00:01 Tangles: could you send me an invite to push to the 3.4.3 repo please? 00:01 OK, so I have written enough of the proposal that spoilers should no longer be an issue. 00:01 FIQ: Message from Tangles at 2017-09-29 23:46 EDT: the 3.6.1 we are running here is vanilla except for livelogging and the achievement recording bugfix we discussed on github a while back, which the devteam still have not merged (or otherwise addressed) 00:01 FIQ: Message from Tangles at 2017-09-29 23:47 EDT: Could probably take that out again until it's fixed upstream, I'm just surprised it hasn't been yet. 00:01 Currently, what I lack in terms of finalizinf the document is general thoughts. This include some tweaks I want to do that affects the entire dungeon, 00:02 that can reasonably be argued part of Dungeon Overhaul 00:03 I did not intend to write a document in first place until this entire thing was brought up again, but felt like I could write one to have my personal thought on things nicely, too. 00:04 Tangles/FIQ: any information on that "achievement recording bugfix"? the word "achievement" doesn't appear in any email sent via the bug tracker 00:04 I haven't send a mail related to this. 00:04 ais523: Did my report regarding sustain ability go through? 00:04 Just checking 00:05 FIQ: yes 00:05 OK, good 00:05 we know of at least one other email that got stuck over the past couple of days, though 00:05 so I fear that this is a really common problem :-( 00:06 ouch 00:10 FIQ: I know, I'm just ribbibg you. 00:10 -!- Grassy has quit [Ping timeout: 240 seconds] 00:10 Tangles: ? 00:10 -!- Grasshopper has joined #hardfought 00:10 -!- mode/#hardfought [+v Grasshopper] by ChanServ 00:10 ribbing 00:10 joking with you 00:10 teasing 00:11 FIQ: About not submitting your dungeon propsal yet :) 00:11 ais523: I submitted a PR on github ages ago (some time before it became widely known that this is not a good way to communicate with the DT) 00:12 ais523: I'll send an email. 00:12 right, I don't even have a github account 00:13 and I'm not sure who has access to the github mirror of NetHack or even whether it's an official devteam mirror or something unofficial 00:13 (the official git repo is on sourceforge) 00:13 paxed assured me it's official. but yes, all the links on nethack.org point to sf. 00:15 well, the sf one is the main one 00:15 the github one is official, but it's just a mirror 00:22 I think DT moved their main repository to nethack.org itself 00:22 https://rodney.nethack.org 00:22 on this place 00:23 oh, could be 00:24 but nethack.org is a wrapper for sourceforge (currently, we're considering migrating away because sourceforge has been unreliable recently) 00:24 Ahh 00:30 K2: I'll check with paxed to see what his thoughts are on 343nao legacy stuff. We can always merge up later if it makes sense. 00:31 rodney.nethack.org is hosted on nao, right? 00:31 nope 00:31 wrong rodney 00:31 err, not rodney.nethack.org 00:31 dtype: awesome 00:31 that's a pdp-8? 00:31 I think *something* was going to be hosted on nao for nethack or something 00:32 K2: but I view paxed as the official maintainer of 343nao, so will want to see what he wants there 00:32 [nd] Jendic (Val Hum Fem Law) had Excalibur thrown at her by some watery tart, on T:2606 00:32 he might hand it off, but just want to check 00:32 understood 00:32 wow, first try 00:33 Jendic \o/ 00:33 mmm bitcoin back up above $4000 00:47 http://home.fiq.se/overhaul.txt 00:48 I am not much for writing essays. But this is what I ultimately came up with. The vast majority of this is ideas I've had for some time but haven't written down concretely anywhere. 00:49 -!- Jendic has quit [Quit: *poof*] 00:51 FIQ: huh, we both put Vlad in a branch accessed from the penultimate level of the Mines 00:51 that's an oddly specific thing to match up on 00:51 hah 00:51 interesting 00:51 ais523: the change in my case was inspired by what dNetHack does 00:52 but I think its surface level thing is a bit silly in a more vanilla NetHack context 00:52 there's a surface level in my version but it's mostly just for flavour and somewhere to put a tutorial 00:52 there's no actual gameplay there if you don't need the tutorial 00:52 (dNetHack has an upstair in the Gnomish Mines that lead to another level of the Gnomish Mines, then to a surface level containing Vlad's Tower with a Cerberus acting as a guardian of sort to guard the tower proper) 00:53 Ah 00:53 Interesting 00:54 hmm, based on what people are saying about the quest unlock 00:55 maybe I should make it viable to do the Valley before the Castle (e.g. by putting the Valley←→Castle stairs on the /left/, the same side as the upstairs) 00:55 I have no issues with such a change 00:55 so that there'd be an option to grind to 14 (if you didn't have the appropriate !oGL / wraiths for some reason) without doing the Castle first 00:56 My solution is to just decrease the XL requirement 00:56 Making it an actual choice rather than "did you find enough GL" 00:56 does your CoMR-for-"sokoban" in your proposal have MC1 or MC3? 00:56 MC2 00:56 if it's the 3.4.3 version of the CoMR I think it's too good as a guaranteed option 00:56 FIQHack swaps MC of displacement and CoMR 00:57 FIQ: the problem there is that although you're adding the choice to do the Quest "early", what you're actually just doing is removing one step in the ascension sequence (finding some way to hit XL14), and also making it so that quest-home and rest-of-quest are not really separate 00:58 Hmm 00:58 you and jonadab both suggested having tributes to things other than Rogue in the rotation 00:58 I can see your point with that 00:58 that strikes me as something that the vanilla devteam might well be in favour of 00:58 ais523: Yeah, we have discussed this earlier pre-proposal, and come up with it independently 00:58 but getting it to work on every platform could be painful 00:59 out of the major roguelikes, ADOM is probably easiest if you want a full interface replacement on tty 00:59 if you just care about the level generation and map symbols, DCSS would be quite easy 01:00 -!- Webmant has quit [Read error: Connection reset by peer] 01:01 [nd] k2 (Wiz Hum Mal Neu), 2678 points, T:3796, killed by a giant ant 01:02 FIQ: haha, three for three on making Fire an enforced speedrun plane 01:02 although somehow, all three proposals for that are different 01:03 Heh :P 01:03 Vanilla is also arguably making Fire a "speedrun" plane 01:03 with the threat of Archonstorms 01:03 right 01:03 maybe that's what inspired us 01:03 perhaps it's just because fire is a good thing to make timers with, though 01:04 right 01:08 Hm 01:08 I seem to have accidentally removed one point in my plane of fire suggestions 01:08 let me address that 01:10 pushed 01:11 -!- greqrg has quit [Quit: Disconnected] 01:11 -!- Grasshopper has quit [Remote host closed the connection] 01:13 ais523: link to your proposal? 