00:38:17 <06m​umra> Making the border at least non-rectangular seems an idea worth immediately trying (and can possibly be done according to the existing depth numbers) 00:45:21 <06m​umra> though i personally find the shoals colour palette quite jarring and would love to see more of a "torch-lit night beach" ambience (being darker doesn't necessarily mean less colourful...) 00:50:09 <06m​umra> where shoals seems a bit incongruent is that it feels like you have portalled to another place whereas all other branches are firmly part of the dungeon structure (except those that are specifically portals, ok and swamp as well sort of, but it's much easier to imagine that there are just some swampy dungeon floors vs an entire ocean with stairs leading to another ocean on the floor below somehow ... yes we can handwave that "a wizard 00:50:09 did it" but it still defies the internal logic of magic seen elsewhere) 00:58:56 -!- indigaz22 is now known as indigaz2 01:12:00 <02D​arby> one caveat I looked into after bringing it up was that endless sea walls do, in fact, currently block LoF, though that can be changed, and matters little if the shape is fully convex 01:22:44 <06m​umra> https://cdn.discordapp.com/attachments/747522859361894521/1423586052718989402/image.png?ex=68e0d953&is=68df87d3&hm=be633271aa52a6f66a96e3e385109de5e900d1ddaaaef77ca59bdb9353532558& 01:23:14 <06m​umra> making open sea below depth -100 makes this kind of effect, the LoF blocking is mostly not a problem but there are occasional spots 01:23:34 <02D​arby> I like it, though probably it shouldn't reach the shallow water? still not sure if that's just me though 01:23:59 <06m​umra> i think that's happening near the map boundaries 01:24:30 <02D​arby> makes sense if so 01:25:29 <06m​umra> one alternative would be to change the feature to "impenetrable fog" and make it block LoS as well 01:27:03 <02D​arby> probably not the alternative I'd pick but I've no idea if it blocking movement but not LoF is significantly harder than it sounds (pretty believable it might be) 01:27:35 <06m​umra> or make it "enormous waves" to explain why you can't shoot thru it (allowing LoF is probably really problematic from what i know about LoF code) 01:27:46 <02D​arby> true 01:28:15 <02D​arby> waves might fit 01:28:52 <02D​arby> re: incongruence, that's an undeniable fact (I had to make another grass palette mainly for shoals). I'm somewhat fond of the way it stands out from the rest of the game, feeling that exceptions to rules are the most important part, though I'll not deny having thought about how I'd retile the floor and rock if it came to it. but I fear even if were done exceptionally well, it would be a controversial change by nature. 01:29:53 <02D​arby> but then I admit I thought I'd get worse for throwing a hundred flowers into Lair 01:30:07 <06m​umra> i am slightly wondering if the "players like shoals because it's bright" thing is actually a "most of the dungeon is too dark on some monitors" problem 01:30:47 <06m​umra> i do find the brightness makes it hard to make out characters and items over the top of the floor tiles, whereas the darker tiles in the rest of the game let things stand out properly 01:31:14 <06m​umra> really all the tilesets should be balanced to a similar brightness/contrast so that alpha or monitor brightness can be applied to the player's taste 01:32:05 <02D​arby> many of the game elements are definitely made to contrast best over much darker tiles, is true 02:47:38 <04d​racoomega> For what it's worth, while the 'invisible wall' of the endless sea is rather gamey, I feel like making them still be 'walls' that are now also irregular with regard to line of fire is just worse - like, the momentary 'organicness' of their appearance is dwarfed by how jarring it is to now have invisible walls that are irregular. imo, there's just a certain degree of "This