. 13:05:56 re troves: it's still not a good idea that you can pay for them. 13:06:06 Sometimes, I want to go down a level, and I think I'm standing on the > 13:06:08 So the prompt should probably go altogether. 13:06:12 when I'm actually on the < 13:06:13 casmith789: ah 13:06:14 cryp71c: in less abrasive terms, why would the fact that there are worse things that can happen if you aren't paying attention mean that merging < and > into the same thing would be more usable or better than having them separate? 13:07:02 I can understand that < and > don't make too much sense for tiles players (but they can use the mouse anyway), but for console, they're extremely intuitive, no? 13:07:11 morik, the fact that there are worse things and that its a good proposal that makes sense (both for new and veteran players) is what makes a design decision. 13:07:25 dpeg, for most, certainly...but not all, I think. 13:07:58 In my case it is a rare but not isolated situation I find myself in. 13:07:59 Cryp71c: never for all 13:08:20 And I'm not proposing the removal of < and > from their current mechanic, but perhaps an additional binding which allows for auto-up/down depending on what the player is standing on. 13:08:39 I'd not like to press > on an < to go up 13:08:48 Anyways, the main issue is the trove thing...which is of higher priority. 13:08:53 cryp71c: I don't think that the state of other things (in terms of usability) should affect whether a new proposal is acceptable... if everything else sucks, that is not a good reason to make another thing suck. If everything else is good, that is also not a good reason to allow something to suck. (I'm speaking generally here--I don't think this particular decision would result in 'suck' per se) 13:09:24 basically, I'm agreeing with casmith789 for the same reasons--if I'm on an up-stair, I would be suprised (in a bad way) if I were to use it by hitting > 13:09:27 morik, the state of other things was simply a counter arguement to "I'm not paying attention so the game should be safe for me" 13:09:49 Cryp71c: the game already has some stuff like that in? 13:09:55 o when in sight of a monster 13:09:58 moving while confused 13:10:00 etc 13:10:01 casmith789, some, but its not a paradigm we're trying to move towards 13:10:03 The surprise of morik and casmith789 is not a good reason: that just means you're used to something and don't want to relearn (a perfectly natural sentiment). 13:10:25 hm? 13:10:33 But the question is whether seperating "use stairs" into two commands < and > has some intrinsic benefit. I think so. 13:10:35 A better question then: What is the benefit to having > and < function the same? More intuitive for tiles players? 13:11:09 pressing > to go upstairs imo is bad, because sometimes I find myself in a situation where is would be detrimental 13:11:15 s/is/it/ 13:11:33 dpeg: I think that suprising existing users IS a critereon that should be used when deciding on UI design. 13:11:40 casmith789, no one is saying anything about mapping > or < to this "use_stairs" functionality. 13:11:53 Cryp71c, at least, not by default 13:11:54 that would be a terrible implementation. 13:11:58 there were proposals to merge them 13:12:02 just voiced 13:12:03 morik: sure. But foremost, one should consider the design in its pure state, i.e. for a new user. 13:12:06 which would have that effect 13:12:08 to be honest, I'm just tired of having to hold shift all the bloody time 13:12:11 Into a single binding, but not to replace the < or > bindings, by default. 13:12:13 how about an helpful error message? 13:12:16 when going in one direction, but not in another 13:12:35 ("You can't go up there, perhaps you wanted to press > to go down instead?") 13:12:37 Zaba: why would you have to hold shift? 13:12:40 dpeg, also I proposed a change in Trove prompting, to be more logical, as proposed upwards. 13:12:46 casmith789, > is shift + < 13:12:56 in addition to both being shift + , and . 