00:01:02 -!- MarvinPA has quit [Read error: Connection reset by peer] 00:01:57 Unstable branch on crawl.develz.org updated to: 0.10-a0-1305-g4f2d919 (32) 00:03:21 -!- casmith789 has quit [Read error: Connection reset by peer] 00:03:57 -!- raskol has quit [Read error: Connection reset by peer] 00:04:03 -!- casmith789 has joined ##crawl-dev 00:04:13 -!- bmh has joined ##crawl-dev 00:04:20 -!- raskol has joined ##crawl-dev 00:07:58 -!- valrus has joined ##crawl-dev 00:13:28 Windows builds of master branch on crawl.develz.org updated to: 0.10-a0-1305-g4f2d919 00:24:26 -!- jeanjacques_ has joined ##crawl-dev 00:28:41 -!- valrus has quit [Remote host closed the connection] 00:43:22 -!- Keskitalo has joined ##crawl-dev 00:43:35 good morning 00:43:47 good morning :) 00:44:16 Keskitalo: I fixed the bug you pointed out. I was only correctly dealing with poison for non-melee sources 00:45:29 bmh: thanks, i'll try out your new patch when i get the chance 00:47:02 Thanks! I'd still like to make it more interesting than a raw 90% chance, perhaps incorporate regen rate 00:48:03 -!- Pacra has quit [Quit: This computer has gone to sleep] 00:49:17 moin! 00:49:35 nice to see Europe waking up when I'm getting ready for bed 00:49:43 crawl is a "follow the sun" outfit! 00:52:38 hehe 00:59:38 -!- bmh has quit [Quit: bmh] 01:03:58 -!- raskol has quit [Read error: Connection reset by peer] 01:04:22 -!- raskol has joined ##crawl-dev 01:05:45 !tell kilobyte The Spider portal vault is meant as just a testbed for the branch, so it can go as the branch is implemented. We can think about repurposing the maps somewhere else later. 01:05:46 Keskitalo: OK, I'll let kilobyte know. 01:07:25 especially the maps by 7hm and mumra :) 01:25:57 -!- ainsophyao has joined ##crawl-dev 01:51:45 -!- evilmike has quit [Read error: Connection reset by peer] 02:04:05 -!- raskol has quit [Read error: Connection reset by peer] 02:04:25 -!- raskol has joined ##crawl-dev 02:10:33 -!- syllogism has joined ##crawl-dev 02:10:45 -!- Zaba has quit [Ping timeout: 255 seconds] 02:12:04 -!- Keskitalo has quit [Quit: leaving] 02:15:27 -!- Zaba has joined ##crawl-dev 02:21:45 Unstable branch on crawl.akrasiac.org updated to: 0.10-a0-1305-g4f2d919 (32) 02:52:56 -!- dtsund has quit [Read error: Operation timed out] 02:55:14 -!- ainsophyao has quit [Remote host closed the connection] 03:03:07 -!- monqy has quit [Quit: hello] 03:03:51 -!- raskol has quit [Read error: Connection reset by peer] 03:19:19 -!- ainsophyao has joined ##crawl-dev 03:19:51 -!- raskol has joined ##crawl-dev 03:27:19 -!- Pacra has joined ##crawl-dev 03:30:42 -!- Pacra has quit [Client Quit] 03:40:36 -!- ainsophyao has quit [Remote host closed the connection] 03:41:46 -!- ainsophyao has joined ##crawl-dev 03:58:10 -!- ainsophyao has quit [Remote host closed the connection] 04:03:54 -!- raskol has quit [Read error: Connection reset by peer] 04:13:00 -!- Wensley has quit [Ping timeout: 258 seconds] 04:19:57 -!- raskol has joined ##crawl-dev 04:21:47 -!- blabber has joined ##crawl-dev 05:06:39 -!- alefury has joined ##crawl-dev 05:08:13 hi 05:08:34 the targeting display seems to be broken in windows tiles 05:09:30 maybe its related to recent webtiles changes? 05:09:51 everything looks like its smite targeted, theres no yellow shading to show the projectile path 05:19:01 Targeting display broken (no yellow shading) (https://crawl.develz.org/mantis/view.php?id=4754) by Galefury 05:19:08 -!- syraine has quit [Ping timeout: 244 seconds] 05:28:00 -!- Pingas has joined ##crawl-dev 05:32:20 -!- ixtli has quit [Quit: Computer has gone to sleep.] 05:33:54 -!- mikee_ has quit [Quit: HydraIRC -> http://www.hydrairc.com <- Organize your IRC] 06:03:48 -!- raskol has quit [Read error: Connection reset by peer] 06:04:08 -!- raskol has joined ##crawl-dev 06:10:31 03kilobyte * rbdc6d141efbe 10/crawl-ref/source/Makefile: Don't split the "greet" spam. 06:10:41 03kilobyte * r5222aba20020 10/crawl-ref/source/dat/des/test.des: Update the "feat_test" vault. 06:10:41 03kilobyte * r2db17388c3dc 10/crawl-ref/source/tilemcache.cc: Add weapon attachment for 4/5 of tiles added in c20d067c. 06:34:58 -!- Pingas_ has joined ##crawl-dev 06:37:39 -!- Keskitalo has joined ##crawl-dev 06:37:54 -!- Pingas has quit [Ping timeout: 258 seconds] 06:48:13 -!- ainsophyao has joined ##crawl-dev 06:59:29 -!- RichardHawk has joined ##crawl-dev 07:04:09 Game froze after triple drawing legendary card of experience (https://crawl.develz.org/mantis/view.php?id=4755) by Tonfa 07:04:13 -!- raskol has quit [Read error: Connection reset by peer] 07:04:14 -!- raskol has joined ##crawl-dev 07:15:08 -!- Torokasi has joined ##crawl-dev 07:41:48 -!- Brannock_ has joined ##crawl-dev 07:46:20 -!- krel has quit [Ping timeout: 252 seconds] 07:46:26 -!- krel has joined ##crawl-dev 07:46:26 -!- Mu_ has joined ##crawl-dev 07:57:03 -!- st_ has joined ##crawl-dev 08:03:53 -!- raskol has quit [Read error: Connection reset by peer] 08:04:14 -!- raskol has joined ##crawl-dev 08:09:15 -!