is actually a video game" that one sometimes has to accept 03:08:26 03SentientSupper02 07https://github.com/crawl/crawl/pull/4751 * 0.34-a0-775-g1aaafc8384: Moth: 3.0 You Can (Not) Powercreep 10(74 seconds ago, 24 files, 269+ 225-) 13https://github.com/crawl/crawl/commit/1aaafc83847d 03:09:30 03SentientSupper02 07https://github.com/crawl/crawl/pull/4751 * 0.34-a0-776-gb636df3df9: Unbrace 10(50 seconds ago, 1 file, 2+ 0-) 13https://github.com/crawl/crawl/commit/b636df3df928 03:22:20 03SentientSupper02 07https://github.com/crawl/crawl/pull/4751 * 0.34-a0-776-g45d42ba4bb: Unbrace 10(14 minutes ago, 2 files, 6+ 0-) 13https://github.com/crawl/crawl/commit/45d42ba4bbd2 03:33:45 Experimental (bcrawl) branch on underhound.eu updated to: 0.23-a0-5261-gd9800d219b 03:48:54 03SentientSupper02 07https://github.com/crawl/crawl/pull/4751 * 0.34-a0-776-gc390d0ac26: Reformat attempt_tabcast_spell 10(40 minutes ago, 2 files, 9+ 11-) 13https://github.com/crawl/crawl/commit/c390d0ac26f3 03:54:14 03SentientSupper02 07https://github.com/crawl/crawl/pull/4751 * 0.34-a0-777-gafb2fc5c75: Use calc_spell_range over spell_range 10(28 seconds ago, 1 file, 2+ 2-) 13https://github.com/crawl/crawl/commit/afb2fc5c75ce 04:17:58 <06m​umra> i wonder if the moth concept might not be better as a talisman rather than species? 04:52:43 -!- Changesite1 is now known as Changesite 05:11:30 Unstable branch on crawl.akrasiac.org updated to: 0.34-a0-901-g85a6fcb (34) 06:05:11 03SentientSupper02 07https://github.com/crawl/crawl/pull/4751 * 0.34-a0-778-g8dcf67588e: Don't tabcast when miscasting 10(29 seconds ago, 1 file, 2+ 1-) 13https://github.com/crawl/crawl/commit/8dcf67588e25 08:34:28 <08n​icolae> as the designer of desolation_temple, i gotta admit, i'm not entirely sure i still like the way i did the impassible terrain on the edges, with Undulations and Waves instead of just being straight lines. but it's been around for ages at this point and i've never seen any complaints, so, maybe it's fine? as for shoals, i do like the bright palette as a contrast to the rest of the dungeon, despite the odd flavor mismatch of stacks of 08:34:29 islands in a dungeon. also, whoever put "sun demons" and "sun moths" in shoals just because of the beach theme is no doubt some kind of psychopath. 08:35:20 <08n​icolae> yeah, this is part of what i think is weird re desolation_temple, you can stand on one side of a... impassable wall bulge? and see to the other side of it but you can't shoot through it 08:35:21 <02D​arby> yeah, I'm more than fine with dropping it after I tested and found it blocked LoF, that made it too significant to me 08:35:28 <08n​icolae> ah, yeah 08:36:04 <08n​icolae> as i've mentioned before, i have... a lot going on, sadly, and might be out of serious vault efforts until November, but if one of the other devs with vault skills wants to straighten out the walls in desolation_temple, you have my blessing 08:36:24 <02D​arby> I was originally of the impression it blocked LoS only and liked the more natural appearance and very mild usability 08:36:30 <02D​arby> but yeah, could do that 08:38:22 <08n​icolae> i like the flavor of the impassable seas in shoals, for what it's worth. given that the dungeons are supposedly created by the insane being known as Zot, i think it's okay if we paste over some flavor disconnects with "yeah, that's weird, huh?" 08:39:12 <02D​arby> yeah same, it's weird in a good way, makes the place memorable 10:24:50 <06m​umra> so many questions is the ocean infinite? what holds the ceiling up? (presumably, infinite magical power would be required to hold up an infinite ceiling) 10:25:21 <02D​arby> probably not literally infinite, just bigger than a level can hold 11:08:23 Could make the stairs portals of some kind, implying one large ocean plane not a stacked set (making the NetHack player in me ask what happens if I dig down from