00:15:07 Unstable branch on cbro.berotato.org updated to: 0.30-a0-779-g7971cf19c3 (34) 00:54:19 Monster database of master branch on crawl.develz.org updated to: 0.30-a0-779-g7971cf19c3 01:34:55 Fork (bcrawl) on crawl.kelbi.org updated to: 0.23-a0-4884-gcf1f1e11f7 02:47:05 -!- nan00k_ is now known as nan00k 04:23:49 Experimental (bcrawl) branch on underhound.eu updated to: 0.23-a0-4884-gcf1f1e11f7 05:06:38 Unstable branch on crawl.akrasiac.org updated to: 0.30-a0-779-g7971cf1 (34) 07:39:14 03advil02 07* 0.30-a0-780-g1c7e38fe9f: fix: remove sprintf 10(13 minutes ago, 10 files, 36+ 48-) 13https://github.com/crawl/crawl/commit/1c7e38fe9f16 07:39:14 03advil02 07* 0.30-a0-781-geb7024b798: fix: prevent some `random_var` warnings 10(8 minutes ago, 1 file, 3+ 0-) 13https://github.com/crawl/crawl/commit/eb7024b798fa 08:00:00 <09gammafunk> wow, had no idea sprintf was deprecated, but makes sense 08:03:07 <06advil> from some googling I think it is deprecated only by apple 😄 08:03:18 <06advil> i.e. it's not deprecated in any standard 08:03:29 03advil02 07* 0.30-a0-782-ge64f9170bb: fix: resolve a couple more warnings 10(2 minutes ago, 2 files, 3+ 2-) 13https://github.com/crawl/crawl/commit/e64f9170bb78 08:04:21 <06advil> but it still seemed reasonable to change (also I get to see current macos warnings now so...) 08:10:13 <09gammafunk> yeah, seems reasonable to put some crawlcode limits in place anyhow 08:11:17 it's disrecommended in favor of snprintf and friends, though 08:11:26 (buffer overflows) 08:26:25 Unstable branch on crawl.kelbi.org updated to: 0.30-a0-781-geb7024b798 (34) 08:39:25 Unstable branch on crawl.kelbi.org updated to: 0.30-a0-782-ge64f9170bb (34) 12:57:01 <08nlavsky> re: the ring of Gozag's Touch. It doesn't look like a good idea to have a strategic effect on a ring. 12:57:09 <08nlavsky> (the tactical vs. strategic distinction here is about being useful when fighting enemies vs. being useful outside of combat) 12:57:17 <08nlavsky> If the unrand ring doesn't provide a tactical benefit, it just increases the amount of ring-swapping. 12:57:23 <08nlavsky> So it might be better to convert this unrand into an amulet. 13:28:46 <10PleasingFungus> it’s fragile. 13:31:46 <08nlavsky> With another amulet-of-nothing, the dreamshard, the player chooses between a one-time (strategic) effect and between all other, tactically useful, amulet types. This ring's drawback is more ring-swapping, which is tedious and unfun. 13:40:08 <10PleasingFungus> i don’t understand 13:40:15 <10PleasingFungus> what do you mean by more ring swapping? 13:40:54 <10PleasingFungus> are you saying that the player would swap this new unrand ring on and off? 13:41:24 <08nlavsky> The player would change the other ring more often. 13:41:27 <10PleasingFungus> why? 13:41:40 <08nlavsky> 1 slot < 2 slots 13:41:46 <10PleasingFungus> i understand that. 13:41:58 <10PleasingFungus> does sac hand also increase ring swapping? 13:42:11 <08nlavsky> yep 13:42:17 <10PleasingFungus> do octopodes and macabre finger users do less ring swapping? 13:42:22 <08nlavsky> yep 13:43:09 <08nlavsky> Let's say, you go into Snake with two ring slots and you have a ring or rF and rPois. 13:43:28 <08nlavsky> now imagine you have only 1 slot available 13:44:02 <08nlavsky> as a result, you switch rings more often 13:44:06 <10PleasingFungus> if i go into snake with two ring slots, i will generally have something generically useful, like AC, on one if not both fingers. 13:44:22 <10PleasingFungus> i'll swap to a resist ring as needed, but that's true whether i have one slot open or two. 13:44:42 <08nlavsky> AC -> relec, AC -> poison, AC-> SInv, whatever 13:44:54 <10PleasingFungus> sure. my point is that i would do that with or without this ring. 13:45:22 <08nlavsky> locking down 1 slot out of 2 doesn't introduce a choice, it introduces tedium 13:45:25 <10PleasingFungus> in fact, i'd do more swapping if i didn't have this ring, in cases where i'm fighting e.g. a salamander and a naga at the same time 13:46:01 <10PleasingFungus> if there is tedium in ring swapping, and there probably is, increasing the number of ring slots only makes it worse 13:46:11 <08nlavsky> (ru gives a tactical bonus for losing a ring slot, by