00:59:20 -!- indigaz20 is now known as indigaz2 03:34:56 Experimental (bcrawl) branch on underhound.eu updated to: 0.23-a0-5261-gd9800d219b 05:33:58 Unstable branch on crawl.akrasiac.org updated to: 0.34-a0-649-g9db2fe2 (34) 08:14:26 <08n​icolae> frankenmap 08:14:40 <08n​icolae> re: temple map for 22 altars getting made out of previously-lost troves 08:14:56 <08n​icolae> perhaps when i will get home i will whip up a quick 22er for temple, there's gotta be some pleasing way to arrange 22 08:16:00 <08n​icolae> lmao, tbh i could probably just add a potential altar to nicolae_temple_bog_church and call it a day 10:37:07 03DracoOmega02 07* 0.34-a0-650-g1aad040a61: Make Dissipation work properly with Slow / Sticky Flame (sentientsupper) 10(34 minutes ago, 1 file, 4+ 1-) 13https://github.com/crawl/crawl/commit/1aad040a61be 10:37:07 03DracoOmega02 07* 0.34-a0-651-g041a10fefa: Don't pre-reveal transporter destinations in Necropolis/Gauntlet 10(10 minutes ago, 1 file, 3+ 1-) 13https://github.com/crawl/crawl/commit/041a10fefa45 10:37:07 03DracoOmega02 07* 0.34-a0-652-g878998d1fe: Include the base type of randart staves in their name 10(26 seconds ago, 2 files, 3+ 3-) 13https://github.com/crawl/crawl/commit/878998d1fe1d 11:53:37 <07a​crobat> @dracoomega found one problem regarding tesseract. > While it is technically possible to lure them away from the orb vault to another staircase, the fact that they are all patrolling (and fast) means that it's nearly impossible to 'cheese' anything in this way. when I first saw the recent changes, I thought the same. however, when I actually played the game, I found that when orb guardians forgot the player's location, they just 11:53:37 returned to their original location instead of patrolling. so, I could cheese it. to be specific, I cheated in this way: enter zot:5 -> explore alarm trap happen -> monsters including orb guardians rush toward me -> take upstair -> take other downstair -> orb guardians forget my location and returns to its original location (but other monsters don't) -> kill monsters freely -> find and step into alarm trap -> do the same thing in this way, I killed 11:53:37 about half of monsters in zot:5 without tesseract activate and without interference from patrolling orb guardians. (after the game, I confirmed once again that this was possible in wiz mode.) 11:57:35 <04d​racoomega> By the way, what patrolling means is 'will return to its original location when not engaging a target'. I... hadn't considered the fact that the other monsters not being patrolling meant you could separate them that way. I tried a variety of ways to lure orb guardians away and then try to sneak into the vault, but that always failed. (And generally noise isn't good at making monsters wander out of the lungs into the rest of 11:57:35 the level, without a player to directly chase, due to the level geometry, so I think it may require Mark to be practical? But getting marked is quite possible, of course) It's possible this could be solved by just making all monsters placed inside the orb vault patrolling themselves? 12:05:23 <07a​crobat> > so I think it may require Mark to be practical? I think maybe very loud noise also can do this? because they will eventually get out if I wait for a long time. or maybe not. I don't know how things work exactly. > It's possible this could be solved by just making all monsters placed inside the orb vault patrolling themselves? I'm not sure which one is better, making all monsters placed inside the orb vault patrolling themselves or 12:05:24 making orb guardians not being patrolling. 12:06:17 <04d​racoomega> Making orb guardians not patrolling makes it actually fairly easy to sneak into the vault using stairs, though. Gather up the ones on the outside to one staircase, go up to leave them there, go down elsewhere, and then just waltz on in and start shouting slowly from inside the first room of the orb vault 12:24:42 <03i​mplojin> I kind of want to raise the possibility here that the tesseract experiment might not be working out very well -- the tesseract commit mentions shout luring as the rationale†, and in order to combat this tesseracts have been introducing an increasingly bizarre series of mechanics that need to be explained to the player, turning zot:5 into an increasingly on-rails puzzlebox of "you must fight this in the one specific way we intend" to 12:24:42 players who, on balance, probably weren't commonly luring zot:5 to begin with. Then we have this issue where players who are either piloting particularly weak characters or are themselves not very good players can unintentionally run up the tesseract generation to the point where they don't feel they can overcome it: I've seen at least three or four entirely