- MarvinPA has joined ##crawl-dev 08:36:10 !tell bmh It would be nice to get 90% res for poison clouds too. I tried just removing the immunity check in clouds.cc, but the player and monsters (if resistant) then receive damage in poisonous clouds, just not the poison status. 08:36:11 Keskitalo: OK, I'll let bmh know. 08:38:48 !tell bmh _actor_cloud_damage in cloud.cc probably just needs a case for CLOUD_POISON. 08:38:48 Keskitalo: OK, I'll let bmh know. 08:44:39 !tell bmh Actually, _actor_cloud_immune could return true 90% of the time, or _actor_cloud_damage could reduce the damage by 90% (but then it'd be nice to add some kind of "partially resist" message. And this just for the player. 08:44:40 Keskitalo: OK, I'll let bmh know. 08:44:50 spam spamity spam 08:55:26 !tell bmh We still want the poison suspectibility mutation too -- mentioning just in case it would change the code for 90% resistance. 08:55:26 Keskitalo: OK, I'll let bmh know. 08:58:27 -!- blabber has quit [Remote host closed the connection] 09:01:33 -!- edlothiol has joined ##crawl-dev 09:03:49 -!- raskol has quit [Read error: Connection reset by peer] 09:04:07 !tell bmh _actor_cloud_immune is used (indirectly) for travel, so returning true for 90% time is a bad idea. oops. (spam enough already?) 09:04:07 Keskitalo: OK, I'll let bmh know. 09:04:17 -!- raskol has joined ##crawl-dev 09:19:26 -!- bmh has joined ##crawl-dev 09:19:26 -!- bmh has quit [Changing host] 09:19:26 -!- bmh has joined ##crawl-dev 09:19:56 treasure trove portal doesn't accept what should be valid (requests +3 dragon armor, won't accept +5 dragon armor artefact) (https://crawl.develz.org/mantis/view.php?id=4756) by petzl 09:23:42 -!- Torokasi has quit [Ping timeout: 258 seconds] 09:25:23 -!- valrus has joined ##crawl-dev 09:37:19 Should troves accept items that are strictly better than those that it asks for? 09:37:20 bmh: You have 5 messages. Use !messages to read them. 09:38:39 zounds Keskitalo, enough messages? :) 09:39:59 hehe :) 09:40:08 I'm going to leave clouds alone for the travel reason 09:41:02 What should a rP- mutation actually do? 09:43:40 -!- Sequell has quit [Ping timeout: 260 seconds] 09:45:00 -!- Sequell has joined ##crawl-dev 09:46:27 -!- Textmode has joined ##crawl-dev 09:50:09 <|amethyst> what does poison vulnerability do to ants etc? 09:53:44 browsing the source quickly, it looks like it modifies poison beam damage 09:54:11 <|amethyst> ah, it also increases the damage from poison itself 09:54:24 <|amethyst> in apply_enchantment 09:54:41 That wouldn't make for a particularly interesting mutation. 09:56:10 <|amethyst> in addition to 1 + 50% chance of 1d(poisonval+1), you also get 2d(poisonval)-1 09:56:47 -!- valrus has quit [Remote host closed the connection] 09:56:51 <|amethyst> so about 5 times as much damage on average 09:57:11 ??rF 09:57:13 resistance[1/2]: Fire and cold deal 50%, 33%, 20% with positive resistance or 150% with negative resistance. Negative energy deals 66%, 33%, 0% with positive resistance. Acid deals 50% with rCorr. 09:57:22 ??resistance[2] 09:57:23 resistance[2/2]: Monster resistances. Fire and cold: negative: 1.5x melee, 2.0x ranged; +: 1/2x; ++: 1/5x; +++ or hellfire: 0x. Elec and poison are the same, except +: 1/3x, ++: 1/6x, +++: 0x. 09:57:57 |amethyst: turning poison into something hilariously deadly is probably a terrible idea. 09:59:28 <|amethyst> hm 10:01:19 <|amethyst> for level 1 poisoning you go from 0.75 damage on average to 1.75 10:02:49 <|amethyst> for level 2, from 1 damage to 4 10:03:03 <|amethyst> err, from 1 to 3 10:03:18 <|amethyst> yeah, that would be bad on low-level players 10:03:46 -!- raskol has quit [Read error: Connection reset by peer] 10:03:50 <|amethyst> but what's a low-level player doing with rP- ? 10:04:22 -!- raskol has joined ##crawl-dev 10:04:43 <|amethyst> It doesn't have to be a mutation, though there's been talk about a rP- enchantment-type effect, and it could be interesting on randarts 10:04:44 |amethyst: dying. swiftly. 10:05:12 <|amethyst> s/enchantment/enchantment or halo/ 10:05:17 -!- blabber has joined ##crawl-dev 10:05:22 rP- on players doesn't sound like something that can ever be fun or interesting, only frustrating 10:05:30 but that might be just me 10:05:54 <|amethyst> I thought one problem with poison now is that it is irrelevant past the early mid game 10:06:17 Snake can still poison you to death. 10:06:24 03dolorous * r0e360740f189 10/crawl-ref/source/dat/descript/monsters.txt: Remove a gender reference from the deep elf blademaster description. 10:06:28 <|amethyst> if there are creatures who give you temporary rP, around poisoners, that could make it more relevant 10:06:34 03edlothiol * rd264d1f036ed 10/crawl-ref/source/directn.cc: Fix the targeting beam not showing up in Tiles (#4754). 10:06:34 03edlothiol * r67b72ca73f68 10/crawl-ref/source/cio.cc: Show the targeting cursor in Webtiles. 10:06:37 -!