the way) 13:46:18 <10PleasingFungus> so? 13:46:23 <10PleasingFungus> it's a strategic decision 13:46:31 <10PleasingFungus> both for the sac and the unrand 13:46:35 <10PleasingFungus> the reward is sort of irrelevant 13:47:07 <10PleasingFungus> anyway, think this is a playstyle difference. if you aren't playing to win, then yes, this ring will increase swapping 13:47:34 <10PleasingFungus> (but note that it dramatically decreases ring swapping if you wear it after sacrificing hand! :P) 13:49:29 <08nlavsky> "if you aren't playing to win, then yes, this ring will increase swapping" I'm not sure I understand this 13:49:41 <08nlavsky> it just increases ring-swapping, period 13:49:51 <08nlavsky> in all cases 13:50:06 <10PleasingFungus> i'm going to try to de-escalate. 13:50:12 <10PleasingFungus> let me ask a question. 13:50:47 <08nlavsky> sorry for escalating, it's just a discussion about game design, right? 13:50:51 <10PleasingFungus> 🙂 13:51:01 <10PleasingFungus> no worries, i was getting a little contentious myself 13:51:15 <10PleasingFungus> > Let's say, you go into Snake with two ring slots and you have a ring or rF and rPois. Are you saying that, when you enter Snake, you would default to having rF on one hand and rPois on the other? (assuming you lacked both otherwise) 13:52:49 <08nlavsky> which will allow me to not switch to other rings most of the time. And if I have only one non-locked slot, I'l have to switch rF <-> rPois when fighting nagas/salamanders (the scary ones with Eruption) 13:52:56 <10PleasingFungus> yes, i get that. 13:53:20 <10PleasingFungus> ok. let's say you're in snake, and you see a naga warrior. only guy around. 13:53:28 <10PleasingFungus> are you keeping that rF ring on, or are you swapping to AC? 13:54:24 <08nlavsky> there's always a mystic around the corner, so switching to rpois is OptiMal when you see a naga 🙂 13:54:53 <08nlavsky> (a mystic, another naga, more naga warrors, etc.) 13:55:10 <10PleasingFungus> i don't understand at all, i'm sorry. 13:55:19 <08nlavsky> If I were an optimal player, I'd switch to AC, probably 13:55:19 <10PleasingFungus> why are you talking about switching to rpois? you already have rpois and rf on, right? 13:55:22 <10PleasingFungus> yes, ok. 13:55:33 <10PleasingFungus> that is the point i'm trying to get at. 13:55:43 <10PleasingFungus> if you're trying to play well, you would already be doing more ring swapping than we're discussing here. 13:55:51 <10PleasingFungus> if you choose not to do so, that's fine! it's a game 13:56:01 <10PleasingFungus> if you want to say ring swapping sucks, hey, you are very far from the only one on the dev team who thinks so 13:56:33 <08nlavsky> there's a balance between having enough defences/resits and having the best defences/resists. the cost, sometimes, is tedium 13:56:38 <10PleasingFungus> > [losing a ring slot] just increases ring-swapping, period. in all cases this just isn't true, though. 13:57:03 <10PleasingFungus> 'enough defenses'? I don't understand what that means 😛 13:57:44 <08nlavsky> for snake, enough defenses would be rF and rPois 13:57:52 <10PleasingFungus> those are resists!!! 13:58:02 <10PleasingFungus> (which are arguably a subtype of defense? enh) 13:58:24 <10PleasingFungus> i usually see 'defenses' used to mean ac, ev, sh... arguably hp 13:58:36 <10PleasingFungus> anyway i gotta go back to work. if you want to have a broader discussion about ring swapping, i think that'd be valuable and worthwhile - i would really love to see any ideas you have 13:59:24 <09gammafunk> curious, do you "require" rF for Snake nlavsky? because I certainly don't 13:59:42 <09gammafunk> I don't require rPois either, but admit that not having it can be pretty annoying 13:59:57 <09gammafunk> but I would put rF in the "nice to have" category even more so than rPois 14:00:04 <08nlavsky> rF is good to have if you want to worry less about salamanders 14:00:14 <10PleasingFungus> the fundamental idea of gaining flexibility in combat is cool, the cost is just maybe a bit too low. not sure what a good cost would be - time, something else..? 14:00:30 <09gammafunk> well, I mention it because you said it in terms of "enough", which basically implies that it's required, because