separate frustrated, long-winded posts about this! The fact that we're sticking players into a 12:24:43 surprise negative powerspiral right at the end of the game and the design of that spiral focuses its effects on players who probably aren't very good just seems misguided to me. Then there's the generate_awake change which I'm also not fond of, because we really don't need to be reducing the efficacy of stealth/stab chars as a playstyle! Those chars are already hideously complex to the point newer players don't touch them or understand how to make 12:24:43 them work. To say at least something positive here, the tesseract tiles are beautiful work and I really like them. †(Was shout-luring even a common approach that players had been using recently? Surely simply fighting in or near the lungs was most common, with an honorable mention to intentionally stepping on an alarm trap and killboxing with dig, but that carried its own issues that you had to fight a lot more stuff than one lung's worth of mons 12:24:44 and wasn't clearly superior to just leaving more of the floor asleep to begin with.) 12:27:08 <08n​icolae> maybe some kind of Super Patrolling where they won't leave their spot if they can't see you personally so they can't be lured, and then put it on the orb guardians et al. 12:28:54 <06r​egret-⸸nde※> It is fundamental to the concept of there being a clock that people can't overcome the end point of it, and the last person who kicked up a fuss played so badly he was teleport trapped to beside the tesseract before entering the vault at all, ignored it, teleported out, and then spent nearly a thousand turns off-level clearing Zot:4 instead. 12:29:18 from at least *one* player, I can confirm that shout-luring the lungs was something that I felt compelled to do often on weak characters to survive old Zot 12:33:44 <06r​egret-⸸nde※> People being frustrated about things killing them because they didn't understand them is a recurring experience with nearly every new monster we've ever added, and people have talked about Zot being very dull for not having anything mechanically unique to it plus encouraging slow luring tactics for probably over half of Crawl's development. 12:37:49 short-term clocks are, I feel, qualitatively different from long-term ones. long-term clocks have historically made an entire game stressful (or been trivial to the point of ignoring), but short-term ones allow for tension spikes and a chance to breathe in between. regardless of the exact details of tesseracts, I think this is good in principle, 12:37:49 though otherwise rare in present crawl. I think this is as much or more of a reason why portals have historically worked out as the "positive incentive versus negative disincentive" part. 12:40:34 the 'on-rails puzzlebox of "you must fight this in the one specific way we intend"' aspect can be (and has been) argued of any number of anti-grinding and anti-scumming strategies, in that it's only really disincentivizing taking an inordinate number of turns to accomplish the same task 12:41:48 <06r​egret-⸸nde※> (While double-trapped lungs were removed a while ago, the fact it had decent odds of happening for nearly all of Crawl's history made quite a few people get used to the idea of shout-spamming; even without such circumstances coming up, it was always valuable for being by the only natural chokepoints in the whole vault.) 12:42:56 there's only one way, remove the shout command 12:45:01 <06r​egret-⸸nde※> People would just cast Mephitic Cloud in place instead. 12:45:46 actually, when I talk about shout-luring the lungs, I'm usually using the word "shout" more generally to mean things like "cast fulminant prism" 12:47:31 I should have included some kind of joke marker, sorry 12:48:38 (I realized but I felt compelled to be unambiguous just for record's sake) 13:01:47 <03i​mplojin> re: People being frustrated about things killing them because they didn't understand them: Here is a link to a post in the roguelikes discord by acrobat, who is certainly in contention for being one of the best crawl players ever, talking about how zot:5 has become potentially a "big problem" for Mu because they can't prebuff. Consider who's saying this, and consider that the playerbase is much, much worse at the game: 13:01:47 https://discord.com/channels/205277826788622337/205316046230388737/1407734258424086640 Introducing mechanically-unique things is great! More gameplay diversity. Consider how this is going to come across to new players, though: By introducing a disincentive for poor play that manifests as a negative power spiral for not immediately understanding mechanics introduced right at the end of what is, for new players, a several-hour game that involves 13:01:48 splatting