00:05 -!- rikersan has joined #hardfought 00:13 -!- Tarmunora-m has quit [Read error: Connection reset by peer] 00:24 -!- rikersan has quit [Ping timeout: 248 seconds] 00:25 !tell ais523 Uploaded my comments on your proposal: https://nethackwiki.com/wiki/User:Phol_ende_wodan/DungeonOverhaul/ais523 00:25 Will do, aosdict! 00:36 !tell rikersan Set your terminal size to the largest your screen will support 00:36 Tarmunora: Message from rikersan at 2017-10-01 23:21 EDT: I got curses to work (set up on asc.run atm not here), but the inventory perm_inv isnt' working :/ 00:36 Tarmunora: Message from rikersan at 2017-10-01 23:22 EDT: I just copied your rc but it sadly does not work 00:36 Will do, Tarmunora! 00:36 !tell rikersan About 160 by 70 is a good size for curses+perm_invent 00:36 Will do, Tarmunora! 00:36 !tell rikersan 180 width would be best 00:36 Will do, Tarmunora! 00:45 !tell ais523 re your ludious proposal, why not make each square have the number of adjacent mine times 1000 so it's still significant? Or maybe make it more interesting -- the number of digits in the number is the number of adjacent mine squares but the number is still random? I think I like this approach best 00:45 Will do, Elronnd! 00:47 !tell ais523 re your more special levels proposal,, if you make them have a very low chance of generating, there should be quite a few. Maybe, there's rne(10)-ish special levels on levels 10-20, picked from a "level 10-20 special level" pool, and same for 20-30? 00:47 Will do, Elronnd! 00:48 -!- yuring has joined #hardfought 01:21 -!- rikersan has joined #hardfought 01:25 -!- rikersan has quit [Ping timeout: 240 seconds] 01:33 !tell ais523 https://elronnd.net/proposal.txt here's my dungeon proposal (unfinished, obviously 01:33 Will do, Elronnd! 01:33 -!- nht has joined #hardfought 01:49 aosdict: upper mines lit, lower dark, is actually from my proposal :P but it's not important 01:51 -!- LarienTelrunya has joined #hardfought 01:52 !tell ais523 unrelated, but do you know why in struct light_source, id, which is a void*, is sometimes treated as an int, and sometimes as a pointer? This is still in nethack4, btw. Wtf?? 01:52 Will do, Elronnd! 01:57 -!- nht has quit [Ping timeout: 264 seconds] 01:58 -!- raisse has joined #hardfought 01:58 -!- mode/#hardfought [+v raisse] by ChanServ 01:59 -!- nht has joined #hardfought 02:18 !tell K2 slex update (non-savebreaking) ready *bundlebundlebundle* it's the current master branch :) 02:18 Will do, LarienTelrunya! 02:23 ooh Beholder can be queried for monster entries now? 02:23 @le?bundle monster 02:23 bundle monster ~3~ purple monster (J) | Lvl: 30 | Diff: 37 | Spd: 12 | Res: poison | Confers: nothing | MR: 70 | Generates: special | AC: -7 | Attacks: 5d8 bite disease, 5d8 gaze disease | Alignment: 0 | Flags: genocidable, carnivore, flies, nopoly, infravisible 02:39 -!- elenmirie has quit [Read error: Connection reset by peer] 02:47 -!- khoR has joined #hardfought 02:57 -!- rikersan has joined #hardfought 02:58 -!- yuring has quit [Quit: Connection closed for inactivity] 03:01 -!- rikersan has quit [Ping timeout: 240 seconds] 03:05 [nd] Raisse (Wiz Gno Fem Neu), 1998 points, T:6308, killed by a Green-elf 03:07 This stuff in my country is pretty spoopy. 03:07 Oh Raisse is still going at it. Some things don't change. 03:07 :P 03:07 [slex] bugsniper (Nec Spe Fem Neu) killed Antje, on T:4040 03:07 hm 03:07 I must have missed the eternity proposal by ais somehow 03:08 but in any case 03:08 I must stay away from such a branch at any cost 03:08 lest I yasd 03:08 :P 03:08 it generates items that work as dnh garnet rod's time stop 03:19 -!- elenmirie has joined #hardfought 03:19 -!- mode/#hardfought [+v elenmirie] by ChanServ 03:27 -!- rikersan has joined #hardfought 03:29 "They used to say nobody cared about morals, but now they say times change." 03:29 yes 4k, I know, thank you very much 03:31 -!- rikersan has quit [Ping timeout: 240 seconds] 03:31 Ah, you found one of my new fortune cookie rumors. 03:31 jonadab: Message from K2 at 2017-10-01 23:25 EDT: have you watched star trek: discovery at all? and if so, what do you think so far? 03:32 K2: What's that? 03:33 A new star trek series I think. 03:33 Also: if it involves J.J. Abrams in any way, I refuse to acknowledge even the possibility that it might exist. 03:44 How does to-hit of an artifact weapon work if the artifact's damage bonus only works against a specific type of monster (e.g. +d40 damage to humans)? Does it mean the to-hit bonus also works against those only, or will the weapon always have the extra to-hit no matter what it attacks? 03:44 I think to-hit always works against everything 03:44 Not sure 03:44 hmm, I think I'll wizmode this 03:46 seems that it doesn't; a +20 to-hit artifact that does bonus damage versus S_HUMAN can hit an AC0 human with no trouble, but against an AC3 ogre it's "miss, miss, miss" 03:47 -!- rikersan has joined #hardfought 03:47 of course that's slex, so it might be different in vanilla, but I'd expect it to be the same since I didn't touch that code 03:51 -!- rikersan has quit [Ping timeout: 240 seconds] 03:54 Slash'EM might be different from 3.4.3 though. 03:54 yeah, possibly 03:56 jonadab: you should've added BOFH alongside the other zapm stuff :P 03:56 for hallumons 03:56 it's a thing outside ZAPM too, of course 03:57 -!- rikersan has joined #hardfought 03:58 I'm familiar with the BOFH in a non-ZapM context, yes. 03:59 in ZAPM it is basically the Rodney equavilent 03:59 but not nearly as bad 03:59 no teleport shenanigans, no harassments 04:00 likes to make himself invisible (no see invis in ZAPM) and spam ranged attacks at you 04:00 Unix Daemons are more dangerous 04:01 also I am very disappointed in that ZAPM didn't go the nethack pun (they're even based on where the pun comes from!) and made them & 04:01 (they're U) 04:01 -!- rikersan has quit [Ping timeout: 240 seconds] 04:01 if slex implements the ZAPM monsters, unix/ftp/whatever daemons will be & :D 04:02 I don't like having monsters on punctuation, though, in a game where monsters are mostly on letters and punctuation is terrain and items. 04:02 for being the final boss before endgame, bofh is a wimp 04:02 "oh, he is dead already, oh well" 04:02 EPI: the Bastard Operator From Hell gets a data delete attack, which can e.g. delete random items from your inventory, or spells, or reset some of your stats to what they were at the start of the game. 04:03 If it were authentic to the stories, he'd delete your account and arrange for the police to arrest you before you even got to his office. 04:03 That would make him an actual threat, yes 04:03 But yes, giving him a "format disks" attack could be interesting 04:03 but it'd just make you drop your floppies before pursuing him, I guess 04:03 Also arrange for your boss to fire you. 04:04 jonadab: no, that's boring 04:04 arrange for you to send your retirement mail :) 04:04 Or that, yes. 04:04 In one of the stories, he tricked somebody into asking their boss to fax something nasty to a bigger boss. 04:05 LarienTelrunya: FTP daemons' megabyte breath hoses your connection which is -50 speed, which is *nasty* (base speed is 100) 04:05 Oh, also, he electrocuted a number of people in the stories. 04:05 Bind daemons breathe bugs IIRC, cursing your inventory 04:05 FIQ: well, in slex the ftp daemon will breathe inertia, which makes you super slow; it doesn't stack though, additional hits from it will just make it last longer 04:05 Mail demons viruses which makes you ill (stops hp regen) 04:06 EPI: a monster with an attack that gives you _negative_ HP regen per turn. 04:06 Ntpd daemons is just a fire attack (but it's the only attack in the game that destroys floppies (besides lawyers seizing evidence, which only takes cracked ones) 04:06 jonadab: sounds like what poison should IMHO be doing 04:06 ) 04:07 What the fuck did I just read. 04:07 LarienTelrunya: You should add lawyers to slex :P 04:07 FIQ: rest assured that once I get around to adding the ZAPM monsters, lawyers will be among them :) 04:08 with a cease and desist attack (paralyze+fear), suing for damage (stealing gold) and seizing evidence (take items you stole from shops) 04:08 LarienTelrunya: Combine all the ZapM Unix Daemons into a single monster whose breath does all the things to you. 04:08 LarienTelrunya: inertia is actually a really good approximation to the idea of what ftp daemons do, yes 04:08 FIQ: Don't forget Summon N^HWitnesses. 04:08 it's basically the same idea 04:09 jonadab: that's basically just a create monster attack that ignores depth/etc 04:09 Eh, breathing megabytes at you in slex should probably cause lag effects. 04:09 FIQ: oh? well, what slex's inertia does is that it steals your intrinsic speed, temporarily disables your extrinsic speed, and sets your actual speed to half of what it would otherwise be 04:09 it's the only way the game can generate unix daemons outside the mainframe (endgame) 04:10 I think 04:10 unless they can generate if the player is sufficiently highlevel 04:10 but I've never seen it 04:10 The package/installation system in Cygwin is so much worse than WSL's. 04:11 But, Cygwin can run Windows-native things, and WSL cannot. 04:11 jonadab: actually that is an interesting idea, what LarienTelrunya replied @ your EPI 04:12 I know that ADOM flavours regen as how fast your blood course through your body 04:12 Which? The poison thing? 04:12 thus, logically, if poison is what negated the regen 04:12 faster regen = faster poison actually makes sense 04:12 Faster blood pumping -> faster regen does not make sense if you're healing from physical wounds, especially the kind that bleed. 04:13 YANI: we need an anticoagulation intrinsic. 04:13 Eeep. 04:14 So any time you are at less than full HP, you bleed profusely, and if you bleed too much, you die? 04:14 well, we'd also need to have a "bleeding" attribute; with anticoagulation it would not time out ;) 04:15 Ah. 04:15 in Angband, getting hit by bites and such will cause your character to bleed 04:15 caves of qud has that 04:15 a negative mutation trait in that game is hemophilia 04:15 which makes the bleeding debuff take forever to go away 04:16 (not literally forever, but it's much more dangerous) 04:16 EPI: implement the mushroom-takes-over-your-brain thing from Earthbound. But don't implement the character who buys the mushroom from you, removing the effect. 04:17 haha 04:17 -!- rikersan has joined #hardfought 04:17 that debuff was amusing 04:17 EPI: hostile nurses will use their scalpels to cause bleeding wounds to you, and are armed with anticoagulant syringes to ensure your wounds keep bleeding. 04:17 FIQ: It was _annoying_. 04:17 jonadab: have you ever installed a cortex crossover in ZAPM? 04:17 it's kind of similar 04:17 I don't think I have. 04:18 My last several games of ZapM all ended in radiation sickness. 04:18 jonadab: it's basically a tamer version of confusion (which in that game is *horrible*) that messes with direction keys 04:18 So I think I have lost my touch. 04:18 which I always thought was more funny than anything 04:18 jonadab: don't drink from vats 04:18 FIQ: I didn't. 04:18 don't carry buggy gamma ray guns 04:18 or x-ray specs 04:18 I think it might've been the goggles, yes. 04:18 this is why I ID ray guns ASAP 04:18 Or in one game a cannister of mutagen. 04:18 that's just bad luck 04:19 you could technically play around it by not picking up risky items 04:19 but it generally makes the game much more tedious 04:19 EPI: add items to dnethack that increase your weeping angel contamination just from being carried in open inventory 04:19 Not picking up any cannisters would be pretty limiting. 04:19 at least for the canister case 04:19 yeah 04:19 EPI: items that cause amnesia just by being carried in open inventory. 04:20 slex has that 04:20 also some roles have it intrinsically 04:20 EPI: an item that, if you pick it up, causes permanent incurable hallucination. 04:20 don't think slex has that 04:21 but it would not surprise me 04:21 -!- rikersan has quit [Ping timeout: 248 seconds] 04:22 it has a permahallu gamemode, but fourk does so too ;) 04:27 -!- rikersan has joined #hardfought 04:27 [slex] bugsniper (Nec Spe Fem Neu) killed the Oracle, on T:6518 04:28 You murderer! 04:32 -!- rikersan has quit [Ping timeout: 240 seconds] 04:32 [slex] Yuring (Mah Dro Fem Neu), 1175 points, T:1119, killed by a dart 04:37 -!- rikersan has joined #hardfought 04:42 -!- rikersan has quit [Ping timeout: 248 seconds] 04:54 [slex] bugsniper (Nec Spe Fem Neu), 40191 points, T:7448, killed by a monster (Undead Katcher), while paralyzed by a monster attack 04:57 -!- rikersan has joined #hardfought 05:00 -!- nht has quit [Ping timeout: 248 seconds] 05:00 -!- nht has joined #hardfought 05:02 -!