00:44 [hdf-us] [un] SuzuneTakada (Con Vam Fem Cha), 1457 points, T:1188, killed by a rothe 01:11 -!- unsound has quit [Ping timeout: 265 seconds] 01:13 -!- unsound has joined #hardfought 01:14 [hdf-us] [evil] dingotron (Con Hob Fem Cha) completed Sokoban, acquiring an amulet of reflection, on T:17229 01:19 -!- unsound has quit [Ping timeout: 240 seconds] 01:21 -!- NAOrsa has joined #hardfought 01:25 -!- unsound has joined #hardfought 01:25 -!- raisse has joined #hardfought 01:30 -!- unsound has quit [Ping timeout: 250 seconds] 01:31 -!- unsound has joined #hardfought 01:35 -!- unsound has quit [Ping timeout: 256 seconds] 01:36 -!- unsound has joined #hardfought 01:49 -!- porkman has quit [Ping timeout: 265 seconds] 02:02 -!- eLtMosen1 has joined #hardfought 02:05 -!- remirol has joined #hardfought 02:05 -!- mode/#hardfought [+v remirol] by ChanServ 02:06 -!- lorimer has quit [Ping timeout: 264 seconds] 02:14 -!- Tangles has quit [Ping timeout: 264 seconds] 02:14 -!- Tangles has joined #hardfought 02:14 -!- mode/#hardfought [+o Tangles] by ChanServ 02:32 -!- Bliss56 has quit [Remote host closed the connection] 02:37 [hdf-us] [nh370] Wolfechu (Sam Hum Mal Law), 10820 points, T:4539, killed by a troll 02:39 [hdf-us] [nh370] dbanethacker (Val Hum Fem Neu) entered the Gnomish Mines, on T:18228 02:41 [hdf-us] [nh370] Wolfechu (Val Hum Fem Law) entered a shop, on T:576 02:46 [hdf-us] [nh370] Wolfechu (Val Hum Fem Law) entered the Gnomish Mines, on T:998 02:47 [hdf-us] [nh370] Wolfechu (Val Hum Fem Law) signified by divine providence, was chosen to carry Excalibur, on T:1196 02:49 [hdf-us] [nh370] Wolfechu (Val Hum Fem Law) made her first wish - "blessed fixed greased +3 cloak of protection", on T:1867 02:55 -!- amateurhour has joined #hardfought 03:03 -!- Kontroller has quit [Quit: Leaving.] 03:07 [hdf-us] [nh370] dbanethacker (Val Hum Fem Neu) reached Mine Town, on T:18732 03:07 [hdf-us] [nh370] dbanethacker (Val Hum Fem Neu) entered a temple, on T:18797 03:10 [hdf-us] [nh370] oh6 (Hea Gno Mal Neu) had Magicbane bestowed upon him by Hermes, on T:49966 03:14 -!- Umbire has quit [Remote host closed the connection] 03:14 -!- Umbire has joined #hardfought 03:15 [hdf-us] [spl] SuzuneTakada (Sam Gul Nbn Law) had Excalibur thrown at them by some watery tart, on T:2909 03:24 [hdf-us] [spl] SuzuneTakada (Sam Gul Nbn Law) made their first wish - "blessed +2 gray dragon scale mail", on T:3890 03:45 -!- Umbire has quit [Remote host closed the connection] 03:51 -!- tacco\unfoog has joined #hardfought 04:01 [hdf-us] [spl] SuzuneTakada (Sam Gul Nbn Law) completed Sokoban, on T:8328 04:11 [hdf-us] [spl] SuzuneTakada (Sam Gul Nbn Law), 26515 points, T:9420, taken down by a priestess of Susanowo 04:24 -!- LarienTelrunya has joined #hardfought 04:24 -!- mode/#hardfought [+v LarienTelrunya] by ChanServ 04:29 !tell Umbire you got it wrong, the real benefit of backwards long jumping is that it lets you get into Moloch's sanctum without having to collect the invocation tools :D 04:29 Will do, LarienTelrunya! 04:29 also, who wants to go to atlantis aquaria? :D (yeah I know the english version gives it the boring name of "wet-dry world" but the german names for the courses are so much better) 04:33 Indeed :o 04:34 there's more examples. one of the courses is called "tall tall mountain" in the english version (how uncreative can they get) and in the german one it has the absolutely cool name "amanita fiasco"! 04:35 and "dire dire docks" (yaaaaaawn) became "wild water wharf" in german :D 04:35 Is it alliterated in German too? 04:35 yes, "Wilde WasserWerft" 04:36 Ah 04:49 -!- ProzacElf has quit [Ping timeout: 240 seconds] 05:05 -!- ProzacElf has joined #hardfought 05:16 KAABLAMM!!! Miezekatze triggers a land mine! Miezekatze falls into a pit! Miezekatze is killed! 05:16 poor kitty 05:18 <@Suzune Takada> https://www.hardfought.org/userdata/S/SuzuneTakada/splicehack/dumplog/1586415176.splice.txt 05:18 <@Suzune Takada> previous play: previous bone was angry priest after the convert altar 05:19 <@Suzune Takada> so, I just wanted to protection, but priest became angry so I dead 05:19 <@Suzune Takada> If you find that bone, there are GDSM, amulet of reflection, Excalibur and Snickersnee 05:20 -!- bug_sniper has quit [Ping timeout: 264 seconds] 05:21 yes, priests with a converted altar are angry (not hostile until you talk to them though) and can't sell you protection 05:21 not even if you convert the altar back to the priest's alignment 05:28 -!- mobileuser has joined #hardfought 05:29 <@Eternal Guardian> It's spelled roguelike. 05:29 but it's a red ink marker (read one) so rougelike is appropriate 05:29 <@Eternal Guardian> It's spelled roguelike. 05:29 I read that the first time, that's why I replied 05:29 <@Eternal Guardian> It's spelled roguelike. 05:30 * LarienTelrunya sprays the bot's face with rouge :P 05:30 oh sorry I thought it was an obstinate human 05:31 heh no, we found out a while ago that it's a bot programmed to just give that answer every time one "misspells" the word roguelike 05:31 annoying 05:32 at the moment I still think it funny to trigger it, but it will wear off ;-) 06:45 of course the 200iq play would be to make a roguelike called rougelike 06:45 <@Eternal Guardian> It's spelled roguelike. 07:01 -!- theRaisse has joined #hardfought 07:01 -!- raisse has quit [Ping timeout: 260 seconds] 07:09 -!- eLtMosen1 has quit [Ping timeout: 265 seconds] 07:16 Pro-tip: don't put Magicbane into your BoH. 07:16 -!- theRaisse is now known as raisse 07:16 does it do that in evilhack? 07:17 I was just going to say "that's nonsense" but I don't know the full extent of K2's evilness 07:17 Yes, but at least only the potions are destroyed 07:18 in vanilla Magicbane also destroys your BoH but then you lose all the contents as well. 07:19 they pulled that into vanilla??? oh god 07:19 uhm 07:19 if I remember that correctly 07:20 tested with 3.6.1 it doesn't blow up the bag, but who knows what was changed since then 07:20 no my mistake 07:21 I just *assumed* magicbane would blow up your BoH in vanilla 07:22 so I appreciate the logic of that but it is still just an evil evilhack thing :-) 07:26 because magicbane has cancellation magic, there's a small chance it can blow up a BoH if put inside. this is only an evilhack thing 07:31 small chance???? so I got really unlucky then. :-) 07:31 si 07:35 I suppose magicbane itself blew up as well? 07:35 yes 07:35 darn 07:36 cancellation is like the antimatter of the network world. its volatile stuff 07:39 Yeah, I slapped my forehead hard after doing that. Plundering bones, at least I got their BoH blessed again. 07:39 only sometimes? K2 you're so merciful! 07:40 hah 07:47 You probably don't need to be putting Magicbane in a bag of holding anyway though? 07:47 It won't blow it up in vanilla though 07:48 No, but I was plundering bones and getting burdened 07:48 only an actual wand of cancellation with charges; not even a spellbook of cancellation 07:49 Still, it sounds reasonable that Magicbane can blow up your BoH. It's a bane of magic. 07:50 But not in vanilla, learned something again... 07:50 a scroll of charging charges an empty bag of tricks in unnethack 07:50 -!- stenno has quit [Ping timeout: 258 seconds] 07:51 if you put the scroll into the empty bag of tricks 07:52 neat trick! 07:52 I wouldn't have a problem with that if it didn't destroy the items 07:52 it should just hang on to its cobj and make the items inaccessible until the bag is again emptied of charges 07:59 It eats them 07:59 Which makes some kind of sense. 08:02 it's just a stupid gotcha like putting cancellation wands in a bag of holding 08:02 Doing that in UnNetHack doesn't destroy the other items in the bag though. 08:03 in evilhack blowing up a BoH, the items in the bag that get destroyed are only those that can break easily (potions, anything made of glass that isnt fooproofed) 08:03 Mm 08:03 everything else is scattered 08:04 You can shatterproof glass? 08:04 you can, such as glass weapons or armor 08:04 but not potions themselves 08:04 shows as 'shatterproof' 08:05 so, its not a total loss 08:05 Can you have a shatterproof crystal ball? 08:05 i believe so yeah 08:06 i'll verify 08:06 I don't believe so, this isn't fourk where you can erodeproof anything 08:07 using a bag of tricks as a container is not the standard use of it. 08:07 Sometimes you don't find any other bag though. 08:08 hmm i guess not... 08:09 oh wait 08:09 * K2 needs to wake up 08:09 K - an uncursed shatterproof crystal ball (0:1) 08:10 boop 08:11 K2 some of my items like scrolls got scattered into lava. Retrieving them by zapping wand of cold and digging. But I am surprised they didn't burn. 08:12 hah yeah me too 08:12 * K2 takes notes 08:13 fireproof scrolls? 08:13 nope 08:14 no doubt they can be fireproofed in evilhack :-) 08:14 those are again possible in fourk 08:15 nah you cant fooproof potions/scrolls/spellbooks in evilhack 08:15 not when wielding and reading confused scroll of enchant weapon? 08:15 but pretty much everything else can be 08:15 ah 08:16 I keep learning 08:16 if you fooproofed scrolls or spellbooks, they'd become too stiff to open up 08:16 YANI: oerodeproof wands and rings should be immune to breaking from electricity damage 08:46 can we get another way to erodeproof things than needing 10 million enchant scrolls? :| 08:47 every valid reason to erodeproof things that gets added is more pressure on scroll supply 08:47 depending on who you talk to, that's the point 08:48 I believe jonadab's perspective on it is if you want fireproof full healing potions you can carry through Gehennom openly, then it's worth about the same as a use of enchant weapon 08:49 in variants that have things like destroy armour eroding things to the point of destruction 08:49 it's more a need than a want 08:53 I think it's quite difficult to balance the amount of repairing and erodeproofing resources. 08:54 But would err on the side of scarcity. 08:55 [hdf-us] [nh370] samiam (Mon Hum Mal Law) hit with a wielded weapon for the first time, on T:13981 08:56 -!- nabru has quit [Read error: Connection reset by peer] 08:58 -!- nabru has joined #hardfought 08:58 [hdf-us] [nh370] Karnac (Pri Hum Mal Neu) made his first wish - "2 blessed scrolls of charging", on T:21 09:00 [hdf-us] [nh370] Karnac (Pri Hum Mal Neu) made his first artifact wish - "blessed rustproof +2 Mjollnir", on T:24 09:00 [hdf-us] [nh370] Karnac (Pri Hum Mal Neu) wished for "blessed +2 silver dragon scale mail", on T:27 09:01 [hdf-us] [nh370] Karnac (Pri Hum Mal Neu) wished for "blessed fireproof +2 pair of speed boots", on T:35 09:05 [hdf-us] [nh370] Karnac (Pri Hum Mal Neu) entered a shop, on T:298 09:06 -!- LarienTelrunya has quit [Quit: Connection closed] 09:06 [hdf-us] [nh370] Karnac (Pri Hum Mal Neu) entered the Gnomish Mines, on T:342 09:10 [hdf-us] [nh370] Karnac (Pri Hum Mal Neu) reached Mine Town, on T:583 09:12 -!- LarienTelrunya has joined #hardfought 09:12 -!- mode/#hardfought [+v LarienTelrunya] by ChanServ 09:12 -!- remirol is now known as lorimer 09:13 [hdf-us] [nh370] Karnac (Pri Hum Mal Neu) entered a temple, on T:674 09:13 aosdict: My thinking is, you're using up a resource (enchant weapon scroll) for _each_ item you fooproof. So it's limited. 09:14 Nobody's going to fooproof fifty scrolls or fifty potions or whatever. 09:14 jonadab!!! I'm in need of your assistance, you said something about polka music in gehennom back in the day? 09:14 Oh dear. 09:15 <@Pacra> someone found a t:20 WoW? 09:15 Maybe I should go start packing up the car, in case we suddenly need to drive to Indiana, to help my sister move into her new apartment. 09:15 And by "suddenly", I mean my mom and sister have been planning this for two months. 09:15 since I can't force the game to output actual polka music, I'd like to at least make it display random quotes from polka music; jonadab: are there songs that you can recommend for this? :) 09:16 Polka has lyrics? 09:16 I mean, I know Weird Al takes pop-song lyrics and sings them to a polka background, but that's parody. 09:17 damn... but my roommate has some foreign polka songs and they do have lyrics 09:17 IIRC they have names like "särkki järven polkka" 09:18 No idea about that. Polka to my knowledge is just that extremely repetitive accordion music they play at Octoberfests. 09:20 This junk: https://www.youtube.com/watch?v=h1TnAPNBV7Y 09:21 <@Pacra> @ jonadab rough time to travel 09:23 haha thanks, maybe if I ever make it to gehennom in non-slex again I'll put that on an endless loop :P 09:27 the only polka music quote i can think of is 'eww... polka music' 09:27 <@Suzune Takada> https://www.hardfought.org/userdata/S/SuzuneTakada/splicehack/dumplog/1585275313.splice.txt 09:27 however, id' rather listen to it than country music 09:27 <@Suzune Takada> It's my first meet Warden Arianna 09:27 [hdf-us] [spl] hyperion67 (Arc Hum Mal Law) became literate by engraving "Elbereth", on T:4248 09:28 <@Suzune Takada> I had nice weapon but she was too powerful 09:29 -!- tonehack has joined #hardfought 09:30 [hdf-us] [evil] Grouchy (Wiz Hum Mal Cha) killed Asmodeus, on T:47336 09:30 -!- porkman has joined #hardfought 09:30 hi 09:30 Hello porkman, Welcome to #hardfought 09:31 @Suzune Takada AC -8 is not good enough against her :-( 09:32 Very hard hitting, but she is easy to defeat if your AC is low enough 09:32 in evilhack anyway... 09:34 LarienTelrunya: https://en.wikipedia.org/wiki/Ievan_polkka 09:35 add leeks too if you haven't already 09:35 ah yes, Loituma 09:35 now I remember 09:35 so we can spin them to the lyrics 09:35 <@Suzune Takada> Ievan polkka? https://www.youtube.com/watch?v=jrgO_9ey53M 09:36 same song, yeah 09:36 :) 09:37 <@Suzune Takada> I think convict in spl is a little easier than other variants 09:37 [hdf-us] [spl] hyperion67 (Arc Hum Mal Law) made his first wish - "blessed greased +3 gray dragon scale mail", on T:4515 09:38 <@Suzune Takada> Because of Bat from Hell 09:38 <@Suzune Takada> (Convicts can reach expert for club) 09:38 quicksilver 09:38 is the best convict weapon in DNH 09:38 not sure if it got nerfed in later versions 09:39 Nerfed 09:39 <@Suzune Takada> of course, just a little. convict is the most difficult role 09:39 <@Suzune Takada> even in spl 09:39 Two attacks now, not 4. So it's a little better than a 2x weapon, not comically so. 09:40 [hdf-us] [nh370] Karnac (Pri Hum Mal Neu) acquired the Mines' End luckstone, on T:2527 09:41 [hdf-us] [nh370] eraserhead97 (Cav Hum Fem Neu) entered the Gnomish Mines, on T:374 09:43 heh 09:44 [hdf-us] [nh370] nabru (Val Dwa Fem Law) entered a shop, on T:292 09:44 [hdf-us] [nh370] nabru (Val Dwa Fem Law) entered the Gnomish Mines, on T:307 09:44 I ascended a foodless atheist true polyselfless...maybe genocidelss too vampire convict in unnethack at one point 09:44 -!