hundreds of characters to even make it to that point, aren't you kind of sending a very player-hostile, IVAN-like message? They've made it to the end, finally, but it's not enough for us to follow the rules of the game thereto, we now want to throw a bunch of cute unique challenges at them at the finish line and punish them if they don't intuit the design requirements. Come on. If you want to do weird mechanics, great! But maybe do them 13:01:48 within the first ~10 floors of the dungeon and tutorialize them and then use them. Was zot:5 with the lungs and the orbs and the traps truly so dull that we need to be busting out "do this right or die slowly and frustratingly" gotchas last-minute? (If there was any pre-implementation discussion re: tesseracts I'd appreciate a link? The idea just feels undercooked.) 13:04:20 <07a​crobat> personally, I don't feel that it is bizarre mechanics. it is interesting that there is a unique system in the final stage of the game that distinguished from the previous stages. I don't know it is bad that weak characters are greatly affected by tesseract. to me, it seems natural that the final stage of the game feels sufficiently challenging and that characters who are not sufficiently prepared to reach that point struggle. (but I 13:04:20 think mummy is only one who are really affected by tesseract.) imo you have a point regarding generate_awake. if players simply fight inside or near lungs instead of using shout-luring, I don't think tesseract made zot:5 more difficult at all. because that's how player are supposed to kill tesseract in the first place. 13:04:28 <12g​e0ff> just need to add a save point when the player enters Zot:5 for the first time (ducks and covers) 13:05:24 <06p​leasingfungus> Extra life! 13:11:28 <07a​crobat> > More gameplay diversity. Consider how this is going to come across to new players, though: By introducing a disincentive for poor play that manifests as a negative power spiral for not immediately understanding mechanics introduced right at the end of what is, for new players, a several-hour game that involves splatting hundreds of characters to even make it to that point, aren't you kind of sending a very player-hostile, IVAN-like 13:11:28 message? this is maybe a legitimate concern though... 13:12:41 <07a​crobat> It's really hard for me to predict how new players will feel. I'm already too familiar with this game 13:13:29 <06r​egret-⸸nde※> (If there was any pre-implementation discussion re: tesseracts I'd appreciate a link? The idea just feels undercooked.) Unavoidably, when the two devs doing the vast majority of work for the game live in the same house, a fair bit of developer conversation can't be easily recorded and referred back to. One of the other active devs originally proposed it a fair number of months back: 13:13:29 https://ptb.discord.com/channels/205277826788622337/205316046230388737/1359924011584327924 This was then followed up a few days before release with a different conversation: https://ptb.discord.com/channels/205277826788622337/205316046230388737/1389855373594329225 13:13:37 at minimum I do feel compelled to quote zot.des' first line from 18 years ago here 13:13:39 # zot.des: The ultimate goal. The Orb must be placed (Z). These maps can and 13:13:44 #          should be serious challenges! Players dying on Zot:5 are okay :) 13:14:46 now, the exact mechanics by which they die are the argument, and there could be one for tutorializing such a challenge earlier in addition 13:17:25 <06r​egret-⸸nde※> New players already have to deal with unexpected and unforeshadowed endgame things like Vaults:5 being an ambush, or not having See Invisible for Zot ghost moths. 13:20:28 <12g​e0ff> @Memoria, by the way, Discord shows your lines with the # at the start as headers (in a really big font) 13:21:06 <07a​crobat> yeah but I think tesseract is maybe a little different from those. ambushes exist in other places as well, and you can also experience losing mp to monsters in other place (like spider or crypt) 13:21:12 ouch, sorry 13:21:23 <06r​egret-⸸nde※> Cognitogaunts aren't invisible! 