01:13 I want to read it :P 01:16 this is probably the most convenient one atm: https://nethackwiki.com/wiki/User:Ais523/Dungeon_Overhaul_Proposal 01:16 thanks 01:17 (it was originally posted in a different format but aosdict put it on the wiki) 01:18 interesting, we both suggested making bigroom guranteed 01:19 I don't see why you wouldn't 01:19 but letting it appear at a bigger depth range makes sense 01:19 this is the kind of reason as to why I avoided specifically reading your proposal -- on more than one occasion have I found that we agree on a lot of things 01:19 ais523: Mhm 01:19 so that you can try it with many different monster sets, and so it's more of a surprise when it happens 01:19 Yeah 01:26 ais523: Leon has a town generator 01:26 on bilious 01:26 And 3.6.x DT has proven that they like his work (more than one of his patches has been incorporated into vanilla) 01:27 Leon? 01:28 Leon Arnott 01:28 goes under L on bilious 01:29 ais523: we both mess with thrones, allthough me more extensively 01:29 you only to vlad throne 01:30 oh 01:30 L is alive?? 01:32 ais523: you can push boulders while flying 01:32 Elronnd: not that I know of 01:32 it's an old patch 01:32 flying's sufficiently hard to get that I'm OK with someone sequence-breaking Sokoban with it 01:33 same with controlled levport 01:34 Are you aware of ADOM's Arena? 01:34 While the general mechanics are different, it's a similar idea in practice 01:37 ais523: Is there a particular reason as to why you can't use the crater clearing as a means of accessing Plane of Earth? 01:37 Or siilar 01:37 *similar 01:37 FIQ: flavourwise? or gameplaywise? 01:37 flavourwise at the moment it's just that the portal isn't there 01:38 I can't see a particular balance reason to either allow or disallow it 01:38 The game remarks that you should return to the surface 01:38 although if for some reason it's considered to be an optimal way out compared to just using Dungeons, it'd force people to solve Sokoban 01:38 (the Arena lets you up there but only once) 01:38 So I think it would make sense to allow it from both surface entrances (because why not? :P) 01:38 but sure, I have no problem with that 01:45 Perhaps it's an antimagic gas that drains charges from items and makes noncharged items nonmagical; a bag of holding suddenly becoming a sack could be a real problem. 01:45 So basically slash'em lethe water 01:45 Except less amnesia and more potent vapours 01:47 right, I was inspired by lethe 01:47 but it's not exactly the same 01:48 [nd] sluggoman (Wiz Hum Mal Neu), 6008 points, T:5630, killed by a leocrotta 01:50 ais523: does tower of madness allow praying? 01:51 or does it count as being in gehennom 01:51 it's not in gehennom 01:51 branches off gehennom are outside in this proposal 01:51 mostly so that you have somewhere to pray 01:51 if you badly need to for some reason 01:52 no free hand probably being the obvious case 02:00 ais523: One thing I've considered at times (but never wrote as part of overhaul) was to make Moloch give random divine retributions while you carried the Amulet in Gehennom 02:01 I'm not sure it really matters for gameplay purposes who's doing the retributing 02:01 (note: I'm not 100% sure that's a real word, but the only other way to form a gerund would be "retribution" and although that's probably correct, it'd not give the impression I want in this case) 02:06 (because I needed an unambiguously active verb) 02:10 ais523: you could make Riders wander around close to the temple rooms and the temples themselves sealed until a Rider is killed for the first time 02:10 which unseals a *different* temple from the one the Rider was lurking around 02:10 if youi want the player to traverse more 02:10 *you 02:10 FIQ: but then you'd backtrack as soon as you killed the Rider 02:10 which seems unrealistic somehow 02:11 -!- raisse has joined #hardfought 02:11 -!- mode/#hardfought [+v raisse] by ChanServ 02:14 -!- ais523 has quit [Quit: quit] 02:32 -!- nht has joined #hardfought 02:37 -!- elenmirie has quit [Ping timeout: 258 seconds] 02:40 -!- elenmirie has joined #hardfought 02:40 -!- mode/#hardfought [+v elenmirie] by ChanServ 02:40 -!- elenmirie_ has joined #hardfought 02:41 -!- elenmirie has quit [Client Quit] 02:41 -!- elenmirie_ is now known as elenmirie 02:41 -!- mode/#hardfought [+v elenmirie] by ChanServ 03:07 [slex] Yuring (Mah Dro Fem Neu), 2879 points, T:2913, killed by a monster (dancing flame) 03:08 [slex] Yuring (Mah Dro Fem Neu), 0 points, T:21, killed by a river moccasin 03:52 why isn't a river moccasin a monster (river moccasin)? 03:55 [nd] Raisse (Wiz Gno Fem Neu) had Magicbane bestowed upon her by Thoth, on T:1297 03:55 that's quick 03:55 one sacrifice to convert the altar and one to get magicbane 04:07 lol, the minetown deli had both holy and unholy water 04:07 (but no plain water) 04:15 [nd] Raisse (Wiz Gno Fem Neu), 3746 points, T:5065, killed by a soldier ant 04:53 -!- Grasshopper has joined #hardfought 04:53 -!- mode/#hardfought [+v Grasshopper] by ChanServ 05:45 -!- Grasshopper has quit [Read error: Connection reset by peer] 05:45 -!- Grassy has joined #hardfought 05:45 -!- mode/#hardfought [+v Grassy] by ChanServ 06:01 -!- raisse has quit [Ping timeout: 258 seconds] 06:17 -!- raisse has joined #hardfought 06:17 -!- mode/#hardfought [+v raisse] by ChanServ 06:22 ooh 3x13 hardware store 06:24 7 tin whistles and not a single magic whistle, oil lamp and brass lantern but no magic lamp 06:24 but a 100zm bag, tinning kit, stethoscope, towel and key 06:24 and a magic marker 06:25 can only afford part of the stuff but I have a trusty cat 06:25 * raisse puts everything in the bag 06:26 lol, dropping an unpaid bag with unpaid stuff in doesn't void the bill 06:27 oh, scratch that, I was carrying stuff from chests 06:28 apart from the 7 tin whistles, there are also 5 lock picks (and 2 keys which I've got 1 of) 06:31 argh, pesky hobbit 06:31 with a +4 dagger or something 06:39 !tell ais523 What happens if a player with flying goes < from the Crater? 06:39 Will do, jonadab! 06:39 [nd] elenmirie (Mon Hum Fem Neu), 12419 points, T:7468, killed by an air elemental 06:59 FIQ: Oooh, taking away existing protection any time you enter the Valley is devious, and players will complain bitterly, but I think I like it. It will certainly change the character of the late game. 06:59 It does make Gehennom-floor gold even more completely worthless, though. 07:01 jonadab: Is the protection masked or removed? 07:01 Mandevil: Message from FIQ at 2017-09-29 11:07 EDT: K2 pushed changes 07:01 Mandevil: That is unclear. 07:01 I can see arguments either way. 07:02 Although merely masking it means players _will_ want to still buy as much as possible after all, for the Planes. 07:02 I think it should be masked. 07:02 Prayer is not permanently removed either. 07:02 That makes good flavor sense. 07:03 "Your surface diety cannot help you down here in my realm, puny mortal." 07:03 Mandevil: Are you going to write up a Dungeon Overhaul proposal? 07:13 I don't think so. 07:13 :-( 07:13 I am working on the scoreboard now. 07:13 Ah. 07:13 Valid point. 08:13 [nd] Raisse (Wiz Gno Fem Neu) had Magicbane bestowed upon her by Thoth, on T:2539 08:18 [nd] Raisse (Wiz Gno Fem Neu) polymorphed her first object, on T:4028 08:25 would angband's vaults make sense in nethack? 08:25 -!- eady has joined #hardfought 08:27 good, not too much backscroll today 08:28 !tell ais523 Ugh, I personally really hate the proposal for alternate tribute levels and think Rogue isn't even that necessary. It's lots more development work for something that either doesn't affect or actively annoys players. 