if you don't have it, then you wouldn't have enough 14:00:43 <09gammafunk> obviously you can have multiple sources of defenses 14:00:44 <10PleasingFungus> i don't think that was what nlavsky was saying 14:00:49 <09gammafunk> that muay independently add up to enough 14:01:34 <08nlavsky> yeah, I know about "rPois is optional" meme and even done S-branches w/o one mysself 🙂 14:01:36 <10PleasingFungus> think the implication is 'rF is good in Snake and is helpful for not dying [aka having 'good enough' defenses] 14:01:40 <10PleasingFungus> meme...!?! 14:01:44 <10PleasingFungus> (ok actually going now.) 14:01:53 <09gammafunk> don't let your memes be dreams 14:02:22 <06advil> I didn't follow all the discussion here but simply by the pigeonhole principle it would be really surprising to me if there was less or the same amt of swapping with one slot than with two 14:02:32 <08nlavsky> yeah, it's a meme. there's a huge gameplay difference when you have rPois and then you don't 14:04:25 <09gammafunk> sure, huge gameplay difference, but that's more in the camp of "it's less fun to rest more". And if you're doing lots of ranged damage it's less painful 14:04:26 <08nlavsky> time to google "pigeonhole principle" I really like learning from you all! 14:04:41 <09gammafunk> fr: shrikehole principle 14:05:57 <06advil> any player has a mostly fixed n slots and m properties that can (via possibly complex combinations) be put in those slots, where at some point in the game m always exceeds n and is entirely a function of the dungeon, not the player, so I think it's always going to be on average a smaller swap distance for larger n to the optimal combo 14:06:02 <09gammafunk> it's more dangerous to not have rpois and even when you're playing more carefully to compensate, you do have to spend more time doing less fun things (mainly resting), but I think it's important to realize you can do the branch without rpois. Do agree with the strong preference to have it over not having 14:06:49 <06advil> it's true that the optimization problem is simply more complicated with more slots and a sufficiently larger m though, so maybe I'm overestimating how often it will be ok to not swap everything 14:07:18 <09gammafunk> we need to add a "ring swap" milestone so we can query in sequell 14:07:27 <06advil> lol 14:07:32 <06advil> add it to the action table 14:14:12 <06advil> I think I've unconvinced myself of my pigeonhold principle argument 14:15:49 <06advil> because that argument depends on the assumption that more scenarios than not will have an optimal set of properties that is small enough to fit on a single ring 14:15:55 <06advil> which I have no idea if is true 14:28:17 <08nlavsky> Anouther solution might be adding some base type (AC, Will, or another tactically useful effect) to the Gozag's ring, which will alleviate the need for ring-swapping a bit. 14:28:21 solution is to obviously give all races tentacles to reduce ring swapping 14:28:22 <08nlavsky> It was kinda lost in the discussion, but my point was that the unrand would work better as an amulet instead of a ring. 14:28:31 <08nlavsky> If it's an amulet, there's a choice between this unrand and all other amulets, which you can't swap in each fight. If it's a ring, the player essentially doesn't have to choose, they can just swap rings more often. 14:29:31 <08nlavsky> giving all species more ring slots would work too 🙂 14:30:36 Personally I just favor sub-optimal but less tedious choices so I'll usually stick to one set of rings for a given branch 14:31:02 I do agree that it'd make more sense for this ring to be an amulet, though 14:31:02 <08nlavsky> well, an obvious solution is to make ring-swapping slower, similar to what was done to amulets 14:31:32 that too 14:55:03 <08nlavsky> Hmm, there could be the same (!) amount of ring swapping for my Snake example: 14:55:18 <08nlavsky> 2 slots 1 slot Threat rPois + AC rPois nagas no swapping no swapping salamanders rPois -> rF rPois -> rF jorgrun rPois -> Will rPois -> Will shock serpent rPois -> rElec rPois -> rElec S.Roka rPois -> AC rPois -> AC invisible S.Roka rPois -> SInv rPois -> SInv ...other 14:55:19 cases for switching to another ring type... 15:00:09 <09gammafunk> invisible S.Roka :wfrShock: 15:03:00 <08nlavsky> Since the game has always useful AC rings, the hypothetical optimal player would do exactly the same number of ring swaps while having N and N-1 slots, right??? 