- rikersan has quit [Ping timeout: 264 seconds] 05:09 hm 05:09 too bad chris has vanished again 05:10 would be cool if next dnh version fixed this: https://nethackwiki.com/wiki/User:FIQ/DevTeam#Major_pet_AI_bug 05:11 !tell NeroOneTrueKing "book of Infinite Spells can grant the specifically AoE versions (which are still only castable at Skilled+)" this makes no sense 05:11 Will do, FIQ! 05:13 aosdict: "I disagree with the level 14 barrier on the grounds that it annoys players to have to put off the Quest and usually compels them to grind in the Dungeons for wraiths or potions of gain level. Doing the Quest before passing Medusa, unless forced to do otherwise, is ingrained in most people." <--- exactly. Instead of making up excuses why the limitation stays in the game, all developers should remove it IMHO :P 05:17 -!- rikersan has joined #hardfought 05:18 pushed a PR to dNAO, at least 05:18 maybe he'll see it eventually 05:18 [nd] elenmirie (Mon Hum Fem Neu) completed Sokoban, on T:9474 05:18 LarienTelrunya: you should add it too :3 05:18 this should apply cleanly to slex, if you want to do it: https://github.com/Chris-plus-alphanumericgibberish/dNAO/pull/40 05:18 FIQ: you mean the pet behavior change? pets are working as intended in slex, they're not supposed to be ultra smart :P 05:19 but I'll take a look at it anyway 05:20 hmm, this is only in dogmove.c? so hostile monsters scared by Elbereth will still stand next to you in a corridor? 05:20 [slex] Yuring (Mah Dro Fem Neu), 3191 points, T:2717, killed by a monster (duck captain) 05:20 LarienTelrunya: yes, this is to make pets not become stuck going *away* from the hero 05:20 due to how the mtrack thing works 05:21 when they want to follow you 05:21 basically, this is what happens 05:21 the player either lingers, or the pet outspeeds 05:21 it only has 1 tile to go when next to you in a corridor -- backwards 05:21 so it does 05:22 -!- rikersan has quit [Ping timeout: 240 seconds] 05:22 now mtrack has a memory of 4 tiles, and the tile closer to you was more recently walked on -- so it will, if it moves at all, favour going further backwards 05:22 and from there it snowballs 05:22 ah, the "pet is too stupid to simply follow down a completely straight corridor" thing 05:22 this is why it's nearly impossible to make pets follow you realibly in NH3 05:22 and now that I know what causes this 05:23 I should be able to "abuse" it 05:23 to make pets actually follow 05:23 and do as I want 05:23 heh :D 05:23 usually, my solution if the pet is being dumb consists of leading the pet through the corridor before me, i.e. using my character to physically block the way back, so the pet can only go forward 05:24 basically 05:24 never linger 05:24 and if the pet ends up next to you and you are slower than it 05:24 swap with it 05:25 and it'll scurry away before you 05:25 yeah, kinda 05:25 that's more or less what I am doing :) 05:25 LarienTelrunya: either way, you should try the "patch" (it just removes some lines) 05:25 see if you like it or not 05:26 well I can give it a try and see how it changes slex's pet AI 05:27 LarienTelrunya: but yeah 05:27 YASI: add an extrinsic property that can appear on some pieces of armor, "of pet mastery", which disables the line of code that causes pets to fail following you through a corridor. 05:27 -!- rikersan has joined #hardfought 05:27 under no circumstance let the pet have a turn in a corridor when it is next to you unless you want to follow it running away 05:27 that's basically how you work-around it 05:28 yeah, and I guess it gets more annoying if there's more than one pet 05:28 because they also block each other 05:28 oh yeah that'll make things even more annoying 05:28 :P 05:28 glhf making pets do what you want if you have >1 05:29 hmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmm do I want to be "evil" and make it so that the annoyance code stays in slex, and is deactivated if the player's petkeeping skill reaches Basic? :P 05:29 well 05:29 is it evil if it's in vanilla? :) 05:29 that way, dedicated "petmaster" roles like Healers, which get that skill, will enjoy a better pet AI while those who aren't good at using pets will have stupid pets 05:29 you could also let tin whistles give temporary "smart AI" 05:30 like it was always intended to (it's broken) 05:30 ah, that sounds like a plan! noted! 05:30 (but yes, petkeeping at basic, or at least skilled, should also do it) 05:30 alright, it's on my to-do list now, thanks :) 05:31 lunchtime now, bbl 05:31 -!- rikersan has quit [Ping timeout: 246 seconds] 05:34 -!- bug_sniper has quit [Ping timeout: 258 seconds] 05:38 [slex] Yuring (Mah Dro Fem Neu), 1283 points, T:689, killed by disturbing their deity 05:39 [slex] Yuring (Mah Dro Fem Neu), 2 points, T:1, killed by an unsuccessful polymorph 05:47 -!- rikersan has joined #hardfought 05:51 -!- rikersan has quit [Ping timeout: 258 seconds] 06:07 -!- rikersan has joined #hardfought 06:07 -!- nht has quit [Ping timeout: 246 seconds] 06:08 -!- nht has joined #hardfought 06:12 -!- rikersan has quit [Ping timeout: 258 seconds] 06:17 -!- rikersan has joined #hardfought 06:21 -!- rikersan has quit [Ping timeout: 255 seconds] 06:37 -!- rikersan has joined #hardfought 06:41 -!- rikersan has quit [Ping timeout: 240 seconds] 06:47 -!- rikersan has joined #hardfought 06:52 -!- rikersan has quit [Ping timeout: 240 seconds] 07:07 -!- rikersan has joined #hardfought 07:07 jonadab: star trek:discovery is a brand new TV show series that came out last week. episode 3 aired last night (new). ep 1&2 were back to back last sunday. you can only watch it online via CBS all access app 07:07 K2: Message from LarienTelrunya at 2017-10-02 02:18 EDT: slex update (non-savebreaking) ready *bundlebundlebundle* it's the current master branch :) 07:07 no JJ abrams 07:08 it takes place before the original star trek but after star trek: enterprise 07:08 timeline is basically the start of the first war between the federation and the klingons 07:11 -!- rikersan has quit [Ping timeout: 258 seconds] 07:17 -!- rikersan has joined #hardfought 07:17 Slash'EM Extended updated 07:18 woohoo! thanks K2! :) 07:19 bhaak: for next Junethack, slex's gehennom trophy has been changed to "entered through the front entrance", i.e. in order to get the trophy, one must use the downstair on the valley (similar to how it works in Nethack Fourk), levelporting into Gehennom will not grant it anymore 07:21 -!- rikersan has quit [Ping timeout: 246 seconds] 07:37 -!- rikersan has joined #hardfought 07:41 -!- rikersan has quit [Ping timeout: 240 seconds] 07:46 -!- rikersan has joined #hardfought 07:51 -!- rikersan has quit [Ping timeout: 264 seconds] 07:56 -!- rikersan has joined #hardfought 08:01 -!- rikersan has quit [Ping timeout: 248 seconds] 08:04 -!- rikersan has joined #hardfought 08:05 -!- rikersan has quit [Client Quit] 08:06 -!- rikersan has joined #hardfought 08:06 k2: yeah my dns settings let me connect now 08:06 rikersan: Message from Tarmunora at 2017-10-02 00:36 EDT: Set your terminal size to the largest your screen will support 08:06 rikersan: Message from Tarmunora at 2017-10-02 00:36 EDT: About 160 by 70 is a good size for curses+perm_invent 08:06 rikersan: Message from Tarmunora at 2017-10-02 00:36 EDT: 180 width would be best 08:06 heh 08:06 also, sorry for msg spam, somebody closed my laptop while things were running 08:06 using googles? 08:07 nah just default 08:07 instead of opendns 08:07 hmm thats odd 08:07 although 08:07 I mean I would keep it like this forever it's just there are suprise inspections -.- 08:07 i was using opendns several months ago at the router level and one day no one could connect to the server, so i ditched it 08:08 !tell tarmunora: hm ok will try that 08:08 Will do, rikersan! 08:08 yeah i understand 08:08 k2: lol nice 08:08 moon is almost fully waxed 08:08 !pom 08:08 rikersan: The moon is waxing gibbous; full moon in 3 days. 08:08 i wonder if opendns has hardfought on a filter somewhere 08:10 why would it 08:10 k2: it's a blacklist I think, they probably noticed somebody connecting to it a lot 08:10 it might be on a game site filter list 08:11 thats happened before 08:11 lol 08:11 hardfought has been up for almost 17 yrs so its 'known' per se 08:12 ah - https://domain.opendns.com/hardfought.org 08:12 not on a blacklist at least 08:13 yeah perm_invent still isn't working when I copy tarm's thing 08:13 I'll just give up 08:13 still trying os x default terminal? 08:14 oh 08:14 I'm dumb yeah 08:14 press O 08:14 sorry I had bad memory yes 08:14 fiq: what about it? 08:14 toggle it ingame 08:14 oh good idea 08:15 Evil Patch idea: monsters with MS_BONES, like skeletons, which can rattle noisily when chatted to and paralyze the player, will now also do so periodically while close by. 08:15 so hardfought.org isnt blocked by opendns by default, but you can certainly block/unblock sites if you are using their service to block different classes of sites on your network 08:15 i use opendns in this way for one of my clients 08:16 and universities tend to have their own custom black lists 08:16 fiq: tried that, it just didn't do anything :/ 08:16 k2: long story but this is a high school of sorts 08:16 same thing 08:16 hm 08:17 in regards to discussion ;) 08:17 let me spec 08:17 yep 08:17 just likely not a custom filter 08:17 while you do it 08:17 they block reddit and imgur too ;-; 08:17 fiq: attend this high school? my mom makes me 08:17 it would have to be custom - opendns doesnt have my site blacklisted 08:17 why else would I learn latin 08:17 it's horrible 08:17 and its not officially under any of their categries 08:17 no I mean 08:17 categories 08:17 Proxies are your friend 08:17 Tarmunora: Message from rikersan at 2017-10-02 08:08 EDT: hm ok will try that 08:17 let me spectate 08:18 while you toggle 08:18 brb 08:18 whee it works with iterm2 :D 08:18 ;) 08:19 thanks k2 08:19 for reminding me >.> 08:19 no prob heh 08:19 -!- Jonathanhanes has joined #hardfought 08:19 -!- Jonathanhanes_ has quit [Ping timeout: 240 seconds] 08:19 ahh 08:19 nice 08:19 hm, doesn't show any hp indicators 08:19 just --- 08:20 there we go reset that 08:22 why the hell does he have an amulet of lifesaving ;-; he's a leprachaun 08:23 No body armour? 08:23 what 08:23 LarienTelrunya: why, oh why? 08:24 LarienTelrunya: here I am trying to figure out how not to make Healers a dedicated petmaster role and then you just go and say it 08:24 rikersan doesn't have any torso armour on 08:24 did someone break my fancy hp pw bars 08:24 yes why 08:24 tarmunora: I had some and it blocked spellcasting too much 08:24 who is responsible for this audacity 08:24 I needed it for healing 08:24 LarienTelrunya: also, I'm just going to nab that EPI as a YANI, since it seems realistic and not evil 08:24 bhaak: it would otherwise be too easy to get the trophy, you could even get it randomly if something zaps a wand of banishment 08:24 fiq: ? what 08:24 (the MS_BONES thing) 08:25 FIQ: Good point, where did they go? 08:25 aosdict: sure, go ahead :) and I said that "healers are petmasters" because I feel that's how you survive as one in vanilla, considering how much all the available weapon choices suck 08:25 inb4 I did 08:25 Well, yes, it is. 08:25 If he copied my rcfile, after I copied Nero's, after he copied yours... 08:25 I took them out 08:26 But the Bard proposal is going to rely on healers not being kind of bad petmasters. 08:26 because they didn't show for me I think 08:26 or are there supposed to just be --- 08:26 no 08:26 then yeah no show 08:26 but you are not alone 08:26 I all I saw was HP:--- 08:26 PW:--- 08:26 they are broken 08:26 so there needs to be _something_ for Healers to do that isn't continually heal up their starting pet 08:26 hold up configing 08:26 Mine work fine 08:27 whose fault is this 08:27 rikersan: Try a 1 line taller terminal and re-enable the bars 08:28 hm ok 08:28 nope 08:29 Huh 08:29 I wonder why they're breaking 08:29 shrug 08:29 what is your hp 08:29 Is the rest of the file copied from mine? 08:29 and pw 08:29 including max 08:29 I think it might be the 15k power that's breaking it 08:29 the hp/pw thing... forgot about that 08:29 will need to let tangles know 08:30 hp -- pw -- 08:30 FIQ: this was just in dnethack correct? 08:30 tihs is my code 08:30 right. thats why i mentioned Tangles 08:30 :P 08:30 tarmunora: yes 08:30 FIQ: kidding 08:31 fiq: it's like 118 max hp, current is that 08:31 power? I don't know it's somethign really big 08:31 like 20k 08:31 -!- ais523 has joined #hardfought 08:31 inc barb so 08:31 ais523 - be ready for the messages when you type next :) 08:32 hm 08:32 I will look when home 08:32 K2: oh dear, perhaps I should just stay silent then :-) 08:32 ais523: Message from jonadab at 2017-09-30 18:41 EDT: 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? 08:32 ais523: Message from jonadab at 2017-10-01 13:50 EDT: Could the whole of netclient.c being ifdeffed out be somehow related to http://192.168.0.35/pastebin/881.txt ? 08:32 ais523: Message from jonadab at 2017-10-01 13:51 EDT: Err, I mean netgame.c sorry. 08:32 ais523: Message from jonadab at 2017-10-01 13:52 EDT: Also netplay.c ? 