08:28 Will do, aosdict! 08:30 aosdict: Well, you know, we could give the tribute level a bit more gameplay relevance. Maybe the Moria level could have a Balrog, the Angband level could have Morgoth, etc. <-- EPI 08:30 FIQ: it's not clear whether you intend for the lava on Fire to be levitatable over? I'm treating the 1HP/turn damage as negligible, so I'm not sure what prevents the ring of levitation from being used to ignore the special mechanic completely. 08:31 FIQ: Yes, that could stand to be more clear; I wasn't sure if maybe you meant 1 _max_ HP per turn? Otherwise it's pretty meaningless. 08:32 (Also, problems arise when you make gold detection not work and put a timer on one of the planes. If you have to navigate by Amulet, there's not enough time. That can take a _long_ time.) 08:33 Personally, I'm not in favor of nerfing portal detection. Hunt For the Magic Tile is already done on vibrating square level, does not need to be repeated. 08:38 Actually, now that I say that, I'm tempted to do the opposite: make the portals on the Planes visible to the naked eye, so as soon as you can see the tile, you see the portal. 08:38 The tricky part is _getting_ there. In one piece. 08:44 YANI: being under water starts a short suffocation timer similar to strangulation; monsters with a grab-and-drown attack now grab and pull you under in a single turn, wetting your inventory and starting the timer. You must somehow get free of the attacker and get out of the water before you suffocate. 08:44 Lifesaving resets the timer. No more enexto() shenanighans from that quarter. 08:44 Pretty sure I already have one like that, but maybe not. Enough people seem to like it that I can see it catching on. 08:45 Players would complain loudly about the one-turn inventory wetting, but on the whole I think this makes water more _interesting_. 08:46 * jonadab might have a slight evil streak down the middle of his back. 08:48 jonadab, and a streak of epis with no nice patch idea 09:22 FIQ: I do agree with you that quest room-and-corridor fillers are boring. I'm wondering if it's worth creating some new filler algorithms to make them more thematically appropriate. 09:23 Like, for Archeologist, upper fillers would probably be jungle levels, and lower fillers would be dungeon/tomb levels. 09:26 But a more interesting tomb-like level, not room and corridor. 09:28 And the Tourist quest is supposed to be all in a city, so its cavern fillers don't make any sense. 09:29 Hmm, it should probably be populated with peaceful "human", dwarf and troll monsters, since it's Ankh-Morpork. 09:38 -!- nht has quit [Ping timeout: 240 seconds] 09:38 aosdict: Oh, yes, please, six levels full of peacefuls you have to wait for to move out of the way so you can progress. That will be so much fun. 09:39 EPI: Gehennom is full of peaceful monsters with passive disintegration attacks. 09:39 Well, there don't have to be that many of them. Just enough for flavor. 09:40 And if I were writing a city filler, it would not have single-wide streets everywhere, just the occasional alleyway. 09:43 -!- nht has joined #hardfought 09:45 aosdict: Ok, fair. 09:59 jonadab: I made it clear that the amulet was to be made more realible 10:01 and you can always use crystal balls 10:02 -!- noty has quit [Quit: WeeChat 1.9] 10:04 back in present time 10:04 jonadab: L's town generator patch made Tou filler into towns 10:06 aosdict: Oh. I seem to have mangled Fire's description in general. At least one remark vanished beforehand too (The Fire timer) 10:07 so I probably accidentally cut that out or similar 10:07 i intended to make it *possible* to levitate over the lava, but mae it give severe damage (10d8 w/o fire res, 10 with, per turn) 10:07 and inventory burns 10:08 I will (re)add this when back home 10:10 -!- Grassy has quit [Ping timeout: 258 seconds] 10:15 jonadab: I was thinking to remove it permanently. However, I can see the merit either way. 10:15 jonadab: Protection, that is 10:17 FIQ: was this for lava in general, or just on the Plane of Fire? Because, realistically, that would happen by hovering a few feet over lava, or when standing next to it. 10:17 aosdict: It was for Fire in particular 10:17 But I can see the merit in making it so everywhere. Moreso with fireproof WW 10:18 I think that all levels with a "lava plain" generation algorithm, like some Quest levels, should expand the floor spaces out by 1 tile in each direction, especially if lava becomes dangerous to levitate or walk over. 10:18 aosdict: Ooh, what if this stacked? 10:19 Like, if you had 6 lava tiles next to you 10:19 you'd take 6d8 or 6 damage 10:19 at endturn 10:19 also see http://tvtropes.org/pmwiki/pmwiki.php/Main/ConvectionSchmonvection 10:19 -!- raisse has quit [Ping timeout: 248 seconds] 10:19 You could also design the Plane of Fire with this mechanic in mind too 10:19 then 10:20 (like, having 1-wide "corridors" seperated with bigger rooms you can hang around in to wait out the occasional boiling 10:20 ) 10:21 I think I like this 10:23 This is already what I had in mind, roughly, but it gets much more interested if more close lava tiles give more damage 10:28 K2: I can't find the ais proposal talk in backlog 10:28 er, in irclog I mean 10:28 It was there previously 10:28 the little bit from yesterday or the longer discussion when he first posted it? 10:29 the longer 10:30 found it 10:35 bhaak: I don't mind Orctown per se. I mind that it destroys people's expectations of a nice place to stock up. With that said, I intend to go with my own fix that I came up with a while ago in fiqhack. Make Orctown a thing, but compensate by moving MT resources elsewhere. 10:41 regarding the talk on gnomes getting hostile remotely 10:42 ADOM has a nice feature IMO where if you eat a corpse of something in the view of a monster of that type, it becomes angry 10:48 05:10 I wanted to expand the "special thrones" idea more than I did 10:48 heh 10:48 I messed with them quite significantly 10:56 !tell ais523 Speaking of hall of fame bugs. jonadab managed to introduce a hilarious bug in Fourk a while ago that completely broke player speed. You would get progressively faster over time (*very fast*), and at T:10 you would have around 9 free turns with normal speed. However, every 11th turn, you would become paralysed for 11 turns. 10:56 Will do, FIQ! 11:09 -!- raisse has joined #hardfought 11:09 -!- mode/#hardfought [+v raisse] by ChanServ 11:33 FIQ: If you could guarantee a friendly Mines' End by having Orctown generate (and it generated _only if_ Orctown generated), I'd be on board with that. But I'm not sure how technically feasible it is to guarantee one level based on another level generating. 11:33 What if you levelport to Mines' End first, or fall through a hole past Minetown? 11:34 FIQ: spork has a feature where if a monster sees that you're immune to something, it stops using that attack (if possible) 11:34 eg a winter wolf breathes on you and you're frost resistant at 100%, it won't bother trying again 11:35 lorimer: Clever. 11:35 Makes monster use their turns more efficiently. 