13:13:01 Some layouts (swedish) have > on shift-< 13:13:05 oh right, different keyboards 13:13:06 (this keyboard happens to have a separate key) 13:13:07 (next to z) 13:13:11 this also allows to have stairs/portals that go both up and down to different places 13:13:17 even though it's not swedish, it has such a key 13:13:18 Zaba: we clearly need special Crawl keyboards 13:13:19 dpeg: I'm not sure I would agree with that statement in general. It depends on a lot of things. In this case, perhaps you are right, as RLs tend to be hard enough for new users, and so existing users getting a little confused may be more desirable than new users having a slightly higher learning curve 13:13:29 03dolorous * r02a51a935f5d 10/crawl-ref/source/mutation-data.h: Show the thin skeletal structure mutation's stealth effects differently. 13:13:29 casmith789: I saw, but I believe that the trove mechanic should be rethought. 13:13:35 Luca__, I hope that never happens. 13:14:03 Troves should always be something gettable in the game, but hard to get 13:14:03 morik: only the radical approach has enabled us to redo monster/item colours, key commands etc. 13:14:22 no +0 elven robes, no 6 potions of experience 13:14:22 morik: there is _always_ an outcry by the old users, but if you listen to them in advance, there'd be no progress. 13:14:38 casmith789: neither of these can come up currently. 13:14:41 03dolorous * r35b78ff6ac2e 10/crawl-ref/source/mutation-data.h: Typo fix. 13:14:43 good 13:14:51 well why does it need to be rethought? 13:14:52 dpeg: sure, surprising existing users isn't the ONLY criterea. If a new design is noticeably better enough to more than offset that, then of course it makes sense 13:15:20 casmith789, paying gold is too easy an option for entering troves. its also very bland. 13:15:30 morik: of course we try to alienate existing users as little as possible (for example, there were these options to reset colours and key commands). 13:15:48 the flavour reason imo is a bit weird 13:15:55 why not just have troves permanently locked 13:15:56 and perhaps in this game's case, any benefit to new players could cause existing user alienation to be weighted much lower, since there is a large desire to flatten the learning curve 13:15:57 it doesn't help some trove layouts have tons of gold 13:16:04 with some items to get in 13:16:05 casmith789: what Cryp71c said. Also, I am not so sure anymore on the troves being timed first, and then you pay to make them untimed. Perhaps there's a better way to set this up. 13:16:29 syllogism: the gold piles in there are essentially bugs. 13:16:32 perhaps troves should be always permanent, and always require item(s) to enter? 13:17:00 the original idea of troves was that a long timer made people dive to get gold before the trove disappeared, wasn't it? 13:17:01 casmith789: the real reason is that we want players to part with something, so they feel more obliged to come up with whatever the trove actually asks for. 13:17:02 Zaba: that was my suggestion on mantis 13:17:14 hm, weird 13:17:23 i dove for gold once and used acquirement to get in; it was fun, but it doesn't generally really work 13:17:25 I think that's the best way: No imp, no locking, no nothing. Bring in stuff, enter. Simple! 13:17:37 I certainly wouldn't pass up my +5 robe of resistance if I'd given 130 gold 13:17:43 syllogism: yes, it is only fun when it works. If not, it is quite disappointing. 13:18:00 One could explicitly tell the time limit, but I am not sure we want that. "2592 turns to go!" 13:18:35 the issue is, the best strategy may just be to ignore the things, but people wouldn't do that because it wouldn't be any fun 13:18:41 and it would be a huge disappointment to try and fail 13:19:09 -!- ais523 has quit [Remote host closed the connection] 13:19:51 I indicated on Mantis that there could be two types of trove quests: strategical (collect some not so rare item, e.g. identify, recharging etc.) and sprint (have to hand over something that exists in the dungeon real quick). 13:20:12 tick, tock, tick, tock, tick, tock ... tick, tick, tick, tick 13:20:23 Luca__: for some reason monster_info std::sort in directn.cc requires monster_info::operator= 13:20:39 03dolorous * r89197f24359c 10/crawl-ref/docs/changelog.txt: Document Jiyva changes in the 0.7 changelog. 