08:32 ais523: Message from jonadab at 2017-10-01 14:12 EDT: Err, make that http://jonadab.jumpingcrab.com/pastebin/881.txt 08:32 ais523: Message from FIQ at 2017-10-01 21:01 EDT: so messing with occupation stuff, I'm having some trouble understanding what exactly is going on in postcommand stuff in allmain, because it seems to kill flags.occupation/flags.incomplete unconditionally, apparently to fix some kind of save issue? 3ac6b6a02d1cad3209fd37cb79ac76867014e16a introduced it 08:32 ais523: Message from FIQ at 2017-10-01 21:03 EDT: I have new eocc struct working properly but the way allmain thrashes incomplete/occupation means I'll have to check for player vs monster for occupation stuff which made this all pointless. I'd like to fix this 08:32 ais523: Message from aosdict at 2017-10-02 00:25 EDT: Uploaded my comments on your proposal: https://nethackwiki.com/wiki/User:Phol_ende_wodan/DungeonOverhaul/ais523 08:32 ais523: Message from Elronnd at 2017-10-02 00:45 EDT: re your ludious proposal, why not make each square have the number of adjacent mine times 1000 so it's still significant? Or maybe make it more interesting -- the number of digits in the number is the number of adjacent mine squares but the number is still random? I think I like this approach best 08:32 ais523: Message from Elronnd at 2017-10-02 00:47 EDT: re your more special levels proposal,, if you make them have a very low chance of generating, there should be quite a few. Maybe, there's rne(10)-ish special levels on levels 10-20, picked from a "level 10-20 special level" pool, and same for 20-30? 08:32 ais523: Message from Elronnd at 2017-10-02 01:33 EDT: https://elronnd.net/proposal.txt here's my dungeon proposal (unfinished, obviously 08:32 ais523: Message from Elronnd at 2017-10-02 01:52 EDT: unrelated, but do you know why in struct light_source, id, which is a void*, is sometimes treated as an int, and sometimes as a pointer? This is still in nethack4, btw. Wtf?? 08:32 ais523: please add an option to change the font size to jettyplay, it seems to default to 8 pt on my old 1024x768 monitor and I can't change it :D 08:32 lol 08:33 LarienTelrunya: I tried that a while back 08:33 wait, I thought you were talking about libuncursed 08:33 LarienTelrunya: jettyplay scales the font size so that the whole ttyrec fits in the window at once, so you should just need to make the window larger 08:33 is there a way to get the #loot windows n stuff to appear in the center on my screen tarmunora? 08:33 Beholder: you can't just crush a man under the weight of so many messages 08:33 ais523: I pushed a PR to DT withpet stuff 08:33 ais523: can't, because it already fills the whole screen and the font is still tiny! :( 08:34 LarienTelrunya: size the window larger than the screen 08:34 you can do that on windows? 08:34 rikersan: Maybe? Who knows 08:34 hm ok 08:34 LarienTelrunya: I think so, probably by moving the window partially offscreen and then sizing the edges that are onscreen 08:34 feel free to look if I did something stupid 08:34 you can move a window with "alt-space m" if you can't get the mouse to the title bar 08:34 it's already annoying if I try to watch someone play on the server with a huge terminal; Yuring for example uses 177x41, but the x dimension of this monitor does not go beyond 157 08:35 LarienTelrunya: how easy is easy? 08:35 jonadab: just a reminder that site-local IPs probably won't work for anyone else ;-) 08:35 and moving the window off the screen "helpfully" stops at 157 too because the stupid operating system knows that's the maximum that fits on the screen, argh! 08:36 bhaak: "easy" as in "lol i startscummed for a wand of banishment and got gehennom trophy 3 turns into the game kind of easy" :P 08:36 also, IIRC that error occurs when something outside libnethack tries to link to something inside libnethack without using the API 08:36 replace with jonadab.jumpingcrab.com 08:36 aimake gets much further than it logically should, but the build can't be completed 08:36 LarienTelrunya: I mean, the role that we had to filter for the Medusa trophy, that was only necessary because it was an important piece of the clan trophies. 08:37 FIQ: incomplete and interrupted in NH4 work like flags.move in vanilla; they're set to a default value before running the code for the command, and the command can override them 08:37 LarienTelrunya: sonst ist es eigentlich egal, gibt genug andere Trophäen, die man leicht holen kann. wenn es die Clan-Trophäen nicht betrifft, kann man es so lassen IMO 08:37 the "better" fix would be to have something like a struct multiturn_state that got returned from the implementation of every command, but that would take a while 08:38 LarienTelrunya: oops. my mental language switch toggled on its own 8-| 08:38 bhaak: it's OK, LarienTelrunya knows German 08:38 and I can read enough German to get a sense of what you're trying to say, although I can't grasp the details 08:38 ais523: I know, but still an odd thing to happen. 08:38 bhaak: naja, ich bin der Meinung, man sollte für seine Trophäen schon arbeiten müssen; es ist dieses Jahr in Junethack öfters passiert, daß Spieler durch ein zufälliges Monster mit Wand of Banishment in Gehennom gelandet sind und "schwupps" hatten sie die Trophäe! 08:39 oh come on, there's no way "wand of banishment" translates to "Wand of Banishment" in German :-P 08:39 you're just taking English and capitalising the nouns :-P 08:39 heh 08:39 LarienTelrunya: he, you don't need to make them suffer all the time, throw them some bones sometimes 08:39 Zauberstab der Verbannung? 08:40 re your ludious proposal, why not make each square have the number of adjacent mine times 1000 so it's still significant? ← I was planning something along those lines, yes 08:40 if they randomly get banished to Gehennom:42 then they're free to use the upstair and go back down again, which would also give them the trophy ;) 08:40 bhaak: but don't bones make the game harder sometimes? 08:41 ais523: that came out of nowhere. the dual meaning in a channel of nethack didn't occurr to me :) 08:42 -!- Tarmunora is now known as Tarmunoraway 08:43 Elronnd: I don't believe I've made any changes to struct light_source 08:43 a variable that can be used sometimes as an int and sometimes as a pointer /should/ have a type of intptr_t, but it's probably best to replace it with a union 08:43 ais523: what will break if I choos not to reset them 08:44 changing something that works to be clearer is a low priority, anyway 08:44 FIQ: the "interrupted" flag doesn't get set back to false elsewhere, IIRC (likewise "incomplete") 08:44 so, for example, multiturn commands will never end until you get interrupted, then you'll get interrupted every turn 08:44 this might not actually be the case in the current code, though 08:48 gotta go now, might be back later ;) 08:48 -!- LarienTelrunya has quit [Quit: Amy suddenly disappears!] 08:51 ais523: incomplete is set to false as part of action_complete 08:52 FIQ: action_complete isn't called in every codepath that actually completes an action, though 08:52 e.g. single-turn actions like "pick up item" don't call it, IIRC 08:52 and incomplete could otherwise be set if the client aborted an incomplete action for reasons the engine doesn't know about (e.g. keypress) 08:53 that sets interrupted though, doesn't it? 08:54 no 08:55 interrupted is engine-side, pressing a key is client-side 08:55 and the information about interruption is only communicated one way (engine to client) 08:56 I commented out the unsetting of these things 08:57 It doesn't seem to have broken occupations 08:57 and I can still interrupt commands 08:57 hmm 08:57 maybe they're unset again somewhere else 08:58 try this test: run a long-running command, interrupt it via keypress, then perform a quick, repeatable action (e.g. dropping 1 gold piece) 08:58 that's the case that's most likely to break 08:58 and it makes this code do what i want: 08:58 if that doesn't break either the code's probably dead nowadays 08:58 if (youmonst.interrupted || busy(&youmonst) != occ_food) 08:58 print "resume food" 08:58 (the 2nd line looks different, I wrote it by hand after goofing up the paste, but yeah) 09:04 -!- nht has quit [Ping timeout: 248 seconds] 09:04 anyway, it's OK for you to change the semantics here if you want to 09:04 ooh but it did break things 09:04 just be clear on what the new semantics are (e.g. which actions have to explicitly set completed/interrupted)? 09:04 not the way described though 09:05 you might need to change a bunch of code to allow for them 09:05 if I eat something 09:05 stop eating 09:05 and then walk a direction 09:05 it's treated as go/run 09:05 you end up farmoving? 09:05 yes 09:05 you can actually do this in vanilla too, although not the same way 09:05 I think, at least 09:05 I forgot the details as soon as I realised they weren't important for the TAS 09:06 haha 09:06 -!- nht has joined #hardfought 09:10 ais523: btw, you said flags.occupation was redundant 09:10 it isn't -- the game uses it in various places to figure out if the player is busy doing something 09:10 not quite, it's used if you want to special-case a reaction to someone doing something in particular 09:10 for example, for watchman_on_duty 09:10 I think it's redundant with incomplete, though (in the sense that you don't need incomplete if you have occupation) 09:10 Yes, that might be true 09:12 ais523: but IIRC you said that not all occupations have occupation enumerates 09:12 it's possible 09:12 for stuff like repeating a certain action several times 09:12 but you could just make an occ_busy as a catchall 09:13 that isn't a special case (like searching) 09:13 also, repeating a single-turn action several times isn't technically an occupation 09:13 it sets complete every turn 09:13 it's just that the client ignores the flag in that case 09:13 ah 09:13 if you type something like 9d1$ that's handled entirely client-side 09:13 K2: Hmm, I was not aware of it. Does it maintain consistent canon with the rest of (non-Abrams, aired-on-screen) Trek? 09:19 ais523: is interrupted only ever used in occupation context? 09:19 I think it might also be used during stuff like 9d1$? 09:19 I made interrupted a monst flag rather than part of eocc, I think this was why 09:19 Maybe I misunderstood 09:19 FIQ: it's used in any situation where the client needs to determine whether to continue something that has the same UI behaviour as a multi-turn action 09:19 OK 09:19 So it is correct then 09:20 right, although it's up to the client (rather than the engine) precisely which actions are "presented as" multi-turn 09:20 monsters don't necessarily need to be using the same UI as the player 09:20 come to think of it, interrupted should probably be in turnstate 09:20 but that might be hard from a technical point of view, not sure 09:23 !tell LarienTelrunya There may be a way to make your desktop larger than your screen/viewport, so that when your mouse would otherwise move off the edge of the screen, the viewport moves in that direction. (This is a standard capability on X11, and a lot of Windows graphics drivers have the option to emulate it; I used to use the Matrox version of it when I had a Matrox Mystique and was stuck on a 17" 09:23 Will do, jonadab! 09:23 monitor, back before the prices of 19" ones came down out of the stratosphere, and was still multi-booting into Windows 95 OSR2 sometimes. 09:24 it would be simpler to make it work equavilent to monsters and players here 09:24 !tell LarienTelrunya ... I used to use the Matrox version of it when I had a Matrox Mystique and was stuck on a 17" monitor, back before the prices of 19" ones came down out of the stratosphere, and was still multi-booting into Windows 95 OSR2 sometimes. 09:24 Will do, jonadab! 09:24 and make maction_interrupted(mon) just set interrupted 09:25 ais523: hmm, location is just coordinates 09:25 ais523: NOte that I'm _only_ getting that error on Windows, not on Linux. I guess I'll have to git bisect it. That'll be fun. 09:25 I think unless they are unset on levelchange, you can continue an action on an equal xy on another level 09:25 * aosdict gives jonadab the Tsurugi of Muramasa 09:26 jonadab: some quick sanity checks: make sure that there are no object files or libraries anywhere in the source tree (other than prebuilt/ for things like SDL), then do a clean build 09:26 (although the potential problem I'm thinking about almost certainly can't affect Windows, it's more likely to affect Linux but IIRC I fixed it on both platforms) 09:27 Elronnd: The proposal wording allows for multipliers; it says something like "make the amount of gold _depend on_ the count", not "be the same as the count". 09:28 jonadab: for minesweeper? 09:29 aosdict: right 09:29 YANI on ais523's proposal: if you lose at minesweeper and the gold "falls through cracks in the floor", it actually falls into Moloch's Sanctum 09:30 why would the Sanctum be directly below Ludios? 