- Wensley has joined ##crawl-dev 10:07:20 the problem with poison is that it's too relevant if you don't find rP and mostly irrelevant if you don't 10:07:34 Zaba: rP- as a halo effect might be non-frustrating, just make it remove rP+ and nothing else 10:08:09 err, irrelevant if you _do_, of course. 10:08:55 bmh, maybe 10:09:08 bmh: give it to green deaths 10:09:47 Eronarn: I enjoy killing players, but not that much. 10:09:55 or maybe a kind of imp, which they summon - 'sickly imp' 10:10:13 <|amethyst> you could make rP- on players just decrease the likelihood of reduce_poison_player 10:10:19 <|amethyst> in effect increasing the duration 10:10:45 <|amethyst> but, more importantly, cancelling rP+ 10:11:45 <|amethyst> hm... but that might be a little too much like multiple levels of rP 10:12:11 <|amethyst> "one level to prevent 90% of poison, two levels to prevent 90% of poison while a poison moth is in view" 10:13:01 <|amethyst> I guess the question there is whether it should be possible to overcome the temporary rP- enchantment with multiple rP+ items 10:13:47 Eronarn: dpeg had floated the idea of a mutation that randomly boosts or reduces magical abilities. Does this sound good or bad to you? 10:14:18 bmh: if it's just a per turn chance of switching, bad, because you can just not start fights until your magic is strong 10:14:58 i thought it was to reduce spell success but increase spell power 10:15:16 MarvinPA: oh. That's much simpler (and less scummy) 10:15:24 yeah, i kinda like it 10:15:28 that doesn't sound mutation-y tbh 10:15:35 <|amethyst> and the opposite mutation too? 10:15:36 seems like more of a staff effect 10:15:49 <|amethyst> or are we still talking random 10:15:52 i think it's much more interesting as something you can't just opt out of easily, though 10:21:56 -!- Pacra has joined ##crawl-dev 10:25:02 bmh: is the poison patch in? what's the percentage of resistance? 10:27:29 Wensley: it's written and on mantis. 90% 10:27:41 has anyone else looked at the hive portal vault patch? it seems kinda cool, bee larvae upgrade to killer bees by eating things and the queen sends bees berserk at you 10:27:49 bleh, I reiterate that 90% sounds like a completely arbitrary number :) 10:27:55 Wensley: I agree. 10:28:08 might be worth pushing and trying although i don't think the old hive vaults work very well for it 10:29:29 -!- Varsovie_crawl has left ##crawl-dev 10:29:35 -!- valrus has joined ##crawl-dev 10:30:23 buki (L1 MuNe) ASSERT(y >= 1 && y <= sz.y) in 'libutil.cc' at line 942 failed. (D:1) 10:30:26 buki (L1 MuNe) ASSERT(y >= 1 && y <= sz.y) in 'libutil.cc' at line 942 failed. (D:1) 10:30:28 buki (L1 MuNe) ASSERT(y >= 1 && y <= sz.y) in 'libutil.cc' at line 942 failed. (D:1) 10:30:33 buki (L1 MuNe) ASSERT(y >= 1 && y <= sz.y) in 'libutil.cc' at line 942 failed. (D:1) 10:30:39 buki (L1 MuNe) ASSERT(y >= 1 && y <= sz.y) in 'libutil.cc' at line 942 failed. (D:1) 10:31:30 buki (L1 MuNe) ASSERT(y >= 1 && y <= sz.y) in 'libutil.cc' at line 942 failed. (D:1) 10:31:32 buki (L1 MuNe) ASSERT(y >= 1 && y <= sz.y) in 'libutil.cc' at line 942 failed. (D:1) 10:31:33 buki (L1 MuNe) ASSERT(y >= 1 && y <= sz.y) in 'libutil.cc' at line 942 failed. (D:1) 10:33:12 Poison suspectibility could make the level of poisoning higher when the player gets poisoned, instead of directly increasing the damage (especially the immediate damage from poison arrow, curare or clouds). 10:33:30 That would lead to taking longer to get rid of the status by waiting, which would be a bit annoying. 10:34:28 i think level and duration of poison are tied together at the moment too, so it might not be easy to have same duration/stronger poison 10:34:53 unless of course bmh wants to rework it all :P 10:35:21 MarvinPA: oh sure, I don't have a job or anything :-P 10:35:26 * bmh starts work tomorrow 10:35:26 heh 10:35:52 And yeah, 90% is arbitrary. 10:36:08 Rather than a 90% cutoff I would prefer to use a function that accounts for regeneration rate 10:36:46 Hell, we could discard all the randomness involved in "A scorpion bit you, try to poison the player" and make every attack be a poisoning attempt 10:37:09 Then set rP0 to confer non-zero resistance and have a single consistent block of code instead of this spaghetti mishmash. 10:38:03 I might like that, but at the same time we'd have to reduce the severity of each poisoning attack, since early players would be annihilated by snakes 10:38:30 bmh: inconsistent with all other resistances, but actually sounds tempting. Would simplify stuff. 10:38:32 kilobyte: You have 1 message. Use !messages to read it. 10:38:57 on the other hand, sometimes it is fun to go toe-to-toe with an early snake and feel the tension of hoping you don't get poisoned 10:40:32 Wensley: he said "poisoning attempt", not "poisoning and then doing partial damage" 10:44:38 kilobyte: yeah, I agree with the inconsistent bit, but I've been on a push to make things in crawl more continuous (ex. skill levels) 10:44:47 <|amethyst> hm, but then you'd have the same chance of avoiding a snake's poison as a sea snake's 10:46:42 kilobyte: What if we split poison into two axes -- duration and potency. That way small snakes could never push you into super-red poison, but could extend the duration of such a poisoning 10:47:42 <|amethyst> hm, how would you specify that in an attack, though 10:47:56 <|amethyst> the duration I mean 10:48:12 standard duration, variable strength? 