13:20:49 bhaak: <3 13:21:02 but then you'd have to explain why there's a big grandmother clock in the trove :) 13:21:18 by: strange, works for me, I'll just add an operator= 13:21:23 bhaak: we don't explain anything. "In the beginning, there was Marduk." 13:21:46 tja, my people want to know what they are doing :-P 13:22:08 "Marduk collects souls of humans and big clocks of their grandmothers." 13:22:21 for reference, the error is here: http://sprunge.us/NeSa 13:23:41 yes I understand 13:23:58 (ouch, is g++ 4.0 the newest available on osx 10.4?!) 13:24:04 I guess the g++-4.0 library implements sorting differently? 13:24:11 probably 13:24:12 dpeg: some things it can ask for now are kinda over the top (e.g. 8 potions of magic) 13:24:24 not sure, this is osx 10.5, but crawl chooses to compile with the 10.4 sdk 13:24:35 don't really care 13:24:45 g++ 4.0.1 is the newest on 10.5 13:24:54 so you're not that far behind :) 13:25:08 uh? really? 13:25:14 no way to get 4.4 or 4.5? 13:25:30 xale: !magic has been unrarified. 13:26:16 apparently current xcode on 10.6 has 4.2.1 13:26:21 oh, here is also a binary for 4.2.1, but it is not linked to g++ 13:26:41 xale: the main issue with !magic is that some characters don't need it at all, whereas items like ?id, ?recharge, ?mapping, !hw etc. are useful for everyone (so comprise a choice of the player). 13:27:37 hm, I guess for stackables, the trove could ask for a first payment (one or two, depending on item) real quick in order to lock it 13:28:28 by: can I help with the timer descriptions? 13:30:01 Kilobyte, just fyi, I'm working on spriggan tiles. 13:31:17 dpeg: if the code in dat/clua/lm_timed.lua makes sense to you, maybe 13:33:54 by: should be fixed now 13:34:26 by: just looking at it. No, it doesn't ;/ 13:40:49 -!- Poor_Yurik2 has joined ##crawl-dev 13:44:01 Luca__: compiles, yay! 13:44:22 !log jcole 13:44:22 48. jcole, XL16 HENe, T:44414: http://crawl.akrasiac.org/rawdata/jcole/morgue-jcole-20100526-184321.txt 13:45:06 heh 44413 | Swamp:4 | HP: 4/110 [vampiric drain backlash (42)] 13:45:40 that "jcole's ghost" was a clone, I assume 13:48:11 !tv jcole 13:48:11 48. jcole, XL16 HENe, T:44414 requested for FooTV. 13:50:20 hrm 13:51:01 by: what's wrong? 13:51:13 footv didn't play the right part 13:52:32 syllogism: draining works properly in master it looks like 13:54:00 there was something wrong with it? 13:54:04 anybody wants to look at my patch? it's a one-liner. 13:55:34 illusions have the same undead/demonic state as the player in master, so no draining backlash 13:55:51 I guess it could make sense to treat them as demonic 13:56:21 but an illusion being considered undead seems like the illusion == ghost implementation detail leaking 13:58:44 syllogism: eb06c7ae729, sets clone holiness to player holiness 13:59:05 so in master, you can drain the clones 13:59:35 xale: what patch? 14:03:09 http://sprunge.us/gOcB 14:03:29 Luca__: btw, did you see Enne's tile rendering global state access clean-up? (f786a607b0) 14:04:23 by: yes, that's nice 14:12:48 03by * r9acb35545138 10/crawl-ref/source/stash.cc: Do not search for spells inside books player cannot understand. (xale) 14:13:33 xale: thanks; is this related to a mantis issue that should be closed? 14:13:50 not that I know of. 14:13:55 ok 14:17:17 btw, If I type "book" to the search, shouldn't it list ~all the spellbooks? 14:17:21 what happened to the idea of making those books known instead? (I think there was even a patch for that) 14:17:52 https://crawl.develz.org/mantis/view.php?id=624 14:21:20 -!- Luca__ has quit [Remote host closed the connection] 14:21:39 -!- Luca__ has joined ##crawl-dev 14:22:42 Luca__: there simply was no decision on the matter. 14:25:28 xale: considering the high-level books can actually be browsed in master, should the patch still be applied 14:25:34 ? 14:25:43 they can? 14:25:49 yes, looks like it 14:25:55 in shops at least 14:26:08 hold on! 14:26:23 the patch says * If you find these books lying around in the dungeon, reading them will _not_ identify them (or give away their spell list) unless you meet the requirements. [1] 14:26:24 oh, it wasn't a shop item, I just created it. 14:26:33 but I assume unidentified books are already excluded, right? 