11:35 and relatively subtle, as it turned out. a lot of stuff doesn't have any special, so it just wails on you 11:36 you only noticed when elemental-type monsters quit using breath weapons 11:36 yeah, it didn't make sense for the monster AI to be -that- dumb. it still has to be dumb, i mean it -is- 10,000 on 1, but 11:36 Making the AI too good makes for awful gameplay. 11:37 "smart kobold", by jeff lait 11:37 -!- nht has quit [Ping timeout: 240 seconds] 11:37 Like fast ranged monsters staying out of melee 11:37 great example of that. players managed to beat it anyway, but for someone who isn't an inherently strong gamer and willing to invest a -lot- of effort in, the game is outright no fun 11:37 Crawl had this thing and it was removed. 11:37 -!- nht has joined #hardfought 11:38 Centaurs actually do try to stay away from you in spork, too, but they're only "fast" and the player can easily become 'fast' or 'very fast' 11:38 In Crawl it's worse because of being able to shoot at non-cardinal directions. 11:38 things with ranged weapons won't run up to you necessarily, either. again though, there's no way for them to -really- stay out of reach 11:39 Yeah, in Crawl it's messy. 11:39 Well, was. 11:41 lorimer: Hmm really? 11:41 FIQHack has that too, kind of 11:41 But not by monsters learning it, rather they do it more rarely 11:42 because adding monster memory was awkward IMO 11:42 yeah, don't look at the actual code in spork for how i did it 11:42 If a variant has already solved this problem, I might look at it 11:42 it has the oogly 11:42 Oh :P 11:42 maybe less oogly than yours, i dunno, but 11:42 it wasn't pretty either 11:42 lorimer: Also, let me guess. It only works for players? 11:43 yep. nightmarish to make it remember every possible critter it's seen 11:43 they'll remember trap -types- they saw other things hit (that was built in vanilla, but only if they hit the trap themselves) 11:43 lorimer: As said, FIQHack currently doesn't store memory of it, but the "API" makes it possible without changing the code that checks for monsters knowing resists 11:43 ahh, yeah 11:43 mine is much more hackly 11:44 Instead, FIQHack just makes it return "yes we know that the target is resistant" 75% of the time, unless it's an obivous property like displacement, speed or invisibility 11:45 aosdict: This is what I want to do, yes 11:45 aosdict: Also, the game has already decided what variant to generate in NH4+friends before you visit it. 11:47 FIQ: there are still some minor issues with it even if that's true 11:47 lorimer: Smart kobold isn't bad. It's a cool gimmick that makes for an unique game 11:47 it's a very well executed concept 11:47 But yes, applying Smart Kobold-esque AI to monsters in NetHack would be a terrible idea 11:47 but the gameplay itself is not "fun" :) 11:48 because the AI is human-grade, which humans really don't want in a heroic fantasy sort of thing 11:48 namely, the player has to traverse several more, more difficult levels before they get those resources 11:48 One thing I do is make *dragons* in FIQHack behave in ways somewhat similar to Smart Kobold's kobolds 11:48 Because they're supposed to be intelligent 11:48 So it makes sense for them to abuse all tricks in the book 11:51 Hmm, what if you could quantify monster intelligence? 11:51 for reference, this is how fiqhack dragons behave: http://sprunge.us/jCIh 11:51 Because right now, there is no way to keep kobolds intelligent for muse purposes, but cut them off from advanced intelligent AI behaviors. 11:51 aosdict: what do you mean? 11:52 FIQHack intends to do this by introducing attributes for monsters 11:52 True attributes, that is 11:52 Currently it's faked 11:52 with a bunch of special cases 11:52 "right now" = vanilla code 11:53 which is basically just flags, yes 11:53 aosdict: for intelligence 11:53 I intended to make it so that 1 is mindless, 2 is animal, 3-25 is general baseline 11:53 and the higher it is, the more clever the monster is 11:54 when ais523 actually does the fiqhack internals rebase, I might try to port it onto 3.6 11:54 this will likely fail 11:54 one issue with this is that intelligence is also tied to spellcasting 11:54 so maybe it should be a seperate per-monster value 11:55 *per-species 11:55 or perhaps you could have it as AI flags 11:55 one could even call it... permonst :P 11:56 kinda like how fiqhack has M3_LINEUP atm 11:56 which is set on spellcasters w/o melee attacks 11:56 to make them not pursue melee 11:58 (it's what makes fiqhack gnomish wizards surprisingly nasty) 12:01 lorimer: I should look more at spork mechanics at some point 12:01 I disagree with a lot of its gameplay changes (orctown!!) 12:01 hee 12:01 Nobody will ever agree with everything I did. 'cept me. 12:01 but like all variants it has its fair share of good ideas and stuff 12:01 and even me not always 12:02 It had a lot of subtle things that made it much better quality than it looked to people who only saw the broad-brush stuff 12:02 [IMO] 12:02 Right 12:02 The broad changes is what put me off sporkhack 12:03 But I should be fair and look into it more carefully regardless 12:03 All variants make different changes, and not everyone agrees with what changes people make (otherwise why would there be as many variants?) 12:03 :) Your call, of course 12:04 I could have marketed it better, but that'd have made me seem like I wanted to replace Nethack, instead of provide impetus for change 12:09 heh 12:11 -!- eady has quit [Remote host closed the connection] 12:17 hi 12:17 Hello K2, Welcome to #hardfought 13:10 ah 13:10 m_seenres 13:14 jonadab: pop quiz! what does monstseesu do? 13:15 lorimer: The code isn't that bad but ^ is a rather peculiar function name 13:15 Not entirely sure if I can even understand the name after knowing what it does :P 13:16 monst-sees-u 13:16 or the other was monster-seen-resistance 13:18 this is just for resistances yeah 13:18 713: monstseesu(M_SEEN_ELEC); 13:18 I think if I were to add this, I would probably do similar except how I currently do trinsic checks 13:18 yeah. monstseesu was intended to potentially also include anything a monster saw you do and needed to remember 13:19 it ended up only being resistances but 13:19 could've been ranged weaponry (it knows you have a bow, it won't stand on line with you) 13:19 Ooh, that's interesting yeah 13:19 or a bunch of other things, really. but $time/$energy in this case 13:20 For trinsic reminders, I think implementation-wise I'd make a linked list with a head in mextra with an array of timers (to make the memory expire eventually) and monid 13:21 This seems sane enough to track memory regarding monsters, not just the player but also other ones 13:21 Bonus points if you add the tracking for players too and reveal it for pets if asked for 13:22 yep. eventually most monsters -do- expire, but 13:22 pets don't 13:22 and benefit as well 13:22 (well, they hopefully don't :-) ) 13:28 [dnh] groovejumper (Wiz Hlf Fem Cha), 759 points, T:1744, killed by a homunculus 13:28 -!- nht has quit [Ping timeout: 258 seconds] 13:32 -!- greqrg has joined #hardfought 13:37 -!- rikersan has joined #hardfought 13:44 !tell ais523 How sane would this be, particularily with respect on save size effect? http://sprunge.us/AGYh 13:44 Will do, FIQ! 13:51 -!