09:30 right after it becomes useless 09:30 It's not. The cracks are just really, really deep. 09:31 it actually crosses my mind that I don't know where Ludios is, physically 09:31 ais523: but I do want to know whether you intended spaces with 0 neighboring mines to be completely empty of gold? 09:31 I think of it as being nowhere because it's a floating level 09:31 aosdict: thanks. 09:31 aosdict: I hadn't decided, it's plausible though 09:31 maybe this is the sort of level that needs playtesting 09:31 ais523: Right, yes, I checked for object files. 09:31 ais523: I don't see a way the game can be played if not. 09:31 I'll check for libraries as well, but I don't _think_ that is the issue either. 09:31 Because otherwise, you get almost no information and might step onto a mine first thing. 09:31 aosdict: the way a very similar level is done in Enigma is to ensure all the tiles touching the border are safe / unmined 09:32 that gives enough information to get started 09:32 hmm 09:32 but only to spoiled players 09:32 it was obvious in Enigma but only because all the numbers were revealed there to start 09:33 something like this will need spoilers anyway, I guess? or it can be something that unspoiled players can figure out over multiple attempts, like they have to do with everything else 09:33 maybe put in an Oracle consultation on the subject 09:33 figuring out how the gold relates to the mine count is probably harder 09:33 If it's a power-of-ten multiplier, that should be obvious to most folks. 09:33 if Ludios is Fort Knox like it's referred to in the code, does that put it in Maine/Kentucky? 09:34 I think of Ludios as being _somewhere_ on the surface and open to the sky. 09:34 ais523: I was assuming you would total up all the "numbers", and then divide 69105 by it 09:34 aosdict: I'm not sure Maine even exists in the NetHack universe 09:35 also, I didn't make any assumptions about the exact formula to use 09:35 ais523: hmm, commands take an int but never return anything else than 0 or 1, right? 09:36 Elronnd's idea about the number of digits being the number of adjacent mines is absurd if mines can be clustered together at all 09:36 FIQ: the way it works in vanilla (unchanged for NH4, I believe) is that a command's timeconsuminess is returned via flags.move 09:36 Well, you want the formula to be straightforward to caluculate in your head once you know what it is. 09:36 I was thinking of making it have this convention: 0 = took no time, 1 = took time, 2 = the user died, 3 = occupation is still going 09:36 flags.move can be set to 0 explicitly, and is also set to 0 if the command returns 0 09:36 in retrospect this API is fairly silly (and has lead to a number of bugs) 09:36 my suggestion is that if you want to make a more logical API, use an enum as the return value 09:36 the above could easily be made an enum, yes 09:37 anyway, and make it so that if the command returns anything else than 3, eocc->current (flags.occupation) is set to occ_none 09:37 aosdict: not really, a 5 is pretty rare in Minesweeper itself, and you need a few five-digit numbers (or lots of 4-digit numbers) to reach the gold target for Ludios 09:38 ais523: you can kick gold coins to work-around the mines 09:38 oh, hmm 09:38 aosdict: I've seen 7 in Minesweeper. 09:38 highest is 8 09:38 -!- rikersan has quit [Ping timeout: 240 seconds] 09:38 but it's not that common 09:38 right. It's absurd on the low end (no mines are next to each other, you get about 1000 gold) and absurd on the high end (7 or 8 mines gives you millions in gold) 09:38 FIQ: If it's 8, you don't see it. 09:39 it's really hard to see an 8, though, because if the mine density is high enough to generate them you're almost never going to actually find the square to click on it 09:39 not in nethack, no 09:39 but in minesweeper it can happen 09:39 s/nethack/ludios/ 09:39 aosdict: What if the number of mines were the _last digit_? 09:39 aosdict: I think the idea is that you use a biased level generator that places mines in such a way that there's the right amount of gold 09:39 jonadab: ooh, interesting 09:39 jonadab: that's closest to my original plan, where a gold quantity is (large round number) + (number of adjacent mines) 09:40 repeatedly seeing values like 2003 and 5001 should tip people off that the last digit is relevant 09:40 but I'm worried that that would be too restrictive 09:40 I think you could tweak the formulas to work if you decided how many mines should be generated 09:41 mines should instead of launching the user in a pit where it originally was, it should launch the user in a random direction 09:41 also: hey, future me! there's enough discussion on this to put in a YANI ^ 09:43 -!- rikersan has joined #hardfought 09:44 hmm, come to think of it… if mines can't be on the borders, each mine has 8 neighbours 09:45 so the total amount of gold would be (gold from significant digits) + (number of mines × 8) - (number of distinct pairs of mines that are mutually adjacent × 2) 09:45 the obvious problem with this is that it's necessarily an even number 09:46 so probably we would need mines on the edges too 09:46 perhaps just the border nearest the player is unmined? that'd be analogous to shops and the like 09:46 but it should be fairly easy to control the total value of all numbers on the board via careful mine placement 09:47 The reason I like the zero adjacent mines = zero gold thing is that it's more immediately recognizable as a minesweeper level 09:47 yes 09:47 True. 09:47 perhaps you could start off with one mine identified, too? 09:48 and a description that made it clear that stepping on it is a bad idea 09:48 Are you counting the 1 or 2 gold pieces placed on each mine in the total? 09:48 What if no mines could be placed on the _near_ edge, guaranteeing the player some numbers to start from? 09:49 is this an idea for a minesweeper special level? 09:49 rikersan: Integrated into Fort Ludios, yes. 09:49 ah cool 09:49 Ugh, the Quest portal can be placed in the Rogue level? Whose terrible idea was that? 09:50 aosdict: branches can be placed on special levels by default in vanilla 09:50 jonadab: I already suggested that, it makes sense 09:50 aosdict: actually I wasn't, that should fix the odd/even issue 09:50 ais: including the big room? 09:50 err, * I was 09:51 yes but it's awful for actually finding the trap 09:51 wait, now I'm confused 09:51 I wasn't but I shouldn't have been 09:51 rikersan: yes 09:51 huh 09:51 doesn't happen very often but it's possible 09:51 aosdict: that is correct 09:51 be glad you haven't had to find it on the big room though then 09:51 !whereis aosdict 09:51 rikersan: aosdict is not currently playing. 09:51 how does !whereis work if someone's playing multiple variants at once? 09:51 aosdict: Bigroom Sokoban entrance happens in vanilla. I fixed this in Fourk by making the minimum Big Room depth one level deeper. 09:52 rikersan: I was wiztesting, quest portal was on Rogue, did ^F to find traps, but they're all colorless 09:52 ais: I think it's only for currently playing a game 09:52 aosdict: lol rip 09:52 rikersan: you can log in on two connections at once 09:52 oh really? TIL 09:53 I'll have to try that later 09:53 note: the vast majority of variants don't support this 09:53 it works in NH4-derived variants but with some visual glitches 09:53 most notably, messages greying earlier than they should 09:53 However, no variant cares if you're also currently playing _another variant_. 09:53 I think dgl will kill the older process if you log in on a newer one, but probably only if they're both the same variant 09:54 I have no idea what dgl does. 09:54 I glanced at that codebase once, never got deep enough to know anything beyond what language it's in. 09:55 I wrote a patch against dgl once but it was a very localized patch 09:55 I didn't need to understand the rest of the code to write 09:55 *write it 09:56 oddly enough, I'm not sure it ever got applied 09:56 (it added a keypress you could use to ask dgl to attempt to reconfigure your terminal so that you could view the character set the game you're watching was playing in; sort-of like a reverse s) 09:58 no idea where it is now, I'm not sure if I wrote it on my own computer 10:13 -!- bug_sniper has joined #hardfought 10:21 !whereis 10:21 FIQ: Yuring [slex] FIQ [fh] FIQ [4k] 10:21 !whereis FIQ 10:21 FIQ: FIQ [fh]: No details available 10:21 FIQ: FIQ [4k]: No details available 10:22 ais523: seems to work reasonably well 10:22 oh, neat 10:22 NH4 variants have no extrainfo, hence no... well, extrai nfo 10:22 hmm, how often do we update line 2 of the save file? 10:22 too rarely 10:22 I had a plan at one point to use that for !whereis but I don't think it's updated often enough yet 10:23 an "unclean" exit will have it be out of date 10:23 only actual saves update it 10:23 right 10:23 we should probably update that periodically 10:24 maybe on level change (only if there hasn't been a level change recently)? 10:24 Isn't whereis info usually a separate file? 10:24 you can update a whereis file as you do that change 10:24 right, we could have a separate whereis file 10:24 or maybe some sort of whereis API, a socket or something 10:25 ais523: is there something wrong withi these changes you think? I wanted to optimize the success rate of having it merged https://github.com/NetHack/NetHack/pull/56/files 10:25 FIQ: I'm at work atm, can't easily check 10:25 ask me later 10:25 Sorry 10:25 Ok 10:25 But when you aren't at work, doesn't that mean no IRC too? :P 10:25 I was on holiday 1-3 weeks ago 10:26 still managed to IRC a lot 10:26 ah 10:26 I thought you just didn't do internet stuff when not at work 10:26 maybe I misunderstood 10:26 I used not to own an Internet connection 10:26 so it was always borrowing of someone, but not always work 10:26 You got an internet connection at home now? 10:26 nowadays I have a connection via tethering, which is limited in bandwidth 10:27 -!- rikersan has quit [Ping timeout: 258 seconds] 10:27 Ah. 10:27 but lets me get online for things like IRC without much trouble 10:27 it turns out that there are mobile phone providers in the UK who are happy to sell tethering-only contracts 10:27 which is a nice contrast to the stories I've heard of people not beig able to legally tether at all 10:28 That is more an US thing 10:28 -!- noty has quit [Quit: WeeChat 1.9] 10:29 -!- NeroOneTrueKing has joined #hardfought 10:29 (OK, it's not /quite/ tethering-only; I also get 10 text messages per month, which I can send via the router's management interface as I don't have a mobile phone) 10:29 The US thing would be that it's not economically advisable to tether your regular computer to a cellular connection while at home, because cellphone data usage costs more than home internet. 10:30 over here the price is comparable but wired internet is faster and doesn't have bandwidth caps 10:30 ais523: I send text messages using email-to-SMS gateways, without having a cellular contract.' 10:30 jonadab: that only works because you're in the US, where texts are paid for by the recipient 10:30 in the UK texts are paid for by the sender 10:30 I think both sender and recipient pay, typically. 10:30 If they have cellphones. 10:31 s/have/use/; 10:31 the UK tends to have fairly large line rental charges because of the people who use cellphones like pagers (i.e. to receive only), meaning that they never pay any call charges 10:31 Although most cellphone users have some number of text messages per month covered by their standard monthly bill. 10:31 In some cases this number is "unlimited". 10:32 yes, it's fairly easy to get unlimited text contracts here too 10:32 And I _think_ the government finally put a stop to "unlimited" meaning "unlimited until you hit the limit, and we charge crazy fees after that without warning you." 10:32 text messages basically don't cost the company anything 10:32 they're a) really small (160 bytes), and b) can piggyback on data that needed to be sent anyway 10:32 Compared to stuff like downloading updates, right. 10:32 -!- rikersan has joined #hardfought 10:32 so the only actual cost to the company is routing it between source and destination 10:32 SMS was designed back in the pre-G era. 10:33 (That is, before cellular connections were described as being some-number-G.) 10:33 yes, and it was effectively free to the company even then 10:34 it was originally thought up as a "ooh, look what we can use this spare space in the communication packets for", then it became inordinately popular so companies started charging for it 10:34 Ah. 10:34 ais523: can we maybe bump that up to 161 bytes, to store message order? When you send a long text, it gets split up and could get received in any order, and my phone isn't smart enough to stitch it back up in the right order 10:34 this is 2017, it's ridiculous 10:34 aosdict: I'm not in charge of the limit 10:35 just complaining :) 10:35 aosdict: THe sender's phone should probably wait until the second ticks over before sending the next portion. 