- ais523 has joined #hardfought 09:44 -!- mode/#hardfought [+v ais523] by ChanServ 09:47 porkman congrats on your 2nd slex asc 09:47 is that a streak also? 09:47 aye 09:48 cool 09:49 there's also a scum on the dumplog list after the streak, that was when i just logged in and because of the options file (which autoselects the role and race), it created a lvl 1 wizard, at which point i went #quit (so it wouldn't overwrite a backup save) and went WTF my wizard disappeared on IRC, 09:49 only for Amy to say something like "bwahaha i nerfe 09:49 d slex more, your game is now in old slex, new nerfier version is on server!) 09:50 heh 09:50 butthat's a moot point, cause i can't start any more games in the current version 09:50 lol, I just updated the server to the new version 09:50 and i refuse to play the newest one 09:50 Anyone wanna help me variant developing? :) 09:51 LarienTelrunya there's a new version of slex ready? 09:51 porkman: what specific is wrong with the newest version that prevents you playing it? 09:51 hardfought is still on 2.6.0 09:51 K2: not ready for hdf; I plan to feature freeze on May 1st, and then I'll probably bundlebundlebundle you ;) 09:51 also, variant developing is an incredibly big field, depending on what you want to do, that changes what you'll have to learn 09:51 I shall make... Pork'em! a SLEX pork! er, Fork! 09:52 * K2 doesnt want to be bundlebundlebundled 09:52 things like "minor adjustments to the stats of items", and to some extent "new monsters", are very easy to learn, other things can be much harder though 09:52 aye, I'm starting with teh easy ones 09:52 forking slex seems like a bad idea because the larger the codebase, the harder it is to find things in it 09:53 porkman: start with a one level nethack and then expand from there. 09:53 [hdf-us] [nh370] eraserhead97 (Cav Hum Fem Neu) reached Mine Town, on T:1753 09:53 [hdf-us] [nh370] eraserhead97 (Cav Hum Fem Neu) entered a shop, on T:1753 09:54 bhaak: well, if you're working from an existing NetHack codebase, removing branches can be hard 09:54 because some parts of the game's code will assume they exist 09:54 so you need to find and remove the assumptions 09:54 <{Demo}> porkman: do u have a setup to compile slex from source? 09:55 not really 09:55 I don't want to remove anything 09:55 ah right, if you're planning to change anything other than level or branch layouts, you'll need to be able to compile the game from source before you can do anything 09:55 that said, I have a suspicion that anyone who wants to fork slex, but not remove /anything/, is beyond help :-D 09:55 lol "I don't want to remove anything --porkman", that's going on your variant's grave once it finished turning into a second slashthem :P 09:56 I'll...unnerf lots of things 09:56 but i won't straight out *remove* much at all 09:56 * aosdict is not sure what pork'em will do that slex wouldn't also incorporate 09:56 [hdf-us] [nh370] eraserhead97 (Cav Hum Fem Neu) entered a temple, on T:1943 09:56 making a SLEX variant is probably as easy as making a marble statue. just chip away what is not good and you will land at perfection. 09:56 for example, make spells work again 09:56 what happened to them? 09:56 a little better than they currently do 09:57 and are we talking abour player spells or monster spells? 09:57 player 09:57 where to begin 09:57 magic mapping maps a few tiles, like a cursed scroll in vanilla 09:57 map level fails something awful, like 1 in 12/14/16 or so depending on skil llevel? 09:57 * aosdict is eyeing a player monster spellcasting merge for xnethack 6.0 09:57 spells backlash far too often 09:58 so while i won't *remove*, say, data delete (the high priest can still have it), i'll not make it possible for it to happen as a spell backlash, etc 09:58 ais523, so i'll need some sort of software for source editing? 09:59 porkman: yes, although almost anything will do, some programs are more convenient than others 09:59 use wordpad, it's the slex approved editor 09:59 what you will need is a text editor (as opposed to a word processor); although Notepad is in theory enough, getting one with more features is probably a wise idea 09:59 (actually, Notepad can't open some of slex's files, they're too large) 09:59 any suggestion? 09:59 I still remember very well how Soviet5lo originally conceived SLASHTHEM because he wanted to change one aspect of one particular playable ace, and in the end it turned into basically SLASH'EM + some patches, removing all trace of SLEX over the years. 10:00 porkman: what OS are you on? 10:00 i remember there was an... ultraedit or something 10:00 *race 10:00 win 10 10:00 porkman: vim 10:00 I vote Notepad++ :-) 10:00 porkman: there's a tradeoff between "powerful" and "easy to learn" 10:00 notepad++ is a pretty good point on the scale, I think 10:00 (and free) 10:00 porkman: more serious answer. maybe sublime 10:01 most serious programmers will put the time into learning a particularly powerful editor, but that's not cost-effective unless you want to make it a career, I think 10:01 * K2 loves notepad++ 10:01 Right, i see 10:01 porkman: or maybe even visual studio code 10:01 I should also mention Visual Studio Code because it's one of the most popular Windows source editors at the moment, although I don't personally like it much 10:01 let's try notepad== 10:01 ++ 10:02 [hdf-us] [spl] hyperion67 (Arc Hum Mal Law) had Demonbane bestowed upon him by Quetzalcoatl, on T:5276 10:03 In 2015 Stack Overflow conducted a worldwide Developer Survey, and Notepad++ was voted as the most used text editor worldwide with 34.7% of the 26,086 respondents claiming to use it daily. 10:03 that's a /lot/ of developers 10:03 everyone knows the real choice is between vim and emacs 10:03 It's my editor of choice when I'm stuck on Windows 10:03 Okay, I have notepad++ 10:04 ais523: also, I think sublime has become disproportionately more popular since 2015 10:04 nabru: I used to carry around an emergency USB stick in case I was stuck on an unfamiliar computer, notepad++ was one of the programs on it 10:04 but I didn't use it in any other context 10:04 ais523: that's very outdated 10:05 i was going to edit the source directly on github 10:05 vimmmm 10:05 but i guess that would be a fool's errand 10:05 Emacs > * 10:05 bhaak: did VS Code even exist in 2015? 10:05 I've been playing with VS code, it's a nice editor actually. But vim is more convenient 10:05 <{Demo}> when i am on windows i just use vim 10:06 tried emacs with evil, couldn't nicely replicate a tabbed workflow within 30 secs and so moved back 10:06 ais523: initial release was in 2015. that's also why it's outdated. but 5 years in IT are a long time 10:06 https://insights.stackoverflow.com/survey/2019#development-environments-and-tools 10:06 vim is in a really weird place among editors; it focuses almost exclusively on the actual text editing part of things, whereas most editors are more concerned about the ancillary things around it 10:06 let's set aside the vim-or-emacs debate for now and bask in our combined superiority over the other editors 10:06 aosdict: right, the real editor war is vim+emacs vs. everything else 10:07 <{Demo}> that's not a war 10:07 tried vs code with vim plugins, some features were missing or buggy so went back to (neo)vim 10:07 Visual Studio Code 50.7%, Visual Studio 31.5%, Notepad++ 30.5%, IntelliJ 25.4%, Vim 25.4% 10:07 <{Demo}> that's a slaughtering 10:07 10:07 10:07 and emacs at only 4.5% 10:07 amateurhour: that's pretty much me on my work computer 10:07 there was one bug where 'u' would undo _everything you had done in that session_ one at a time, slowly before your eyes, unstoppably 10:07 amateurhour: was there at least a redo? :-D 10:07 amateurhour: in vscode? that happened to me :D 10:08 amateurhour: what's wrong with using mulitple tools? when I was still doing Java for money, I was using vim and eclipse concurrently 10:08 yeah, but there wasn't a symmetric bug, you had to spam redo lol 10:08 I guess if it worked like Emacs's default undo/redo interface, then the redo would redo everything then undo everything again 10:08 (assuming the same bug) 10:08 it was some interaction between vs code's concept of history and how the plugin tried to implement it 10:09 ais523: i use emacs for mail, usenet and irc, and I don't even use it for development 10:09 <{Demo}> lol 10:09 nothing wrong with multiple tools bhaak imo 10:09 bhaak: that actually makes sense given Emacs' history 10:10 it's a little weird that GNU Emacs became best known as a text editor, it was originally intended to be a desktop environment 10:10 (which explains all the "Emacs is an OS" jokes) 10:10 ok, so 10:10 but I think it was a little too radical in terms of how it works as a DE 10:11 i have notepad++, how would I make it compile slex? 10:11 amateurhour: the real reason I can't use vscode is that it would freeze up on files of only a couple megabytes 10:11 oh yeah I had that problem too aosdict 10:11 porkman: notepad++ doesn't do the compile, it's just used for editing 10:11 <@Suzune Takada> I think binder in dnh is only for experts of nethack 10:11 on a computer with what ought to be a ton of memory 10:11 for the actual compile you'll need a separate toolchain, I'm not 100% sure how that works on Windows for Slash'EM 10:11 <@Suzune Takada> memorize of binding souls are not easy 10:11 [hdf-us] [nh370] nabru (Val Dwa Fem Law), 219 points, T:854, killed by a giant bat 10:12 aosdict: I looked into making language plugins for VS Code, the API seems oddly designed 10:12 [hdf-us] [evil] krm26 (Bar Hum Fem Neu) wished for "blessed greased fireproof +2 dragonhide cloak of protection", on T:49620 10:12 e.g. you have to indicate offsets within a file as though it were encoded in UTF-16, which in most cases means internally re-encoding the whole file because UTF-16 isn't a very common format for source code 10:12 ais523: do you have any citation for that claim? it got "editor" in its name, so it would be a strange if it wasn't originally intended to be an editor. being a *programmable* editor of course opens up a can of worms that slowly usurps all possible usecase of a user (and therefore it's almost inevitable to land at a DE) 10:13 bhaak: the early docs for the GNU project talked about having programs written in "C or Scheme", Emacs was the only Scheme interpreter around at the time 10:13 err, not Scheme 10:13 Lisp 10:13 implying that they planned to use it as an interpreter to run their programs 10:14 (they later switched to Guile, which is a Scheme dialect, rather than elisp) 10:14 ais523: ah, okay. but then that's GNU emacs, not the earlier emacs variants 10:15 Very long time ago I programmed in lisp and prolog in emacs, it was very much like its own os 10:15 bhaak: I qualified as GNU Emacs specifically 10:15 emacs predates GNU and IIRC what would become GNU emacs also predates GNU 10:15 most versions of Emacs were, indeed, intended primarily as editors 10:15 although the original, original Emacs was a plugin for TECO 10:15 right, I didn't notice the GNU 10:15 what was that wiki article with the basics of variant developing... 10:16 you mean the one about like nethack 3.1.0? 10:16 Amy, which software did you use? 10:16 mingw 10:16 actually that's a good point, you probably want to match LarienTelrunya's toolchain, because it wouldn't surprise me if nothing else worked at this point 10:16 [hdf-us] [nh370] nabru (Val Dwa Fem Law) entered the Gnomish Mines, on T:281 10:16 it's easy to break a project on one toolchain if you test only on a different one 10:17 but, you use XP 10:17 ais523, UTF-16 isn't that unusual because Visual Studio itself would save C source code in UTF-16 with unicode comments in it 10:17 <{Demo}> i have built slex on win10 before 10:17 NCommander: wow 10:17 [hdf-us] [dnh] SuzuneTakada (Bin Elf Fem Non), 25 points, T:185, killed by a goblin 10:17 I know most C compilers can handle it, but I'm not sure they do by /default/ 10:18 Aosdict, not sure which version it talks about 10:18 ais523, most Windows tools working with unicode use UTF-16 because internally Windows is entirely UTF-16 10:18 is there a BOM or not? 10:18 ais523, TBH, no idea. I just remember that bit of weirdness when I was dealing with a offshored project coming home with comments in various languages. 10:18 -!- eLtMosen has joined #hardfought 10:18 -!- eLtMosen has quit [Changing host] 10:18 -!- eLtMosen has joined #hardfought 10:19 porkman: you might as a first milestone, get it compiling and then change some des files. if you get that working consistently, you are ready for the more gruesome changes 10:19 looks like gcc doesn't accept utf-16 by default, even with a BOM 10:19 if you want true horror, try cross-compiling NetHack :) 10:20 ais523, feel free to try VS. I admit this was a long time ago, but I'm near certain it would take it. I can't imagine it using codepages because even in Windows NT it was all UTF-16 10:20 even installing mingw comes with a lot of possible variants:p 10:21 The problem with building SLEX isn't so much getting the environment, it's watching your machine turn its mind inward as it deals with the horror that's most.c 10:21 *monst.c 10:21 NCommander: with the linux subsystem that might be easier than you think. also unnethack's win32 version gets cross-compiled reguarly. but then, unnethack has a sane build system. as almostthe only variant 10:21 right 10:21 bhaak, I dunno, I like magic :) 10:22 $ gcc -finput-charset=utf16 hworld16.c \ cc1: error: failure to convert utf16 to UTF-8 \ hworld16.c:1:19: error: failure to convert utf16 to UTF-8 10:22 OK, that's a new one on me 10:22 o_o; 10:22 Deep magic is deep magic 10:22 I didn't even know gcc could or would do that 10:22 the conversion failure was at the space between #include and 10:22 bhaak, I managed to get nethack 36 building with openwatcom (abiet with a hand made makefile), but I never got the entire thing cross compiling to DOS like I planned 10:22 and it's curious because iconv can convert the same file correctly 10:22 so i don't need fortran, C++, ADA or Objective C, i guess 10:23 porkman: for NetHack or Slash'EM, you don't 10:23 ais523, if you want to see deeper magic, look how internationalized domain names work :) 10:23 is there a how-to anywhere that shows how to compile nethack for windows? 