- rikersan has quit [Quit: Lost terminal] 13:54 aosdict: what I wanted to do if I made a friendly minesend 13:55 was to give the alignmed priest the luckstone and a "quest" that consists of you disposing of all the orcs in the old minetown 13:55 and when you have, reward the luckstone and a protection freebie 14:08 I was just thinking earlier about how there don't seem to be any standard "kill 20 monsters and return to me" miniquests in nethack 14:09 but a mines' end still wouldn't be "friendly" for an elf, for example, right? 14:10 and I wonder how you would get the priest to give you that guidance 14:11 by having him tell you? 14:11 ideally it would be some sort of mayor of minetown who asks you to do it, but he ought to be a gnome king or something 14:11 also 14:11 minetown is friendly for elves 14:11 don't see why minesend should be different 14:11 well, Gh aren't friendly 14:11 but the place in general is neutral 14:12 would he just tell you like a quest leader? 14:12 yes 14:12 kinda 14:12 (if responses are slow, it's because I'm IRCing from a bus. The wonders of modern technology.) 14:13 I am IRCing from a car 14:13 :P 14:13 I have about .5 to 2 second lag per keystroke 14:13 Heh 14:13 Get a handheld with a real keyboard :) 14:13 i've used this OP for over 4 years 14:13 I'm using my laptop. 14:14 wow 14:14 and yet you type slower 14:14 interesting 14:14 the lag is network latency, not actual typing speed 14:14 ah 14:14 ssh? 14:14 yeah. pretty good considering all the hops it needs to make. 14:15 I see 14:15 i use ssh to my VPS which is in the uS (I live in EU) every day more or less :P 14:15 It's where i have my weechat session 14:16 hmm, u.usa could be a variable name :) 14:16 Haha 14:16 not sure who decided that every player stat needed a u prefixed to it 14:17 Uh.. 14:17 To not have a million globals? 14:17 no I mean, why can't you just have u.hp instead of u.uhp 14:18 Oh 14:18 That is due to old compilers 14:18 Which reuqired field names across different structs to be different 14:18 That's why you have mx/my, ox/oy, ux/uy, etc 14:19 ...and of course, we probably still want to suport those old compilers, don't we. 14:19 :) 14:21 At some point, someone on the dev team is going to have to decide that the point of pride of supporting so many decades' worth of systems is not worth the rigidity it brings to the code. 15:05 What? My DEC Vax won't be supported anymore? 15:05 There'll be riots! 15:08 Mandevil: it's not even like you can't play nethack on those systems. You can still play any version up through . 15:19 Unacceptable. 15:19 That's right there with Windows 10 not being available (for my Vax) 15:23 -!- Grasshopper has joined #hardfought 15:23 -!- mode/#hardfought [+v Grasshopper] by ChanServ 15:40 What do you want to eat? [suB or ?*] 15:43 I think I'll stick with my beEF 15:44 you can have beEF on the suB! 15:44 and an egG with it 15:44 or if you're Swedish, ost (which is cheese in most scandinavian languages) 15:45 how about some fisH and chiPS? 15:45 yes! 15:49 that's nice to eAT 16:14 FIQ: well, in unnethack, you have the town which slightly compensate a missing minetown 16:16 -!- Grassy has joined #hardfought 16:16 -!- mode/#hardfought [+v Grassy] by ChanServ 16:19 The spell misses the owlbear. The owlbear turns into a giant mimic! 16:19 -!- Grasshopper has quit [Ping timeout: 255 seconds] 16:25 heh 16:27 -!- raisse has quit [Ping timeout: 248 seconds] 16:46 -!- ais523 has joined #hardfought 16:52 bhaak: Right, I don't have an issue with it as much as I have with vanilla 16:53 Since there's still a place you can restock and stuff in un 16:53 -!- Grassy has quit [Ping timeout: 248 seconds] 17:00 -!- callforjudgement has joined #hardfought 17:00 -!- ais523 has quit [Remote host closed the connection] 17:00 -!- callforjudgement is now known as ais523 17:00 FIQ: context? 17:00 ais523: Message from jonadab at 2017-09-30 06:39 EDT: What happens if a player with flying goes < from the Crater? 17:00 ais523: Message from aosdict at 2017-09-30 08:28 EDT: Ugh, I personally really hate the proposal for alternate tribute levels and think Rogue isn't even that necessary. It's lots more development work for something that either doesn't affect or actively annoys players. 17:00 ais523: Message from FIQ at 2017-09-30 10:56 EDT: Speaking of hall of fame bugs. jonadab managed to introduce a hilarious bug in Fourk a while ago that completely broke player speed. You would get progressively faster over time (*very fast*), and at T:10 you would have around 9 free turns with normal speed. However, every 11th turn, you would become paralysed for 11 turns. 17:00 ais523: Message from FIQ at 2017-09-30 13:44 EDT: How sane would this be, particularily with respect on save size effect? http://sprunge.us/AGYh 17:01 jonadab: presumably the same thing that happens if a player with flying goes < from home:1 17:02 FIQ: why is typ an array? also, I'd recommend a different word, "typ" is IIRC already commonly used with a different meaning (although maybe I'm remembering wrong) 17:02 oh right, in the phrase "otyp" 17:02 ais523: Mhm 17:03 Also it is an array because you can store memory about more than one thing 17:03 -!- Grassy has joined #hardfought 17:03 -!- mode/#hardfought [+v Grassy] by ChanServ 17:03 in that case shouldn't it be a bitfield? 17:03 ais523: The idea is to store the move you learned about it 17:03 you're unconditionally allocating space for an ordered subset of memtyps 17:03 So that it is obsoleted if it was long enough ago 17:03 oh, in that case the type of the array is wrong 17:04 it should have turncounts as elements, not typs 17:04 Er what 17:04 Yeah I messed that up 17:04 (I'm not sure if NetHack is consistent on whether turncounts are ints or longs) 17:04 SHould be an int 17:04 The *indexing* is by enum but the actual content is just an int 17:04 No idea what happened there 17:04 brainfart I guess 17:05 I'd say it could be fairly safe save-wise if it's uncommonly used (i.e. few monsters would have a long emem) 17:05 however you probably want an mtag for it 17:05 because the differ isn't going to like it otherwise 17:05 Hm 17:06 Yeah that makes sense 17:06 and I'm not sure the existing mtag code is precise enough to record something indexed by two different monster IDs 17:06 What do you mean? 17:06 right now an mtag is meant to be unique in the entire save file 17:06 Uh... 17:06 and I can't see a way to use the existing fields of struct mtag to uniquely define a specific emem argument 17:07 Ah I see what you mean 17:07 *specific emem structure 17:07 Inventory chains should be a similar case? 17:07 Just that it holds more data generally 17:07 if an item is on minvent it isn't anywhere else 17:07 so you can uniquely describe the struct obj via the object ID 17:08 ditto for other sorts of inventory chains 17:08 Well this wouldn't be anywhere else either besides on the monster? Not sure if I am understanding 17:08 apart from nexthere but luckily that doesn't need to be saved/restored at all 17:08 FIQ: the object ID of an object is unique across all objects 17:08 It's basically stored the same way as an inventory chain 17:08 whereas the referenced-monster-ID of an emem is not unique across all referencing monsters 17:08 it's simply a case of having different information in the struct itself 17:09 Oh 17:09 So the issue is a lack of unique identifier for memory chains? 