10:36 Then they'd have different timestamps, which should make it easy for the recipient's phone to put them in order. 10:36 ooh, this reminds me, I thought of an interesting compromise solution to randomized Sokoban 10:36 (err, not /this/, something else I was doing) 10:36 the idea is that we have a batch of maybe 10-20 (not sure on the ideal number yet) possibilities for each Sokoban level, which are the same on everyone's game 10:36 but we generate a new batch every month 10:37 I think you have mentioned it before 10:37 Deterministically, so it works the same in local play? 10:37 this gives time for people who like it to generate spoilers for the people who like spoilers 10:37 -!- rikersan has quit [Ping timeout: 240 seconds] 10:37 jonadab: right 10:37 personally, I think this would annoy everyone 10:37 or most people, at least 10:37 although that would lead to people getting at levels in advance, but it's probably still better than the version that downloads them from nethack.org 10:38 hmm, any idea which file the game uses if a file exists both inside and outside nhdat? 10:38 If someone wants to predict future Sokoban levels in advance, I don't see the harm in it. 10:38 well, I was thinking that there'd be a race to complete the spoilers for the month 10:38 but people would probably just automate it 10:38 However, on the whole, I think it's probably better to just make a yes-or-no decision on random Sokoban. Possibly leaving the _first_ level of Sokoban as a designed easy level for tutorial reasons. 10:38 anyway, one nice thing about this is that we could probably make it work with unmodified vanilla binaries just by giving people a replacement sokoban.lev to drop into their data folder 10:39 also sokogen can generate easy levels just fine 10:39 if you set the difficulty level to something like 10 it's hard to get wrong 10:39 (still possible, but easier than the bottom level in vanilla) 10:39 Ah. 10:40 isn't there currently some bias to sokogen that makes it less than releasable at the moment? 10:40 aosdict: It doesn't generate pits/holes. 10:40 aosdict: it assembles levels from portions according to a template 10:40 but so far it only has one template so the levels all end up looking much the same 10:40 Or the room with the stair to the next level. 10:40 (and it has no way to do the goal area) 10:40 -!- rikersan has joined #hardfought 10:41 I'm working atm on code to allow more flexible templates 10:42 ais523: Doesn't the game always use one or the other depending on compile time options? 10:43 This is for NH3 10:43 NH4 d oesn't have the option anymore 10:43 FIQ: the install script will only create one or the other 10:43 that's unrelated to what the actual file loading code does, though 10:43 and I'm pretty sure it can load files from outside nhdat (e.g. license) 10:57 -!- rikersan has quit [Ping timeout: 240 seconds] 11:13 -!- rikersan has joined #hardfought 11:18 -!- rikersan has quit [Ping timeout: 248 seconds] 11:20 -!- elenmirie has quit [Read error: No route to host] 11:20 -!- elenmirie has joined #hardfought 11:20 -!- mode/#hardfought [+v elenmirie] by ChanServ 11:23 -!- NeroOneTrueKing has quit [Ping timeout: 240 seconds] 11:23 -!- rikersan has joined #hardfought 11:28 -!- ais523 has quit [Quit: meeting] 11:28 -!- rikersan has quit [Ping timeout: 248 seconds] 11:43 -!- rikersan has joined #hardfought 11:48 -!- rikersan has quit [Ping timeout: 248 seconds] 11:54 -!- rikersan has joined #hardfought 11:58 -!- rikersan has quit [Ping timeout: 240 seconds] 12:06 -!- raisse has quit [Ping timeout: 258 seconds] 12:06 -!- raisse has joined #hardfought 12:06 -!- mode/#hardfought [+v raisse] by ChanServ 12:13 -!- rikersan has joined #hardfought 12:15 FIQ: YASI: in fiqhack, if you die and a nearby pet zaps your corpse with a wand of undead turning before you start to rot, you come back to life and your game is not lost. 12:18 -!- rikersan has quit [Ping timeout: 260 seconds] 12:25 aosdict: I actually seriously proposed that at one point. 12:26 The problem is it's hard to implement because it complicates the question of how the code knows when the game is over. 12:28 You could make it a one-turn immediate-reaction thing, I guess. 12:28 YANI: unless there is a ghost above a corpse, if the corpse's monster has an undead form, zapping turn undead at it will turn it into that undead 12:28 (adding a third meaning to the phrase "turn undead") 12:32 YANI: also do this for living monsters, subject to monster MR 12:33 aosdict: Yes, I want to do that at some point 12:34 -!- rikersan has joined #hardfought 12:34 The first thing you said, that is 12:35 -!- NeroOneTrueKing has joined #hardfought 12:38 -!- rikersan has quit [Ping timeout: 246 seconds] 12:54 -!- rikersan has joined #hardfought 12:58 -!- rikersan has quit [Ping timeout: 255 seconds] 12:59 -!- raisse has quit [Ping timeout: 248 seconds] 13:14 -!- rikersan has joined #hardfought 13:17 -!- raisse has joined #hardfought 13:17 -!- mode/#hardfought [+v raisse] by ChanServ 13:18 -!- rikersan has quit [Ping timeout: 258 seconds] 13:34 -!- rikersan has joined #hardfought 13:38 -!- rikersan has quit [Ping timeout: 246 seconds] 13:40 -!- tacco has joined #hardfought 13:43 -!- ais523 has joined #hardfought 13:44 -!- rikersan has joined #hardfought 13:48 -!- rikersan has quit [Ping timeout: 248 seconds] 13:52 -!- khoR has quit [Quit: Connection closed for inactivity] 14:04 -!- rikersan has joined #hardfought 14:06 -!- nht has quit [Ping timeout: 258 seconds] 14:09 -!- rikersan has quit [Ping timeout: 260 seconds] 14:12 -!- nht has joined #hardfought 14:14 -!- nht has quit [Client Quit] 14:14 -!- rikersan has joined #hardfought 14:19 -!- rikersan has quit [Ping timeout: 240 seconds] 14:20 -!- ais523 has quit [Remote host closed the connection] 14:21 -!- ais523 has joined #hardfought 14:34 -!- rikersan has joined #hardfought 14:35 jonadab: re your recent wiki edit: what would your opinion be on changes that made scrolls safer to identify without a ?oID or +oID? 14:36 as it stands, wands and stones are the only item classes with an "inherent" safe ID method, which in both cases has a fairly low cost 14:36 Armor. 14:36 whereas even if use-IDing scrolls were safe, it would be a bad idea due to the risk of wasting a powerful scroll 14:37 Well, most armor. Helms are arguable. 14:37 you need a pet for that, don't you? 14:37 Well, yes. A pet or an altar. 14:38 Making scrolls safe to read-identify would weaken my argument against orctown considerably.' 14:38 really, out of all the item classes 14:38 the most problematic ID game comes for wands and scrolls 14:38 Apart from that, it seems like an unnecessary change otherwise. 14:38 ais523: Potions are also generally safe to use-ID 14:38 Even though it's technically not completely safe 14:38 It's as safe as in brogue, more or less 14:38 Potions are _unimportant_ to use ID early in the game. 14:38 FIQ: yes, potions also have some of the best indirect use-ID 14:38 Right, mosters use them. 14:38 I wouldn't call potion ID inherently safe, but I would call it one of the best-balanced classes for ID 14:38 Well, the important ones. 14:39 -!- rikersan has quit [Ping timeout: 255 seconds] 14:39 as you can viably use-ID them with precautions, dip-ID them, or observe monsters using them (in addition to the methods that work universally) 14:39 (like +oID) 14:39 And books... well, if you're willing to use a TON of holy water, you can bless them all. 14:39 ais523: Scrolls can also be use-IDed. Very carefully 14:40 jonadab: I'd argue that with vanilla mechanics, IDing books is mostly unimportant 14:40 ais523: That's because you never play Monk. 14:40 But yeah, it's something I only do as last-resort 14:40 what you care about is whether you can safely read them 14:40 Or Healer. 14:40 not what they actually are 14:40 Or Priest. 14:40 ais523: Right, but you pricecheck for that. 14:40 Or, as noted, use a boatload of holy water. 14:40 like, you don't make the decision to read a book or not based on the contained spell, but on the spell /level/ (i.e. price group) 14:41 Yes. 14:41 it's rare for players to say "well I'd take a 60% risk on reading this book if it were this spell, but not if it were that spell) 14:41 Although it's not just "can I read it" but also "does that price group have anything I want in it", because most of them don't. 14:41 ais523: I think aosdict, or someone else, made a better solution for how to deal with spellbooks rather than have their appearance be more complicated the more difficult they are to read 14:41 By using weight 14:41 ais523: if it's a spell you want, you can bless it. 14:41 You can easily afford enough holy water for all the spells you _care_ about. 14:41 jonadab: you're acting like holy water is a limited resource :-P 14:42 at least one top player used to just unconditionally blank all potions they came across 14:42 and bless them 14:42 But doing so for _all_ the spells is a LOT more holy water, because there are like 30 spells nobody ever uses for anything ever in the history of ever. 14:42 this is the best argument I've seen for empty potion bottles, incidentally 14:42 because at least that way you might use the original potion before refilling and blessing the botlte 14:42 (the plan is basically for quaffing to leave the bottle intact, but dipping to destroy the bottle) 14:43 ais523: I suggested a while back, I don't think you was around, to make alchemy result a 1+1=2 14:43 I prefer giving the less useful potions better uses. Starting by making throwing the negative ones a viable tactic. 14:43 FIQ: I was around 14:43 1+1=2 is definitely better than 1+1=1 14:43 You dip full healing into gain energy. You get 2 gain ability 14:43 And lose FH and gain energy 14:43 I think I would like to see a variant do that, and see how players deal with it. 14:43 You dip 10 FH into gain energy. You still get 2 gain ability 14:43 jonadab: Brogue does it by making basically all potions AoE 14:43 and now have 9 FH left 14:43 but that's very weird from a flavour point of view 14:43 ais523: Anyway, what I was getting at was 14:44 ais523: I mean I'd like to see a variant do 1+1=2 14:44 What if holy water worked the same? 14:44 ais523: Thrown potions can't work like Brogue in NetHack without a ton of work. 14:44 -!- rikersan has joined #hardfought 14:44 1+1=2 on holy water still gives you holy+regular = 2 holy 14:44 But they should at least do something meaningful to the monster they hit. 14:44 ais523: Oh, that's an interesting point. 14:44 ais523: No, you'd *lose* the holy here 14:44 with the current alchemy rules 14:44 I guess 14:45 Although, holy water doesn't follow the normal alchemy rules necessarily anyhow. 14:45 holy water + uncursed gain level = 2 blessed gain level seems problematic 14:45 ais523: OK, I guess I was unclear 14:45 although it's the logical result from 1+1=2 rules 14:45 Not *exactly* the same 14:45 Maybe dipping into holy water could continue to work as now. 14:45 But rather, use up HW on *one* potion/scroll/whatever 14:45 that said, it could potentially be made interesting, if holy water were more limited 14:45 holy water would then become a sort of "potion of wishing" 14:45 that you can use to duplicate nonblessed potions 14:46 unholy water would be a lot more valuable in this situation 14:46 floppy disk of optimization? 14:46 jonadab: Not quite 14:46 I don't really think I like the direction this is going though 14:46 What ais523 is suggesting is what would happen if the 1+1=2 rule is consistent 14:46 Oh, right. 14:46 ais523: But you aren't understanding what I am trying to say :P 14:46 For HW in particular 14:46 Not sure if 1+1=2 should apply to holy water dipping. I don't really consider that alchemy as such. 14:46 I suggest only being able to dip *1* item 14:46 Not 30 14:47 OK, that's an entirely different rule though 14:47 1 X + 1 Holy Water = 2 blessed diluted X 14:47 NeroOneTrueKing: Message from FIQ at 2017-10-02 05:11 EDT: "book of Infinite Spells can grant the specifically AoE versions (which are still only castable at Skilled+)" this makes no sense 14:47 NeroOneTrueKing: oh, I like that 14:48 although it implies that diluted holy water should exist 14:48 Worth noting that alchemy results are always diluted 14:48 Anything that prevents you from dipping 30 potions of uncursed water into one holy water and getting 30 holy water is going to be quite a significant balance change (not necessarily BAD, but significant). 14:48 (which presumably has the same quaff/throw effects as holy water, but is incapbale of blessing things) 14:48 FIQ: you'd change that in this scenario 14:48 Right 14:49 to produce diluted results if either input was diluted or if either input was water 14:49 -!