shoals:1, have I just pulled the plug out the bath?) 11:13:05 <02D​arby> that's another for my list of "I hypothetically prefer it from a flavor perspective, but expect others would consider it unnecessary complexity from a gameplay perspective, and therefore have not brought it up" 11:14:40 <02D​arby> (mostly with regard to formicids having an implied exception for the branch) 11:16:48 <02D​arby> (I would not hesitate to make magic portalstairs if it turned out to be an unexpected uncontroversial point though) 11:19:28 <06p​leasingfungus> yeah, it's a natural thought, but... 11:25:12 <06d​olorous_84348> "One of the reasons merfolk water magic was mostly lost is because of how it can go wrong. There are legends of a dungeon where aquamancers tapped into an extradimensional ocean plane and somehow covered several floors in the dungeon with infinite seas. They turned it into a beach where they could live, but there was no putting back the ocean..." 11:36:04 <06m​umra> Also ignis, also probably a bunch of other little wording details all over the place that assume the structure of the dungeon 11:36:45 <02D​arby> yeah, mostly a case of "too sad if certain things just Don't Work here" 11:38:21 <06m​umra> Escape hatches would also need to be flavoured as "also a portal, but a randomised one". Basically things will just keep getting more awkward, 11:58:35 <09h​ellmonk> We have no choice but to remove shoals for the sake of structural integrity of the dungeon 12:39:41 <06m​umra> Was thinking about how/where you would drain all that water. Then it hit me ... when the dungeon generates with Swamp, that's because they tried to drain Shoals 12:58:58 <12g​e0ff> @hellmonk, re:anti-luring, fr: add an annotation when a unique returns back home, so the player could see the updated location of the unique on the Ctrl-o screen (instead of manually checking the notes to see what was the unique's home floor) 12:59:05 <12g​e0ff> (also, the last gohome commit's updated the contrib submodules) 15:12:03 03dolorous02 07* 0.34-a0-902-ge5e6b2775c: Disallow acquiring foul flame brand if holy. 10(9 minutes ago, 1 file, 3+ 0-) 13https://github.com/crawl/crawl/commit/e5e6b2775c30 19:23:18 03SentientSupper02 07https://github.com/crawl/crawl/pull/4751 * 0.34-a0-779-g0cf2fa2676: Attempt to fix a build testfail 10(47 seconds ago, 1 file, 2+ 2-) 13https://github.com/crawl/crawl/commit/0cf2fa26763a 20:28:15 03Implojin02 07* 0.34-a0-903-g3e737eca7e: Give baubles a console glyph 10(18 minutes ago, 6 files, 9+ 5-) 13https://github.com/crawl/crawl/commit/3e737eca7e42 22:04:34 <04d​racoomega> So, it turns out that I'm pretty sure the numbers with give for damage rating are a bit weird and wrong (imo). It uses the maximum value for weapon base damage and enchant/slaying, but the average value for the bonus from skills (which is rolled randomly on each attack). This means that the number it gives is neither your average damage or your maximum damage (the latter of which monster damage on xv - the semantically closest 22:04:34 other thing we show - tries to be). Instead, it's some weird hybrid. This is arguably even weirder if you have negative slaying, since it now gives your lowest roll for slaying instead of your highest, which is virtually a 4th thing that is neither average nor maximum damage. I had been thinking of changing it to just be maximum damage (for parity with xv), but realize that doing so creates a different source of potentially misleading information 22:04:35 in the complete breakdown. If we look at an example string of: Damage rating: 42 (Base 17 x 137% (Dex) x 147% (Skill) + 8 (Ench)) the percentage bonus from stats is never randomized, but the bonus from skill is randomized. What is currently shown is the average, but each hit (in this example) gets a bonus that varies from x100% - x194%. If I made damage rating properly be 'maximum', that