01:09:35 <11O​dds> These are just mockups, but in principle how do people feel about these interfaces for spell failure skill guidance? Personally, I'd find the "1 skill level" mostly useful for working out how much my armour/shields training matters, and the "target" one mostly useful for deciding whether to train up a far off spell. The 10% threshold is a bit arbitrary but seems like it conveys most of what you want. 01:09:35 <11O​dds> https://cdn.discordapp.com/attachments/747522859361894521/1559840574633283666/image.png?ex=6ad08a3d&is=6acf38bd&hm=7035e3db5a3faa9fbb44088b3d68543e59722a89128dd3013bd57d99174afa90& 01:09:36 <11O​dds> https://cdn.discordapp.com/attachments/747522859361894521/1559840575086141480/image.png?ex=6ad08a3d&is=6acf38bd&hm=9cb236e82971d0818024efb198cf6e8a0cd96e0541b1722e0acc0bbfada3ebbd& 01:09:36 <11O​dds> https://cdn.discordapp.com/attachments/747522859361894521/1559840575535194122/image.png?ex=6ad08a3d&is=6acf38bd&hm=269973132009a6ece956c466ebd073cc63dec202cf7d8a659fb2d4dadc800456& 01:22:01 <04d​racoomega> The skill one just looks straightforwardly useful and I can't think of any quibbles with it at the moment. (Even if it's less useful than the other one, in theory.) The 'training required to get this to 10%' one is potentially far more useful at answering a relevant gameplay question, but I admit I find myself uncertain about how much the 'optimal path' here comes off sounding like coaching advice - especially with it 01:22:01 incorporating non-spell skills in the mix (and presumably armour, too, in many cases). I don't know if I can fully articulate why that feels a little more 'hands on' about telling the player what they should be doing than usual, but it does to me. I'm not sure I have any immediate suggestion of how to approach that differently, though. I do think that 10% is a fine arbitrary cut-off for such things, though, for what it's worth. 01:24:01 <04d​racoomega> I think when I was first mulling over such a feature myself, it probably wouldn't have even considered spellcasting skill, but purely the school skill(s) in question. (Which isn't optimal training advice, of course, but can at least give a sense of approximately how far away something likely is.) 01:24:16 <04d​racoomega> This is, of course, a less fully 'correct' answer in some ways 01:24:50 <11O​dds> Yep, certainly I understand the feeling in the second paragraph 01:24:51 <04d​racoomega> I guess maybe what I had been looking for was just more of a sense of "Approximately how long would this take" and less "Exactly how I should approach that", if that makes any sense 01:25:22 <11O​dds> Yeah, I think restricting this to only spell schools would keep most of the utility (for me) 01:25:32 <04d​racoomega> (And anything over a few XLs might already be losing some accuracy with potential equipment changes and stat gains and so forth) 01:25:44 <11O​dds> Usually I'm wondering "is the translocations investment for gavotte from scratch mild, moderate or silly" 01:25:56 <04d​racoomega> Yeah, similar here, kinda 01:26:04 <11O​dds> And for that I don't really need to be told about details of spellcasting and shields 01:31:33 <11O​dds> (Will leave this here for a bit to gather any more opinions) 02:33:22 <12g​e0ff> The "To lower your failure rate with this spell to 10%, it would be quickest" screen is missing "or just join Chei" / :kobold: 02:33:41 <12g​e0ff> But seriously, all this looks extremely useful. Although, the optimal advice feels like a 💰 subscription-tier feature, in a good sense. 02:33:57 <12g​e0ff> It'd be great to have "press (s) to set these targets" there. Because otherwise the player needs to jump back and forth from the spell description to the m screen to set all of them. 02:36:15 <11O​dds> Chei would not like being described as a quick option 02:36:25 <11O​dds> What do you mean by the second sentence? 02:38:39 03kate-02 07* 0.35-a0-1290-gf171d4b144: Fix typo and tweak wording 10(3 minutes ago, 1 file, 1+ 1-) 13https://github.com/crawl/crawl/commit/f171d4b1441e 02:42:33 <12g​e0ff> If DCSS were a pay-to-win game, this feature would be worth a mention on the list of perks provided by a paid account. 02:49:57 <12g​e0ff> Anyway, it is useful and I definitely would be interested to know the optimal path to the 10% failrate for a spell. The training costs are already displayed on the skills screen, so you kinda have a way to know what to train. But you need to check these costs after each new skill level, so the utility of them alone is not that great. On the other hand, combining these training costs and the encumbrance formulas and giving the 02:49:57 player the optimal skill targets feels more user-friendly. 