- rikersan has quit [Ping timeout: 240 seconds] 14:49 ais523: but this should work for water in general 14:49 this probably also implies that water + any nondiluted potion = 2 diluted potions 14:49 1 GL + 1 water = 2 diluted GL 14:49 which makes a lot of sense, logically 14:49 I assume you rebalance this by making diluted potions half as strong, or water harder to obtain 14:51 YASI: dipping potions into fountains or water produces diluted potions of sickness 14:52 actually the correct rebalance is probably to make potions in general rarer 14:52 and/or generate diluted sometimes 14:52 actually that makes a lot of sense 14:53 now a diluted GL gains you one level, and an undiluted GL gains you two levels via mixing it with a blanked potion you don't want 14:53 FIQ: The spellbook weight thing was ais523's idea. The spellbook description complexity thing was either ais523's idea or got created through discussion 14:53 The scroll label length thing was my idea IIRC. 14:53 the spellbook weight thing wasn't my idea, I don't think 14:53 -!- NeroOneTrueKing has quit [Ping timeout: 248 seconds] 14:54 Meh 14:54 Doesn't matter 14:54 Either way it's a good idea 14:54 my spellbook plans were related to adjectives that became visible upon price-ID 14:54 Much better than complicating descriptions 14:54 so "a red spellbook" becomes "an expensive red spellbook" or whatever when you find a store that can ID them 14:54 FIQ: it's only good if weight is obvious (and only fiqhack has that right now) 14:54 No 14:54 aosdict: NH4 has that. 14:54 NH4 in general does 14:54 aosdict: all NH4-derived variants show weight 14:54 -!- rikersan has joined #hardfought 14:54 also all Slash'EM-derived variants 14:55 As well as Acehack, DynaHack and all variants based off SLASH'EM 14:55 Doesn't dnethack also? 14:55 No 14:55 Ok, wasn't sure about that one. 14:55 I think a read-IDing buff could be that the spellbook will only actually crumble to dust if it's cursed. Otherwise, you always keep the book (unless it literally explodes) 14:56 aosdict: SLASH'EM makes uncursed reading safe 14:56 aosdict: in 3.4.3, if you read a spellbook, it becomes cursed while you're reading it, and you get the contact poison result 14:56 then the spellbook will initially be intact but will disappear the next time you save and reload 14:56 which reminds me, I'm not 100% sure I ever reported that one 14:56 That sounds like in_use bugs 14:56 and I'm also not 100% sure I got the details right 14:57 Basically 14:57 in_use not being unset correctly 14:57 yep, there's two functions which are meant to be called as a pair 14:57 And when you reload, the game figures you tried to sighup cheat 14:57 and removes it 14:57 "Finishing off %s..." 14:57 but only one gets called if the spellbook becomes cursed during reading rather than starting out cursed 14:58 Incidentally, NetHack4 has a manual-recovery-forced bug here 14:58 ais523: I mean, you will still be able to suffer the negative failure effects if the book is uncursed, but you will never actually lose the book 14:58 FIQHack, and 4k via fh, addressed that 14:58 aosdict: one thing that bothers me about reading spellbooks is that the lowest-level failure effect is one of the most dangerous, possibly /the/ most dangerous 14:58 pretty much the only precaution that works against teleport+paralyze is being on a no-teleport levle 14:59 *level 14:59 ais523: The 2nd is also more dangerous than 3rd 14:59 Blind rather than aggravation 14:59 isn't the paralysis for only about 2 turns thoguh? 14:59 *though 14:59 Oh and that blindness lasts *forever* 14:59 Level 4 IIRC is confusion 14:59 -!- rikersan has quit [Ping timeout: 260 seconds] 14:59 aosdict: I thought it was pretty long 15:00 Which is even less dangerous than 3 15:00 also I guess it's technically helplessness, not paralysis? (the only difference is that free action blocks paralysis) 15:00 FIQ: I guess part of the reason I don't like spellbook weight-ID is because I think the thick spellbook should have a bumped-up weight and the thin and parchment spellbooks should have a bumped-down weight 15:00 :P 15:01 Simple fix 15:01 Remove those descriptions 15:01 :) 15:01 oh wow, this is more broken than it looks 15:01 the helplessness isn't paralysis, it's /spellbook read time/ 15:01 Right 15:02 i.e. the number of actions the book would take to read are converted into turns of helplessness 15:02 If it is read successfully it is an occupation 15:02 If not, it's a helplessnes 15:02 which means that this is an action ordering bug 15:02 s 15:02 You didn't know this? 15:02 I thought it was intended...? 15:02 even if you're simulating occupation as helplessness, the teleport should happen /after/ the helplessness, not before 15:02 FIQ: so did I 15:02 but then I actually read the code 15:02 nomul(context.spbook.delay); /* study time */ 15:03 that comment makes it fairly clear to me that this isn't meant to be some sort of paralysis-scaling-on-difficulty 15:03 What is nomul? 15:03 ugh, it's hard to summarize in a sentence 15:03 The function name looks suspect 15:03 the direct effect it has is to place a cap on the value of multi, which is a variable that does a ton of different things 15:04 if multi is positive, it counts the number of remaining turns of a multi-turn action, except that COLNO (of all things) is a special case for movement 15:04 That... didn't really helpl me understand :P 15:04 if multi becomes 0, that interrupts a multi-turn action but has no other effects 15:04 Ah, naturally 15:04 if multi is negative, thenn minus multi counts the number of turns of helplessness you have 15:04 So 15:05 the main effects of nomul are therefore: nomul(0) interrupts actions; nomul(-x) causes x turns of helplessness (but not stacking with other helplessness sources) 15:05 nomul(5) is 5 turns of occupation 15:05 nomul(-5) is 5 turns of helplessness 15:05 ? 15:05 this should really be split up 15:05 no, occupations don't set multi; logically they should but they don't 15:05 aosdict: it is in NH4 15:05 yes, I hear quite a lot about nh4 around here 15:05 oh, and just to confuse things further, multi is also the variable used to store keyboard input for a count, as in n20s, the 2 and 0 go into multi as they're being typed 15:06 I assume this was its original purpose and it sort-of grew out of control from there 15:06 don't know for certain though 15:06 I guess the vanilla devteam likes having if(multi) checks 15:06 if(multi) would be weird, it's normally if(multi < 0) 15:06 there is a broken check on multi in the lifesaving code, though 15:06 which is really handy in the TAS 15:07 (note: this is also responsible for "You survived that effect on your life." occasionally being printed out of sequence, several turns later) 15:10 I think I would like to see a variant do that, and see how players deal with it. <-- sounds like my development philosophy :) 15:12 do what? 15:14 -!- rikersan has joined #hardfought 15:17 FIQ: 1+1=2 alchemy 15:19 -!- rikersan has quit [Ping timeout: 240 seconds] 15:21 jonadab: once I finish working on all this design stuff I might take a crack at that 15:34 -!- rikersan has joined #hardfought 15:39 -!- rikersan has quit [Ping timeout: 260 seconds] 15:45 I'd be tempted to do it in Fourk, but I think players are already about to shoot me over the damage reduction nerf. 15:45 So I think I need to do Pw rebalance first, to give things a chance to cool down. 15:46 ah yes, the problem with rebalancing is that nerfing is often the way to go, and it's never popular 15:47 Also, I'm thinking about nuking MC from orbit, and I need to preserve some of my emergency PR for dealing with the player reaction to that. 15:52 jonadab: btw, I'm one of the people who makes my own Sokoban spoilers 15:53 when I started getting into NetHack, I decided I was willing to spoil myself on almost everything, except for Sokoban solutions 15:53 in fact I didn't look up spoilers for that for over a year I think, and then it was just to see what other peoples' solutions were like 15:53 ais523: I made my own Sokoban spoilers for some of the variant Sokoban levels, before the Wiki had spoilers for non-vanilla ones. 15:53 getting the same level more than once was interesting there because I could remember what went wrong the previous times 15:54 Oh, when making my own spoilers, I copy-paste the level over to a text editor, do my spoiler there, copy-pasting the level down after each step to edit for the next step, etc. 15:54 -!- rikersan has joined #hardfought 15:54 So occasionally I do mess up and create a "solution" that has a problem. 15:55 But if so, I can make a note of it at the top and revisit the question next time I get that level. 15:55 jonadab: I didn't write my solutions down 15:55 just learned them through repeatedly doing them in game 15:55 Ah. 15:55 I think I could probably do that no with levels of difficulty similar to _most_ of the 3.4.3 levels. 15:56 With certain notable exceptions, like 2a/2B. 15:56 (Wiki 2a, Steelypips 2B) 15:56 with vanilla nightmare, it's mostly just a case of remembering the first move 15:56 everything is pretty much forced from there 15:56 err, the first couple of moves 15:57 (the reorganisation at the top-right, and the entry to the main area) 15:57 Oh, I can do it flipped and rotated now. 15:57 I've played through NetHack Sokoban a lot of times. 15:58 I guess I'm better at Sokoban (the game) than most NetHack players 15:58 when I was ill several years ago and basically unable to do anything 15:58 I solved over 1000 Sokoban levels to pass the time 15:59 jonadab: you want people shooting you iover 4k changes 15:59 -!- rikersan has quit [Ping timeout: 248 seconds] 15:59 mess some with alignment record ;) 16:00 FIQ: I think with a couple of exceptions most folks have accepted that now. (The race- and role-specific ways to gain alignment helped a lot, once I finally implemented them.) 16:00 Or did you mean add traps and monsters that reduce your alignment just to be evil? 16:01 EPI: add a role that loses alignment every time a monster dies. 16:02 the whole "neutral atheist pacifist non-healer" meme is probably related to that 16:04 Yes, but you don't _lose_ alignment when your pet kills things. You just don't gain any either. 16:04 indeed 16:05 -!- rikersan has joined #hardfought 16:05 Actually, the way vanilla alignment works, if we flipped the gains when you kill things to losses, you could probably wrap a 64-bit alignment around through negative to positive. 16:06 Because I don't think there's a bounds check on the low end, unlike the high end. 16:09 -!- rikersan has quit [Ping timeout: 240 seconds] 16:15 -!- rikersan has joined #hardfought 16:19 -!- rikersan has quit [Ping timeout: 240 seconds] 16:26 jonadab: 64-bits worth of kills is going to be basically impossbile unless there's soem way to double your negative alignment 16:27 even 32-bit maxint is an almost unimaginably huge number 16:27 Hmm. Well, there's a big multiplier (it's _way_ more than one point of alignment per kill), but you may be right. 16:29 jonadab: how big is big? if it doesn't scale on itself, it isn't big enough 16:29 less than about 15-16 digits is insignificant at this size 16:29 Right, it only scales on permonst features, I think. 16:30 So yeah, wouldn't actually work. 16:30 an easy way to understand how big 64-bit maxint is is that Berry has scored more points in NetHack than everyone else put together 16:30 and in fact, unless there's a 64-bit maxint ascension I don't know about, has scored more points in a single game of NetHack than everyone else put together 16:31 Right. 16:34 9 quintillion is the sort of number that humans simply can't perceive how big it is 16:34 -!- rikersan has joined #hardfought 16:36 2 billion – a 32-bit maxint, /is/ the sort of number you can reasonably perceive, e.g. it's approximately equal to the number of seconds in 63 years, and thus even more approximately the number of seconds in a typical human lifespan 16:36 but 64-bit maxint is just out of the realm of what people can meaningfully visualise 16:37 if everyone currently living on Earth dedicated 24 hours a day, for the rest of their lives, working on the same problem, the total number of person-minutes worked would still be less than a 64-bit maxint (about one sixth, so it's of the right sort of order) 16:38 but that doens't really let you visualise it because most people don't have a good idea of what the entire population of tge planet is like 16:39 -!- rikersan has quit [Ping timeout: 264 seconds] 16:44 -!- rikersan has joined #hardfought 16:49 -!- rikersan has quit [Ping timeout: 240 seconds] 17:04 -!- rikersan has joined #hardfought 17:07 -!- Jonathanhanes has quit [Ping timeout: 240 seconds] 17:08 -!- NeroOneTrueKing has joined #hardfought 17:09 -!- rikersan has quit [Ping timeout: 240 seconds] 17:14 ais523: how did he do that anyway 17:14 gold duping? 17:14 yes 17:14 it's the only practical way to reach numbers that high 17:14 because it doubles 17:15 Right 17:15 Berry's actually got 32-bit maxints via linearly scaling methods (i.e. no doublings) in the past, because he's crazy 17:15 Makes sense. 