14:26:42 but maybe it's because I'm in wiz mode 14:28:55 once it's identified, you can read it 14:28:59 hm.. 14:30:02 but I assume you ran into some actual bug? 14:30:12 else I can just revert, no harm done 14:32:00 ...grrr, I can't seem to push the rebased version of the unarmed_effects branch. 14:32:02 I'm going to try all this again and let you know. 14:32:29 it is definitely broken in 0.6 though! 14:33:05 Cryp71c: maybe you haven't specified a '+' before the branch name to force? 14:33:11 Luca__, yeah, I did that. 14:33:46 http://pastie.org/978661 14:38:59 -!- ais523 has joined ##crawl-dev 14:40:17 by: yep, it probably needs to be reverted 14:41:06 by: you can't identify a book that's beyond your level of understanding, but you can use a scroll and I'm assuming it will automatically identify in a shop. 14:41:44 after that you can see a list of spells and they become searchable. 14:41:56 03by * r5e1565eefe28 10/crawl-ref/source/stash.cc: Revert "Do not search for spells inside books player cannot understand. (xale)" 14:44:34 okay, I'll move the enchantments FR 14:47:35 ??git 14:47:35 git[1/2]: git clone git://crawl-ref.git.sourceforge.net/gitroot/crawl-ref/crawl-ref 15:15:05 what's with trunk inscribing pears and meat rations with {unknown} after I died 15:16:08 nice, bug 15:16:34 is there any reason for having show_def instead of putting the LOS info in map_knowledge as well? 15:19:04 i.e. would you accept removing env.show and show_def and using map_knowledge instead? 15:19:56 yes, sounds sensible 15:21:12 there's a small open question re coordinates: currently crawl tries to hide absolute coordinates from the player 15:21:22 isn't that orthogonal? 15:21:36 anyway, I'd propose setting the point where the level is first entered as (0, 0) 15:21:43 and giving out coordinates in that coordinate system 15:21:46 yes, that would make sense 15:21:53 (this doesn't give more information than now) 15:22:17 it's mostly orthogonal, just that env.show is currently in player-centric coordinates 15:23:12 yes, but you can just do show2grid internally, right? 15:23:17 sure 15:31:58 -!- Luca__ has quit [Remote host closed the connection] 15:32:18 -!- Luca__ has joined ##crawl-dev 15:33:57 -!- syllogism has quit [] 15:45:03 -!- casmith789 has quit [Remote host closed the connection] 15:46:00 by: could you take a look at this: http://sprunge.us/BWMj 15:48:59 move accomplished 16:01:26 xale: too tired now to figure out what that's meant to accomplish 16:01:42 hopefully someone else will have a look 16:02:12 -!- by has quit [Quit: night] 16:02:16 -!- Anym has joined ##crawl-dev 16:02:44 Anym: hi! 16:02:56 Hi dpeg. 16:03:17 Anym: just twenty minutes ago I moved some comments of you to the wiki :) 16:03:32 Excellent. 16:05:00 is there an easy way to create a lot of item on the floor? 16:06:11 nevermind. 16:06:43 xale: Repeat (0) should also work for wizmode command IIRC. 16:07:25 |, &o: all 16:07:35 &| 16:38:48 -!- casmith789 has joined ##crawl-dev 16:39:24 -!- casmith789 has quit [Remote host closed the connection] 16:40:00 -!- Luca__ has quit [Remote host closed the connection] 16:40:17 -!- Luca__ has joined ##crawl-dev 16:41:29 Hello Anym! 16:41:43 Hi sorear. 16:44:34 sorear: I mentioned you when by asked whether 4.0 combat used the same basic approach to EV (compare to-hit with EV) and AC (compare d(dam) with d(AC)). Do you know? 16:45:34 not offhand 16:45:52 I beleive 4.0 had more hidden factors in EV 16:46:09 like the ones I recently removed from SH 16:47:39 monster dodging was very much as now - sum a bunch of random factors, compare to mob EV 16:48:57 player dodging: 1d(mob to-hit) < random2limit(EV, 40) + random2(dex) / 3 + (melee evasion fudges) 16:52:23 melee dodging is quite different now - just a to-hit/EV check (with random factors) 16:52:26 no dex 16:53:15 -!- ogaz has quit [Ping timeout: 240 seconds] 16:53:40 sorear: thank you 17:12:54 -!- Cryp71c has quit [Quit: going home] 17:26:34 Bye everyone. 17:26:59 bye! 17:27:08 -!- Anym has quit [Quit: ChatZilla 0.9.86 [Firefox 3.6.3/20100401080539]] 17:39:31 -!- pointless_ has joined ##crawl-dev 17:41:58 pointless_: Hi! 17:42:03 Hi 17:45:19 pointless_: What do you think about ballisto colonies reacting particularly sensitive if they're down to one ballisto? 