would make the skill bonus there be 194%, which could give the 22:04:35 impression of skills having an outsized effect compared to stats (ie: players might think they were twice as good as they actually are. Of course, at the moment players might instead think they are worse than they actually are....) Average damage might be less misleading in that regard, though of course it means lowering the scale of numbers shown for damage rating a lot. But also might be confusing if a +8 weapon says that the bonus from its 22:04:36 enchantment is +4. Or would that just be educational? (Maybe rename 'Damage Rating' to 'Average Damage' or something?) 22:04:48 <04d​racoomega> Kind of felt I should ask for others' thoughts on this 22:06:52 <06p​leasingfungus> hello! 22:07:26 <06p​leasingfungus> it's me! the damage rating guy! 22:08:02 <06p​leasingfungus> > But also might be confusing if a +8 weapon says that the bonus from its enchantment is +4. This is close to the main reason it uses something like 'max' instead of 'average' - to make enchantment line up nicely. 22:08:41 <06p​leasingfungus> > Maybe rename 'Damage Rating' to 'Average Damage' or something? I really would not want to commit to anything like that. I specifically named it 'damage rating', not 'max damage', because I did not want to commit to anything nearly that specific in Crawl's nightmare labyrinth of a combat damage system. 22:08:53 <06p​leasingfungus> damage ratings are only intended to be, very vaguely, comparable to other damage ratings. 22:09:03 <06p​leasingfungus> they're a way to vaguely compare weapons against each other. 22:09:32 <06p​leasingfungus> they're not a way to see how much damage you will deal, either on average or at max. trying to accomplish that seems really hard. (though correct me if I'm wrong!) 22:10:18 <06p​leasingfungus> > a bit weird and wrong I'd say they're weird, but not wrong. IMO! 🙂 22:11:39 <04d​racoomega> I actually think what we have right now is actually very close to average, except for the part where weapon base damage and slaying both use their full value instead of half. (Since we're already abstracting away AC and brand and all that) 22:12:34 <04d​racoomega> It's the mix of using 'full value' for some parts and 'half value' for other parts that I think leaves the results a little misleading. (Certainly I didn't realize it worked that way until yesterday) 22:12:41 <06p​leasingfungus> iirc the repeated/multiplied rolls also mean that average is way below half of max, which could be confusing to people 22:12:50 <06p​leasingfungus> > a little misleading in what way? 22:13:52 <06p​leasingfungus> i really don't feel strongly at all about the original question of skill bonus, honestly. i'd be totally fine with doubling it 22:14:18 <06p​leasingfungus> i feel more strongly about keeping 'damage rating' decoupled from 'max/avg damage', until or unless crawl combat is drastically simplified 🙂 22:18:51 <04d​racoomega> I guess I myself had assumed (and I had seen other people assume) that damage ratings represented (approximately) the max damage the weapon could deal (ignoring brand). And when I realized it was neither average nor maximum, but valuing different parts of the equation somewhat differently, that felt odd to me. I genuinely hadn't realized you could roll a number considerably higher than your damage rating. Like, my understanding is 22:18:51 that it was using the term 'damage rating' to save it from having to ensure that it accurately accounted for every possible combat effect (which is fair, of course). Though I'd also been planning to do some refactoring on the subject to separate temporary and permanent sources of slaying, since currently damage rating includes a bunch of temporary things while excluding a bunch of other temporary things (and it felt more consistent to me to not 22:18:55 include transient things