17:15 Indeed. 17:15 although it takes a very long time, it's doable because some actions give so many points 17:15 but you can't even make a scratch on 64-bit maxint that way 17:18 FIQ: re +oInfiniteSpells: it's a "this is a feature not a bug" thing. Yes, special cases could be made on the book to never select those spells, but the book already can select normally impossible spells (poison spray, acid blast). 17:18 NeroOneTrueKing: my point was 17:18 why not let unskilled-basic cast it? 17:18 why let IS generate spells you literally cannot cast 17:18 that is silly 17:18 and I don't mean because fail rate is too high 17:20 you do have a point there 17:20 The "cannot cast" part should be in failure rate, not memory, for one thing. 17:21 -!- Jonathanhanes has joined #hardfought 17:23 -!- jonadab has quit [Ping timeout: 248 seconds] 17:24 -!- rikersan has joined #hardfought 17:25 -!- Elronnd has quit [Ping timeout: 276 seconds] 17:25 NeroOneTrueKing: have you seen FIQ/ais523's idea to do away with failure rate in favor of power cost? 17:26 I'm not sure that was mine, but I at least do remember discussing it 17:26 not yet, but it sounds interesting 17:26 it might have been mine, it'd have come out of a longer conversation in that case 17:26 -!- Elronnd has joined #hardfought 17:26 the basic idea is to take the failure rate formula and divide the Pw cost by the calculated rate, rather than using it as a fail chance 17:26 err, success rate, not fail rate 17:26 so 50% fail doubles pw cost, 99% fail multiplies pw cost by 100 17:27 https://nethackwiki.com/wiki/User:Phol_ende_wodan/YANI/IRC_archive#Making_utility_spell_failure_matter 17:27 I like that 17:27 -!- Jonathanhanes has quit [Ping timeout: 240 seconds] 17:27 it still only makes utility spell failure matter in a situation where your maxpw is too low, just makes things a bit less grindy 17:28 one thing it does do is buff !oGE, but probably not by that much 17:28 -!- rikersan has quit [Ping timeout: 240 seconds] 17:28 !oGE need a use outside of alchemizing anyway 17:29 It's harder to deliberately spend resources to increase your maxPw so you can grind than it is to just grind because you only need 20 maxPw anyway. 17:29 it always struck me as weird that !oH works for either maxHP increase or curHP increase, but !oGE affects only maxPw, not really curPw 17:30 aosdict: I'm thinking more of a situation where you have a level 1 spell at 90% fail but you actually have 50 maxPw 17:30 that's pretty grindy either way 17:30 Yes, I think there should be something like an equivalent of extra healing for Pw 17:30 Your options are always either gain energy, which is akin to !oH, or confused charging. 17:33 aosdict: gain energy is nothing like !oH 17:33 it always increases the max 17:34 I mean in terms of "items that give you a quick immediate boost" 17:34 it's akin to healing in terms of current increase magnitude 17:35 right, but !oH is only really useful for instant healing and for alchemy (and for some status cures if blessed) 17:35 whereas for !oGE, the maxpw increase is its main purpose for existence 17:36 well, that's what I want: something that restores a good chunk of Pw, doesn't have to change maxPw, that isn't quite as drastic as confused charging 17:36 because there is not much else right now if you run out in the middle of a fight 17:37 and plain old !oGE isn't that good for a quick power boost 17:37 hmm, I wonder if an item that let you cast from HP would be balanced and/or used 17:38 although it'd probably have to be an amulet and some amulet alternatives are so good that they knock out other competitors 17:38 actually the amulet slot is kind-of weird at the moment, it's a mix of really powerful effects and really niche effects 17:39 Sounds like an idea for a dnethack artifact 17:39 apart from amulet versus poison, which would be the only one that really functioned if not for the fact that you rarely ID it early enough to be able to use it usefully 17:44 -!- rikersan has joined #hardfought 17:45 Amulet versus poison is fairly good as an early non-cursed non-bad amulet you can't identify but probably should wear anyway. 17:45 of course, with poison immunity, it'd be even better. 17:45 right 17:45 I don't have any real issues with the amulet versus poison 17:45 it's all the others that have issues 17:46 basically because they're either so niche that they're necessarily outcompeted, or else so powerful that you wouldn't use anything else 17:46 or polyfodder 17:46 (this latter category includes life saving and, if you don't have it in another slot, reflection) 17:47 YANI: wearing a cheap plastic imitation of the Amulet of Yendor incurs the sucker markup in shops. 17:47 spork added the amulet of power, at least. 17:48 And extrinsic ESP is nice since the only other source is some artifacts, but yes, it is probably outcompeted by life saving or reflection. 17:48 aosdict: cpiot"oY currently doesn't do anything (apart from refuse to go into bags in some variants), that doesn't seem like the best way to give it a use :-P 17:48 I can see a direction in which amulet choice is a playstyle thing 17:48 e.g. amulet of power for a caster 17:48 bug_sniper had an idea that you could use amulets of strangulation *offensively* 17:49 and amulet of unchanging for polyform play 17:49 aosdict: ah, the Shiren method of balancing items 17:49 let anything be used on an enemy no matter how little sense it makes 17:49 Nah, it was just for that one. 17:49 I guess you could potentially manage to land the amulet on someone else's neck? presumably an amulet of strangulation would be particularly attracted to those 17:50 -!- rikersan has quit [Ping timeout: 240 seconds] 17:50 Yes, the idea was that you could apply or otherwise get the amulet around their neck (doesn't work on headless of course), and then they'd die in 5 turns 17:51 Maybe there could be a chance of breakage so it's not infinitely reusable 17:52 ooh, I didn't think of it being reusable 17:52 that said, an attack that kills most enemies in 5 turns wouldn't be particularly broken if it kept you occupied during that time 17:53 the only reason that this might be broken is that you can strangulate one enemy while meleeing another 17:54 well, the current behavior of the amulet establishes that it can kill by itself 17:56 I think the big balance thing here is that it could be used as a cheap way to kill any hard boss, Rodney, Demogorgon etc 17:56 well, not Juiblex I guess, since he's headless 17:57 just generate those bosses already wearing an amulet 18:00 -!- ais523 has quit [Remote host closed the connection] 18:01 -!- APic has quit [Quit: Scheduled Downtime. See You later, Gals and/or Pals. ☺] 18:01 -!- ais523 has joined #hardfought 18:06 -!- rikersan has joined #hardfought 18:07 -!- raisse has quit [Ping timeout: 246 seconds] 18:11 -!- rikersan has quit [Ping timeout: 240 seconds] 18:19 aosdict: helm of telepathy 18:20 oh right 18:21 also 18:21 wouldn't it be a bit mean to make monsters able to strangle you by throwing amulets of strangulation? 18:25 The original idea never talked about throwing them. We're talking grab the head and force it onto the neck. 18:26 -!- rikersan has joined #hardfought 18:27 size/STR roll to resist? 18:28 -!- eady has quit [Remote host closed the connection] 18:28 hmm, if you had to hit the monster with a wielded "oS 18:28 then you'd at least have the disadvantage of a) having to hit, and b) not having your regular weapon wielded 18:29 (although b is less of a problem with AceHack/NH4 mechanics) 18:31 -!- rikersan has quit [Ping timeout: 258 seconds] 18:34 NeroOneTrueKing: yeah, seems reasonable. You shouldn't be able to do that to a MZ_HUGE or maybe even a MZ_LARGE monster. 18:37 (hmm, also, choking via rope golem should be lumped into all the "breather timer" yanis) 18:46 -!- rikersan has joined #hardfought 18:51 -!- rikersan has quit [Ping timeout: 248 seconds] 18:55 -!- ACTPic has joined #hardfought 18:57 -!- rikersan has joined #hardfought 19:01 -!- rikersan has quit [Ping timeout: 260 seconds] 19:02 -!- ACTPic has quit [Quit: leaving] 19:12 [sp] k2 (Pri Hum Mal Law), 757 points, T:1791, killed by a straw golem, while frozen by a monster's gaze 19:16 -!- rikersan has joined #hardfought 19:21 -!- rikersan has quit [Ping timeout: 258 seconds] 19:22 [sp] k2 (Pri Hum Mal Law), 157 points, T:490, killed by a falling object 19:36 -!- rikersan has joined #hardfought 19:41 -!- rikersan has quit [Ping timeout: 248 seconds] 19:46 -!- rikersan has joined #hardfought 19:47 [sp] k2 (Pri Hum Mal Law), 7337 points, T:3343, poisoned by a locust 19:50 -!- jonadab has joined #hardfought 19:50 -!- mode/#hardfought [+v jonadab] by ChanServ 19:50 -!- rikersan has quit [Ping timeout: 248 seconds] 19:53 -!- greqrg has joined #hardfought 19:57 -!- rikersan has joined #hardfought 20:02 -!- rikersan has quit [Ping timeout: 260 seconds] 20:04 -!- Tarmunoraway is now known as Tarmunora 20:16 -!- rikersan has joined #hardfought 20:21 -!- rikersan has quit [Ping timeout: 240 seconds] 20:23 -!- ACTPic has joined #hardfought 20:25 -!- ACTPic has quit [Client Quit] 20:26 -!- APic has joined #hardfought 20:32 -!- tacco has quit [] 20:36 -!- rikersan has joined #hardfought 20:41 -!- rikersan has quit [Ping timeout: 248 seconds] 20:45 !tell K2 yo 20:45 Will do, FIQ! 20:45 oops 20:45 wut up dog 20:45 K2: Message from FIQ at 2017-10-02 20:45 EDT: yo 20:46 !tell K2 can you give me MTF's save? it's MTF_1506231166 20:46 Will do, FIQ! 20:46 yo 20:46 K2: Message from FIQ at 2017-10-02 20:46 EDT: can you give me MTF's save? it's MTF_1506231166 20:46 oh you're here 20:46 cool 20:46 i'm actually a figment of your imagination 20:46 ok 20:46 i'm not the droid you're looking for 20:46 * FIQ punches himself 20:46 no I am awake 20:46 there is no spoon 20:47 you know what would be even nicer though 20:47 if you could actually give me the save so I can fix the bug MTF is having ;) 20:49 ah 20:49 ok one sec 20:50 ais523: I was also going to change the tame color to green. Do you have some kind of objection to this? 20:50 The main reason being that object piles being blue is pretty ingrained in people's minds at this point 20:50 and green meaning tame has predecent in other RLs (DCSS use it, for example) 20:51 FIQ: blue was chosen for the best possible contrast against monsters that are generally tame 20:51 green makes some foregrounds hard to see 20:51 apart from that I don't mind 20:51 Yeah I figured there was a reason you picked that color 20:51 FIQ see pm 20:57 -!- rikersan has joined #hardfought 21:02 -!- rikersan has quit [Ping timeout: 260 seconds] 21:06 -!- noty has joined #hardfought 21:07 -!- rikersan has joined #hardfought 21:12 -!- rikersan has quit [Ping timeout: 258 seconds] 21:15 -!- ais523 has quit [Ping timeout: 248 seconds] 21:15 !tell K2 pushed fixs 21:15 Will do, FIQ! 21:15 *fixes 21:19 -!- rikersan has joined #hardfought 21:19 !tell K2 pushed fiqs 21:19 Will do, aosdict! 21:29 [dnh] rikerw (Bar Inc Mal Cha) was given their Quest, on T:11772 21:30 -!- ais523 has joined #hardfought 21:37 ais523: did you see my memo a while back regarding base64 xlog entries? 21:37 for proper escaping of them 21:37 FIQ: I saw a mention of it but can't remember the details 21:37 [dnh] rikerw (Bar Inc Mal Cha) killed Thoth Amon, on T:12347 21:37 basically 21:37 I wanted to do away with the stupid escape problems 21:37 so I figured I'd just add base64 versions of the fields 21:37 since NH4 already has base64 code anyway 21:38 ah right, in xlog 21:38 yep 21:38 I think there's some argument for trying to stick close to vanilla's code and not making Mandevil's life even harder 21:38 and I also think it's beneficial for the xlog to be human-readable 21:38 hmm I wonder if 4724c82 is something I fucked up or from nethack4 or even vanilla 21:38 ais523: I set it up in such a way to be backwards compatible 21:39 I did not change the semantics of existing fields 21:39 I just added, for example, dumplog64 21:39 in addition to dumplog 21:39 So Mandevil's thing, K2's stuff, etc, still works fine 21:39 base64-aware or not 21:39 just they'll keep suffering from the escape issues 21:39 there's also some argument for reducing the amount of junk in the xlog :-P 21:40 shhhh 21:40 also 4724c82 has a surprisingly large range of people who could have messed it up, including dtsund 21:41 who isn't normally a potential culprit for this sort of thing 21:41 just looked at NH3, looks like it was all good back then 21:41 probably happened when unweapon changed to bashmsg 21:41 git blame with lots of options to ignore whitespace, renames, etc. is probably the best way to find out if you really care 21:42 I dunno about vanilla barbarians but dnh thoth amon is a pushover 21:42 it took 1 turn, 2 hits 21:42 this is also true in vanilla 21:42 the only thing it has going for it really is that it's a spellcaster 21:42 it's guarding the Heart of Ahriman, an actually decent monster would be wasted :-P 21:42 Haha 21:43 But this is dNetHack 21:43 Heart of Ahriman is actually useful there 21:43 It doesn 21:43 't have any fancy invocation effects like most neutral artifacts 21:43 but it gives 3 very good resistances 21:43 MR, reflection, drain res 21:43 wow, that is pretty crazy for slotless 21:43 This puts it pretty much on par with the other neutral quest artifacts 21:44 also, lev on invoke? 