10:23 including C++ is often a decent idea to get at its dependencies 10:23 <{Demo}> yeah there is k2 10:23 i dont think there's one on the wiki, or if there is i havent seen it 10:23 K2: it'll be different for vanilla and not-vanilla 10:23 <{Demo}> i think there is one on the wiki 10:23 because the vanilla build depends on a Visual Studio project file that variants basically never update, so it won't work on them 10:24 whereas most variants attempt to do the Windows build the same way as the UNIX build 10:24 I thought there was a makefile for the win32 tiles 10:24 oh wow 10:24 ok, this is a bit ...d ated ... 10:24 A copy of Microsoft Visual C V6.0 SP3 or later || A copy of Borland C 5.5.1 command line tools 10:24 SLEX is slashem based, I don't know if they ever had a VS project file 10:25 also, if they did, it likely won't work on modern VS versions 10:25 Those compilers are old enough to drink ... 10:26 i wouldnt mind compiling evilhack for windows. spread the love as it were 10:26 now I'm thinking there's a pretty good chance nethack37 still compiles on Windows 95/NT 3.1 10:26 i compiled nethack 3.4.3 for win32 a long time ago but i have since forgotten what i did 10:26 Slash'EM 0.0.7E7F3 (the "3.4.3" equivalent) ships with NetHack 3.4.0's instructions for compiling on Windows, but has its own competing set of instructions for compiling on Windows alongside them 10:26 K2, if you want to see evil, compile for Windows CE :) 10:26 heh 10:27 its competing instructions use mingw, and are probably the instructions that Amy uses 10:27 K2, it involves serious magic because its a cross-compile across architectures 10:27 porkman: I recommend starting by reading sys/winnt/Install.gcc; it's probably a bit out of date but should give you an idea 10:27 -!- raisse has quit [Ping timeout: 256 seconds] 10:27 K2, https://github.com/gittup/nethack/blob/master/sys/wince/Install.ce - breath in the evil 10:28 ignore the GTK-related sections because a) GTK has changed a lot since 2004, b) you have no chance of getting a reasonable tiles version anyway because I bet most of SLEX's monsters don't have tiles 10:28 Aren't there fallback "letter" gylphs? 10:28 NCommander: that looks surprisingly nonevil to me 10:28 evil is relative 10:28 <{Demo}> slex and slashem are compiled on windows the same way 10:29 <{Demo}> just get the mingw tools in your path and run a couple batch scripts or something 10:29 It would be nice if SLEX had tiles 10:29 It might slow the rate of crazy 10:30 [hdf-us] [dnh] malor (Arc Dwa Mal Law) wished for "blessed fixed +2 Tie-Dye Shirt of Shambhala", on T:23219 10:30 <@Suzune Takada> slex has too many monsters/items so won't be any tailset 10:30 -!- eLtMosen has quit [Ping timeout: 256 seconds] 10:31 NCommander: why do you blow into the same horn as AntiGulp? I decided against tiles on purpose, not only because I can't stand them but because they'd also slow my creativity to a crawl when it comes to adding new stuff :P 10:31 <{Demo}> just use autogenerated tiles 10:31 LarienTelrunya, I don't like ties either, I'm just trying to stop SLEX's monst.c from becoming blackhole.c and the city of Frankfuit from becoming a crater 10:31 *tiles 10:31 [hdf-us] [nh370] klintos (Mon Hum Fem Law) acquired the Mines' End luckstone, on T:17262 10:32 too late :P 10:32 or to quote jonadab, "that ship has sailed long ago" 10:33 how big is monst.c, in bytes? 10:33 no one shall ever stop me from adding to monst.c! 10:33 <{Demo}> ais523 im more concerned about how much memory all the permonst structs use during runtime 10:33 [hdf-us] [evil] k2 (Pri Cen Mal Neu) had a mace entrusted to him by Camaxtli, on T:15573 10:33 I think there comes a point at which you want the monsters in some sort of external structure that's loaded at runtime rather than being hardcoded in at compile time 10:34 ais523: 15.083.392 10:34 OK, that's a) clearly too high to be sane, but b) not as bad as I thought 10:34 LarienTelrunya: you're still on XP? then you won't get past the 4GB barrier :-D 10:34 and the permonst structures should be smaller than the corresponding source file 10:34 well I did switch to the nethack 3.60 method of loading them into a new structure that can also be modified during gameplay 10:35 <{Demo}> yeah that's not as bad as I thought either 10:35 15 MB of permonst structures doesn't seem ridiculous, especially as they'll be in a swappable segment 10:35 LarienTelrunya, there's an upper limit of how much abuse you can put a compiler through before software services will come and take it away 10:36 <{Demo}> i would watch out for software protective services, yeah 10:36 [hdf-us] [evil] krm26 (Bar Hum Fem Neu) had a scale mail entrusted to her by Crom, on T:50256 10:37 * NCommander notes yesterday, I did clean out my old spare keyboard from storage 10:37 https://pbs.twimg.com/media/EVG9PhNU4AEmKLg?format=jpg&name=large 10:37 There's nothing more amazing than a clean keyboard 10:37 * K2 wonders why he's saving these saddles while playing as a centaur 10:38 [hdf-us] [nh370] Qwesti (Tou Hum Mal Neu), 166 points, T:359, killed by a water moccasin 10:38 K2, so your pet can mount you? 10:38 that just sounds... dirty 10:38 so 10:38 what can i do 10:39 NCommander having a humanoid pet that can ride you if you're a centaur would be neat. not sure how to make that happen though, nor do I think even want to make that happen 10:39 K2, yeah I realize after I said it, "wrong variant author" :) 10:40 since i've decided to jump down this rabbit hole... 10:40 hmm 10:40 smop for sure 10:40 porkman: climb back out while you still can? ;) 10:41 [hdf-us] [un] SuzuneTakada (Con Vam Fem Cha), 4881 points, T:2738, killed by an orc shaman 10:46 [hdf-us] [nh370] eraserhead97 (Cav Hum Fem Neu) destroyed etai's ghost, the former Fighter, on T:5856 10:48 [hdf-us] [nh370] Karnac (Pri Hum Mal Neu) entered Sokoban, on T:3604 10:48 [hdf-us] [evil] k2 (Pri Cen Mal Neu) had Demonbane bestowed upon him by Camaxtli, on T:16118 10:48 boop 10:49 [hdf-us] [nh370] Qwesti (Tou Hum Mal Neu) entered a shop, on T:576 10:49 -!- nabru has quit [Read error: Connection reset by peer] 10:49 [hdf-us] [nh370] eraserhead97 (Cav Hum Fem Neu) destroyed Grandolar's ghost, the former Thaumaturge, on T:6197 10:50 -!- nabru has joined #hardfought 10:51 [hdf-us] [spl] hyperion67 (Arc Hum Mal Law), 5006 points, T:6711, sent to the next life by an Uruk-hai called Aiulog of Ai-baneagris 10:54 "Narrow weapons like short swords aren't as good as broad weapons like broadswords." lol aosdict! actually, Demonbane is a narrow weapon while Frost Brand is a broad weapon, despite the latter not being a broadsword at all :P 10:55 Larientelrunya, i am not climbing nowhere : 10:55 :p 10:56 [hdf-us] [dnh] Pacra (Wiz Orc Mal Cha), 117 points, T:289, killed by a system shock 10:57 [hdf-us] [nh370] dbanethacker (Val Hum Fem Neu) acquired the Mines' End luckstone, on T:22764 10:57 the nice thing about false rumors is you can say practically anything in them as long as it isn't good nethack advice 10:58 [hdf-us] [nh370] Karnac (Pri Hum Mal Neu) entered the Bigroom, on T:4557 10:58 [hdf-us] [nh370] nabru (Val Dwa Fem Law), 227 points, T:649, killed by a sewer rat 10:58 heh yeah, I had great fun designing new ones for slex :D 10:58 "They say that tossers gain an alignment bonus by tossing an item into a pool of water." 10:59 "They say that bleeders can cure their hemophilia if they quaff 20 blessed potions of full healing." 10:59 "They say that experts of the Ataru form can triple-wield lightsabers and also wear three rings." 10:59 and so on :D 10:59 false rumours, I take it? 10:59 yeah 10:59 though in xnh I got rid of the false rumors that were true if you looked at them in a certain way, and the true rumors that were so opaque or inscrutable as to be useless 10:59 [hdf-us] [nh370] nabru (Val Dwa Fem Law) entered a shop, on T:292 11:00 the general principle for true rumors is that a true rumor should be helpful 11:00 because of the start-of-dungeon engraving 11:00 I don't think a false rumour has to be misleading, just amusing 11:01 in this most recent commit I moved the Nth Law of Hacking false rumors into rumors.tru because ... there isn't actually anything false about them? 11:02 are they useful? 11:02 I forget what most rumours say, it normally makes more sense to memorize the actual game mechanics than to memorize the rumours 11:02 [hdf-us] [evil] k2 (Pri Cen Mal Neu), 65193 points, T:16856, killed by a dwarf king 11:03 damn 11:03 that dorf king was decked out 11:03 that's gonna be a good bones file if one was generated 11:03 two of them allude to leaving the dungeon with the amulet and another one says "the last blow counts most" which is true about killing bloes 11:03 blows 11:04 [hdf-us] [nh4] viviana (hackedhead) (Rog Hum Fem Cha), 12363 points, T:16, quit 11:04 start-of-dungeon engraving?! 11:05 LarienTelrunya: yeah there's a random true rumor appearing as graffiti in the first room. it was my less intrusive answer to splicehack tip of the day and unnethack scroll of hint 11:05 heh 11:05 no such thing in slex of course, the player needs to get lucky to find true rumors, or bless their fortune cookies :P 11:06 What's the chance that the player will find it if they don't know it's there? 11:06 100% because xnh highlights them 11:06 [hdf-us] [nh370] nabru (Val Dwa Fem Law), 241 points, T:999, killed by a coyote 11:06 ouch K2 :-( 11:06 [hdf-us] [evil] Grouchy (Wiz Hum Mal Cha) destroyed Vlad the Impaler, on T:48335 11:06 unforced error 11:06 ah 11:06 [hdf-us] [evil] Grouchy (Wiz Hum Mal Cha) acquired the Candelabrum of Invocation, on T:48336 11:11 [hdf-us] [nh370] nabru (Val Dwa Fem Law) entered the Gnomish Mines, on T:544 11:13 [hdf-us] [dnh] malor (Arc Dwa Mal Law) destroyed Vlad the Impaler, on T:25164 11:14 Andrio: yes, and since graffiti is either bright green, bright blue, or bright purple, your natural impulse is to go "ooh what's this colorful thing" 11:16 [hdf-us] [nh370] nabru (Val Dwa Fem Law) entered the Gnomish Mines, on T:120 11:19 [hdf-us] [evil] krm26 (Bar Hum Fem Neu) killed Asmodeus, on T:51272 11:22 is there any meaning to the different colors? 11:23 [hdf-us] [nh370] nabru (Val Dwa Fem Law) entered the Gnomish Mines, on T:259 11:27 while hallucinating, it should change color in real time :D 11:29 [hdf-us] [nh370] nabru (Val Dwa Fem Law), 421 points, T:620, killed by a homunculus, while sleeping 11:30 [hdf-us] [nh370] samiam (Mon Hum Mal Law), 59534 points, T:16927, killed by a wand 11:30 -!- bug_sniper has joined #hardfought 11:32 LarienTelrunya: does hallu change colors/symbols in real time in SLEX? 11:32 unfortunately not 11:32 it'd be fairly easy to make NH4 do that, I think, not sure about vanilla 11:32 Would that even be possible without significant code changes? 11:32 yeah nh4 is coded to have events where your keypress timing matters, which has no place in nethack IMHO :P 11:33 Andrio: you just modify the interface to send a redraw command every second or so 11:33 LarienTelrunya: only if you use the commands where the keypress timing matters 11:33 bhaak: no, I just decided that multicolored graffiti was more interesting 11:33 on the client side? 11:33 you always have an equivalent command that doesn't 11:33 ais523: why is eating a corpse a command where keypress timing matters? 11:33 but most people don't use them because it turns out that exploring one step at a time / eating one bite at a time is very tedious 11:34 LarienTelrunya: because you can see what happens while you're eating, and might want to abort when you see what happens 11:34 huh so you did make a command that allows you to take only a single bite? how do I access that? 11:34 1e 11:34 wtf 11:34 I seem to recall trying that and it still eating the whole corpse unless I very quickly press a key 11:34 n1e? 11:34 really, maybe it's buggy 11:35 hackedhead: that nearly always isn't correct in NH4, you'd have to explicitly reconfigure the controls for that 11:35 main keyboard numbers are interpreted as counts by default (and numeric keypad numbers as movement commands) 11:35 oh interesting 11:35 [hdf-us] [nh370] Qwesti (Tou Hum Mal Neu) entered the Gnomish Mines, on T:2325 11:36 although on some terminals you have to turn numlock off because they send the same codes otherwise (even though I asked them to send different codes) 11:36 * aosdict is not a fan of the ability to take just one bite of a corpse unless something is done to reduce the effective lifetime of lizard corpses 11:37 huh, that's a good point, in vanilla it's optimal to eat lizard corpses in places where you get interrupted the turn after 11:37 1e, does seem to work 11:37 only at 53w of the 700w gnome corpse 11:37 in slex I made it so that lizard corpses always get consumed in one bite, nyah-nyah! :D 11:37 LarienTelrunya: I think that's actually the correct fix 11:37 :) 11:37 LarienTelrunya: considering they're MZ_TINY that is sensible 11:37 either that, or make it so that it's the last bite that cures the stoning 11:37 Gulp 11:38 considering you only have 3 turns at best to do so, that'd be kinda evil 11:38 if it takes several turns, starting to eat one while you're one turn away from turning to stone may result in you dying anyway 11:38 ais523: that would make them almost useless for when you're fighting a cockatrice and are certainly going to get interrupted 11:38 Hmm... trapped a monster in soko 11:38 You can make it so that such interruptions don't happen when you're eating a lizard corpse. IIRC there's been a similar change for tins. 11:38 aosdict: is that a bad thing? 11:39 will poly wand hit the boulder and the monster? 11:39 Insofar as it would incentivize tins and acid potions, no. 11:39 Probably, yes. 11:39 as it is cockatrices are basically irrelevant in NetHack, except when used as a weapon 11:39 ... to spoiled, experienced players. 11:39 to spoiled inexperienced players too 11:39 ^ 11:40 (who tend to regard cockatrices as incredibly, mindbogglingly dangerous and stay well away from them, which is very easy given how slowly they move) 11:40 or just shoot them. 11:40 or both! 11:40 they're most dangerous as corpses, honestly 11:41 !PoM 11:41 APic: The moon is full for 1 more day. 11:41 new moon + no lizard in inventory, I have had YASD with that 11:41 I've had a live cockatrice spawn near the Minetown temple once and got it to petrify the priest using conflict 11:41 even unspoiled inexperienced players who are playing NetHack in a remotely intended way (e.g. look up any monster you don't know in an encyclopedia intended for something other than NetHack) will probably get a clue 11:41 either groping them, other things wielding them, or falling down stairs with them. 11:41 what about the NetHack encyclopedia itself? 