10:48:21 <|amethyst> hm 10:48:28 <|amethyst> that could work 10:48:33 -!- Torokasi has joined ##crawl-dev 10:48:53 <|amethyst> though I was thinking it would be neat to have early monsters with extremely low duration POISON_NASTY 10:49:05 <|amethyst> early-ish 10:49:24 <|amethyst> but a standard duration is much simpler 10:49:43 buki (L1 MuNe) ASSERT(y >= 1 && y <= sz.y) in 'libutil.cc' at line 942 failed. (D:1) 10:49:48 buki (L1 MuNe) ASSERT(y >= 1 && y <= sz.y) in 'libutil.cc' at line 942 failed. (D:1) 10:52:25 buki (L1 MuNe) ASSERT(y >= 1 && y <= sz.y) in 'libutil.cc' at line 942 failed. (D:1) 10:52:32 He's going to keep doing that, I imagine. 10:52:34 buki (L1 MuNe) ASSERT(y >= 1 && y <= sz.y) in 'libutil.cc' at line 942 failed. (D:1) 10:57:25 hmm, +10% power per level for the wild magic mut seems kinda low, even at level 3 that's less than a single enhancer 10:57:47 -!- ais523 has joined ##crawl-dev 10:57:53 might try it at 20% per level, since that's pre-stepdown also 10:59:21 wild magic? 10:59:26 yes 10:59:30 is it intended for ds or is it a general mutation? 10:59:34 sounds really cool 11:00:00 i think just as a general mutation, the idea is it's -spell success +spell power 11:00:31 very cool 11:00:34 then again DS could do with more double-edged muts... 11:00:41 would need to be more carefully balanced then, though 11:00:56 yeah i just wondered if it was aimed for ds since i've heard people talking about them needing more caster focused mutations 11:01:22 powered by pain was one such 11:01:59 there's some other interesting ideas on the wiki actually, one fairly nice one is a boost to spellpower the lower your hp is 11:02:28 MarvinPA: That sounds tedious. 11:02:38 buki (L1 MuNe) ASSERT(y >= 1 && y <= sz.y) in 'libutil.cc' at line 942 failed. (D:1) 11:03:11 <|amethyst> that sounds more like a PbP replacement 11:03:23 it was proposed as a PbP counterpart 11:03:30 "Spellpower boost based on the percentage of hp remaining as a counterpart to powered by pain – in this case taking damage increases the strength of your spells but doesn't guarantee that you have the mp to cast them." 11:03:46 -!- raskol has quit [Read error: Connection reset by peer] 11:03:50 might be cool with a similar effect for melee 11:04:30 having mutations that are too melee or spellcasting focused could prove an issue i think 11:04:33 * bmh watches buki play 11:04:42 could be extended to a general berserker-y thing maybe, yeah 11:04:43 remaining or lost? 11:04:55 PbP? 11:05:05 for the spell boost thingy 11:05:07 i guess the problem with that one is that it really is entirely useless for non-casters, PbP has a few uses at least 11:05:18 yeah 11:05:21 gs, hybrids 11:05:25 a boost to both damage and spell power? 11:05:34 oh, should be percentage lost i imagine, just looks like a typo 11:05:55 -!- blabber has quit [Ping timeout: 260 seconds] 11:06:33 FR: Allow spectators to view inventories and other informational screens during gameplay :) 11:06:51 -!- blabber has joined ##crawl-dev 11:07:13 bmh: we have that! next time you're spectating, use "m hey open your inventory" 11:07:22 :-P 11:08:28 bmh: imagine that would be a cool feature for webtiles especially 11:08:48 that's something I plan to implement :) 11:08:50 pity there isn't a trunk build on webtiles or i'd play more on it 11:08:59 edlothiol: cool! 11:09:01 yes, I think it might only work in webtiles :P 11:09:01 many other things to do before that, though 11:09:27 edlothiol: i stopped playing webtiles because i had stability and lag issues a while back 11:09:31 edlothiol: so how does one go about watching webtiles ttyrecs? 11:09:35 has any of that been adressed? 11:09:57 Wensley: currently, you can't. it will be possible in trunk/0.10 11:10:07 I see 11:11:02 Xiberia: that depends on what you mean by "a while back" 11:11:52 wtf? I just ran into a vampire on D:1 11:12:20 That's OOD with a vengeance. 11:12:23 edlothiol: 2-3 weeks i believe 11:14:06 not much change since then... are these lag issues only on webtiles? how's your ping to CDO? 11:14:43 yeah i play regularly on cdo console without any issues 11:15:03 i had really bad issues with firefox where it would just freeze up and i had to reload the page to play 11:15:17 then i switched to chome, which worked fine for a few weeks 11:15:31 now i freeze up on chrome as well however 11:15:51 freeze up during loading, or while playing? 11:15:58 while playing 11:16:08 the recurring pattern seems to be while aiming something 11:16:14 when the tile highlighting appears 11:16:38 then it just freezes 11:16:48 if i try to reload it usually just goes black 11:17:00 takes 4-5 tries befeore it refreshes properly 11:17:34 what's your username? 