21:44 Yeah IIRC that was kept 21:44 But not sure 21:44 it has remove curse on invoke I think 21:44 that looks like an early wish for realtime speedruns, to me 21:44 except that they probably wouldn't be playing neutral 21:44 Too bad dNH disallows artiwishing 21:44 er 21:44 quest artiwishing 21:44 artiwishing is huge in dnh lol 21:45 ais: yeah I was wondering how hard it was in vanilla with a useless artifact 21:45 ais523: I actually did a vampire barbarian speedrun in dnh once 21:45 ais: dnh is blessed rmv curse on invoke 21:45 it's insanely nice, you can uncurse your two-handed weps 21:45 Back when it had an OP artifact 21:45 That was supposed to have drawbacks 21:45 ? 21:45 but a caveat with them made them not apply to undead 21:45 including player vampires 21:45 rikersan: Ah, it was changed then 21:45 yep 21:45 no more lev 21:46 I still think the vampire changes were bad ;-; 21:46 the no blessed armor 21:46 I can get no silver weps but blessed is just annoying 21:47 YASI: sometimes Rodney interventions bless your items rather than cursing them 21:50 Hm 21:50 Apparently using polearms isn't very easy to abort 21:50 it is set to be forced 21:50 Is there a reason for this? 21:50 Otherwise I'll just get rid of that 21:50 almost certainly someone wrote that backwards 21:50 -!- rikersan has quit [Ping timeout: 248 seconds] 21:51 I see no reason to force that 21:51 the blessed penalty is small, 4 per item 21:51 while the silver penalty is 20 per 21:51 (all divided by 2) 21:51 in general I'm against forcing direction prompts; the only advantage I can see to forcing position prompts is in the case of things like involuntary controlled teleport 21:51 which could plausibly be cancelled by mistake 21:51 and which have a clear way to cancel them intentionally 21:52 hmm 21:53 if you have teleportitis + teleport control 21:53 maybe I should make it so that if you are busy i.e. searching 21:53 you automatically dismiss telportations 21:53 not sure 21:53 bad idea, many players do long searches to trigger teleportitis when they can't teleport-at-will for some reason 21:53 I find that prompt rather annoying at times 21:53 but unsure how to properly fix this 21:54 I can see wanting to improve the UI but it should perhaps work some other way 21:54 ais523: I explicitly said "+ teleport control" 21:54 but it's still dubious 21:54 FIQ: controlled itis doesn't necessarily let you /trigger/ the teleport 21:54 FIQ: waiting for it to trigger doesn't cost 20 pw + some hunger 21:54 if you're low level you have to wait for random itis and then control the destination 21:54 and even if you're high level you might be low on Pw or not want to spend it 21:55 there are three relevant trinsics: teleportitis, teleport control, teleport at will 21:55 which control the presence, destination, and timing of teleports respectively 21:55 right 21:55 (although teleport at will is IIRC not represented as a trinsic in vanilla; not sure about NH4) 21:55 it is not an intrinsic in NH4 either 21:55 if you have the first two and not the third you can't control the timing of teleports, so have to wait for them 21:57 another thing that would be nice 21:57 if, while travelling, teleportitis procs 21:57 the getpos should default at travel pos 21:59 the main reason it doesn't atm is because it'd make the engine/interface API rather more complex 21:59 (in vanilla default positions are kind-of broken; in NH4 they're consistent but not as user-friendly as they could be) 21:59 so it's a lot of effort to go to for a gain that would be comparatively minor 22:00 well 22:00 if travel is interrupted 22:00 it remembers the position 22:00 I don't see why teleportitis prompts can't use this data too 22:02 hmm 22:03 I wonder what exactly I did to break covetous behaviour in such a way that peaceful covetousness is... weird 22:03 a peaceful covetous monster 22:03 will teleport to a random location 22:03 every single action 22:03 this looks a bit silly with monster detection 22:03 and annoying if you're trying to off demon lords, too 22:04 yeah I think someone mentioned that to me at some point 22:04 asmodeus/baalzebub being a pain to deal with becahse rather than demanding a bribe 22:04 "stand still Asmodeus so I can hit with magic missiles!" 22:04 they warp all over the place 22:04 Yeah. 22:05 it looks a bit amusing with monster detection 22:05 but yeah 22:05 it's silly 22:06 -!- rikersan has joined #hardfought 22:07 ais523: what would a "menucolors" configuration UI look like? 22:07 FIQ: have you seen my suggested autopickup configuration UI? 22:07 no 22:07 it'd be similar to that 22:07 I'm not sure if I posted it anywhere 22:07 I asked jonadab once to suggest a design for msgchannel option UI 22:07 it kind-of got stuck behind the same problem that held up the channelization UI, only it's even harder to write 22:08 it was pretty nice IMO but it depended on things which I don't think is possible in the current windowing system 22:08 anyway, my idea behind autopickup was to configure it via the \ menu (which would be a menu now, not just a list) 22:08 one issue 22:09 some people like to also pickup based on item name/etc 22:09 it'd allow you to select single item descriptions or (for when you have the items identified) identities, or groups that were related in some way 22:09 that usable amulets of strangulation idea was one of my nice patch ideas 22:09 for example 22:09 then you'd customize the entire selected group at once 22:09 maybe pickup items in general 22:09 but never artifacts 22:09 (blast danger) 22:09 groups would display with menucolor/autopickup properties shown if they were consistent in the group, otherwise a "(varies)" or the like 22:10 but yes, artifacts would probably want a separate category here 22:10 not sure if I understand 22:11 so in my \ list, I have, say 22:11 -!- rikersan has quit [Ping timeout: 240 seconds] 22:11 a - potion of oil (red) - autopickup, cyan 22:11 b - unidentified potion - autopickup, dark cyan 22:11 acehack-style inventory format :D 22:11 c - potions - autopickup, color varies 22:12 d - show unidentified potions 22:12 something like that 22:12 *inventory color format 22:12 the details can vary 22:12 anyway, I see 22:12 but how do you make an exception 22:12 would you press c? 22:12 and you do it from there? 22:13 FIQ: if you wanted an exception for a particular potion class, say oil, you could do it from a 22:13 or d if it wasn't currently IDed 22:14 wouldn't I do that from a? 22:14 nice patch idea: cancelling yellow lights should let you use them as a lantern for a while 22:14 if the exception is more complex, arguably allowing customization of it is the sort of thing we don't want to encourage (as requiring complex customization is bad UIwise), but it'd probably be under the item you're checking 22:14 FIQ: this is the \ list 22:14 so items that aren't currently IDed don't show up at all 22:14 unless you expand the list of unidentified items 22:14 ah 22:15 bug_sniper: ooh, cantrip idea! Summon a tame yellow light 22:15 now, it's fairly common to want to adjust autopickup options mid-game 22:15 but probably not to want to differentiate between different currently unIDed items in a way that varies game to game 22:15 aosdict: as a temporary light source? :-D 22:16 or, hmm, that actually gives a combat advantage since it'll blind stuff... maybe should be something else that emits light but won't kill itself? 22:16 ais523: what I want to be able to do 22:16 ais523: yes 22:16 is to add more special rules 22:16 not as part of the game, but for configuration 22:16 for example 22:17 I want to not autopickup empty lakps 22:17 *lamps 22:17 -!- rikersan has joined #hardfought 22:17 FIQ: ah right 22:17 I think my original suggestion had empties as a separate category 22:17 a - empty wand, etc. 22:17 I'm not sure if I still have it written down, though 22:18 I vaguely remember not being able to find it anywhere after writing it, and recovering some of it from a backup but never finding the original file 22:18 I wonder if it's committed into a non-default branch of the NH4 repo or something 22:19 also while pondering about momster memory stuff earlier 22:20 it came to my mind that if one wanted to, one could add per-monster hostility in there 22:21 -!- rikersan has quit [Ping timeout: 240 seconds] 22:23 ais523: but yes, doing *inventory* "menucolors" as part of discovery list makes sense 22:23 but what about spells 22:23 IMO spell menucolors should be a separate feature from inventory menucolors 22:23 yes, I can see the merit i nthis 22:23 and possibly is minor enough to not even need customization 22:23 but how would you do it 22:24 FIQ: customizably? in the options menu, in a similar form to 3.6.1 statuscolors 22:25 I haven't seen that interface 22:25 I should tak e alook 22:25 *take a look 22:34 -!- rikersan has joined #hardfought 22:42 -!- greqrg has quit [Quit: Disconnected] 22:45 -!- Menche has quit [Quit: NO NO NO MY UPTIIIIIME] 22:47 -!- rikersan has quit [Ping timeout: 240 seconds] 22:52 cantrip yani: light a radius 3 area 22:52 that said I'd fight against making it customizable unless I really had to 22:52 adding options is the lazy way out 22:53 it means that your existing players get less annoyed, but it makes life harder for new players (who haven't seen your game yet and thus won't shout at you) 22:59 -!- Menche has joined #hardfought 23:02 cantrip yani: mage hand, you specify a direction and the nearest item on the floor in that direction, assuming it's under 100 weight or so, is brought to you. 23:03 (I'm reading through d&d spells looking for cantrip ideas.) 23:03 -!- rikersan has joined #hardfought 23:04 oh, that's a clever idea 23:07 mage hand would also be a nice way to get stuff out of pits while levitating. 23:07 -!- Jendic has joined #hardfought 23:07 -!- rikersan has quit [Ping timeout: 248 seconds] 23:08 cantrip yani: create an illusion of yourself at your spot. Monsters will assume it is you, and if they see both you and the illusion, they'll move towards whichever is closer. 23:11 that would be a nightmare to code 23:11 even in fiqhack 23:13 -!- rikersan has joined #hardfought 23:18 -!- rikersan has quit [Ping timeout: 240 seconds] 23:19 hmm, a lot of these are just plain attacks 23:20 there's a cantrip for 1d8 of every damage type, it seems 23:22 tfw you have to take a -3 luck penalty in sokoban 23:23 okay, so here's something I want to add: some way to put a protection charm on a corpse, so that it is not touched by pets and will not rot. But I'm not sure what the delivery mechanism should be. 23:23 reading a scroll of enchant weapon while confused and wielding the corpse :P 23:24 Having consumables for it seems like the way to go, but then they'd be too rare to see much use, or else people would hate that these mostly useless things are eating up the object probabiities 23:25 What if it transformed the corpse into a new food item that could still be eaten? 23:26 NeroOneTrueKing: the main impetus for this is for bards: if one of your tame player monsters dies, you probably don't have the resources to resurrect them, but since you can't exactly get more player monsters, you should be able to preserve their corpse until you can 23:26 aosdict: icebox 23:27 ais523: that requires your chosen corpse to die conveniently near an icebox 23:27 aosdict: ah, so a tinning kit is not what you're looking for 23:27 right, the problem is that iceboxes aren't very portable 23:27 I don't see a huge balance problem with making the box itself lighter 23:27 maybe around 200 23:27 even so 23:27 a bard should not be toting around an icebox as part of their normal play :P 23:28 just give them a /oUT if that's an issue 23:29 that could work, although it should have fewer charges than usual 23:29 but I think it's more strategically interesting to have them not be able to resurrect their comrade until they can actually get the resources 23:29 being able to do it from the get-go kind of implies you're expecting them to die :) 23:30 the thing about NetHack is, there's only one way to resurrect tame monsters 23:30 and if it randomly generates it's with a huge number of charges 23:30 /oUT is nearly useless for most characters 23:31 I was thinking of adding a song which can also resurrect tame monsters, but yes, only one exists now 23:33 -!- rikersan has joined #hardfought 23:38 -!- rikersan has quit [Ping timeout: 258 seconds] 23:48 -!- rikersan has joined #hardfought 23:56 -!- rikersan has quit [Quit: Lost terminal] 23:56 -!- rikersan_ has joined #hardfought