11:41 IIRC that warns about cockatrices 11:42 also the Oracle if you're lucky 11:42 It has the reputed ability to kill people by either looking at them—"the death-darting eye of Cockatrice"[5]—touching them, or sometimes breathing on them. 11:42 LarienTelrunya: you know what to do :-D 11:42 [hdf-us] [nh370] nabru (Val Dwa Fem Law) entered the Gnomish Mines, on T:249 11:42 cockatrice breath attack when 11:42 clearly we need all three versions in SLEX 11:42 @le?master mason 11:42 master mason (D) | Lvl: 16 | Diff: 24 | Spd: 9 | Res: petrification | Confers: nothing | MR: 20 | Generates: special | AC: -3 | Attacks: 3d8 breath petrification, 4d6 bite physical, 3d3 lashing physical, 3d3 claw physical | Alignment: 7 | Flags: genocidable, carnivore, flies, thick hide, oviparous, seeinvis, lithivore 11:42 Not sure I'd trust any source not associated with a game for specific details about that game 11:42 (it's also mentioned that cockatrices and basilisks frequently get conflated in mythology, the only real difference seems to be that basilisks don't have wings) 11:42 ^^ breathes petrification :D 11:43 @v?cockatrice 11:43 cockatrice (c) | Lvl: 5 | Diff: 8 | Spd: 6 | Res: poison petrification | Confers: poison | MR: 30 | Generates: gehennom dungeons | AC: 6 | Attacks: 1d3 bite physical, 0d0 touch petrification, 0d0 passive petrification | Alignment: 0 | Flags: genocidable, omnivore, oviparous, touch petrifies, infravisible 11:43 due to harry potter, basilisks are more associated with an instadeath gaze 11:44 due to d&d actually 11:44 basilisk has been around a long time 11:44 Andrio: the point is not to trust those sources, it's to get clues 11:44 Not sure I'd trust them for clues either 11:44 well, NetHack's the sort of game where you're supposed to be able to reason things out using outside knowledge 11:44 K2: in pop culture 11:44 which is incredibly rare among games 11:44 mm 11:45 it possibly doesn't work in general because many of the sources are somewhat offbeat 11:45 [hdf-us] [dnh] Pacra (Brd Elf Mal Cha), 1755 points, T:3360, killed by a small aklys 11:45 (Kingdom of Loathing is another game that runs on the same principle, which may be why it was part of /dev/null/nethack for many years; I don't really like KoL though) 11:45 Like the fact that not only can you use a credit card to unlock things, that's the only use for it 11:46 -!- amateurhour has quit [Ping timeout: 240 seconds] 11:46 Barring the Platinum Yendorian Express Card of course 11:46 I think that's more a joke based on "what would a credit card do in a medieval society?" 11:46 and its only purpose is being small, thin and somewhat rigid 11:47 you could order things online instead of having to walk to the nearest village 11:47 Why are there tourists? 11:47 https://nethack-yanis.github.io/yanis/1013.html 11:47 ^ the one credit card idea not associated with unlocking things 11:48 nethack incorporates credit scores 11:48 -!- raisse has joined #hardfought 11:48 [hdf-us] [nh370] Karnac (Pri Hum Mal Neu) had Magicbane bestowed upon him by Odin, on T:6467 11:48 * aosdict is enjoying the ability to share a link to an idea 11:48 Not really, though, right, as it's specific to each shop? 11:48 Mm 11:48 'oh you want to purchase this wand of wishing in monthly installments?' hmm no, your credit rating is not up to standard' 11:49 aosdict: are you copperwater? 11:49 yes 11:49 for some reason I thought there were two separate people involved 11:49 then I saw the footer of that .io page and thought it was worth asking 11:49 fwiw, many of your pull requests did in fact get pulled 11:49 although it's making it hard to keep up with the pace of development as a result 11:50 K2: I still think it's amusing that a wand of wishing contains enough charges to wish for enough gold to buy the wand of wishing you just used 11:50 so wands of wishing should cost more? got it 11:50 this strikes me as an obvious economic issue, a bit like selling $10 bills for $5 11:50 Well, most of the ones I was really hoping for haven't gotten much traction 11:50 That only takes one charge though 11:50 (namely, the ones that remove interface screw) 11:51 Andrio: exactly 11:51 [hdf-us] [nh370] nabru (Val Dwa Fem Law), 249 points, T:1093, killed by a coyote, while frozen by a monster's gaze 11:51 And you can still wish for a wand of death to murder the shopkeeper. 11:51 aosdict: you have to understand how controversial the issue of interface screw is 11:51 up the amount a shopkeeper charges for using a charge from a wand of wishing 11:51 many people get upset when you remove it, some people probably think there isn't enough atm 11:51 ^ YANI 11:51 although many other people want it removed entirely, of course 11:51 You only have yourself to blame for the increased shopkeeper death toll then 11:52 it's not like it's the devteam's job to keep the shopkeepers alive, they knew what they were doing when they moved in 11:52 everytime a shopkeeper is killed, a kitten is killed also. think of the kittens 11:52 I would like to see the people who think there needs to be MORE interface screw 11:53 * K2 looks at LarienTelrunya 11:53 i.e. Amy? 11:53 also, the devteam has taken some recent steps towards decreasing it 11:53 amnesia no longer forgets objects and maps 11:53 Mm 11:53 I want the old amnesia behavior back! 11:53 That was a very interesting change 11:54 thanks 11:54 I shall never forget what amnesia was like in 3.4.3 11:54 back then it still *meant* something to get hit by amnesia :P 11:54 I feel responsible for it because I wrote the original code, although I wasn't responsible to putting it into vanilla 11:54 I find it kind of interesting that with the new amnesia, you can lose a single skill level and enough training points that the amount you have left 'should' not even be enough to keep your remaining skill levels 11:54 [hdf-us] [nh370] Qwesti (Tou Hum Mal Neu) destroyed nabru's ghost, the former Stripling, on T:2607 11:55 e.g. you can be skilled in saber with 120 uses, and have that reduced to 1 but still remain basic in saber if I understand correctly 11:55 [hdf-us] [dnh] malor (Arc Dwa Mal Law) killed the high shaman of Moloch, on T:27728 11:55 oinkz 11:55 Not a major issue though because those numbers aren't seen except in wizard mode 11:55 Shaman him 11:56 UnNetHack does it in an interesting way too; mind flayers don't make you forget maps and identities any more, but Cthulhu does. 11:56 -!- amateurhour has joined #hardfought 11:56 heh 11:57 Don't forget your cursed potion of gain level. 11:57 -!- Chris_ANG has quit [Ping timeout: 240 seconds] 11:57 [hdf-us] [nh370] nabru (Val Dwa Fem Law), 60 points, T:204, killed by a water demon 11:57 what does cross-compiling mean? 11:58 What I do with my IRC bot 11:58 when you compile the thing on one system in order to run it on a different one 11:58 Andrio: no 11:58 Something like that 11:58 [hdf-us] [dnh] malor (Arc Dwa Mal Law) killed Lungorthin, on T:27841 11:58 compiling a piece of software on one _architecture_ for a different _architecture_ 11:58 ah 11:59 I always thought cross-compiling meant "compile software for platform X on platform Y" 11:59 creating an x86 binary on my machine for your x86 machine is not cross compiling 11:59 right, a good example is using Linux to generate the Windows version of UnNetHack 11:59 ok 11:59 hackedhead: it is if they're different operating systems 11:59 A cross compiler is a compiler capable of creating executable code for a platform other than the one on which the compiler is running. 11:59 what does compiling mean? 11:59 ais523: I suppose. 11:59 porkman: converting source code into an executable 11:59 porkman: building an executable from souice 11:59 how do I shot web? 12:00 *source 12:00 [hdf-us] [nh370] Qwesti (Tou Hum Mal Neu) murdered Bucica, his faithful dog, on T:2848 12:00 e.g. converting allmain.c + friends into nethack.exe 12:00 I know...nothing about programming, so, it'll be a bit of an impossible task. But, streaking slex seemed like an impossible task when i started it. so... 12:00 and btw amy, i hate those monsters 12:00 LarienTelrunya: there isn't a ranged version of that, but you can use #monster while polymorphed into a spider to place a web at your current location 12:00 that's the spirit 12:00 heh, in slex, monsters with webbing gazes exist :D 12:00 porkman: they obviously work as intended, because they're meant to annoy you to hell 12:01 ok, so there's modifying the source, and then you have to compile it into an .exe 12:01 [hdf-us] [nh370] Qwesti (Tou Hum Mal Neu), 1342 points, T:2853, killed by a magic missile 12:01 right 12:01 ok, so two different things 12:01 what are the obstacles to compiling? 12:02 you normally need two different programs to do the compile, although both are included in mingw 12:02 one program reads the .c files individually and converts them each into machine code, the other (called the "linker") combines all the files together into a single .exe 12:02 -!- Bliss56 has joined #hardfought 12:02 and the whole process can be very complex for a large project like NetHack 12:03 (especially because NetHack generates some of its .c files during the compile, meaning it needs additional .exe files to generate them) 12:03 so normally you use a separate program to automate the whole thing because it's too complex to do by hand 12:03 mkay 12:03 damn flex and bison to hell :P 12:03 the main obstacles, thus, normally, are a) getting all the programs you need, and b) configuring the program that automates the build process 12:03 LarienTelrunya: we got rid of them in 3.7, shame you don't like it 12:04 heh, aosdict you got lua instead, which isn't really better 12:04 NetHack is mostly written in C, but as LarienTelrunya mentions, partially written in lex and yacc, so you need compilers for those languages too 12:04 (with flex and bison being the most readily available, typically) 12:04 I feel like making a nethack variant is a little out of reach for someone who doesn't know what compiling code is 12:05 [hdf-us] [nh370] nabru (Sam Hum Fem Law), 110 points, T:447, killed by a water moccasin 12:05 Eh, i'll figure it out 12:05 aosdict: lots of people learned C by trying to patch NetHack 12:05 so it's not impossible 12:05 there are two complicating factors here, though: a) doing it on Windows, b) doing it on SLEX rather than vanilla 12:05 even so it should still be doable though 12:06 bear in mind that this is someone who's ascended SLEX, so even though there's a lot to learn when it comes to compiling, there was a lot to learn to do that too 12:06 time to rewrite nethack in a friendlier language :^) 12:07 Time to rewrite Nethack in Python! 12:07 Time to rewrite Nethack in Scratch! 12:07 amateurhour: I actually think C is one of the friendlier languages when it comes to build process 12:07 [hdf-us] [evil] krm26 (Bar Hum Fem Neu) killed Juiblex, on T:52467 12:07 the problem is more that C /developers/ tend to make the build process for their own programs unnecessarily complex 12:07 Did you say from scratch? 12:08 due to C not being able to do much at compile time 12:08 for 3.7 there's been a real effort to simplify the build process, but 3.4.3's is a nightmare 12:08 amateurhour: I think nethack has a 100% success rate at foiling every "let's rewrite it in X" attempt due to people either giving up or realizing it's easier to make their own roguelike 12:08 all those who ever started a "rewrite nethack in foo language" project have abandoned it, so I can't recommend trying that :P 12:09 aosdict: well, there's Pathos, which started as a "rewrite NetHack in C#" and ended up having to deviate mainly for copyright reasons, but also the author doesn't seem to understand what makes NetHack NetHack 12:09 and removed most/all of the special cases, which completely changes the nature of the game 12:09 so, i can start by compiling slex from the source as it is 12:09 porkman: yes, that's a good way to start 12:09 once i manage to do that, i can then make some changes and see if it compiles 12:09 [hdf-us] [nh370] nabru (Sam Hum Fem Law) entered a shop, on T:212 12:09 right 12:10 ais523: I classify that, along with ZAPM, as attempts that ended up making their own roguelike 12:10 <{Demo}> zapm is epic 12:10 aosdict: well ZAPM is clearly different 12:10 whereas I keep on having to correct people on Reddit who try to persuade people that Pathos is a version of NetHack 12:10 so i need a compiler 12:10 porkman: yes 12:10 Amy, which one ddi you use? 12:10 actually zapm didn't start off as an attempt to rewrite nethack, it was to make a roguelike similar to nethack but sci-fi 12:10 MingW? 12:10 porkman: she used mingw 12:11 and which text editor did you use? 12:11 porkman: why do you keep asking me, I already told you that it's mingw 12:11 and you know that I edit with wordpad and proton 12:11 porkman: you really don't want to have asked that second question :-D 12:11 I think it was generally agreed in most NetHack-related channels that LarienTelrunya's choice of editors was a mistake, and editor compatibility is fairly unimportant 12:11 okay 12:12 -!- mobileuser has quit [Quit: Connection closed] 12:12 So, nethack is in C, if i compile it into an .exe, would that then work on windows or do i have to emulate another OS? 12:12 or install another OS 12:12 porkman: it'd work only on Windows 12:12 (which would mean having to learn to run another one) 12:12 by default, a compiler will output for the operating system you're currently running on 12:12 so if you're on 64-bit Windows you'll get an executable that only works on 64-bit Windows 12:13 ok 12:13 (often the version matters too, if you aren't paying attention, but most people are on Windows 10 by now anyway; NH4 makes an attempt to work perfectly back to Windows Vista and be runnable back to Windows 2000, but I don't think most people care about backwards compatibility) 12:13 (also I haven't actually /tested/ this) 12:14 [hdf-us] [nh370] nabru (Sam Hum Fem Law) entered the Gnomish Mines, on T:679 12:14 I would be surprised if somebody was still running Vista today. :P 12:15 from the point of view of modern software development, Vista was the first good version of Windows 12:15 everything before was kind-of hopelessly insecure internally 12:15 although this caused a lot of problems when Vista came out, because lots of existing programs were doing inherently insecure things, which now suddenly didn't work 12:17 aosdict: #649 on your YANI archive is actually a YASI, I think 12:17 maybe I misclassified it when first making it, but it wasn't intended seriously 12:18 [hdf-us] [spl] AntLegs (Dra Hum Mal Neu), 0 points, T:130, slipped while mounting a saddled baby orange dragon 12:18 so, i'm starting from here http://www.mingw.org/wiki/HOWTO_Install_the_MinGW_GCC_Compiler_Suite 12:18 and here http://www.mingw.org/wiki/mingw_for_first_time_users_howto 12:19 ais523: Eh, it's fine, not going to remove it. 12:19 (I did add your suggestion for a "You cannot escape from [ustuck]!" YAFM when you hit esc when engulfed or whatever to xnethack) 12:19 porkman: your latter link is about how to run the compilers manually 12:19 which wou wouldn't normally be doing for NetHack 12:20 aosdict: OK, I still like that one 12:20 <@luxidream> New artifact Imaginary Widget: The base object is randomized among non-polearm melee weapons each game. Deals double damage to monsters that don’t exist in real life (e.g. foxes and mastodons and giant eels are real; wargs and mumaks and krakens are not). 