03:33:14 Experimental (bcrawl) branch on underhound.eu updated to: 0.23-a0-5261-gd9800d219b 06:48:13 03dolorous02 07* 0.35-a0-1291-g44cd51e735: Adjust a spellbook name (powerbf, Predelnik). 10(22 hours ago, 5 files, 6+ 6-) 13https://github.com/crawl/crawl/commit/44cd51e73576 07:03:38 04Build failed for 08master @ 44cd51e7 06https://github.com/crawl/crawl/actions/runs/38574829500 07:23:47 03kate-02 07* 0.35-a0-1292-g70712b7926: Improve Revenant ability description 10(4 hours ago, 1 file, 5+ 1-) 13https://github.com/crawl/crawl/commit/70712b79267a 07:23:47 03kate-02 07* 0.35-a0-1293-g71cad4733a: Fix a changed spellbook name in vaults 10(9 minutes ago, 2 files, 2+ 2-) 13https://github.com/crawl/crawl/commit/71cad4733aee 07:41:04 03dolorous02 07* 0.35-a0-1294-gf90bb6af1c: Rewrap a mutation description. 10(69 seconds ago, 1 file, 2+ 2-) 13https://github.com/crawl/crawl/commit/f90bb6af1c24 07:49:42 03dolorous02 07* 0.35-a0-1295-gfabe31b45b: Fix description of earthen dragon vein. 10(4 minutes ago, 1 file, 1+ 1-) 13https://github.com/crawl/crawl/commit/fabe31b45bfe 10:09:46 <04d​racoomega> Is this supposed to be an endorsement or a criticism? ^^; 10:09:54 <04d​racoomega> Anyway, I think there's a lot of inherent fuzziness regarding the combined utility of each of those skills (and all the other things they affect, which are only partially overlapping) that it feels overly prescriptive to me for the game to be like "Here's what you ought to train." Whether or not this is something that is already mathematically calculable (albeit with considerable difficulty) 10:42:19 <11O​dds> Yeah I’m quite torn. I wonder if there’s just a way to word it that makes it clearer this isn’t prescriptive…. I wouldn’t be totally happy to have a version of this I think is a bit less useful and hides more 11:03:01 New branch created: fishingstaff (1 commit) 13https://github.com/crawl/crawl/tree/fishingstaff 11:03:02 03hellmonk02 07[fishingstaff] * 0.35-a0-1290-gdf154e176a: Require attunement for magical staves 10(5 minutes ago, 10 files, 63+ 27-) 13https://github.com/crawl/crawl/commit/df154e176adb 11:03:02 New branch created: pull/5512 (1 commit) 13https://github.com/crawl/crawl/pull/5512 11:03:02 Branch pull/5512 updated to be equal with fishingstaff: 13https://github.com/crawl/crawl/pull/5512 11:05:51 -!- Slacking is now known as slacking 11:06:52 Re the above, I think the "path to 10%" is far too likely to be an outright fib given how other circumstances may change. What I'd be interested to know is the spell failure / XP ratio of appliable skills _right now_ - trivial to know (if you know the formula) for schools and Spellcasting, but opaque to know how Shields or Armour stack up by comparison. 11:14:11 <11O​dds> That’s the other part, on the skill menu. I think both are useful in different ways; the 10% thing answers “how far away is this spell from being castable” - a question I often want to ask. 11:20:21 love the first two, not sure about the one shown on the map. feels like it plays the game for you a bit too much 11:21:22 would be cool if you could change the number of levels the diff is for with +/- or similar as well 11:22:03 the one shown in the spell description* 11:25:01 <11O​dds> Interested in other options to answer the thing the spell description tries to do (“how far away is this spell”) 11:30:10 <02D​arby> without currently weighing in on the overall concept, it feels like it would make sense for 10% to be an rc-modifiable default 11:33:17 <08o​____0> for whatever it's worth I don't feel like the game doing the numbers for you to show efficient leveling (maximizing for one variable, castrate) is perscriptive. There's obviously other considerations 11:36:26 <08o​____0> I wouldn't take it as a suggestion that I should stop at 10% cast rate or that training these skills is the best way to play. Just that it's the most xp efficient way to reach an arbitrary fail rate with your current stats/gear. 11:39:14 <08o​____0> Like a big thing is that you have multiple spells in different schools so while you often want to focus on one spell at a time you're still thinking about other spells and those sorts of considerations aren't taken into account (in addition to other considerations like the value of non-spell skills) 11:47:20 <04d​racoomega> Trying to articulate my exact reservations about this (which I feel only grow stronger the more I think about it) and admittedly struggling a little. I guess I feel the UI shouldn't have 'opinions' or 'solve' anything for a player, and