at all?) 22:20:48 <04d​racoomega> But also, I'm pretty sure that the way the math works right now, it's possible for a weapon that will deal more average damage to have a lower damage rating than one that will deal less (though I'm not sure how plausible that specific combination of stats/skills/slaying between the two would be to arise in a real game. Plausibly not to a large degree; I didn't do math on that part yet.) 22:26:17 <06p​leasingfungus> mm, that would be misleading, i agree 🙂 22:27:01 <06m​umra> One thing I always wish I could see is damage per AUT 22:27:26 <02D​arby> damage rating was once closer to that, if I recall! 22:27:41 <06p​leasingfungus> yeah, it showed that at one point, but AC just makes it way too misleading 22:27:46 <06p​leasingfungus> qbl damage looks bonkers 22:28:22 <06p​leasingfungus> anyway, i'm digging into this. suspect the skill thing may have been me misreading maybe_random 22:28:29 <06p​leasingfungus> i never remember what that function actually does with random = false 22:29:00 <06p​leasingfungus> doesn't help that maybe_random2 and maybe_random_div are very different, iirc 22:29:28 <04d​racoomega> I think they are, yeah? 22:30:12 <04d​racoomega> (maybe_random2 is at least as straightforward as returning the strict mean if random is false) 22:31:19 <06p​leasingfungus> if (random_factor) return random2(nom + denom) / denom; else return nom / 2 / denom; 22:32:03 <06p​leasingfungus> I don't really know what one is supposed to do with this function. 22:32:25 <06p​leasingfungus> (but i'm getting a bit off-track, i suppose...) 22:33:57 <06p​leasingfungus> anyway, changing the skill part seems fine, and if we feel like we can make this actually accurate as a max damage thing, i'm not fundamentally opposed to renaming it. it seems hard to me, but you've done a lot of hard things 🙂 22:33:58 <04d​racoomega> Yeah, that one is more confusing 22:34:06 <06p​leasingfungus> just trying to give some context on the current state 22:34:42 <04d​racoomega> I do hope my wording didn't come off as trashing it or anything! I was just surprised when a related thing had me dive into the guts of it and I noticed what looked like an inconsistency to me. 22:34:54 <04d​racoomega> I did actually review some of the history of what was included and eventually removed from it 22:35:11 <04d​racoomega> (A very understandable history, really) 22:35:29 Unstable branch on crawl.develz.org updated to: 0.34-a0-903-g3e737eca7e (34) 22:36:32 <06p​leasingfungus> oh yeah, i wasn't offended or anything 22:36:43 <06p​leasingfungus> damage rating is my baby, but is an ugly baby 😛 22:36:54 <06p​leasingfungus> just trying to help 22:37:01 <06p​leasingfungus> %git 08012983ebf 22:37:02 <04C​erebot> kilobyte * 0.10-a0-762-g08012983eb: Give you.skill() a "scale" argument for partial skills, make callers use it. (14 years ago, 41 files, 306+ 305-) https://github.com/crawl/crawl/commit/08012983ebf1 22:37:12 <06p​leasingfungus> - your_to_hit += maybe_random2(1 + you.skill(SK_FIGHTING), random_factor); + your_to_hit += maybe_random_div(you.skill(SK_FIGHTING, 100), 100, random_factor); these are some strange diffs 22:39:14 <04d​racoomega> random_factor seems to be a confusingly named argument for 'calculate randomly or not' 22:39:22 <06p​leasingfungus> yeah that's one of the things i dislike here 😛 22:39:31 <04d​racoomega> I would have expected it to be, like, a number >.> 22:39:31 <06p​leasingfungus> i would expect a 'factor' to be a number, not a bool 22:40:01 <04d​racoomega> Well, fortunately I am pretty sure we don't use that word in any related code anymore 22:40:11 <04d​racoomega> Just in the rng functions themselves 22:41:44 <06p​leasingfungus> say you have 20 fighting. Before the change, if you're doing this randomly (ie in actual combat), you get random2(1 + 20) -> +0-20 to-hit. After the change, you get random2(20 * 100 + 100) / 100 -> (0-2099) / 100 -> 0-20? I guess that's right... hrm. 22:47:12 <06p​leasingfungus> well, i don't like it, but maybe it's fine. 