13:21:33 I can't edit them so those on the other side of the Discord bridge must suffer 13:22:08 (but I will avoid that in the future, I should've guessed it would format like that since that's how it works in general) 13:24:12 <06r​egret-⸸nde※> (The first of these two conversations calls back on another further back conversation: https://ptb.discord.com/channels/205277826788622337/205316046230388737/1150920472050008075) 13:25:59 <08n​icolae> that's from copy and pasting the comment directly from the .des file, comments in .des files begin with #, which is also discord markup for heading text 13:26:40 <08n​icolae> oh, nevermind, i think you knew that part 13:28:52 yeah I just failed to consider that the bot would include formatting 13:33:59 <04d​racoomega> I think it's fair to question whether a particular design achieves its goals. I do this constantly! In fact, I spent three hours just yesterday analyzing a game that someone complained about, looking both at their replay and playing the seed myself (with their exact character) multiple times to see how plausible it was to deal with certain situations and when, and discussing this with Index and others. I think it's worth noting 13:33:59 that this isn't just an 'overly convoluted way to stomp out shout luring'. That was the starting point that led to this design, but in the process of developing it, I felt it became an interesting capstone challenge of its own and more than just 'trying to end shout-luring' (even if it also does that). I realize some people have complained at length about them; I have probably read every single conversation on the subject. And without dismissing 13:34:00 their specific complaints (some of which I have examined at length, as I said), I feel like several impassioned complaints on a topic does not itself mean a design has failed. Okawaru gifting got many times more complaints, even bombardier beetles got more (somehow), and shafts are famously prone to being complained about as bad design. But at the same time, I have also seen multiple people say that they enjoy dealing with tesseracts and were happy 13:34:00 for the change. So much so that multiple people independently have asked for that sort of feature to have a chance of appearing earlier in the game as well: partially for learnability but also because they feel the dynamics they encourage are interesting. I am hardly saying the current state is perfect and complete, but also don't feel like feedback alone suggests it should be discarded outright at this point, especially when there's plenty of time 13:34:01 for more iteration. 13:36:20 <06r​egret-⸸nde※> (I wonder where tutorializing tesseracts would be? A new portal? It sort of thematically fits with some of my old clashing-elemental-planes free-portal-bound-ally portal Conflux plans around Bailey depth....) 13:37:14 <12g​e0ff> could add a tesseract to guard a rune, like in Vaults 13:37:21 <04d​racoomega> Like, I absolutely don't want this to be a 'there's a narrow window to respond properly and then you're fucked' kind of thing, but I genuinely don't think it is that, even now. And that most scenarios of getting overrun and run out of resources involved an extended series of misplays - at any point along which it was still possible to make different decisions and turn things around. The percentage of players actually dying to 13:37:21 tesseracts was fairly low, last I looked (and the changes made a few days ago specifically address scenarios where the tesseract could be activated prematurely, without the player's agency in the matter, which I suspect were responsible for some of the outliers). 13:37:24 being on the defending side of shafting arguments has certainly demonstrated this principle more than anything 13:39:28 the *possibility* of any form "slow death" is difficult to avoid, being how consumable attrition works as a much more fundamental mechanic across the game 13:44:29 <04d​racoomega> I actually spent a while mulling over exactly 'how bad should it be at its worst?'. Since, of course, there's the fear that dealing with sufficiently poorly will turn the game into an unwinnable situation. Capping the number of spawns on the floor was an attempt to dial that back a bit (whether sufficiently yet or not). But at the same time, if there is a capstone challenge that we still want to be beatable by players who have 13:44:29 burned all their resources and made many, many misplays in a row, what is the point? At some point of handling something badly, shouldn't one just die? (Perhaps one key thing that separates this from other situations where the player can get overwhelmed due to being slow or managing resources poorly is that most of those don't allow you to leave to somewhere else safer. Like, Pan respawns can get pretty out of control on some floors if one is playing 13:44:30 slowly (and, say, going back to some hidden corner to regen all the time). We actually had to deal with a crash the other month due to someone generating over 600 monsters on Lom's floor, which is way way way more than tesseracts can ever do. But there, there's nowhere safe to take a breather. You either abandon the rune forever or you die. Here, you got back to Zot:4 and mull over how implausible Zot:5 feels now. I've not been quite sure how I feel 13:44:30 about that part, I will admit.) 