optimizing 'minimally required training' among potentially 7 different skills is actually very non-trivial math. Like, it's true that this is objective information that a player could 11:47:21 theoretically calculate, but also until this point, the math involved usually made it untenable for anyone to actually do so (certainly without external tools, and I'm not even sure any exist that work this well). It doesn't feel like the answer to a simple direct question like "What is my accuracy or success rate at the moment?" where it seems unobjectionable that the game just tell you the answer. I wish I could articulate this more clearly than 11:47:21 what is admittedly close to "It feels wrong.". Like, too close to the game giving you the answers to questions that the player's own skill is supposed to be handling. Obviously it's not entirely doing that, since at minimum, the player is choosing what spell they want to prioritize, which is arguably the more important question in the first place. I dunno.... 12:00:18 <11O​dds> More substantive thoughts in a bit, but is it 8 skills? Or maybe you can’t get shapeshifting and armour and shields to all matter together 12:01:02 I saw the other part on the skill menu but it makes the information I want quite well hidden 12:02:29 <04d​racoomega> Oh, I guess you can in some situations. Jademantle has a skill-based boost at low skill, and doesn't block either body armour and shield, so yes 12:03:07 <04d​racoomega> Dragon Veins success rate could theoretically be boosted by all of spellcasting, fire, ice, earth, air, shapeshifting, armour, and shields 12:08:39 <04d​racoomega> I also thought about Charlatan's Orb for a moment, but you can't have that and a shield on at the same time. (Yet) 12:15:58 <08o​____0> I think it's probably fine if it only considers spell schools.... needs special logic for gn/dj too... and then manuals to take into account... 12:35:42 Gn presumably doesn't care, at least, given it has no way to act on the information short of buying manuals 12:42:42 RepoMan (L23 VSSt) ASSERT(in_bounds(source)) in 'beam.cc' at line 682 failed. (source = (19,69)) (Vaults:3) 12:42:46 <11O​dds> So yeah, what I'm struggling with is: * I think some version of this is useful. I would regularly use it to decide whether to aim for a particular spell. * Totally understand the concerns about the UI "solving" too much. I think I feel them a little less strongly, because I don't think calculating how best to reduce the spell failure is a particularly interesting problem, or actually the one you want to solve most of the time. But if 12:42:47 players read this as "this is the way to skill to head for this spell", that would be a bad thing both in reducing decision making and the recommended decisions are usually bad. * The version that only considers spell schools is a little weird to me. I do think it would just be less informative to me than the complete version, and it seems odd to me to implement that less useful version because we're concerned about whether players will overly rely 12:42:47 on the information in the more useful one (Gnolls and manuals and such I'll leave aside for now as I don't think any of them are dealbreakers) 12:43:12 <11O​dds> !crashlog RepoMan 12:43:17 <04C​erebot> 2. RepoMan, XL11 TrFi, T:4508 (milestone): https://underhound.eu/crawl/morgue/RepoMan/crash-RepoMan-20240121-142745.txt 12:43:48 <04d​racoomega> Oh, 0.31 12:43:54 <11O​dds> That is old though 12:44:03 <11O​dds> Not the version, the crashlog 12:44:07 <04d​racoomega> Oh 12:44:20 <04d​racoomega> So it is 12:44:31 <11O​dds> !crashlog 12:44:34 <04C​erebot> 23865. RepoMan, XL23 VSSt, T:42887 (milestone): https://underhound.eu/crawl/morgue/RepoMan/crash-RepoMan-20261014-194241.txt 12:44:38 <11O​dds> Weirdly that is the one 12:45:26 <04d​racoomega> Oh, I see the issue 12:45:57 <04d​racoomega> In order to get the multibeams to cover the proper spaces, it considers the origin point of them to be one step 'behind' where they are firing. And in this case, it rewound into the level border. 