17:09 Because that would be highly redundant except for save benefits 17:09 (And would have a much larger risk of wraparound than other IDs) 17:10 right; the much better fix here is to change struct mtag to be able to take two ID arguments 17:10 which would use a little extra RAM but otherwise have no drawbacks (in particular, the save wouldn't get larger) 17:10 What if it was "hashed"? 17:10 Which doesn't *eliminate* the risk of overlap 17:10 but reduce it 17:11 how large do mids get in a particular game? 17:11 Easiest way to check is vanquished creatures list 17:11 something like the ID of the referenced monster + 65537 * the ID of the referencing monster (using 32-bit unsigned arithmetic) should work 17:11 65537 is a really convenient number here because a) it's close to 65536 and b) it's prime 17:12 Note that mid can be 0 in case it is referincing the player 17:13 Not sure if that is a problem here 17:13 doesn't matter 17:13 ais523: What happens if there *is* collisions? 17:13 Does things break down? 17:14 Or will the collisions just diff poorly? (and this should be rare, and the struct isn't that big since it has a lot of zeroes) 17:14 it'll cause the differ to incorrectly diff the new version of one struct against the old version of the other, which will probably lead to a much larger diff than should be there 17:14 note that this can happen even if the struct hasn't been changed 17:14 so you'd get a diff for it in every single save diff line, which isn't really great 17:14 it won't be wrong in the sense of not being able to be loaded, though 17:14 Ah, yeah, that sounds less than optimal 17:18 yet another possibility would be to add some sort of "tiebreak" for colliding mtags 17:18 that'd make the save code more complex in another way, though 17:19 ais523: greqrg raised a point I never realized when it comes to the NH3 pet AI 17:19 the fact that NH3 stores monster tracks to eliminate them exploring the space they just walked on 17:19 (usually) 17:19 probably "helps" a lot in making pets far more reluctant at following you 17:20 oh, that is a good point 17:21 it's known to make monsters bad at fleeing, it wouldn't surprise me if it makes them bad at following a backtracking player too 17:21 or in general 17:21 if a monster outspeeds you 17:21 (which kittens/etc do!) 17:21 they will have nowhere to go and might end up going backwards rarely 17:22 and then end up likely going in the wrong direction 17:22 -!- ais523 has quit [Quit: It seems most convenient to apologise for my connection in the quit message, given how often it comes up… If I immediately reconnect, it's probably because I could send but not receive.] 17:22 -!- ais523 has joined #hardfought 17:22 or in general 17:22 if a monster outspeeds you 17:22 (which kittens/etc do!) 17:22 they will have nowhere to go and might end up going backwards rarely 17:22 they will have nowhere to go and might end up going backwards rarely 17:22 and then end up likely going in the wrong direction 17:23 are you sure pets are affected by the track? 17:23 Hmm 17:23 note that some monsters, and I think kittens might be among them, make random moves sometimes when tame 17:23 Let me double-check 17:24 so you'd expect it to be basically impossible for kittens to follow down a corridor even in the simplest cases if they were 17:24 (I guess if the kitten randomly moves both ways it'll get locked on its own track, so maybe would stop paying attention to it then?) 17:25 -!- raisse has joined #hardfought 17:25 -!- mode/#hardfought [+v raisse] by ChanServ 17:26 [nd] Raisse (Wiz Gno Fem Neu) had Magicbane bestowed upon her by Thoth, on T:6045 17:27 which variant is nd? 17:27 /* lessen the chance of backtracking to previous position(s) */ 17:27 Yep, it is in the dogmove code too 17:27 nd is 3.6.1 17:28 oh, nethack-dev? 17:29 yes 17:29 [nd] Raisse (Wiz Gno Fem Neu), 6508 points, T:6497, killed by an ape, while frozen by a potion 17:29 Hmm this code seems really suspect 17:29 http://sprunge.us/FHUO 17:30 I thought the chance of avoidance was like 75% 17:30 But it seems to be much larger 17:31 hmm 17:31 is that if(rn2) test backwards? 17:31 wait, why is it multiplying by MTSZ? 17:32 that code isn't just suspect, it's more or less completely nonsensical 17:32 yeah 17:32 that was the part that made no sense to me 17:32 btw, do you know what order mtrack is stored in? 17:32 0 is newest 17:32 so that part looks OK 17:32 so basically 17:33 MTSZ * (amount of allowable tiles - age of the track) 17:33 also, what happens if all the tiles are rejected via goto nxti? 17:33 Presumably the same thing as for monsters, it gets "stuck" 17:33 and doesn't move at all? 17:34 presumably we don't notice kittens doing that a lot because they're speed>12 17:34 right 17:36 ais523: So it seems like in a corridor where a pet has accidentally ended up going in the wrong direction, the chance of escaping this issue is MTSZ * 2 - 0, meaning 12.5% 17:37 I just find it hard to even think about that code 17:37 it has a units confusion 17:37 it helps if you treat MTSZ as just being the number 4 17:37 Yeah it shouldn't even point to MTSZ 17:38 either 4, or a macro of its own 17:38 ais523: It is possible that it was intended to be an addition? 17:38 FIQ: no idea 17:38 But that still makes no sense 17:38 it's either a typo or thinko of some sort 17:38 -!- raisse has quit [Ping timeout: 248 seconds] 17:39 the logical code would be along the lines of if(rn2(MTSZ) > age_of_track_element) goto nexti; 17:39 or possibly >= 17:40 ais523: However, with this said 17:40 I feel like pets shouldn't even use this code in first place 17:40 agreed 17:41 And in fact I'd argue this change should be in 3.6.x 17:41 right, that's what I'm thinking 17:41 I'm just dreading making the change as the rest of the devteam are probably going to make me playtest it and it's hard to figure out exactly what would count as a success there 17:42 Alternatively, a possibility is to turn the code off when the pet is close only 17:42 if the pet is stuck going backwards down a corridor 17:42 So that pets don't end up oscillating if they have no idea where you are 17:42 then turning it off would seem even worse 17:42 what you actually want is to prevent pets getting stuck in a corner of the room 17:42 due to trying to get to you crow-flies 17:42 but the code doesn't actually do that 17:43 ais523: no I mean 17:43 have the tracking code for pets be turned on when pets are far away from you 17:43 so that if you i.e. monster detect 17:43 you don't see ridiculous corridor movement from them 17:43 does going the wrong way down a corridor and changing direction ~every 8 turns count as ridiculous? 