15:04:32 <08nlavsky> (I'm slowly un-convincing myself that losing a ring slot would lead to more ring-swapping) 15:05:53 <08nlavsky> we need a mathematician here 15:06:11 sadly don't think there are any in here 15:08:20 ebering maybe. and I thought someone else but not finding it in logs 15:09:19 <09gammafunk> dpeg and elliptic 15:09:59 <09gammafunk> ebering definitely 15:10:37 oh I was joking :P 15:17:07 <08nlavsky> "Since the game has always useful AC rings," <-- this condition could be relaxed. For some chars it's AC, for others it's Slay or Int+. The optimal player would do the same number of ring-swaps for the least important ring while having N or N-1 slots. 15:24:05 <08nlavsky> it turns out, PF was right from the beginning 15:25:36 <09gammafunk> he used the oldest dev trick in the book 15:25:58 <09gammafunk> argue with you just enough until you convince yourself that you're wrong 15:29:13 <08nlavsky> https://tenor.com/view/kill-bill-master-gif-10355953 16:06:51 <08nlavsky> and the final, obvious in retrospect, observation would be that more ring slots allow more ring-swapping, and fewer slots allow less ring-swapping 16:24:03 Unstable branch on underhound.eu updated to: 0.30-a0-782-ge64f9170bb (34) 18:08:28 <09hellmonk> Don't think slow swap is a complete solution to equipment swapping fwiw (because stairs exist). This is a problem for at least some orbs currently, as mentioned by acrobat in the main discord. 19:13:48 <10PleasingFungus> idk that we need a complete solution. the perfect the enemy of the good, etc 19:14:14 <10PleasingFungus> improvement can be gradual (though i’m not endorsing 5 turn ring swaps) 20:17:54 New branch created: visiblesign (1 commit) 13https://github.com/crawl/crawl/tree/visiblesign 20:17:54 03hellmonk02 07[visiblesign] * 0.30-a0-783-geb8b792639: feat: soft-remove monster invis 10(4 minutes ago, 16 files, 7+ 118-) 13https://github.com/crawl/crawl/commit/eb8b79263989 20:19:15 Interesting 20:19:56 There is something to be said about "predicting" where a monster will be, but that's such a small thing compared to the annoyance of invis monsters 20:38:15 <09gammafunk> god, it's been many many years since I worked on that monster::unseen_pos stuff 20:38:34 <09gammafunk> one of my earlier projects to fix various monster invis issues, and I just remember it being a nightmare 20:40:20 <09gammafunk> regarding monster invis design: not knowing exactly where a monster will be does arguably add some tactics that are interesting, but I'm not sure the annoyances are worth 20:41:16 <09gammafunk> it's a pretty common piece of monster functionality across these types of games to have invisible monsters where you're not given exact location information 20:41:20 <09hellmonk> yeah, it does matter for perma-invis monsters and in some cases where monsters blink or leave los for other reason 20:41:46 <09hellmonk> but autopickup and autofight not working in all the other cases is quite bad 20:41:58 <09gammafunk> I do generally like how it works in infra-arcana, for example, but that's mostly because of the severity of the to-hit penalty 20:42:16 <09gammafunk> yeah, it has very inherent UI issues 20:42:50 <09hellmonk> so I am hoping to clean up the other cases and then adjust the penalties 20:43:38 <09gammafunk> most of the cooler aspects of tactics that involve changing your attack selection and possibly repositioning the monster to somehow more favorable terrain (which maybe is just upstairs) is all still happening with your approach 20:44:04 <09hellmonk> stairdancing 20:44:09 <09hellmonk> truly crawl's original sin 20:44:32 <09gammafunk> right, it's fairly hard to pick examples of where you have to reposition an invis monster specifically because it's invis 20:44:49 <09gammafunk> seems to mostly be about buffing yourself or choosing a less dodgeable attack 20:44:55 <09gammafunk> or of course swapping on sinv 20:45:06 <09hellmonk> this does make stairdancing invis monsters slightly easier since you don't have to