12:46:20 <04d​racoomega> But like... that actually is where the origin point needs to be for the path to be correct 12:46:42 <04d​racoomega> So I think actually the assert is overly aggressive now 12:48:23 <04d​racoomega> Maybe map_bounds(source) instead (since I don't actually think anything bad happens to fire 'from' the map border) 12:49:50 <04d​racoomega> (This does mean that (0, 0) sources stop asserting here, which could sometimes be indicative of an actual problem, but I'm not sure that warrants more complex handling here?) 12:51:49 <11O​dds> Seems reasonable... bit hard to predict if something elsewhere expects it to be in bounds 12:54:34 <11O​dds> (On the spell skills thing, I think for now I will just do the uncontroversial change, until/unless we work out a design for the other one that we are mostly all mostly happy with) 13:01:39 <04d​racoomega> I don't want to just be a stick in the mud over something where some form of it would definitely be useful 13:07:02 <11O​dds> Definitely not just you sticking in the mud, I was somewhat feeling the same way and others here said similar things 13:35:44 <06p​leasingfungus> it seems like this would be useful, certainly, but a button to instantly reach level 27 would also be pretty useful 🙂 13:35:56 <06p​leasingfungus> it's an affordance that would change the way that players interact with the game 13:36:26 <06p​leasingfungus> the interesting (and harder) question is, would that new mode of play be more fun? 13:41:26 <11O​dds> Yeah... and I'm really very sure it would be for me - because having knowable, important information be opaque either leaves me feeling bad for not being able to make informed decisions or wastes my time piercing the opacity. But I can readily imagine why it would be less fun for some players, if they followed it too hard. 13:44:10 My opinion is not worth much but I can say that I reach out for dcss doodad to get spell failure training informations and I know two other player who do the same. Having it be built in would just be less hassle 13:48:30 <04d​racoomega> This is not meant as judgement on this quandry one way or another, but I think external tools existing which can answer a question and the game itself seemlessly providing that same answer are not equivalent. The additional effort and 'externality' of 3rd party resources is tonally different in a way from what the game itself facilitates in a way that I think can be consequential. (Not just because we go from 'a small subset 13:48:31 of motivated players' having access to this versus 'everyone, whether they are new to the game or not' but that it also comes off as a sort of endorsement of how people ought to be interacting with a system, I think.) 13:49:54 <04d​racoomega> In the same way that there are numbers and formula that we deliberately expose in-game and those we do not (even though the source is open and so those numbers are never hidden, and even the wiki may also provide them.) 13:50:40 <04d​racoomega> As some subjective combination of what we feel are most important, have the best signal-to-noise ratio or what we want to encourage players to be thinking about by default 14:31:13 03DracoOmega02 07* 0.35-a0-1296-gca9efb6cd3: Track flux form in the form XP table 10(54 minutes ago, 1 file, 9+ 2-) 13https://github.com/crawl/crawl/commit/ca9efb6cd3ba 14:31:13 03DracoOmega02 07* 0.35-a0-1297-gc4298273a4: Add a function to check a player's primary form since a given XL 10(4 minutes ago, 2 files, 29+ 0-) 13https://github.com/crawl/crawl/commit/c4298273a49c 15:19:44 Should "Augurer" be "Augur" in rand_wpn.txt? A 1913 dictionary listed the former as being an obsolete spelling of the latter; the modern dictionaries I checked don't include it at all. 15:30:06 M-W has it (albeit that's American and it is listed as obsolete). Wiktionary has it. The OED has it. While I can't see why it _shouldn't_ be changed, I fear your search was not extensive. 15:43:49 Unstable branch on underhound.eu updated to: 0.35-a0-1297-gc4298273a4 (34) 15:44:05 Chambers (1977) has it as from Shakespeare; _Asterix and the Soothsayer_ uses it, although the English translation is from 1975 so neither of these is perhaps _very_ current usage 16:00:02 also does any of this apply to Australian English? 16:01:10 I think I'd be willing to take a bet that the Aussies aren't appreciably different from British English on this specific word :-) 16:01:30 probably 16:02:08 the 1913 on the other hand is probably spurious unless you actually found a British/Australian one instead of the usual web1913 which is US English iirc 16:03:22 (yep) 17:30:43 An online version of the Australian Oxford Dictionary went straight from augur to augury, with no variants listed. It focuses more on contemporary use than something like the OED does. 22:37:39 Unstable branch on crawl.develz.org updated to: 0.35-a0-1297-gc4298273a4 (34) 23:04:34 Windows builds of master branch on crawl.develz.org updated to: 0.35-a0-1297-gc4298273a4 23:41:13 Unstable branch on cbro.berotato.org updated to: 0.35-a0-1297-gc4298273a4 (34) 23:59:39 Monster database of master branch on crawl.develz.org updated to: 0.35-a0-1297-gc4298273a4