12:20 <@luxidream> this is definitely a joke 12:20 @luxidream: why not polearms too? 12:20 I could see splice implementing that 12:21 that said, almost every NetHack monster doesn't/didn't exist in real life 12:21 <@luxidream> it would be very likely to be a polearm and thus terrible 12:21 I guess there are some powerful monster classes that do, such as high priests 12:22 porkman: for what it's worth, when advising people on how to build NH4 (which also uses mingw), my normal advice is to install Strawberry Perl because it bundles a complete mingw C toolchain that's already been fully configured, + a lot of helpful libraries 12:22 I'm not sure how easy the "intended" way to get mingw is 12:23 Ais523, I've actually now streaked slex btw, not just ascended it :) 12:23 Strawberry perl 12:23 thanks, i'll find it 12:24 [hdf-us] [nh370] Karnac (Pri Hum Mal Neu) wished for "blessed Eye of the Aethiopica", on T:7609 12:29 The simplest way to get a working toolchain is to install Strawberry Perl, available at http://strawberryperl.com, which comes with a working C toolchain. You will also need to install Flex and Bison, scanner and parser generators; the versions at GnuWin32 (http://gnuwin32.sourceforge.net/) work (although they don't have a very Windows-like idea of directory structure, and thus can't be installed to paths with spaces in, and produce 12:29 compiler warnings; if you know a better option, let us know). Strawberry Perl's and GnuWin32's executables will all need to be on your PATH (search for "PATH" in Control Panel, on recent versions of Windows). 12:30 porkman: see that; slex's build process is different but it uses the same programs, so those instructions should apply for you too 12:33 [hdf-us] [evil] AntLegs (Mon Gia Mal Law), 57 points, T:51, killed by strangulation, while helpless 12:34 [hdf-us] [nh370] sawbones (Hea Gno Mal Neu) reached Mine Town, on T:2459 12:35 [hdf-us] [evil] AntLegs (Cav Gia Mal Law), 28 points, T:25, killed by a water elemental 12:35 [hdf-us] [nh370] nabru (Sam Hum Fem Law), 1188 points, T:1911, quit 12:36 !lastgame AntLegs 12:36 K2: [hdf-us] https://www.hardfought.org/userdata/A/AntLegs/evilhack/dumplog/1586450033.evil.html 12:37 ah 12:38 [hdf-us] [nh370] sawbones (Hea Gno Mal Neu) entered a temple, on T:2553 12:38 whoa I just looked at my commit graph and wow do I have a lot of activity starting on the day the corona lockdown happened (and continuing up until today)? :D 12:38 [hdf-us] [nh370] Karnac (Pri Hum Mal Neu) completed Sokoban, on T:8590 12:39 -!- Chris_ANG has joined #hardfought 12:39 -!- mode/#hardfought [+v Chris_ANG] by ChanServ 12:39 SLEX 2.66: Powered by Covid™ 12:40 i wouldnt state that 12:40 [hdf-us] [evil] krm26 (Bar Hum Fem Neu) killed Baalzebub, on T:53520 12:40 -!- misha_ is now known as misha 12:42 [hdf-us] [evil] Grouchy (Wiz Hum Mal Cha) had a dagger entrusted to him by Anhur, on T:48873 12:43 [hdf-us] [evil] AntLegs (Mon Cen Mal Cha), 164 points, T:250, killed by a magic missile 12:43 [hdf-us] [nh370] nabru (Sam Hum Fem Law) entered the Gnomish Mines, on T:332 12:44 YANI: when using the Big Idea that price-ID doesn't work at all, make price-ID work for ?oID and nothing else, to preserve part of the essence of the system and to give a way to get started 12:44 is there any non-slex variant that removes price-IDing? 12:45 I like price IDing :'( 12:46 NH4 kind-of does the opposite by making many items formally ID inside shops 12:46 [hdf-us] [nh370] nabru (Sam Hum Fem Law), 131 points, T:426, killed by a gnome lord 12:47 my favourite ID fix is to have price ID bands 12:47 where there's some amount of randomization as to which band an item will be in each game 12:47 doesn't ADOM also ID things in shops? 12:47 and have the bands show up in the item name 12:49 -!- Umbire has joined #hardfought 12:49 how do you show that flavorfully 12:50 "an expensive red potion" 12:50 the band only show up when you take the item to a shop 12:50 so it's still price-ID, just less informative than it used to be and with a more streamlined interface 12:51 ok, installing strawberry perl 12:52 So, strawberry perl, MinGW, Flex and Bison? 12:52 strawberry perl comes with mingw 12:52 you need to install flex and bison separately though 12:52 will it clash if i installed minGW on its own already? 12:52 I don't think so 12:52 <@luxidream> removing price ID altogether simplifies id strategy too much imo 12:52 you need to set the PATH variable to tell Windows which programs to use 12:53 @luxidream: well it makes IDing harder, obviously; do you mean that the new, harder strategy is too straightforward? 12:53 -!- Umbire_ has joined #hardfought 12:54 <@luxidream> there's some value in looking for specific items by price class, something that a newer or less skilled player might overlook 12:55 -!- raisse has quit [Ping timeout: 256 seconds] 12:56 -!- Umbire has quit [Ping timeout: 265 seconds] 12:56 <@luxidream> it becomes completely brain dead when all items have the same price 12:56 I think the main complaint is that it's pretty much completely braindead atm anyway 12:57 hmm… what about making all sell prices identical, but leaving buy prices as they are atm? 12:57 you could still throw your items into a shop to ID them but that would have the obvious downside of having to buy them back 12:57 yeah for someone learning how to do it, it's basically just "here go to the wiki page and follow these steps and look up in this table" 12:58 then label your potion "conf/hallu/sleep/healing/xhealing/restore" 12:58 <@Tone> Price identification makes the ID system more interesting than many other roguelikes where you randomly use items until you find the identify scroll 12:58 ^ 12:58 I think the hope would be to add more indirect methods 12:58 actually, the worst item class in this respect atm is wands 12:58 @Tone I think that goes more for _informal_ identification, of which price ID is a subset 12:58 they have a very simple, direct ID method that hardly costs 13:01 I think I tend to like the general proposed idea of having items' base costs fuzzed by some amount, so there's no concept of price tiers but you can get a feel for what's cheap and what's expensive without saying "well, it sells for 25, I guess I'll be blanking THIS one" 13:01 there's fuzzing already atm 13:01 it's just too easy to reverse 13:01 on the other hand I understand the limits of such a system 13:01 I think some people have even managed to distinguish unIDed glass from unIDed gems by screwing around with details of the fuzzing algorithm 13:02 the "band" proposal was mostly for UI simplification rather than for the gameplay changes 13:02 if such a system were implemented, the existing spoilers would just be transformed into "price 300 scrolls can only be fuzzed down to 150, so if it sells for less than 150 you know it can't be one of these" 13:03 <@luxidream> there's certainly some strategy in choosing what items to formally ID early on based on their price class 13:03 <@luxidream> when resources are limited 13:03 <@luxidream> a character with good resistances probably won't care for rings with base price 150 13:04 aosdict: my suggestion would be more like "every scroll sells for 100, 150, 200 or 300, scrolls are placed in a price class at the start of the game with some randomization" 13:04 YANI for fuzzing systems anyway: each shk derives their own idea of the base cost of every item based on its base cost in objects.c and their monster id hashed together, and then buys and sells that type of object at that fuzzed base cost. so one shk might treat ID scroll as base price 15, another might treat it as base price 28 13:04 <@luxidream> this isn't immediately obvious from the spoiler, or completely thoughtless 13:04 like, at the start of the game, it's randomly decided whether punishment is base 150 or base 200 13:04 <@Tone> putting the consumables into "buckets" is what makes it more interesting than just randomly using them or formally identifying them 13:04 <@Tone> If you fuzz the prices then it just makes those buckets wider 13:05 yes, I think the buckets is the interesting part 13:05 and the fuzzing can still be added for flavour but shouldn't have any gameplay impact 13:05 although, if you see a scroll selling for 251 or whatever and think "wow, I can get this base 300 at a huge discount" it might be worth buying because of that? 13:06 I believe last time this came up people expressed a like of Golden Krone Hotel's system where each unidentified item is said to be definitely one of three types, but there's no clean interface way to do that in nethack 13:06 <@Tone> ha, I was just typing something about GKH 13:06 aosdict: yes, it'd either be out of place if you just stated the options, or horrendously spoilery if you did it in any more indirect way 13:06 <@Tone> was basically about to say that it wouldn't work for nethack imo but is an interesting take on it 13:06 ais523: it's easier when you have tooltips available 13:08 well, the i menu is an obvious place to put it (assuming you have a variant where the i menu exists; is it in vanilla yet?) 13:08 ... wasn't the i menu in 3.4.3 13:08 I don't think that's new 13:08 ok 13:09 my general take on the whole thing is that I'd rather improve the methods of non-price-id informal identification 13:09 now pork'em compiling has my full attention 13:09 aosdict: the menu that comes up after selecting an item after pressing i 13:09 if you do that in 3.4.3 nothing happens 13:09 Oh. 13:09 Not in vanilla yet, no. 13:09 <@luxidream> seems annoying to have to check every item in that way 13:10 it's a little controversial because some people use i as a means of absorbing stray keystrokes, to compensate for network lag 13:10 In slex and xnethack (and maybe splice and evilhack stole it from xnethack) you can use the i menu to get the object stats and encyclopedia entry for that object 13:10 so that would fit, but I agree with luxidream that it seems tedious to use and then manually name your items that 13:11 slex actually has handwritten info for every single item in the i menu ;) 13:11 <@Tone> aosdict: How do you improve informal price ID for consumables? There are certain things we can already do like take guesses based on probability, note what monsters pick up and use them, note what nymphs drop, etc. 13:11 @Tone: natural price ID for consumables is already enough for the whole game 13:12 by which, I mean, around dlvl 20 or so look at which consumables are already formally IDed 13:12 any that isn't probably isn't worth using 13:12 @Tone I broke it down by object class at https://nethack-yanis.github.io/bigideas/Identification.html 13:12 this is despite the lack of any explicit action by the player to ID them 13:12 in a way, this seems even more disappointing than ID for other categories… 13:13 <@luxidream> I don't see monsters iding genocide, charging, gain energy, gain ability, taming 13:13 ah right, I normally ID charging by wishing for it (and if playing non-genoless, genocide too) 13:13 <@luxidream> it's also quite possible to have to formally ID healing potions 13:14 and I don't think I normally use ?oTaming, that's probably a mistake though 13:14 oh that reminds me of a comment to make on ais523's complaint about want engrave ID being too easy, namely: if you remove engrave ID, you can still identify most of the wands you can engrave ID by zapping them at a hostile monster 13:14 <@luxidream> that's probably because it's as rare as genocide or charging 13:14 btw, what are people's opinion on reverse identify effects? 13:14 (i.e. choose an item class and get told its appearance) 13:14 awesome 13:15 I think reverse-ID effects would really add to the game, but am unclear on how powerful they are 13:15 that's included in NethackTNG iirc 13:15 Ais523, very 13:15 at least in slex 13:15 where ID'ing things is a massive challenge 13:15 splicehack has them 13:15 porkman: I'm inclined towards that too 13:15 <@luxidream> playing it in splicehack, which is close to vanilla 13:15 scroll of knowledge 13:15 <@luxidream> they aren't very powerful 13:15 given that in some cases I'll use a wish purely to reverse-ID items 13:15 <@luxidream> I only ever used them to ID genocide and charging 13:15 I think the problem is you have to first identify the scroll of knowledge? 13:15 that said, I think there are huge diminishing returns 13:15 [hdf-us] [nh370] sawbones (Hea Gno Mal Neu) entered Sokoban, on T:4188 13:15 (items resist ID, ID sometimes fails to identify all properties of an item, etc) 13:16 <@luxidream> ais523: this is a product of your playstyle 13:16 and yeah, the amount of useful things you can identify with them is small 13:16 in vanilla, minimal 13:16 even so, making reverse-ID a confused effect for ?oID would probably be broken 13:16 there's little need in vanilla for reverse ID 13:16 you would only really rev ID charging 13:16 maybe genocide 13:16 to be able to marker write them 13:16 oh and spellbooks 13:16 [hdf-us] [nh370] sawbones (Hea Gno Mal Neu) became literate by reading the fortune inside a cookie, on T:4206 13:16 ais523: make enlightenment a confused effect for identify 13:16 <@luxidream> you need it to write genocide and charging 13:16 conflict is the other item that I think might be worth a reverse ID 13:16 why 13:16 aosdict: ooh, interesting 13:16 you get so many regular IDs 13:17 you can't make a ring of conflict if you haven't got it 13:17 and if you have got it, you can just ID it with a scroll 13:17 or spell 13:17 ais523: https://nethack-yanis.github.io/yanis/339.html 13:17 I find it's normally quite late in the game before I can afford to ID all my rings 13:17 <@luxidream> also trivial to price ID and informally identify 13:17 conflict can be wear-identified near a peaceful monster :D 13:17 though I'm surprised I haven't implemented that one in xnh yet 13:17 <@riker> aosdict: that would be nice, and make conflict more common 13:17 <@riker> er, enlightenmet 13:18 <@riker> i'm always surprised at how rare enlightenment wands/pots are (for me at least) 13:18 actually, one thing I dislike 13:18 is blessed ID 13:18 <@luxidream> enlightenment wands are rare as wow 13:18 organizing your inventory to get the full effect of the full identify is really micromanagey 13:18 <@riker> the full invent ID? 13:18 <@luxidream> iirc 13:18 heh, i charged my slex wand of enlightenment 3 or 4 times through 13:18 and then you forget what should and shouldn't be in main ID afterwards 13:18 @riker Note that enlightenment has recently become more common in vanilla from potions of hallucination 13:18 thanks to msearle5 and me 13:18 <@riker> that's true 13:18 (slex wands have a chance to go back to 0:x when charged from 7:x) 13:18 porkman: enlightenment strikes me as being much more powerful in slex than vanilla 13:18 well yes it is 13:19 it showed me that scrolls of wishing could not exist in my game 13:19 But I would still get behind confused ?oID for enlightenment, if only because the existing effect of... nothing... is boring 13:19 <@luxidream> I like lower variance versions of the scroll of identify 13:19 so i had to alter reality to make some appear :p 13:19 <@luxidream> like xnh's implementation 13:19 -!