guess their position once you go up tbf 20:45:28 <09gammafunk> right, but that's a loss of like maybe one turn for monsters that aren't unseen horrors 20:45:39 <09hellmonk> but I don't think it ruins any of the invis monster designs outright 20:45:42 <09gammafunk> in fact unseen horrors have a cool tactical aspect to that invisible positioning 20:45:59 <09gammafunk> where since you are less certain of where they can be, you have to take more care to get them in stuff like halways 20:46:09 <09gammafunk> but that's due to invis + unique movement 20:46:14 <09hellmonk> yeah, unseen horrors specifically probably need some work 20:46:47 <09gammafunk> for normally moving monsters invis doesn't introduce too much repositioning, just means you might attack differently or isolate them or reset the fight to get them un-invis 20:47:05 <09hellmonk> yes, and you count turns more carefully if you duck around a corner or whatever 20:48:08 <09hellmonk> anyway my hope is to fix all the remaining targeter stuff this weekend and then maybe adjust the invis penalties to be harsher, or to scale more severely 20:48:26 <09gammafunk> seems cool, any ideas on what to do with unseen horrors? 20:48:34 <09gammafunk> I guess on a fundamental level they still sort of work 20:48:38 <09hellmonk> not really 20:48:38 <09gammafunk> since they're still hard to hit 20:48:56 <09hellmonk> give them rakshasha illusion idk 20:49:13 <09gammafunk> yeah there's probably a relatively simple tweak you could do if they end up being too weak after this 20:49:31 <09gammafunk> they're still getting in extra hits and hard to hit 20:49:42 <09gammafunk> and hit pretty hard when you first encounter them etc 20:49:56 <09hellmonk> there's possibly room to expand the use of af_shadowstab 20:50:08 <09hellmonk> probably not on unseen horrors though 20:50:09 <09gammafunk> oh god, shadowstab on unseens would be pretty brutal 20:50:10 <09gammafunk> yeah 20:50:26 <09gammafunk> now on shadow yaks it works 20:50:55 <09hellmonk> 2 less base damage than shadows, so uhhh, 5 less after shadowstab multiplier? 20:51:06 <09hellmonk> simply getting shadow stabbed every turn 20:51:11 <09hellmonk> on D:6 21:16:01 <10PleasingFungus> do think the metric for a funseen horror is: if this design was proposed for a totally new mons, would we add it? 21:16:23 <10PleasingFungus> don’t know that ‘megadamage unless you have sinv’ is that exciting 21:16:51 <10PleasingFungus> maybe it is! idk 21:41:56 fun fact: gammafunk and hellmonk are the same color in IRC and similar enough lengths that whenever the two have long conversations, at a skim it looks like one of them just talking to themselves 21:45:10 <09gammafunk> I think we'd probably propose something like it if we continued to have invis and batty movement, probably in part because of how we like new combination of monster abilities/properties, especially if they seem 'emergent', which unseens sort of are. I'd be sad to see them go both because they seem to pose an interesting tactical problem and because their theme is great 21:45:50 <09gammafunk> I guess I'm thinking fondly mostly of how you can fight them in intelligent ways that negates the megedamage in that you can kill them before it 'adds up' too much 21:46:12 I'll be honest, I don't think I have ever in any circumstance been threatened by an unseen horror 21:46:22 <09gammafunk> which is just stuff like using aoe attacks or corridors to narrow where they can be 21:46:31 <09gammafunk> granted hellmonk's branch would basically eliminate that second part 21:46:53 <09gammafunk> unseens do kill a lot of players 21:46:55 And generally when I encounter one the amount of tactics I use to deal with them is basically zero (if it's too late to be a threat) or just a quick jog to a corridor 21:47:07 I'll take your word for it 21:47:52 Just checked, never even been close to death from one 21:48:13 <09gammafunk> taking a quick jog to a corridor describes how you deal with just about any monster, really 21:48:30 <09gammafunk> I don't think it's a question of whether they're dangerous 21:48:41 exactly 21:49:16 <09gammafunk> exactly what? 21:49:19 I don't view unseen horrors as something that results, by necessity, in emergent