11:18:25 getserious mostly i believe 11:19:38 <|amethyst> !lg * place=d:1 killer=blue death -tv 11:19:38 4. Pancakes, XL1 OgCK, T:232 requested for FooTV. 11:19:46 <|amethyst> oops, wrong channel 11:19:51 -!- raskol has joined ##crawl-dev 11:22:16 Xiberia: It's very possible that the freezing bug is fixed 11:23:38 -!- monqy has joined ##crawl-dev 11:24:37 at least, according to the logs you haven't played since I've fixed a bug very similar to what you describe 11:24:40 edlothiol: cool 11:25:05 maybe i'll try playing a bit on webtiles again once i'm done with my current game 11:25:28 ok, let me know if it still freezes 11:27:33 will do 11:32:24 -!- jeanjacques_ has quit [] 11:32:48 -!- blabber has quit [Remote host closed the connection] 11:34:29 If we were to implement rP- aura, how would we convey this information to the player? 11:52:14 could have a new status light, "Frail" 11:53:35 buki (L1 MuNe) ASSERT(y >= 1 && y <= sz.y) in 'libutil.cc' at line 942 failed. (D:1) 11:59:40 bmh: Having had to look at the code :) I too think the idea of every hit poisoning, and rP0 really being a foo% chance to resist, would simplify things nicely. 12:00:18 03MarvinPA * rcede1ecd76ab 10/crawl-ref/source/ (5 files in 2 dirs): Remove the unused teleport-at-will and throw frost/flame mutations and their abilities 12:00:22 Keskitalo: what would we do about poison beams? 12:00:28 03MarvinPA * ra045af5444c9 10/crawl-ref/source/ (enum.h mutation-data.h spl-cast.cc): Wild magic mutation: each level decreases spell success but increases spellpower 12:00:30 <|amethyst> should the % depend on the strength of the poison? 12:00:42 MarvinPA: you++; 12:00:46 There's a rPois status light, -rPois status light would work. "Frail" conflicts with the mutation. 12:00:51 :) 12:01:10 bmh: You mean the direct damage from them? Same question is for the direct damage from poison clouds. 12:01:22 Keskitalo: nah, I mean the poison probability 12:02:09 Ah, they have 100% chance now, right. 12:02:42 (as do the clouds :)) 12:02:55 <|amethyst> as does POISON_NASTY I believe 12:03:44 -!- raskol has quit [Read error: Connection reset by peer] 12:03:58 Too many different systems all over the code. :P 12:03:59 -!- raskol has joined ##crawl-dev 12:04:26 Keskitalo: Perhaps we could have a grand unified theory of resistances! 12:04:35 one code path >> many code paths! 12:04:59 bmh: is >> > >? 12:05:23 Wensley: If we're talking exit codes, that's a 0. If we're talking binary it's a 1. 12:06:30 only if one code path is unsigned, though 12:06:56 bmh: I've put like week of coding into reducing level types from three to one. It turned out to be like an order of magnitude bigger project than I imagined, but the amount of silly hacks revealed/removed is just mad. 12:07:20 kilobyte: you mean removing multi-level resistances across the board? 12:07:22 kilobyte: Tasks like that always are :-\ 12:07:33 over 2000 lines of code removed -- and not in huge blocks but as single lines scattered all around 12:08:06 kilobyte: Ooh, that sounds awesome. 12:08:18 Wensley: no, BRANCH_*, LEVEL_*, portal vaults 12:08:26 buki (L1 MuNe) ASSERT(y >= 1 && y <= sz.y) in 'libutil.cc' at line 942 failed. (D:1) 12:08:35 dammit, I wish he would stop crashing. 12:08:36 oh wow, awesome 12:08:51 what's that assert, did his entry vault spawn outside the level boundaries? 12:08:57 kilobyte: speaking of portal vaults -- how much work is it to add a new barebones portal? 12:09:28 bmh: A new portal vault? Just a des file. 12:09:47 Keskitalo: I want to make a portal with chaos clouds instead of walls. 12:10:01 oh, that's just a bunch of lua -- duh :) 12:10:03 bmh: what if every wall in the abyss was a chaos cloud 12:10:10 until kilobyte implements his plan of killing level_type dead, at which point we'll have to declare each portal vault as a branch :P 12:10:17 Wensley: The thought has crossed my mind 12:10:33 would make tornado an interesting abyss strategy 12:10:37 Wensley: I can make an abyss vault like that. 12:10:44 make it huge! 12:11:00 Zaba: yeah, you'll have to add a BRANCH_ enum and an entry in branch-data.h 12:11:58 recompile cycles when tweaking portal vaults or indeed adding new ones is something we're missing a lot right now ;p 12:11:59 (How dangerous are chaos clouds?) 12:12:30 Zaba: on the other hand, you can refer to the branch in C++ code at your leisure, and can do portal vaults with not a single line of code in either C++ or lua other than the enum+entry. Just DEPTH:Bedlam and you're set. 12:14:02 Zaba: right, I admit, you'll have a whole compilation cycle after adding a new portal vault branch. Which you do once-twice per year, while one in two commits already costs that. 12:14:50 tweaking vault layouts wouldn't require a whole recompile, would it? 12:15:05 Wensley, fortunately, no 12:15:15 phew 12:15:25 Wensley: lua layouts not at all, C++ layouts -- a single file + link unless you change headers 12:15:41 kilobyte, I still wonder what you'd set the branch to for ad-hoc portal vaults created using &P 12:15:49 those that don't normally generate in a branch of their own, like 'layout' 12:16:01 which still make sense together with &P for testing purposes 12:16:35 &P _already_ requires you to answer two questions, one for the branch, one for map 12:16:51 well, no, it doesn't ask for the branch, but a tag 12:17:00 which need not correspond to a branch 12:17:02 "Destination for portal (defaults to 'bazaar')?" 