- eLtMosen has joined #hardfought 13:19 games where you get ganked by rng on blessed id scrolls are really unfun 13:19 psst, xnethack has a popular scroll of identify implementation, psst 13:19 :P 13:20 <@luxidream> I would like dyna's too but it makes buc irrelevant I think 13:20 let me use the YANI tag to record this: YANI: indirect ID is more common, but ?oID has less prospects to multi-ID, and in particular will never ID the whole of inventory 13:20 ais523: how do you feel about thrones still doing blessed ID? still micromanagey? 13:20 ew 13:21 YANI: Identification as a prayer boon (contingent on you having a lot of unidentified objects, or a high fraction of unidentified objects, in your inventory). 13:21 aosdict: as a random thing that can happen, yes, although it's sufficiently micromanagey for sufficiently small benefit that I don't think most people will actually do the micromanagement, so it's not as bad 13:21 if there were a guaranteed full-ID in the game, that'd also be less bad because it'd only happen once 13:22 heh, that's because you haven't seen me micromanage slex 13:22 <@luxidream> castle throne? 13:22 <@luxidream> that would be interesting 13:22 -!- raisse has joined #hardfought 13:22 @luxidream: I had a proposal where there were four guaranteed thrones, each with a specific effect guaranteed for first sit 13:22 full-ID, genocide, wish, I forget the other one 13:23 and I also forget which one was where 13:23 electric shock 13:23 :-D 13:23 if there were a guaranteed full ID in a known spot you would have players returning to their stash and micromanaging things so that they could sit on the throne with exactly the 52 most desired items to identify in their inventory 13:23 <@riker> ais: grant audience? 13:23 randomise the thrones 13:23 :^) 13:23 <@luxidream> that's more of a problem with stashes than it is with the full inventory id 13:23 aosdict: I'm aware, that might not be such a terrible thing if the throne is somewhere really dangerous though 13:24 Pretty much nowhere is really dangerous after it's been cleared. 13:24 <@luxidream> sitting there naked in castle is probably not smart 13:24 aosdict: you're talking about special levels, presumably? 13:24 YANI: A throne effect can tame nearby monsters, but monsters that generate in throne rooms are immune to this effect. 13:24 filler levels actually have more monsters after being cleared than before, in general 13:24 (this is something that most players don't realise) 13:24 <@riker> aosdict: that can or that did? 13:24 I don't think we're using the same definition of the word "cleared" 13:25 aosdict: if you go round the level killing all the monsters 13:25 @riker any monster creatable by thronemon() 13:25 there's an equilibrium between the rate at which monsters spawn and the rate at which the player kills them 13:25 @luxidream For those who don't have 52 high-priority items to identify, you could settle for, say, 45 and keep wearing all your protective gear. 13:25 but filler levels generate with fewer monsters than the equilibrium proportion, so the number of monsters on filler levels goes up over time (asymptotically to the equilibrium) 13:26 My point stands, guaranteeing a full identify just means a lot of players will try to milk that for all it's worth 13:26 in slex I made the full-inventory ID effect go into containers, problem solved :D 13:26 aosdict: guaranteeing a wand in the castle, or a quest artifact in the quest, or … also means that players will try to milk that for all it's worth 13:27 and besides, in vanilla, having around 4 or 5 b?oID is effectively the same as a guaranteed full-ID and is not very hard to achieve 13:27 ais523: as with many things, this is not so much an argument aimed at taking down existing instances of it, it's an argument against creating more instances of it 13:27 even if all of them miss the full-ID effect, you'll probably have IDed everthing naturally 13:27 from the lesser ID effects 13:27 except for when you're unlucky and they all just id 1 item >:( 13:27 this seems to commonly happen to me for some reason 13:27 <@luxidream> have positive luck 13:27 <@luxidream> they will identify at least two 13:28 or just two, whatever 13:28 either way it's a feelsbad moment when you save up a stack to bless and roll low every time 13:29 what does xnh do for them? 13:29 1/4/7 13:29 for c/u/b 13:30 wow, those numbers are much higher than I was expecting 13:30 it's possible they're somewhat high 13:30 YASI: cursed confused identify gives you amnesia 13:30 they could be 1/3/5 13:30 I may end up changing that if I feel that it gives too many ids too early 13:30 I would think 1/1/2 would be more balanced 13:30 overall players are very happy that they know exactly what they're getting 13:30 so how does strawberryPERL run minGW? 13:31 1/1/2?! 13:31 <@riker> ais: I'd prefer 1/2/4, to have a downside to cursed ID 13:31 I think I should let others respond to that 1/1/2 proposal :) 13:31 porkman: it has a bin/ directory that contains all the programs you need 13:31 you need to add that to your computer's PATH variable 13:31 search for PATH in Control Panel; there are two PATH options, you want your personal user PATH, not the system PATH 13:32 then you just add the bin directory from Strawberry Perl to your PATH 13:32 <@luxidream> for what it's worth, the current expected value for a blessed ID scroll if you have exactly 20 items to ID is 6 13:32 and all the C compilers will start working in your terminal 13:32 <@Tone> heh, people were just complaining about how sometimes they only get 2 IDs from blessed scrolls 13:32 so in environment variables, and i'll add a new one 13:32 porkman: well, PATH is an existing environment variable 13:32 so you'll need to edit its value 13:32 yep, when i do, it lets me add new ones within it 13:32 ah, I see, it has a friendly interface 13:33 yes, you add a new part to its value, showing the new directory ("bin" within the Strawberry Perl build) 13:33 err, the Strawberry Perl install 13:33 <@luxidream> the expected value for an uncursed scroll is 2 13:33 once you've done that, you can test it works by opening a new terminal and writing "gcc --version" 13:33 <@luxidream> so I'd say 1/2/6 is a pretty fair distribution 13:34 6 just seems like you'd need to put no effort into IDing at all 13:34 you naturally find a b?oID quite early, normally 13:34 and 6 identifications are more than you need in a typical game 13:34 @luxidream by that logic, 1/3/5 would be decent 13:34 (formal IDs, that is) 13:35 slight nerf to blessed, slight buff to uncursed (which is good as a major problem with id scrolls is that everyone wants to hold them until they can get blessed) 13:35 come to think of it, "no ?oID, no +oID" sounds like an interesting unofficial conduct that is likely very easy 13:36 aosdict: that's more than just a /slight/ buff to uncursed 13:36 the strawbery perl install went into c/strawberry 13:36 c/strawberry/perl/bin 13:36 that one? 13:37 <@luxidream> we seem to be playing very different games if you typically use less than 6 identifications per game 13:37 the expected value for an uncursed scroll is 1 because I have very low expectations 13:37 porkman: I think so, check the files in it 13:37 the important programs are named "gcc.exe" and "ld.exe" 13:37 (there are a number of helper programs but they'll be in the same place) 13:38 <@luxidream> amateurhour: that's not how expected value works 😛 13:38 @luxidream: I'm talking about formal IDs; obviously I identify more items than that over the course of the game, it's just that indirect ID is so strong that formal ID doesn't seem useful for much 13:38 hmm, there's no .exe in it 13:38 I was joking luxidream :P 13:38 porkman: your OS is probably hiding the .exe 13:38 no gcc.exe 13:38 lemme find it 13:39 it might just be strawberry/bin 13:39 <@luxidream> I don't know, I have a hard time informally identifying items like polymorph control or genocide without wasting them 13:39 it's been years since I did his so I don't remember the value offhand 13:40 @luxidream: have you tried just not using them? :-D 13:40 <@luxidream> Right, I usually formally identify them 13:40 <@Tone> a lot of rings in particular are annoying or tedious to identify informally 13:40 ^ 13:40 @Tone: I agree, rings are the category I most often use formal IDs on 13:40 <@Tone> that hallu enlightenment commit helps a bit with that I guess 13:41 and sometimes scrolls when I can't be bothered to price-ID them 13:41 autocursing helms as well 13:41 why are you using random-appearance helmets at all? 13:41 none of them does anything useful, apart from arguably HooA if you want to cheese the endgame 13:41 as in there's no reasonable way to informally id them that I know of 13:41 and you want a hoa without wearing it 13:41 yes, but why not just ignore them and pick up a dented pot or something? 13:42 or an orcish helm, that works too and is available very early 13:42 because you want a hoa? so you have to lug them around 13:42 aha, it's on strawberry/c/bin 13:42 amateurhour: you could just do what I do and die on Astral 13:42 porkman: I guess in retrospect that makes sense 13:42 <@luxidream> YANI: change items that are 90% generated cursed to 100% 13:42 are we talking about different things? I'm saying I carry the helm around to formally id, so it's another category of items 13:43 amateurhour: yes, but why are you carrying the helm around to formally ID? 13:43 I'm proposing an alternative strategy, of not carrying it around and then not formally IDing it 13:43 <@luxidream> because it could be HooA which is very useful 13:43 this has the benefits of a) not bloating your inventory with useless helms, b) saving on formal IDs 13:43 <@luxidream> but not if you put it on at the wrong time 13:43 it also has the disadvantage of not gaining knowledge of an item that has a chance of saving you on Astral 13:44 but this disadvantage doesn't matter through 99% of the game 13:44 gcc (x86_64-posix-seh, Built by strawberryperl.com project) 8.3.0 13:44 Copyright (C) 2018 Free Software Foundation, Inc. 13:44 This is free software; see the source for copying conditions. There is NO 13:44 warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. 13:44 yay 13:44 it worked 13:44 that's mingw, I can tell from the version string 13:44 (mingw is itself a version of gcc) 13:44 so now you can compile C programs 13:45 NetHack isn't 100% written in C, though, so you'll need to install flex and bison too 13:45 Amy says she hadn't 13:45 <@luxidream> ais523: using the minetown altar as the opposite alignment is probably a significant advantage 13:45 but, i've downloaded them 13:45 (correction: NetHack 3.7 is 100% written in C but Slash'EM, and NetHack 3.4.3, aren't) 13:45 gotta figure out how to install them 13:45 <@luxidream> there are also other edge cases where you want excalibur/stormbringer 13:45 porkman: the gnuwin32 versions, which are the only ones I know of, don't like spaces in paths 13:45 so they unfortunately can't go in Program Files 13:46 it would be nice if better versions existed 13:46 <@Tone> Solution: Only try on helms when you're in the quest branch. 13:46 <@luxidream> you still lose protection that way apparently 13:46 <@Tone> ah, never noticed that 13:46 <@luxidream> probably because you've never done that? 13:46 <@luxidream> lol 13:46 @luxidream: I guess our playstyles are very different, using an item you probably won't find early in Minetown isn't something I'd put as part of any strategy 13:47 <@Tone> I've accidentally "put on" a HoOA on a quest level 13:47 Yeah, always have to account for playstyle 13:47 <@luxidream> it's rare but it happens in a nonzero number of games 13:47 Larientelrunya, how did you then do it without using flex and bison? 13:47 anyway the point was that rings and helms are difficult to informally id 13:47 as that would likely require either adapting strategy on the fly in response to IDing a HoOA early (which couldn't happen to me), or else buliding backtracking into the strategy 13:47 by following the install instructions 13:48 run sys/winnt/setup.bat and nhsetup.bat and then mingw32-make -B -f makefile.gcc install 13:48 fwiw I believe that in 3.7 it is probably correct to avoid the mines altogether, unless you need death drops or experience from mines filler levels 13:48 <@luxidream> that's an interesting take 13:48 oh, there may be prebuilt versions of the lex and yacc parsers, that's probably what it's doing 13:49 I've already been avoiding Sokoban in most of my games recently 13:49 not even because I dislike it, just because the time spent isn't proportional to the gain 13:49 <{Demo}> i do not approve of soko 13:49 do you avoid sokoban because it's mirrored and hurts your brain? 13:49 ais523: slex did change the level compiler at one point but I recompiled the parsers on unix and then copied them into the main repo :D 13:49 <@luxidream> where do you find your dwarvish mithril coats then? 13:49 because I'm considering it for that reason 13:49 it may be worth doing level 1 for the random comestibles… 13:50 LarienTelrunya: aha 13:50 <@luxidream> even speedruns have to go the mines for a pickaxe and armor 13:50 porkman: following LarienTelrunya's instructions should probably work, as long as "make" is installed 13:50 it might be installed as "mingw32-make" or just as "make" 13:50 just have positive luck irl and have a stacked dwarf spawn 13:50 (I think Strawberry might have installed it, not sure though) 13:50 <@Tone> The minetown altar can be pretty damn valuable for a lot of characters 13:51 hmmm, run won't find sys/winnt/setup.bat 13:51 <@luxidream> it can also be the only source of price ID, if you're unlucky 13:51 porkman: use cd to change into the base of the slash'em distribution first 13:51 cd? 13:52 <{Demo}> lol 13:52 porkman: it's a command that tells the terminal what directory you consider important for commands 13:52 did you not grow up with MS-DOS? 13:52 not really, i grew up with very early macs 13:52 if you don't give a cd command it won't realise that you're trying to work on slash'em extended, rather than some other unrelated program 13:53 so you'll need to start off with cd to the directory containing all the slash'em files 13:53 do i do that from a terminal? 