17:44 and disable it if you are close (as in <10 tiles) 17:44 Yes, current behaviour is silly as well 17:44 In fact it's worse 17:44 because it actively harms gameplay 17:44 I'll see if I can compile nethack easily here 17:45 not sure if just make will work, nethack 3 tends to be... special 17:45 yeah seems like I have this thing set up properly 17:45 let's see 17:47 it works 17:47 on a base level 17:48 pets are actually following 17:48 now, the only issue is 17:48 does this break things down horribly in some way? 17:48 because if the code removal works great 98% of the time 17:48 and does really weird things 2% of the time 17:48 that's bad 17:49 this is weird 17:49 I am walking in a corridor 17:49 and my pets are actually following me 17:49 like in NH4 17:49 so it's weird seeing it in vanilla too? 17:50 ais523: so yeah, seems like this is definitely the issue 17:50 ais523: no, it's great, just... not used to it, I'm used to it being really frustrating 17:50 if you're in wizmode, I recommend drinking a b!oDM, magic-mapping the level, and teleporting over to the other end 17:50 to see how pets behave when you aren't nearby 17:50 ok 17:51 hm 17:51 need to mess with wizmode permissions 17:51 edit sysconf in the playfield 17:51 I hate the way 3.6 does this mostly because it keeps reverting every time I rebuild 17:52 would be nice if there were some way to remember the new sysconf 17:53 what is the showmap command in NH3? 17:53 M doesn't work, nor does #showmap 17:53 FIQ: Because you asked, I am guessing that monstseesu neither checks whether a monster can see you, nor is it the function that's called when a monster does see you. Hmm. Lemme think. 17:53 oh there is a wizmode command list 17:53 ok 17:54 FIQ: it's ctrl-F in 3.4.3 17:54 might have changed for 3.6, though, wizmode commands are less stable UI-wise than regular commands 17:54 yeah found it 17:54 OK 17:54 so the pet gets stuck in a corridor because of lack of actual pathfinding 17:54 leading to oscillation 17:54 Not sure if this is good or bad 17:55 jonadab: there isn't a monstseesu in either 3.6.1 or NH4 17:55 jonadab: It is used to log monster intrinsic memory in spork 17:55 ais523: Right, I was just curious what jonadab would think it would do 17:55 Because the function name made no sense to me :) 17:56 "checks whether a monster can see you" is too specific for a function IMO; a variant that had that would probably do something more along the lines of NH4's msensem 17:56 a special-case AI routine seems more likely 17:57 -!- Grassy has quit [Ping timeout: 240 seconds] 17:58 ais523: So, what do you think? Is oscillation OK if it allows realible monster following? Or not? 17:58 I personally think oscillation is better but don't think everyone would necessarily agree 17:58 the correct fix is NH4's but that probably won't get into vanilla for performance reasons 17:59 alternatively, make use of the player track for pet following 17:59 does that thing actually work yet in 3.6? 17:59 No 17:59 Mandevil (re fast monsters that stay out of reach and shoot you): Oh, man, Brogue centaurs. 18:00 jonadab: I'm not sure Brogue centaurs are that well designed, the usual way to defeat them is with an AI exploit 18:01 and Brogue doesn't really have the tools to deal with kiting monsters, other than attack staves 18:01 you can't even aim obstruction "behind" them because of the way it targets 18:01 hmm, maybe Brogue needs a potion of obstruction, but even then, you wouldn't want to spend limited resources every time to defeat a fairly common enemy 18:01 FIQ: What if only _kobolds_ got "smart kobold" AI in NetHack? 18:02 jonadab: that'd be a lot less frustrating than in Smart Kobolds, I thinkn 18:02 because unlike in that game, they wouldn't have much opportunity to prevent you gaining exp and items 18:03 ais523: OK so in my experience pets basically never move away from you with this code disabled 18:03 FIQ: even to hunt down a monster elsewhere in the room? 18:03 So what if it was enabled only if a pet is 6+ tiles away? 18:03 Pets don't even do that I think? 18:03 hmm, I'm beginning to think that the correct solution is to turn off the track but to increase the randomness of the pet's movement in general terms 18:03 In fact, it was a thing I added in FIQHack because i Found this annoying 18:03 *I found this annoying 18:06 ais523: I am tempted to file this as a bug report 18:07 It will probably be ignored like most other things I send :-) 18:07 (OK, that's hyperbole) 18:07 that seems like a good idea, although note that this sort of bug report often gets fixed in a particularly bizarre way 18:07 also, many things you send are addressed quickly, it's the rest that we're not sure what to do with so we don't do anything 18:07 Yeah I know 18:07 It was hyperbole 18:07 based on the lack of replies 18:09 ais523: You mean sending a report has a risk of backfiring due to some devs'... peculiar view on things? 18:09 ISTR a case where someone (tungtn?) reported a bug with a suggested fix, and the devteam did the exact opposite 18:09 that was before i joined 18:09 yes 18:09 that very thing 18:09 Iwas thinking of that too 18:09 anyway, the actual active portion of the devteam is very small 18:09 I was the other one that sent the report lol 18:09 (2 bug IDs was referenced) 18:10 the devteam itself has many members but most never do anything (which is reasonable, of course, it's not a day job it's a hobby) 18:10 -!- Grassy has joined #hardfought 18:10 -!- mode/#hardfought [+v Grassy] by ChanServ 18:11 commit 850ed0cba623cb403122a7ab280402a83383e92d 18:11 address #H4247 & #4248 - theft of quest artifact 18:11 this commit is evil 18:12 I can sort-of see the argument about not wanting MRtifact reliance to be 100% safe, but it has the nasty side effect of making asc kits yet more predictable 18:12 I guess right now there's the tradeoff of MRtifact versus CoMR versus GDSM, each of which have trouble 18:12 err, which have problems of some sort 18:12 so it helps make asc kits more diverse in that sense, but it still seems problematic 18:13 (the correct fix is probably to increase the number of "helpful but not necessary" extrinsics that you'd want in an asc kit to above the number of slots you have, so that you need to make choices about which to use) 18:14 -!- ais523 has quit [Quit: qut] 18:18 Mandevil: Vax is no problem it has ANSI C compilers. 18:20 It's more like, you won't be able to compile on System/38 or something. 18:23 hmm ais left 18:23 was going to ask him if http://sprunge.us/QKDj seemed OK 18:23 but I guess I will go for it then 18:24 I clarified the 1st part of that 18:24 as to what i meant 18:24 *I meant 18:31 FIQ: if thinks are really bad in unnethack, there's always the blackmarket. somehow you need to get rid of the gold from fort ludios :) 18:32 there, sent report 18:32 I really want this fixed 18:32 but knowing devteam... 18:33 -!- winny has quit [Ping timeout: 246 seconds] 18:41 !tell ais523 Is that why I haven't figured out how to deal with Brogue centaurs, if I don't happen to have a well-charged staff of lightning or something? 18:41 Will do, jonadab! 18:47 -!- winny has joined #hardfought 19:12 the 18:12 is what I did -- tried to pick slots that had one obvious choice and crammed them full of things which were actually kinda nice 19:17 Amulet of flying, cloak of half damage, deep DSM that grants drain resistance? 19:19 Shifted the Eye to require wearing, added an amulet of thing-i-can't-remember, gave all the DSMs some toys, added shields, helmets, few other things 19:19 On the whole, this seems like a viable approach. 19:20 it worked reasonably well. there was a pretty healthy debate going for a long time over whether GDSM, SDSM, WDSM (froze the floor... and the water plane), or RDSM (+d6 damage to every hit) was best 19:22 -!