12:17:16 then it asks for a bazaar map 12:17:16 yes, and I can enter 'layout' which is not a branch 12:17:56 Zaba: heh, just tried it -> crash 12:18:06 Wensley: CLOUD_SILENCE? :) 12:18:09 I do it all the time 12:18:33 no crash. 12:18:37 bmh: ooh, that sounds interesting 12:18:42 choking clouds! 12:18:53 ERROR in 'dungeon.cc' at line 2545: No maps or tags named 'layout_delve', and no backup either. 12:18:57 or, asphyx clouds :P 12:19:15 Wensley: like calcifying dust? :) 12:19:16 kilobyte, you enter just 'layout' on the first prompt 12:19:20 exactly 12:19:42 kilobyte, because that's the tag 12:19:45 -!- valrus has quit [Remote host closed the connection] 12:20:17 03dolorous * r5caa1a9db473 10/crawl-ref/source/abl-show.cc: Fix indentation. 12:20:41 and jeez, 'no maps or tags named foobar' is such a descriptive error message 12:21:06 Zaba: there exists a map by that name... 12:21:22 it never really tries to pick a vault by name, only by tag 12:21:30 ow -- evilmike is evil -- evilmike_abyss_exit_mutagenic 12:21:39 it prompts for a name later -- only in wizard games 12:21:49 it does that for naturally occuring portals, too 12:22:22 all &P does is create a portal that points to the tag you enter at the prompt 12:23:13 <|amethyst> hm, and inside a layout portal, downstairs go to other layouts, while upstairs and all hatches go back to the portal 12:23:16 in other words, the other functionality that I missed is being able to ask for a tag the next time a level is generated 12:23:39 -!- dtsund has joined ##crawl-dev 12:23:48 |amethyst, yeah, stair behaviour is weird that way 12:24:08 (with a side effect of silly stair behaviour (not nice) and no monster generation (nice)) 12:25:23 I guess it can be replicated as a variant of &^R -- regenerate the level you're on but asking for a tag instead. So you'd be able to do that both in a portal vault, for the same behaviour as currently, or on a normal level. 12:26:40 is there some other functionality of &P I missed? 12:35:38 oh, case in point. Be in Pan, enter a portal vault, try to exit. You land on D:1 with "Ouch, no portal vault exit point!" 12:36:43 kilobyte, I understand the motivation behind killing level_type and indeed support it, the only thing I don't like is the inflexibility with branches that it introduces 12:38:50 it just would be nice if we didn't have to sacrifice anything to get rid of that mess 12:40:36 oh bleh -- bug! iron grates block clouds 12:42:28 buki (L1 MuNe) ASSERT(y >= 1 && y <= sz.y) in 'libutil.cc' at line 942 failed. (D:1) 12:43:01 Zaba: I wouldn't want to kill any functionality, of course... but I don't understand what flexibility you're talking about 12:44:42 it's the same flexibility that lets us redefine properties of monsters in vaults instead of hardcoding every monster definition, except applied to branches 12:44:55 the only thing I see is adding a new portal vault _type_ (like, adding, say, "Ruins" rather than a new map for an existing portal vault), which requires a recompile... which is damn minor 12:45:22 just like for monsters, you can redefine a few properties but face a brick wall if you want to change something else 12:45:51 -!- Torokasi has quit [Ping timeout: 248 seconds] 12:46:12 Iron grates block spreading clouds (https://crawl.develz.org/mantis/view.php?id=4757) by brendan 12:46:42 kilobyte, indeed. It would be nicer if such redefined monsters were first-class. 12:46:58 -!- evilmike has joined ##crawl-dev 12:47:24 -!- bmh has quit [Quit: bmh] 12:47:28 likewise it'd be nice if portal vault "branches" as we have them now were first-class without hardcoding :P 12:48:50 I've been convinced that *that* would probably be overkill at this point 12:49:15 there are so many more kinds of monsters, you know? 12:50:08 is there some flavour reason that nessos has animate dead? 12:50:09 buki (L1 MuNe) ASSERT(y >= 1 && y <= sz.y) in 'libutil.cc' at line 942 failed. (D:1) 12:50:16 buki (L1 MuNe) ASSERT(y >= 1 && y <= sz.y) in 'libutil.cc' at line 942 failed. (D:1) 12:50:18 it seems really weird and unfitting and he barely ever gets to use it anyway 12:50:29 MarvinPA: none I can think of 12:50:34 that is really weird. also, crawl's nessos isn't even that accurate to the story 12:50:36 it's sort of weird, but also hilarious 12:50:47 buki (L1 MuNe) ASSERT(y >= 1 && y <= sz.y) in 'libutil.cc' at line 942 failed. (D:1) 12:50:49 the mythological one had toxic blood, nessos should by all rights drop a hcl corpse :P 12:50:56 because you never expect it, and it's so out of the blue when it happens 12:51:14 evilmike: let's give all centaurs hcl 12:51:19 well when it does happen it just means he's blocking off his line of fire to you :P 12:51:21 Sounds like someone's resizing his terminal mid-game 12:51:27 Zaba: having no first and second class branches is exactly my point... together with removing code duplication 12:51:32 -!