13:53 then, all the commands you run from then on will assume that they should be operating on your slash'em installation by default 13:53 right 13:53 the cd command will only last for one terminal session 13:53 so you do it from a terminal, then you use the same terminal for the other steps of the build 13:54 the Slash'EM build process doesn't have a user interface of its own, so it has to be run from a terminal 13:58 yay, it works 13:59 what, the whole slex build process? 13:59 or just the setup batch files? 13:59 the cd 13:59 haha 13:59 oh :-) 13:59 it'll take me a while to get my head around it 13:59 cd is fairly reliable 14:00 [hdf-us] [nh370] Karnac (Pri Hum Mal Neu) destroyed microlance's ghost, the former Man-at-arms, on T:11880 14:05 this is starting to have an exploding level of complexity 14:05 fun 14:05 -!- raisse has quit [Quit: Suddenly, Raisse disappears out of sight.] 14:05 -!- raisse has joined #hardfought 14:05 if i managed to learn my current profession, I might be able to learn to compile something :) 14:06 i don't think i have make installed though 14:06 compiling something you've never compiled before is normally quite difficult even for experienced developers, because every project seems to do things its own way and has pitfalls of its own 14:06 [hdf-us] [nh370] nabru (Val Dwa Fem Law) entered a shop, on T:167 14:06 you can probably use the mingw package manager to install make, gnuwin32 probably has a version of make too 14:07 the one from strawberry perl? 14:07 it's a fairly commonly seen package, which is probably why slashem depends on it 14:07 strawberry perl won't have optional packages 14:07 you just get what you get with it 14:07 I was hoping make would be there, maybe not though 14:07 so i'll just use regular mingw 14:07 regular mingw will definitely havei t 14:07 * it 14:07 make isn't a compiler of any sort, it's an automation tool that's used to run compilers 14:08 do u even just think that software engineering is piling manure on top of manure? 14:08 yeah, it has make.bin installed 14:08 ever* 14:08 so now i have to set up a new path to the standalone mingw 14:08 [hdf-us] [nh370] nabru (Val Dwa Fem Law) entered the Gnomish Mines, on T:412 14:09 amateurhour: with respect to build systems, yes 14:09 there are so many layers involved 14:09 that was part of the motivation for writing aimake, it merges about 5 or 6 of them into a single program, rather than having a huge number of layers which are each trying to work around deficiencies in the layers below 14:10 manually running gcc is too difficult -> let's create make 14:10 manually writing makefiles is too difficult -> let's create autotools 14:10 [hdf-us] [nh370] Karnac (Pri Hum Mal Neu) killed Medusa, on T:12275 14:11 amateurhour: you might be interested in reading http://nethack4.org/blog/building-c.html which is basically the "this is why I wrote aimake" document 14:11 <@riker> dnethack has "gnumakefile" in the home dir 14:11 it's basically just a list of all the complexity involved in compiling a C program 14:11 <@riker> super easy, just run "make install" :P 14:11 [hdf-us] [nh370] nabru (Val Dwa Fem Law) reached Mine Town, on T:631 14:11 @riker: if you have make installed, and all the compilers installed too 14:11 this has a chance of working on Linux and old Mac OS X, but won't work on Windows or recent Mac OS X 14:11 <@riker> it runs fine on recent macOS 14:11 I am interested in reading it 14:11 <@riker> source: am running on mojave right now 14:12 <@riker> it may or may not work on catalina but I have enough 32 bit apps I hvaen't updated 14:12 but before I read it I'm going to post this just in case it's relevant: https://xkcd.com/927/ 14:12 is that the one about creating new standards? 14:12 yeah, heh 14:12 this is different because it's a tool rather than a standard, and having multiple tools that do the same job isn't really an issue 14:13 maybe I'm interpreting standard a bit more loosly than a formal standard 14:14 sounds interesting 14:14 [hdf-us] [evil] Grouchy (Wiz Hum Mal Cha) killed Baalzebub, on T:49644 14:19 [hdf-us] [nh370] nabru (Val Dwa Fem Law), 611 points, T:2002, killed by a gas spore's explosion 14:21 -!- mobileuser has joined #hardfought 14:22 [hdf-us] [evil] krm26 (Bar Hum Fem Neu) had a leather armor entrusted to her by Crom, on T:55329 14:26 [hdf-us] [spl] hyperion67 (Arc Hum Mal Law), 1620 points, T:4257, killed by a psionic bolt 14:39 -!- raisse has quit [Ping timeout: 265 seconds] 14:42 [hdf-us] [nh370] Karnac (Pri Hum Mal Neu), 124482 points, T:13402, killed by a minotaur, while fainted from lack of food 14:43 -!- elenmirie_ has quit [Ping timeout: 256 seconds] 14:45 -!- elenmirie_ has joined #hardfought 14:50 [hdf-us] [nh370] nabru (Val Dwa Fem Law) entered a shop, on T:746 14:53 [hdf-us] [nh370] nabru (Val Dwa Fem Law) entered the Gnomish Mines, on T:1121 14:53 [hdf-us] [spl] hyperion67 (Arc Hum Mal Law), 240 points, T:1306, mowed down by a small mimic 14:56 [hdf-us] [evil] k2 (Val Gia Fem Law) became literate by reading a scroll, on T:5062 14:58 [hdf-us] [nh370] nabru (Val Dwa Fem Law), 758 points, T:2016, killed by a pony, while frozen by a monster's gaze 15:02 -!- noomy is now known as MoonyTheDwarf 15:03 -!- MoonyTheDwarf is now known as noomy 15:06 -!- bug_sniper has quit [Ping timeout: 265 seconds] 15:11 [hdf-us] [dnh] Pacra (Wiz Hum Mal Neu) entered Minetown for the first time, on T:3777 15:18 rofl on the competing standards 15:18 [hdf-us] [nh370] nabru (Val Dwa Fem Law) entered the Gnomish Mines, on T:267 15:21 [hdf-us] [nh370] nabru (Val Dwa Fem Law) reached Mine Town, on T:744 15:21 [hdf-us] [nh370] nabru (Val Dwa Fem Law) entered a temple, on T:781 15:22 [hdf-us] [evil] krm26 (Bar Hum Fem Neu) wished for "blessed greased +2 dragonhide speed boots", on T:55621 15:23 [hdf-us] [nh370] nabru (Val Dwa Fem Law) destroyed nabru's ghost, the former Stripling, on T:962 15:25 [hdf-us] [nh370] Kevkat (Hea Gno Mal Neu) killed Medusa, on T:24128 15:26 [hdf-us] [nh370] nabru (Val Dwa Fem Law) entered a shop, on T:1347 15:30 <@microlance> i accidently pissed of the watch when lawful and on my way out i took my helm of and killed a watchman and i didnt get any "you murderrererererer!!!" message. does this mean i got away? 15:31 <@luxidream> were you lawful when you killed him 15:31 your helm... of opposite alignment? 15:31 <@microlance> no chaotic 15:31 <@luxidream> if no then you're good 15:31 no, no murder if you killed while chaotic 15:31 murder is assessed when you actually kill them, how you annoyed them in the first place isn't relevant 15:32 <@luxidream> nice patch idea: remove murder 15:32 <@luxidream> everyone in the dungeon is morally ambiguous anyway 15:33 @luxidream: why is a "nice patch" encouraging murder? :-D 15:33 <@luxidream> well, it makes nethack a little less racist 15:34 <@microlance> murdering and stealing are lawful good activities in these universes 15:34 <@luxidream> murder of elves and dwarves is already okay, in fact, it's good play 15:34 <@microlance> most LG paladins are serial criminals of the worst kind 15:34 @luxidream that's more of nethack's thing that only humans are allowed to be shks, priests, watchmen etc 15:35 nethack has some odd hangups about humans 15:35 [hdf-us] [spl] hyperion67 (Arc Hum Mal Law), 1228 points, T:3185, struck down by a gnome lord, while sleeping off a magical draught 15:35 in the sense that most comparable fantasy systems don't treat them in the same special ways 15:36 <@microlance> no non-humans among the paladin orders in earlier dnd 15:36 in variants with, say, an elven shopkeeper, it'll still be murder to kill them 15:37 <@microlance> but sometimes murdering them is justified, if they put large mimics in their d2 shop for example 15:38 <@luxidream> I still don't see how murdering the shopkeeper is worse than murdering a random dwarf 15:38 It's the logical conclusion of the early DnD stuff Nethack was based on 15:39 the dwarf has the potential to hate you, the shopkeeper has only love in his heart 15:39 So it's a design flaw 15:39 Of sorts 15:39 <@microlance> the dwarves dont hold customers hostage or rob and murder them 15:40 <@microlance> murdering a dwarf just going about his way mining randomly like they do, should be seen as worse 15:41 I'd never do that. 15:41 Fido! Stop! Don't get him excited! Fido! Give the dwarf his kidneys back! Bad dog! 15:41 hey this ain't dwarf fortress 15:43 <@luxidream> invisible customers are not welcome 15:43 <@luxidream> can see what you're doing just fine 15:43 <@luxidream> ?? 15:43 when the beer runs out is when the riots start 15:44 <@microlance> they just dont want those creepy stalkers in their shops 15:45 rofl 15:50 [hdf-us] [nh370] Grandolar (Wiz Gno Mal Neu) entered a shop, on T:1123 15:57 [hdf-us] [nh370] hyperion67 (Arc Hum Mal Law), 13 points, T:262, killed by a goblin 15:58 *bangs head on wall some more* 16:01 -!- raisse has joined #hardfought 16:04 [hdf-us] [nh370] hyperion67 (Arc Hum Mal Law) entered the Gnomish Mines, on T:1660 16:06 -!- raisse has quit [Read error: Connection reset by peer] 16:06 -!- raisse has joined #hardfought 16:06 <@Sig> > You hear the roar of the sea. That can't be right.... Dammit dnh I was trying to avoid evil's mines. 16:07 I'm not sure I want to know what evilhack did to the Mines 16:08 [hdf-us] [nh370] Grandolar (Wiz Gno Mal Neu) entered the Gnomish Mines, on T:2104 16:10 [hdf-us] [nh370] hyperion67 (Arc Hum Mal Law), 390 points, T:1921, killed by a gnome 16:10 ais523: it's like grunthack but not quite as bad because the zombies are a little more sane and the water is lit up so you don't surprise wander into it 16:12 "it's like grunthack" was enough to make me know I didn't want to know :-D 16:12 screw grunthack 16:12 "not quite as bad" doesn't mean I think it is good :D 16:13 [hdf-us] [nh370] Kevkat (Hea Gno Mal Neu) wished for "3 blessed scrolls of charging", on T:25082 16:15 C:\>mingw32-make -B -f makefile.gcc 16:15 mingw32-make: makefile.gcc: No such file or directory 16:15 mingw32-make: *** No rule to make target 'makefile.gcc'. Stop. 16:16 lol if you survive the zombies in grunthack you can just drown in unseen water too? 16:16 [hdf-us] [nh370] Kevkat (Hea Gno Mal Neu) wished for "Blessed Fixed Greased +3 Helm of Brilliance", on T:25082 16:17 [hdf-us] [nh370] hyperion67 (Arc Hum Mal Law) entered a shop, on T:1263 16:19 [hdf-us] [nh370] sickopsycho (Val Dwa Fem Law) wished for "blessed rustproof +2 gauntlets of power", on T:26602 16:19 [hdf-us] [nh370] Kevkat (Hea Gno Mal Neu) wished for "Blessed Magic Marker", on T:25087 16:19 [hdf-us] [nh370] hyperion67 (Arc Hum Mal Law) entered the Gnomish Mines, on T:1501 16:20 <@Sig> The water is invisible in grunt??? 16:22 [hdf-us] [nh370] hyperion67 (Arc Hum Mal Law), 716 points, T:1536, killed by a gnome 16:25 porkman: you need to be in the src directory, which is where the setup script copied the makefile to. 16:25 [hdf-us] [nh370] Kevkat (Hea Gno Mal Neu) entered Gehennom, on T:25218 16:25 my advice would be to use a batch script, I have the following in my SLASHEM-Extended folder: https://github.com/SLASHEM-Extended/SLASHEM-Extended/blob/master/compile-all.bat 16:26 and the path needs to point to the folder where your mingw installation has the mingw32-make.exe, obviously 16:26 but, I guess since you don't know what MS-DOS is, you might also not know what a .bat file is :P 16:27 [hdf-us] [nh370] sickopsycho (Val Dwa Fem Law) changed form for the first time, becoming a wererat, on T:27113 16:28 I think i've figured out what a bat file is 16:29 so, i'll smash my head against it until i figure it out 16:29 porkman: I recommend wearing a hard helmet if you're going to do that 16:30 hah 16:30 porkman (Hac Hum Mal Neu), 239349 points, T:1586464255, killed by a bat file 16:31 YES, it's compiling 16:31 LarienTelrunya: ^^ new slex monster? 16:31 oh lol, now I'm remembering 16:31 B class, obviously 16:31 it probably won't finish compiling on windows unless you do a specific change to config.h 16:31 me, NEU? 16:31 i'm CHA 16:32 aosdict: yes, thanks ♥ 16:32 what do i have to change on config.h? 16:33 comment out the awful curses interface 16:33 wait, curses is the tiles one? 16:33 no, curses is a tty-based interface, but different from tty 16:33 slex has no tiles 16:33 you... don't know what a curses interface is?! 16:33 no, would I ever have tried it? 16:34 "the curses interface" is NetHack-specific 16:34 no, because I've disabled it on the esm server due to crashiness 16:34 I don't think it's the default anywhere, but it's a very common non-default setting on NAO and Hardfought for vanillaish variants 16:34 ok, so compiling slex takes some time 16:34 it's still doing it 16:34 compiling most programs takes a few minutes, slex is likely to take much longer 16:34 because the source files are a lot larger than they would be for a more normal program 16:35 LarienTelrunya: what optimization level do you use for slex? 16:35 do i need to buy a powerhouse desktop PC to do this? 16:35 porkman: my guess is no, big compiles can normally be handled by modern laptops, it just takes a while 16:35 uhh optimization level? 