13:46:43 <04d​racoomega> I don't think the current state of tesseracts requires narrow builds to answer it. I feel like virtually any build that isn't terribly weak will have a variety of answers to tesseracts, including 'play relatively conventionally' (using consumables and god powers to rush the tesseract quickly and then retreat is a strategy some people can do, but is absolutely not a required thing. Many people (including those who are not experts 13:46:43 at the game) have reported playing fairly conventionally (if slightly more aggressively than before the change) and not having any particular difficulties with them. (Regarding generate_awake affecting stabbers, I will admit that this was mostly because them being asleep or awake didn't matter for anything now, so it felt needlessly confusing to display a sleeping state. I did not consider this making them less stabbable in certain rare situations, 13:46:44 and am not opposed to changing that back on that basis. Tesseracts deliberately lack most elemental resists (unlike a lot of nonliving things) so that a variety of builds are on more equal footing regarding them.) 15:12:17 03fslottner02 {regret-index} 07* 0.34-a0-653-ge5d32cb860: add new arrival vaults (#4702) 10(2 weeks ago, 1 file, 64+ 0-) 13https://github.com/crawl/crawl/commit/e5d32cb86085 15:12:17 03regret-index02 07* 0.34-a0-654-g40ee31a434: Make piles of debris crumble, not wither 10(2 weeks ago, 1 file, 4+ 1-) 13https://github.com/crawl/crawl/commit/40ee31a43446 15:12:17 03regret-index02 07* 0.34-a0-655-g011df5c81d: Gigantic set of mostly branch tiles (Darby) 10(7 days ago, 177 files, 908+ 647-) 13https://github.com/crawl/crawl/commit/011df5c81d52 15:12:17 03regret-index02 07* 0.34-a0-656-gd42c42303d: Add a few remixes of Darby's tiles 10(4 days ago, 19 files, 151+ 113-) 13https://github.com/crawl/crawl/commit/d42c42303d61 15:12:17 03regret-index02 07* 0.34-a0-657-gcdb8317334: Highlight V:$ and Zot:$ staircases (Darby) 10(3 days ago, 10 files, 54+ 3-) 13https://github.com/crawl/crawl/commit/cdb8317334f3 15:12:17 03regret-index02 07* 0.34-a0-658-gbd457bb66c: Let Mara dual-wield with a new tile 10(3 days ago, 6 files, 21+ 8-) 13https://github.com/crawl/crawl/commit/bd457bb66c68 15:12:17 03regret-index02 07* 0.34-a0-659-g34913017d9: Update some branch / portal descriptions (Darby) 10(25 hours ago, 1 file, 9+ 7-) 13https://github.com/crawl/crawl/commit/34913017d9c6 15:12:17 03regret-index02 07* 0.34-a0-660-gbf16c74b88: Add a little more commentary to portal_generation_order 10(20 hours ago, 1 file, 1+ 1-) 13https://github.com/crawl/crawl/commit/bf16c74b88d1 15:12:17 03regret-index02 07* 0.34-a0-661-g35c39e4809: New monster: bes kemwar, for early D 10(19 hours ago, 39 files, 168+ 28-) 13https://github.com/crawl/crawl/commit/35c39e4809b3 15:12:17 03regret-index02 07* 0.34-a0-662-g0751984624: Name the HP*2 status back to "doubled vigour" 10(18 hours ago, 18 files, 30+ 30-) 13https://github.com/crawl/crawl/commit/075198462499 15:12:17 ... and 2 more commits 15:28:09 04Build failed for 08master @ 5d8561da 06https://github.com/crawl/crawl/actions/runs/17111551395 15:29:18 03dolorous02 07* 0.34-a0-665-g657cd509c9: Fix spelling. 10(6 minutes ago, 1 file, 1+ 1-) 13https://github.com/crawl/crawl/commit/657cd509c9e9 15:32:29 03dolorous02 07* 0.34-a0-666-g56b9c373c2: Fix typo. 10(66 seconds ago, 1 file, 2+ 2-) 13https://github.com/crawl/crawl/commit/56b9c373c2e9 15:41:33 Unstable branch on underhound.eu updated to: 0.34-a0-652-g878998d1fe (34) 15:43:06 03dolorous02 07* 0.34-a0-667-g646a485ab0: Add missing ENDMAP. 10(59 seconds ago, 1 file, 1+ 0-) 13https://github.com/crawl/crawl/commit/646a485ab01b 15:43:21 04Build failed for 08master @ 657cd509 06https://github.com/crawl/crawl/actions/runs/17111856727 15:48:25 03dolorous02 07* 0.34-a0-668-g2cf2f3ce0b: Fix spelling. 10(59 seconds ago, 1 file, 10+ 10-) 13https://github.com/crawl/crawl/commit/2cf2f3ce0b7e 15:59:14 04Build failed for 08master @ 56b9c373 06https://github.com/crawl/crawl/actions/runs/17111898133 16:02:15 03dolorous02 07* 0.34-a0-669-ge8778ee6be: Fix warning. 10(2 minutes ago, 1 file, 1+ 1-) 13https://github.com/crawl/crawl/commit/e8778ee6be92 16:09:51 04Build failed for 08master @ 646a485a 06https://github.com/crawl/crawl/actions/runs/17112088918 18:08:45 New branch created: pull/4740 (1 commit) 13https://github.com/crawl/crawl/pull/4740 18:08:45 03RypoFalem02 {GitHub} 07https://github.com/crawl/crawl/pull/4740 * 0.34-a0-649-g99eb28645c: Remove a duplicate yaml key (no functional changes) 10(19 hours ago, 1 file, 0+ 1-) 13https://github.com/crawl/crawl/commit/99eb28645c94 18:58:40 03RypoFalem02 {gammafunk} 07* 0.34-a0-670-gee87135f2d: Remove a duplicate yaml key (no functional changes) 10(20 hours ago, 1 file, 0+ 1-) 13https://github.com/crawl/crawl/commit/ee87135f2de7