17:47:01 -!- Luca__ has quit [Remote host closed the connection] 17:47:19 I don't know what, what specifically would happen? 17:47:21 -!- Luca__ has joined ##crawl-dev 17:50:09 pointless_: some chance to be hyperactive? 17:51:50 that could work, I was mostly thinking about making hyperactive part of the cap condition i.e. if there are a lot of ballistos they become hyperactive after reaching the cap (which stops the spawn cycle) 17:52:02 pointless_: yes, that is good 17:52:53 I'd like to have some reaction on the other end for two reasons: a single ballisto is not yet a colony, but you can kill as if it was one. And if you manage to cut down an actual colony to the last ballisto, it'd be cool if the ballisto showed that. 18:12:44 03pointless_ * r980150f56de0 10/crawl-ref/source/godabil.cc: Remove some debug output 18:20:49 -!- Textmode has joined ##crawl-dev 18:21:47 morning all 18:23:39 -!- ais523 has quit [Remote host closed the connection] 18:24:08 -!- sorear has quit [Read error: Connection reset by peer] 18:25:09 -!- sorear has joined ##crawl-dev 18:25:17 Textmode: hi! 18:26:23 whats up? 18:26:29 Nuffink! 18:27:15 meep 18:28:13 Luca__: do you know if negative position in item_def for items in inventory also applies to monsters? 18:28:28 what do you mean? 18:28:44 if monster picks up an item, what's the item's position? 18:29:02 not sure 18:32:04 -!- ogaz has joined ##crawl-dev 18:39:59 I think I've figured out #1596, but I can't fix it 18:41:49 it's in the item grid, but it doesn't have a location. 18:42:12 so, a mimic picks up its item when created 18:42:52 then when you list items and have unidentified items the non-existent location is used and it all crashes 18:43:07 and by unidentified items I mean unidentified mimics 18:43:47 heh, yeah 18:43:47 however it seems like an item can't be both it monster's inventory and have a location, so I've got no idea how to fix that 18:44:03 use the monster's location instead? 18:44:27 mimics are non-transitory creatures, after all 18:44:47 Would that lead to a problem when it teleports? 18:46:01 all places that use get_mimic_item get a reference from mitm 18:46:09 but the item in mitm has no location! 18:46:30 so? 18:46:48 just get its holder and then get the holder's position. 18:49:26 where would I do that? 18:49:27 if (item->held_by_monster()) coord_def pos = (item->holding_monster())->pos(); oor something like that 18:50:56 the callers don't necessary want pos, they just grab item_def and do whatever they please with it 18:51:09 so then you'll have to fix every place that uses position from this item_def 18:51:15 hm 18:51:17 unless I'm missing something 18:51:29 I haven't looked at the code 18:52:37 perhaps add an helper in item_def that does that? 18:52:57 but is it actually needed? 18:55:06 -!- lorimer has quit [Remote host closed the connection] 18:56:15 I think get_mimic_item should return a copy with a correct position 18:57:14 helper isn't going to do much good since everything is already accessing "pos" directly. 18:57:38 s/pos/pos()/g on all the source? 18:58:56 or return a version with only observable data and computed position 18:59:12 but anyway, does anything that read the item position get run on mimic items? 19:00:06 sure, otherwise it wouldn't crash 19:01:39 ah I see 19:03:14 bye, it's late over here 19:03:39 -!- dpeg has quit [Quit: zzz] 19:10:01 -!- ogaz has quit [*.net *.split] 19:10:44 -!- ogaz has joined ##crawl-dev 19:10:51 no longer crashes, but I've got no idea whether this is okay: http://sprunge.us/FdOF 19:14:53 -!- Mu_ has quit [Quit: Defecator, may everything turn out okay so that you can leave this place.] 19:32:11 -!- Enne_ has joined ##crawl-dev 19:32:11 -!- Enne has quit [Read error: Connection reset by peer] 19:32:12 -!- Enne_ is now known as Enne 19:47:23 also, it's a duplicate of #1518 19:56:34 is there any way to generate a specific vault? 19:56:48 nevermind! 20:04:32 xale: In what sense? 20:04:35 &L 20:04:37 Hi, Enne! 20:05:06 Hey, due. :) 20:06:16 I almost have a monster list all working. :) 20:06:21 Woohoo! 20:06:37 I know jpeg wants it to be visible all the time or maybe pop up when monsters appear. 