- tacco has joined #hardfought 19:33 -!- noty has joined #hardfought 19:38 [slex] Yuring (Mah Dro Fem Neu), 2641 points, T:2752, killed by a monster (leocrotta), while panic digging 19:42 YANI: instead of a black market, have a craftsmen's guild or some other such place where you can find master craftsmen who will give you the gear you need for very large sums of money. 19:43 These people mainly consume unneeded resources and produce needed ones, and also do some things the player can do normally but don't have the resources to do. 19:45 In general, you can get most anything you need... but the harder it is to produce, the more exorbitant the cost. 19:56 [slex] Yuring (Mah Dro Fem Neu), 635 points, T:1013, killed by a monster (porter tiger) 20:03 Also, it's not as good as wishes, since you can't specify enchantment or beatitude. For things for which enchantment isn't important, the material cost will be high. 20:05 and if u punch him he gets mad 20:27 lorimer: my main issue with your change to eota is that you nerfed eregen for no apparent reason 20:27 because apparently being unable to cast before the quest wasn't enough 20:27 let's make *wizards* unable to cast after quest too :D 20:46 whats lorimer's variant? 20:49 spork? 20:52 -!- tacco has quit [] 21:19 Spork, yes 21:23 Spork is the first variant I remember 21:29 that's interesting. when spork was released, slashem was still going strong 21:30 ah, yeah I guess I never thought of slash em as a variant for some reason 21:48 lorimer: I have to agree with FIQ. the EotA nerf was pretty harsh. 21:48 what was the reasoning behind that? 21:48 the power regen scales with level 21:49 at XL30 you have more regen than the original Eye... at XL14 less 21:50 only for wizards though correct? 21:50 still scales with level for non-wizards, but less regen 21:50 the Eye's base item is the amulet of power so 21:50 right 21:51 i did like that inclusion - you didnt have to wish for an artifact to get pw regen 21:52 the spell casting success rate nerf to wizards wearing *any* armor was pretty harsh as well, but understandable 21:52 err 21:52 any body armor 21:52 that made me Sad™ 21:52 i think i may have excluded crystal plate from that, FWIW, but that's not really much of a change. 21:53 ahhh 21:53 but yeah, the "my wizard wears all the armor your barbarian does and casts all the spells too" problem was ugly in vanilla 21:53 right. understandable 21:54 so spell of reflection, frees up amulet slot 21:54 and protection both stacks and provides solid AC 21:54 can still get intrinstics for everything else 21:54 etc tec 21:54 *spell of protection that is 21:54 etc 21:54 just have to be smart about it 21:54 eh, i wanted to see how players would solve it TBH 21:55 "if you want big regen, your neck is busy" 21:55 kind of like playing a giant/orge/kobold in grunthack. cant wear shit for armor so have to get creative 21:55 ascensing a giant wizard was interesting 21:55 jesus h christ 21:55 i cnt spl for sht 21:56 heh 21:56 ascending 21:56 there we go 21:56 no, really officer. english is my first language. i swear... 21:57 *only* language 21:59 at some point i'm going to make it my mission to ascend every role in sporkhack 21:59 :D 22:00 can't remember if someone did or not. hmm 22:00 i dont believe so 22:00 https://scoreboard.xd.cm/zscore.sh.html 22:00 aaxelb came very close 22:06 [nd] sluggoman (Wiz Hum Mal Neu), 8441 points, T:7387, killed by a jaguar 22:22 -!- Grasshopper has joined #hardfought 22:22 -!- mode/#hardfought [+v Grasshopper] by ChanServ 22:24 -!- Grassy has quit [Ping timeout: 248 seconds] 22:25 [nd] sluggoman (Wiz Hum Mal Neu), 1819 points, T:2395, killed by an invisible gnome king 22:26 lorimer: You mean protection the spell? 22:26 gaaah 22:26 -!- Grassy has joined #hardfought 22:26 -!- mode/#hardfought [+v Grassy] by ChanServ 22:27 well... at least you can fail to complete the digdug challenge 22:27 Protection the spell is broken but unusable in practice due to how painful it is to use UI-wise (outside fh/dnh) 22:29 damn... if you cant complete the digdug challenge, you wont be able to dig *ever* 22:29 heh :) 22:29 -!- Grasshopper has quit [Ping timeout: 248 seconds] 22:29 That makes the game unwinnable 22:30 Apart from abusing enexto exploits 22:30 -!- Grasshopper has joined #hardfought 22:30 -!- mode/#hardfought [+v Grasshopper] by ChanServ 22:30 couldnt you delete a challenge acceptance thru the devnull website? 22:31 i dont remember 22:31 Not temporarily 22:31 -!- Grassy has quit [Ping timeout: 240 seconds] 23:15 [slex] Yuring (Mah Dro Fem Neu), 1029 points, T:957, killed by a monster (mummy) 23:21 K2: I imagine it's theoretically possible to win without digging, by using Xorn polyself, or by taming lots of umber hulks, but it would be painful. 23:22 YANI: Digless conduct. 23:22 meh 23:24 Tangles: if you're willing to do xorn polyself, a lot of the game gets easier 23:30 the magic portal to the digdug level - is that always in minetown? or can it be anywhere in the mines? 23:30 Always minetown. 23:31 The other thing with the challenges int he past is that if you borked a challenge in one game, but wanted to keep that game, you could always just park your game, and fire up a new game on another server to get the challenge done. 23:32 If we're only running one server, this is no longer possible, which increases the difficulty a fair bit. 23:33 has orctown been taken into account? 23:33 and also, couldnt you cancel challenges via the website in the past? 23:33 orctown - if you've accepted the digdug challenge, and the portal is inside the outer walls, you're not going tobe able to access it 23:34 Yes. And I'd like for that to be possible still. 23:34 ugh 23:34 yeah that is bad. 23:34 if orctown then portal spawns outside walls 23:34 perhaps? 23:35 There's already code to make sure you fall into the outside region.. can probably just piggyback on that for the portal creation. 23:35 It's hard to test though. 23:36 But yeah, you did have the option to cancel/ignore challenges via the website, but it was permanent - no-take-backsies. 23:37 fuck me this challenge is hard 23:37 digdug 23:37 you either dont know what to do with the pump or you miss half the time 23:38 [slex] Yuring (Mah Dro Fem Neu), 1136 points, T:1481, killed by a monster (green koopa) 23:39 welp. the digdug challenge appears to work fine 23:39 other than i havent won it yet ;) 23:44 once you lose all your lives to the pookas there's no getting them back is there? 23:45 Nope. 23:46 [gh] phollenback (Mon Elf Fem Cha), 4074 points, T:4336, killed by a dwarvish king 23:47 Though I do wonder if you could use the pacman challenge's fruit token to restore lives in other challenges. 23:47 [slex] Yuring (Mah Dro Fem Neu), 472 points, T:908, killed by a monster (brute) 23:50 just curious 23:50 digging is kind of a Big Deal™ 23:51 if i dont think i could win the digdug challenge, i dont think i'd accept it 23:51 would make game extermely hard to win at that point 23:51 plane of earth, etc 23:53 at least i can try again if i die and start a new game 23:59 the "lifegiving bonus fruit" which only works if it's been formally identified, only restores lives for pacman currently.