😛 22:47:46 <06p​leasingfungus> the function i always wanted was a maybe_div_rand_round. 22:48:25 <06p​leasingfungus> i kept looking at maybe_random_div and being disappointed it wasn't that. unfair, i suppose 22:48:32 <06p​leasingfungus> if only someone could just add a new function... 22:58:39 Windows builds of master branch on crawl.develz.org updated to: 0.34-a0-903-g3e737eca7e 23:07:40 <06m​umra> seems to be something up with vampiric weapon descriptions 23:07:41 <06m​umra> https://cdn.discordapp.com/attachments/747522859361894521/1423914451924680776/image.png?ex=68e20b2c&is=68e0b9ac&hm=0f77e8f0cbb9560ff7db733715b7c9a99854a0e1e7b7f92789319e0ed3d7735f& 23:07:54 <06m​umra> doesn't describe Vamp prop anywhere 23:09:03 <06m​umra> %git b3b7527496cacaa7305a1ddd32922bd3ce1bab30 23:09:04 <04C​erebot> dolorous * 0.34-a0-880-gb3b7527496: Move weapon ego descriptions to the database. (4 days ago, 4 files, 162+ 88-) https://github.com/crawl/crawl/commit/b3b7527496ca 23:09:12 <06m​umra> i'm thinking something in that commit 23:15:04 <06m​umra> https://cdn.discordapp.com/attachments/747522859361894521/1423916313230704735/image.png?ex=68e20ce8&is=68e0bb68&hm=88e14892f920ce7fe4a03ad36587d9f376952567cd984d318369ff5a64ca435f& 23:15:11 <06m​umra> this trove vault seems to be missing some walls 23:20:41 <06r​egret-⸸nde※> Looks like I forgot to use a shuffle to properly clear the door alongside the walls in mixing together enter_trove_1 and enter_trove_3 into one vault during the big trove overhaul. I'm... technically in the middle of vault review, as part of incorporating another five or so new tile sets, so I can fit it into my current docket. 23:29:52 <09h​ellmonk> it looks cool. Breaking into the secret trove by deleting the whole ass room around it 23:34:00 Unstable branch on cbro.berotato.org updated to: 0.34-a0-903-g3e737eca7e (34) 23:40:51 03dolorous02 07* 0.34-a0-904-g9f382c6734: Fix vampirism and spectralising ego names (mumra). 10(4 minutes ago, 1 file, 2+ 2-) 13https://github.com/crawl/crawl/commit/9f382c67348f 23:43:41 <06m​umra> how is bane of warding supposed to work? i've lifted the bane now but i'm still wandering into enemies that have the damage-immune at range buff. I'm thinking they got spawned with it when I arrived on the floor but it hasn't been removed now even tho the bane is lifted? 23:44:34 <06m​umra> hmm, unless this V entrance was placing enemies with buff and it's a coincidence 23:44:35 <04d​racoomega> It's not removed from enemies when the bane is lifted. It has a chance of being applied to enemy groups as you encounter them (but this does mean that sometimes it can apply to connected enemies that were out of LoS of the one you actually saw) 23:45:08 <06m​umra> ok yeah, these were all part of a big group that spawned from the V entrace 23:45:40 <04d​racoomega> (It was originally on a per-monster basis, but it turned out that warding only one monster of a band often felt entirely meaningless, since you'd just kill all the rest first and then it would be in melee anyway, so it wards a bunch of nearby things at once, when it triggers) 23:46:13 <06m​umra> yeah they were offscreen when it happened, but didn't get pulled during the fight 23:46:25 <06m​umra> so then it seemed odd after i'd rested to still see the buff 23:49:15 <04d​racoomega> All of the banes that have a chance to buff monsters never have the buff wear off 23:49:27 <04d​racoomega> The curse triggered once; now you just have to deal with it, etc. 😛 23:50:00 <04d​racoomega> Saves having to worry about dactions, but probably also better gameplay not to give any incentive to park/skip and come back later 23:55:34 Monster database of master branch on crawl.develz.org updated to: 0.34-a0-903-g3e737eca7e