20:06:44 I'm thinking I'll leave that as an exercise to the reader. 20:08:15 * due writes grumpy comments on the tracker. 20:08:28 Argh, David left. 20:18:32 * due kills the DK discussion with a stick. 20:30:08 If we had a monster list in tiles, what would you expect left-clicking on a monster to do? 20:30:15 Attack if adjacent, move one step closer if not? 20:30:17 Something else? 20:30:28 examine 20:30:34 Left clicking, attack/move closer. 20:30:37 Right click, examine. 20:31:00 Possibly swap them? I'm not sure. the default attacking might not be expected. 20:31:09 something I really like in the console version is a list of monster on the right 20:31:17 Right-click is describe pretty much everywhere. 20:31:41 I'm working on the tiles equivalent of that. 20:31:46 that's great 20:34:11 -!- Luca__ has quit [Remote host closed the connection] 20:34:27 -!- Luca__ has joined ##crawl-dev 20:36:40 Enne: local hp, maxhp = you.hp() <-- that's right, yeah? 20:36:56 Ah, apparently. 20:36:59 * due forgot about luaterp. 20:37:00 No idea, sorry. 20:37:09 :( Noo, Enne is my lua deity! 20:37:36 I know about some of it, but there's a *lot* of lua functions. 20:37:37 You're not allowed to disavow knowledge of anything Lua! 20:37:38 ;) 20:37:51 "I don't know anything, but I'm good at reading source code." 20:38:02 Heh, LUARET2 presumably returns two values. 20:38:16 Which I'm pretty sure means you can split it with x, y = z. 20:40:41 -!- morik has quit [Disconnected by services] 20:40:46 -!- morik_ is now known as Morik 20:41:04 Enne: The health bar is tiles-only, yeah? 20:41:06 -!- bfein_ has joined ##crawl-dev 20:41:09 Or not, I suppose. 20:41:22 Health bar on the dungeon itself is tiles-only. 20:41:32 Yea. 20:41:46 but there's that great big one in console, so it's not too unambigious in both contexts :) 20:41:49 * due working on the tutorial. 20:42:42 Hm, is there something else for CMD_DISPLAY_COMMANDS in tiles? 20:43:00 There's a health bar in tiles as well. 20:43:36 Yeah, exactly! 20:43:45 Hooray, the first tutorial message without a tiles check! 20:44:58 <3 20:47:46 One more to go and the first level is complete. 20:47:49 Apart from post-wear hooks. 21:09:07 nope, that patch is broken. 21:10:52 maybe it's better to have another special location for mimic items? 21:11:39 or.. change unlink_item() to handle all this. 21:21:11 03due * rca65751d9005 10/crawl-ref/source/dat/des/tutorial/tutorial.des: Update more of the tutorial messages. 21:21:45 My first commitin weeks! 21:25:06 Hooray! 21:41:12 * due gone. 21:41:47 -!- sorear has quit [Read error: Connection reset by peer] 21:42:13 -!- sorear has joined ##crawl-dev 22:03:57 -!- straydusk has joined ##crawl-dev 22:08:13 -!- straydusk has left ##crawl-dev 22:11:40 I think item_def needs another way to mark items held by a monster 22:12:20 perhaps something like pos == coord_def(-3, -3) or something 22:12:41 -!- xale has quit [Quit: xale] 22:18:26 03Enne * rd9333cb15b05 10/crawl-ref/source/rltiles/ (dc-gui.txt gui/tab_label_monster.png): Adding monster tile tab images. 22:18:27 03enne.walker * rbc13d7a69337 10/crawl-ref/source/ (directn.cc tilereg-dgn.cc): Move special-case cloud logic. 22:18:28 03enne.walker * rbeafec8f2f22 10/crawl-ref/source/ (tilepick.cc tilereg-dgn.cc tiles.h view.cc): Refactor tile animations and variations. 22:18:30 03enne.walker * r5babdf18d246 10/crawl-ref/source/ (14 files): Add monster list for tiles. 22:18:33 03enne.walker * r1f1ceb1cdc83 10/crawl-ref/source/ (tilereg-act.cc tilereg-act.h tilereg-dgn.cc tilereg-dgn.h): Allow left-clicking on monster list to attack. 22:18:34 03enne.walker * r80a4537a175f 10/crawl-ref/source/ (5 files): Merge ActorRegion and MonsterRegion. 22:18:54 I'm amused that the CIA hook translated the first commit, but not the rest. 22:20:53 -!- eith has quit [Ping timeout: 240 seconds] 22:35:10 -!- Madtrixr has joined ##crawl-dev 22:37:03 -!- Enne has quit [Quit: Enne] 23:10:06 -!- pointless_ has quit [Ping timeout: 245 seconds] 23:35:43 -!- Eronarn has quit [Ping timeout: 260 seconds] 23:37:36 -!- Eronarn has joined ##crawl-dev 23:58:45 -!- Luca__ has quit [Ping timeout: 240 seconds]