- bmh has joined ##crawl-dev 12:51:32 -!- bmh has quit [Changing host] 12:51:32 -!- bmh has joined ##crawl-dev 12:51:45 -!- bmh has quit [Client Quit] 12:51:52 dtsund: no, he can't even get past the char select screen, and deleting his save doesn't help 12:52:01 Zaba: Ziggurats for example instead of calling a function to set depth, hold it itself which causes no end of problems 12:52:12 kilobyte, I understand, that is why I'm ultimately in support of the plan 12:52:19 Hm. Is he playing on a nonstandard terminal size? 12:52:26 Wensley: maybe he's confused his shell about the size of his terminal! 12:52:41 having all branches first-class, even without being able to create them dynamically, is better than having two classes of them 12:52:43 SamB: but it only crashes when he plays trunk, he can play older versions just fine 12:52:57 so unless there's some new assert for min termsize 12:53:02 Zaba: heck, we could even have BRANCH_FIRST_LUA 12:53:30 Zaba: except that I very recently removed the same thing for colours due to bugs and useless complexity :p 12:53:41 buki (L1 OpVM) ASSERT(y >= 1 && y <= sz.y) in 'libutil.cc' at line 942 failed. (D:1) 12:53:42 kilobyte: well, colors are different 12:53:47 there weren't many free 12:54:37 kilobyte, it's more of a problem for monster redefinitions, because making _all_ the vault-custom ones first-class by hardcoding them would be madness 12:54:52 even though that's what's been happening lately 12:55:33 if a vault monster is interesting, I think it makes sense to upgrade it to first-class to be used elsewhere :D 12:55:52 Wensley, yes 12:55:57 Wensley: MEGABATS 12:55:59 if it's interesting outside the vault 12:56:01 <|amethyst> in an ideal world you could create first-class monsters without recompiling, but I'm not sure how that would interact with saves 12:56:39 |amethyst, pretty badly, I imagine 12:56:50 |amethyst: a monster isn't really interesting if it has no custom code 12:56:54 <|amethyst> yeah, unless you serialize the monster definitions too 12:56:57 but then, having different classes of monsters might end up interacting pretty badly with one's sanity 12:56:58 <|amethyst> kilobyte: lua I was thinking 12:57:10 Wensley: there are a lot of vault monsters which are basically quasi-uniques.. can't see how those would benefit from being used outside their vaults 12:57:22 <|amethyst> but there aren't the hooks for that now I guess 12:57:25 evilmike: killtracker for example 12:57:38 oh, that 12:58:24 evilmike, the problem is that vault-redefined monsters are a hack, they don't really fit into the system 12:59:58 from that stem all the issues, which get fixed for the most common cases, but which are probably lurking in huge numbers for less common ones 13:01:26 Zaba: to be fair, such issues would probably still exist even if such monsters were not a hack! 13:01:50 that makes sense. I recall the madness with corpse names (gnoll corpse of gnoll shaman) 13:01:51 SamB, yes, but their cause wouldn't be the fact those monsters are a hack 13:02:07 so basically this would lose us a scapegoat 13:02:25 it would lose us a source of bugs 13:02:57 do you have any idea what that would do to our quotas? 13:03:12 no 13:03:43 -!- raskol has quit [Read error: Connection reset by peer] 13:03:46 well, it would make it a lot harder to hit them, that's what! 13:04:01 but it'd also reduce the need for hitting them 13:04:03 -!- raskol has joined ##crawl-dev 13:04:15 * SamB doesn't think Zaba gets the joke 13:04:43 I deferred getting the joke for a time when I am less sleepy :P 13:17:09 -!- dpeg has joined ##crawl-dev 13:17:50 is there any point in troves requiring the item to be identified? 13:17:56 yes 13:18:05 otherwise you can use the trove to id items 13:18:07 the only result I see is a very confusing message 13:18:43 id and lose 13:19:12 dpeg: is that really so much of an issue? 13:20:00 items that could be possibly unidentified: potions of healing, heal wounds, cure mut, scrolls of blinking, acq and id, wands of hasting and healing 13:20:53 practically, cure mut is the only one of those that might plausibly be unIDed past the early game, if the player owned it 13:21:07 unless players were deliberately trying not to ID things 13:22:52 hmm, actually, it might happen for potions on a mummy or wands on a felid 13:27:21 but should that be penalized? 13:27:23 troves that demand potions on mummies are pretty amusing... healing or heal wounds are pretty easy to id though, since they're common. Usually it's just a free trove 13:27:38 unless you're a mummy of nemelex 13:29:02 maybe you should just have to fight some rats if you guess wrong 13:29:20 it worked for LttP! 13:29:37 had to look for a wand of hasting after having sacced everything... 13:29:38 !lg . win race=fe god=nemelex xobeh 13:29:39 1. KiloByte the Eviscerator (L27 FePa), worshipper of Nemelex Xobeh, escaped with the Orb and 3 runes on 2011-02-21, with 1311626 points after 167011 turns and 17:42:01. 