tactics—just something that can be dealt with through emergent tactics if you're feeling fancy 21:50:06 <09gammafunk> oh, you can't hit them very well 21:50:09 <09gammafunk> even in a corridor 21:50:21 <09gammafunk> you can do so if you're a minotaur or something really strong or have amazing melee damage 21:50:43 <09gammafunk> but an average char isn't a mi or doesn't have that; those sorts of chars view most monsters as trivial 21:50:57 I'm not talking about a Mi or a melee brute 21:51:01 <09gammafunk> the emergent part is 21:51:14 <09gammafunk> how you combine invis and batty and get something unique 21:51:22 ah, right 21:52:14 <09gammafunk> that's how players die to them, they can't hit them and choose not to park them. granted you can nearly always park them, but they can put you in serious danger if you get ambushed far from stairs 21:53:00 Sorry, let me try restating: I don't think that unseen horror is that interesting as a monster as it generally is not typically threatening, even to fairly weak melee characters or casters. I haven't found myself ever needing to rely on unique strategies to deal with them because of invis+batty—I just rely on the same thing you would for any other monster. 21:53:34 <09gammafunk> yeah I don't agree with the statement that it's not typically threatening to fairly weak melee characters and casters 21:53:55 Fair! I don't have any evidence besides personal experience for that, anyway. 21:54:06 <09gammafunk> they can be neutralized by stairs fairly effectively, that is true, but you have to bypass them effectively if you do that 21:54:39 <09gammafunk> they are a skill check against using stairs and avoiding an area with a dangerous monster and a skill check against choosing an appropriate attack 22:03:58 <09gammafunk> some comparable monsters in terms of number of kills for recent versions on d:5 and d:6: > 572x Eustachio, 507x a vampire, 468x a bullfrog, 441x an unseen horror, 357x Sigmund, 355x Ijyb, 346x a shadow imp, 338x a gargoyle, 335x a wraith, 303x a shadow, 296x a vampire mosquito, 258x a boulder beetle, 240x a kobold demonologist, 22:04:28 <09gammafunk> they're not topping the list of course, but the dungeon list for this depth is really big and unseens are pretty rare etc 22:05:32 daaaaaang ago 22:05:38 er. s/ago/okay 22:05:47 never mind I totally believe you 22:17:38 <10PleasingFungus> @gammafunk think i was unclear. my comments were about the proposed unseen horror with shadowstab, not about the current design 22:18:20 <09gammafunk> oh, sorry 22:18:34 <10PleasingFungus> i think we pretty much agree on everything 22:18:39 <10PleasingFungus> if i’m reading you right 22:19:36 <09gammafunk> yeah I didn't love the shadowstab proposal for them, but maybe with lower base damage it works? even then it might just be too overwhelming because they do appear pretty early 22:19:45 <09gammafunk> guess you'd probably lower their speed or something 22:20:36 <09gammafunk> yeah I know you and I share a love for weird single eyeball things since you made that one spooky unrand whose name I forget now 22:20:49 <10PleasingFungus> majin-bo? 22:21:01 <09gammafunk> yes! although perhaps you didn't have roctavian's awesome tile in mind when you made it 22:21:20 <09gammafunk> I just like the idea of a weird eyeball stick you hit things with 22:21:37 <09gammafunk> his tile did come much later 22:22:58 <10PleasingFungus> i did not, yeah 22:23:01 <10PleasingFungus> but it’s fun 22:27:13 <09hellmonk> one downside of my current approach is that you still lose access to all of the nice interface stuff (like hex success chance) against invisible monsters, even when you "should" know it 22:27:36 <09hellmonk> but maybe that is just a necessary compromise 22:30:45 <10PleasingFungus> i haven’t had any time to review the pr yet 22:31:08 <10PleasingFungus> but one thing that should happen, if it doesn’t already, is that all neutral enemies should also be visible to you, like allies 22:31:30 <10PleasingFungus> hard to remove attack prompts etc without guaranteeing that an invis is a hostilr 22:34:01 <09hellmonk> Yeah I will change that 23:56:38 Windows builds of master branch on crawl.develz.org updated to: 0.30-a0-782-ge64f9170bb