13:30:53 kilobyte: well, random shop items could have unidentified duplicates 13:31:24 but I agree with you that there isn't much point in requiring identification 13:31:31 elliptic: just checked the log... I remembered wrong, the wand I finally found came identified from a bazaar 13:32:01 currently wand troves always ask for a highly charged one 13:32:17 03MarvinPA * r20382255809c 10/crawl-ref/source/mon-spll.h: Adjust a lot of emergency spells 13:32:17 03MarvinPA * r80109d8837f4 10/crawl-ref/source/mon-cast.cc: Don't let monsters cast Regen while under DDoor 13:32:17 03MarvinPA * r3d2659eb4912 10/crawl-ref/source/mon-cast.cc: Allow monsters to cast their emergency spells at a higher HP threshold, but have them choose it a bit less often 13:32:18 03MarvinPA * rf14abfc00503 10/crawl-ref/source/mon-spll.h: Remove Animate Dead from Nessos 13:32:18 03MarvinPA * r34b472649b53 10/crawl-ref/source/mon-spll.h: Give Mnoleg Malign Gateway 13:32:21 and it is highly unlikely for one to start that highly charged 13:32:27 oh right 13:32:40 you could recharge everything but that has pretty little point 13:33:01 so this is mainly relevant for a mummy who brings potions to the trove 13:33:32 back, children in bed now 13:33:34 I care mostly about the misleading message, though 13:33:40 honestly it would be good if certain trove options didn't happen for certain races anyway 13:33:46 yeah, the message is awkward 13:34:40 MarvinPA: good changes, thanks. Also for wild magic, didn't expect that one at all :) 13:34:52 too bad, trove code is in lua, which makes it a bitch to run such checks :p 13:35:15 dpeg: was nice and easy to implement and i liked the idea, so why not :) 13:37:00 elliptic, kilobyte: feel free to change troves. Just wanted to say that we had some reason for the current status. Doesn't mean it was a very good reason, of course. 13:37:46 ah, I think the point was that you can recharge a wand, try it, recharge some more etc. 13:39:21 not really compelling, though 13:39:44 mutations that sound good to me: fumes (the fog variant), -acc, hydrophobia (slow drinking), tripping, extra legs for Ce, +healing from potions, mana regen, and, blowing my own horn, evolution 13:39:55 Whoo no more teleport self on Asmodeus.. although, that did make it easier to grab the rune. ;) 13:40:36 Keskitalo: sound all good to me. What about casting from HP? 13:41:00 speaking of slow drinking, would it be worthwhile to change blurry vision to a mutation that makes you read slowly? 13:41:10 failing to read scrolls is amusing but it tends to just be extra keystrokes for a lot of stuff 13:41:38 Extra legs could resist trample and give a higher chance of unarmed attacks! 13:42:02 I wonder if 3d2659 makes bears berserk more often (which would be good). 13:42:16 yeah I asked about that too 13:42:21 in ##crawl though) 13:42:22 ( 13:42:52 the emergency spell threshold was changed to 1/3 hp instead of 1/4. So, it will, right? Or is there something really weird about bears? 13:42:55 Keskitalo <--> kilobyte 13:43:11 evilmike: yeah but with a halved chance 13:43:17 evilmike: sounds good to me 13:43:18 oh, I see 13:43:42 -!- Zaba has quit [Ping timeout: 260 seconds] 13:44:09 evilmike: failing to read lets you decide whether you want to retry, which might be important tactically 13:44:14 kilobyte: I promised to put approved-of mutations to the top. I'll use your list? 13:44:40 Maybe they should just have a chance of berserking whenever they take damage. 13:44:42 but yeah, it might be simpler to take more time 13:45:19 dpeg: I can't tell you which ones you like 13:45:23 Any opposition to "cast partially from HP" mutation? 13:45:48 kilobyte: "approved of" = by us, not just by me 13:45:52 would be brutal on a deep dwarf 13:46:51 i'm not opposed to it necessarily, although it reminds me i'd quite like to have makhleb invocations be cast from hp instead of mp (and then remove his mp-on-kills, and maaaaybe increase his hp-on-kills a bit if necessary) 13:47:09 it'd fit with his new bloodshed/mortification/etc theme! 13:47:56 and also means he could lose the mp-on-kills but still be pretty good for casters since using invocations would leave them with all their mp to cast with still 13:48:07 MarvinPA: good points 13:48:09 MarvinPA: it would make him mostly worthless on most chars 13:48:21 how so? 13:48:28 what I like about the mutation is that it's good or bad depending a lot on the circumstances 13:48:41 meleers only... no much point on conjurers 13:49:00 I'd instead get rid of Vehumet's MP for kills for, well, being redundant with Makhleb! 13:49:03 well like i said, they'd be able to conjure just fine and would also have an extra pool from which they can use invoc 13:49:28 Vehumet could get a stronger discount 13:49:34 perhaps even a lot stronger 13:50:12 mp fits veh a lot better imo, and makhleb is already by far and away the best conjurer god, a little less appeal for them seem fine to me 13:50:30