16:35 not sure how I set that? :P 16:35 LarienTelrunya: -O followed by a number when running the compiler 16:35 it tells the compiler how much effort it should put in to the resulting program 16:36 uhh, not sure, probably the default as defined in the makefile 16:36 it strikes me that for slex in particular, telling the compiler not to optimize monst.c might make the compile much faster, without changing much in the resulting code 16:36 given that there's not much to optimize there 16:36 although, the compiler might notice that and not do much wrt optimization 16:37 it's basically just one big constant, after all 16:37 anyway, most people just compile with -O2 or -O3 or -Os or whatever and it makes sense, but slex is an exceptional case and might benefit from an exceptional setting 16:38 [hdf-us] [nh370] Kevkat (Hea Gno Mal Neu) wished for his first artifact, on T:25519 16:38 [hdf-us] [nh370] Kevkat (Hea Gno Mal Neu) made his first artifact wish - "The Eye of the Aethiopica", on T:25519 16:39 OK, found it 16:39 in the SLEX repo 16:39 it is using -O1 16:39 which is an unusual setting but makes sense for a very large program, and which I might have recommended 16:40 it might actually have been kerio who set the optimization settings to what they are now, back in the day 16:40 you're basically asking the compiler to do a half-assed job of compiling, but not a completely terrible one 16:42 if you care about compilation time, tcc is a nice target 16:43 as I recall, it actually segfaulted trying toc ompile slex 16:44 ais523: my linux-debug hints file uses -O0 16:44 otherwise going in with gdb will find a bunch of variables optimized out 16:45 I hate -O0 so much but do realise that it's sometimes necessary 16:45 I'll often compromise with -Og 16:45 (which /shouldn't/ be a compromise, but often turns out to be in practice) 16:46 what's wrong with -O0 16:47 [hdf-us] [nh370] every (Mon Hum Mal Law), 2644 points, T:3967, quit 16:47 [hdf-us] [nh370] hyperion67 (Arc Hum Mal Law), 1089 points, T:2170, killed by a coyote 16:48 [hdf-us] [evil] krm26 (Bar Hum Fem Neu), 1228006 points, T:57638, killed by life drainage 16:48 I once had a project where ... interesting bugs would happen at higher optimization levels 16:48 I just use -Ofast for debugging so it compiles faster :^) 16:48 I've had that too, the solution was 'just compile with O2 and ship it' 16:49 evilhack Demogorgon nearly killed me with ONE acid attack and I have 300+ HP. What a nasty surprise. :-) 16:49 but I wanted to know what happened :| 16:49 [hdf-us] [nh370] Luxidream (Wiz Orc Fem Cha), 2712 points, T:2468, quit 16:49 nabru: ubsan 16:49 -!- krm26 has joined #hardfought 16:49 [hdf-us] [nh370] Luxidream (Wiz Orc Fem Cha), 4 points, T:44, killed by a sewer rat 16:50 That was brutal. I guess I know about the Vlad changes now 16:50 ouch krm26 :'( 16:50 mobileuser: oof 16:50 ok, so, what's optimisation? 16:50 Had absolutely no way to escape, before I knew it he had drained me below use of my offense 16:50 [hdf-us] [nh370] Luxidream (Wiz Orc Fem Cha) entered a shop, on T:477 16:50 <@Pacra> oh dang krm26 16:51 porkman: when you compile your code, you can tell the compiler to spend extra time making it faster 16:51 <@Pacra> I probably would have died to demo, this eh game 16:51 I tried to wand of teleport my way out of his tower and almost made it, I was one step from getting away and level teleporting back to my stash 16:51 <@Pacra> but got lucky 16:51 <@Pacra> ah, were you literally at lvl1? 16:51 Yup, died to level drain to 0 16:51 <@Pacra> oof 16:51 ...extra time making it faster? 16:52 how's it faster, if you spend extra time? 16:52 With I think 11 potions of gain level left behind in my stash for a quick ascent :-) 16:52 see, what I love about slex is that you need no stash 16:52 you ARE the stash 16:52 krm26 I was lucky to wield Stormbringer against Vlad 16:52 I was really surprised how Vlad went to town on my -55 AC 16:52 porkman: the compiler spends extra time making a program that runs faster 16:52 Shoulda wished for a dragonide elven helm of drain resistance I guess -- I had thought about it 16:53 C:/Strawberry/c/bin/../lib/gcc/x86_64-w64-mingw32/8.3.0/../../../../x86_64-w64-mingw32/bin/ld.exe: windows.o:windows.c:(.rdata+0x50): undefined reference to `curses_procs' 16:53 collect2.exe: error: ld returned 1 exit status 16:53 makefile.gcc:671: recipe for target '\games\slam32d\slashem.exe' failed 16:53 mingw32-make: *** [\games\slam32d\slashem.exe] Error 1 16:53 is that what you were talking about, Amy? 16:53 [hdf-us] [nh370] Luxidream (Wiz Orc Fem Cha) entered the Gnomish Mines, on T:1383 16:53 yes 16:53 [hdf-us] [xnh] Bartlem (Val Hum Fem Law) acquired the luckstone from Mines' End, on T:9935 16:53 as I said, config.h, comment out the line that says "awful curses interface" 16:53 I gotta say, not a fan of Barbs at all, but the Ring of P 16:53 P'hul is awesome 16:54 even I still forget that sometimes 16:54 only a p'hul doesn't realise how great that ring is :) 16:54 -!- Andrio has quit [Ping timeout: 260 seconds] 16:54 [hdf-us] [nh370] Luxidream (Wiz Orc Fem Cha) reached Mine Town, on T:1764 16:54 [hdf-us] [nh370] Luxidream (Wiz Orc Fem Cha) entered a temple, on T:1810 16:55 ais523: I've found that -Og will still optimize out some variables, whereas O0 reliably doesn't 16:56 and it's not like O0 makes my build of nethack run horrendously slow or anything. The worst-case algorithms in the game are usually O(ROWNO*COLNO). 16:56 [hdf-us] [nh370] hyperion67 (Arc Hum Mal Law), 134 points, T:1071, killed by a jackal 16:56 -!- krm26 has quit [Remote host closed the connection] 16:56 [hdf-us] [nh370] every (Wiz Gno Fem Neu) entered the Gnomish Mines, on T:878 16:57 Ok, I like notepad++ 16:57 it seems to show different kinds of things in different colours 16:58 it's...fleecy! 16:58 [hdf-us] [nh370] Luxidream (Wiz Orc Fem Cha), 2973 points, T:2766, killed by a rothe, while frozen by a monster's gaze 16:58 <@Pacra> hm, 16:58 after you get a bit more familiar with that, try something with even more features like vscode 16:59 <@Pacra> I'm guessing I can't pass beholder commands via discord 16:59 there are a lot of things you can bundlebundlebundle with notepad++ :-) (ie plug-ins) 16:59 [hdf-us] [nh370] zoya (Wiz Elf Mal Cha), 133 points, T:198, killed by a sewer rat 17:00 [hdf-us] [nh370] Luxidream (Wiz Orc Fem Cha) entered the Gnomish Mines, on T:657 17:00 ais523: I've found that -Og will still optimize out some variables, whereas O0 reliably doesn't ← that's what I meant by "compromise" 17:00 if it didn't do that it wouldn't be a compromise 17:01 porkman: to comment out a line in a Makefile, put # at the start of it 17:01 -!- eLtMosen1 has joined #hardfought 17:01 (just realised that that wasn't mentioned so far) 17:01 oh, this isn't a makefile 17:01 all the other comment lines had a / 17:01 to comment out a line in a .h file, put /* at the start and */ at the end 17:01 [hdf-us] [nh370] Luxidream (Wiz Orc Fem Cha), 761 points, T:772, killed by a kobold zombie 17:02 I don't understand why you hate -O0 so much 17:02 just realised we hadn't defined that 17:02 aosdict: have you looked at the output code? it makes me really upset 17:02 Pacra you might be able to, give it a go? 17:02 the registers spill every statement, that's so much memory bandwidth being used for no good reason 17:02 ais523: I'm not in the practice of compiling and decompiling code for no reason, no 17:02 it's only a constant factor slowdown, but it's probably around ×3 17:02 which is way too much 17:02 [hdf-us] [nh370] Luxidream (Wiz Orc Fem Cha) entered a shop, on T:482 17:02 <3 17:03 I get upset when people don't care at all about the efficiency of their programs 17:03 for a debug build of nethack on a modern computer, it really makes no difference at all 17:03 so, which mode does what? 17:03 porkman: what do you mean by "mode"? 17:03 O0, 1, 2, 17:03 etc 17:03 <@riker> @Pacra yes you can 17:03 <@riker> beholder was configured for that fine 17:03 porkman: higher numbers means the compiler puts in more effort 17:03 -!- eLtMosen has quit [Ping timeout: 240 seconds] 17:04 0 generates the simplest possible code even though it's really stupid 17:04 1 is the lowest level that generates actually usable code 17:04 2 is the balanced option that generates the fastest code it can while being safe 17:04 3 is the wild, safeties-off option that does things that makes some code faster but other code slower and just generally incomprehensilbe 17:04 then there are a few lettered options 17:04 -!- Andrio has joined #hardfought 17:04 s generates the smallest executables possible 17:04 and g generates the fastest code that isn't painful to debug 17:05 ok, so it optimizes the code 17:05 yep 17:05 whether to use -O2 or -O3 or -Os is often a matter of much debate 17:05 [hdf-us] [nh370] Luxidream (Wiz Orc Fem Cha) entered the Gnomish Mines, on T:1612 17:05 the other options aren't very popular, although I can see -O1 making sense for slex 17:05 all of this because it's translating it all into a binary file in the end 17:05 right 17:05 there's more than one way to do that 17:06 C is often described as a low-level language, but it doesn't map that well onto modern processors 17:06 so there's a lot going on to translate the source into a binary 17:06 -!- raisse has quit [Ping timeout: 250 seconds] 17:06 [hdf-us] [nh370] Luxidream (Wiz Orc Fem Cha), 1420 points, T:1930, killed by a bolt of lightning 17:06 (OTOH, on old-fashioned processors, it was a very good match and C was basically just a system-agnostic way of writing your machine code) 17:07 [hdf-us] [nh370] zoya (Wiz Elf Mal Cha) entered the Gnomish Mines, on T:1008 17:07 [hdf-us] [nh370] hyperion67 (Arc Hum Mal Law) entered a shop, on T:2630 17:07 obviously, since I have been compiling and running my debug build with -O0, O1 is not the lowest level that generates actually usable code :P 17:07 [hdf-us] [nh370] hyperion67 (Arc Hum Mal Law) entered the Gnomish Mines, on T:2645 17:07 but nowadays, processors have features like out-of-order execution and parallel instruction dispatch that can make it very hard for a human to figure out what code to write 17:08 aosdict: but think of the environment! think of your electricity bill! 17:08 admittedly, if you're heating your house anyway 17:08 may as well heat it using your CPU rather than your central heating 17:08 ais523: I believe zortech c++, from the late 80s/early 90s was the first compiler to apply actual optimizations 17:08 [hdf-us] [nh370] Luxidream (Wiz Orc Fem Cha) entered a shop, on T:645 17:08 that sentiment is better directed at bitcoin miners than me 17:08 and I don't think ooo was a thing until the late 90s 17:08 moon-child: that late? 17:08 or are we talking about C compilers specifically? 17:09 just c compilers 17:09 [hdf-us] [nh370] Luxidream (Wiz Orc Fem Cha), 168 points, T:670, killed by Mr. Angmagssalik; the shopkeeper 17:09 ah, in that case, because C was officially released in 1989 17:09 so, once i have compiled it, where does the compiled file appear? 17:09 it'd just be a matter of who wrote it first 17:09 I heard an anecdote from its author that a magazine was doing c compiler benchmarks, and his compiler realised the benchmark wasn't doing anything and so removed it 17:09 and was accused of cheating 17:09 ais523: well, sure, but people were using it since the 70s 17:09 [hdf-us] [nh370] zoya (Wiz Elf Mal Cha), 583 points, T:1318, killed by a wererat 17:09 porkman: that'll be based on the details of the makefile, so you'll have to ask LarienTelrunya how hers is configured 17:09 <@Pacra> oh cool, thanks @riker. show me an example? logfile perhaps? 17:10 with some makefiles you write "make install" to have the files copied somewhere else, and with others you just find the file directly and run it 17:10 it's configured to compile it to c:/games/slam32d 17:10 LarienTelrunya: is that with or without a separate install step? 17:10 [hdf-us] [nh370] BLARGO (Val Hum Mal Law) killed Lord Surtur, on T:42512 17:10 if that folder doesn't exist, the compilation may fail, so either create the folder or change the makefile I guess 17:10 ais523: without; the exe and all other files necessary to run the game appear in that folder 17:10 got it 17:11 it did create it 17:11 allll by itself 17:11 LarienTelrunya: you know this is bad form, right? :-D 17:11 porkman: great! does it also run? 17:11 still recompiling it 17:11 [hdf-us] [nh370] Luxidream (Bar Orc Fem Cha) entered a shop, on T:651 17:11 if you're installing into the root of C then you need admin rights, but most of the compile doesn't need admin rights 17:11 ais523: yes; this is actually a leftover from how SLASH'EM is configured 17:11 i have admin rights :) 17:11 [hdf-us] [nh370] BLARGO (Val Hum Mal Law) acquired the Bell of Opening, on T:42515 17:11 so if you do it all at once you need to turn on the admin rights mid-compile, which is awkward 17:12 that's why the install step is normally separate 17:12 the slashem devs did make some questionable design decisions 17:12 I guess it's OK because this is Windows, which is used to programs requesting admin rights at random times 17:12 How long did it take you to create a workable game, Amy? 17:12 [hdf-us] [nh370] Luxidream (Bar Orc Fem Cha) had Cleaver bestowed upon her by Set, on T:1403 17:13 LarienTelrunya: they switched to autoconf in 0.0.8 17:13 porkman: what do you mean with workable? you mean how long it took me to set up mingw and stuff? don't remember, was so long ago 17:13 which...I won't say it's good, but it's obviously better than nethack's whole...deal 17:13 porkman: the issue is not creating the workable game originally – Slash'EM itself is workable – rather, it's getting the game back to a workable state after you start changing it 17:13 no, i mean, to then edit the source to make slashem into the first version of slex 17:13 moon-child: autoconf solves the wrong problem 17:13 (first version that worked) 17:13 ais523: as a user, autoconf is easy 17:14 huh? I made a change, recompiled, and the game still ran and I did see the effect of my change :P 17:14 [hdf-us] [nh370] Luxidream (Bar Orc Fem Cha) entered the Gnomish Mines, on T:2082 17:14 moon-child: not always, it depends on the quality of the makefiles 17:14 then did that a couple more times and quickly I had my first panic :P 17:14 if they're well-written, you'll often discover that it'd be just as easy without autoconf 17:14 fair enough 17:14 LarienTelrunya: huh, panic before segfault? that's a little surprising 17:14 (I should know--I wrote just such a makefile for slashem9) 17:15 ais523: the generation chances for rings (probabilities) exceeded 1000, and there was a panic that caught it. 17:15 in NH4, I discovered I needed an autoconf equivalent for precisely two things, neither of which were actually implemented by autoconf 17:15 [hdf-us] [nh370] hyperion67 (Arc Hum Mal Law), 556 points, T:2682, killed by a wand 17:15 which things? 17:15 [hdf-us] [nh370] Luxidream (Bar Orc Fem Cha) had Dragonbane bestowed upon her by Set, on T:2556 17:15 a) how to switch the compiler into C11 mode, b) the correct spelling of _Noreturn for the compiler in use 17:16 [hdf-us] [nh370] Luxidream (Bar Orc Fem Cha) had Ogresmasher bestowed upon her by Set, on T:3376 17:16 btw porkman, if you made a minor change, you'll want to run mingw32-make -f makefile.gcc install 17:16 [hdf-us] [nh370] Vigorix (Tou Hum Mal Neu) had Vorpal Blade bestowed upon him by The Lady, on T:2706 17:16 the -B in mingw32-make -B -f makefile.gcc install makes it so that *everything* gets recompiled; if you just changed one file, leave the -B out so that it only recompiles the one file, which is muuuuuuuuuuuuch faster 17:16 ais523: #if __STDC_VERSION__ >= 201112L #include #else #define noreturn #endif 17:16 moon-child: that will never generate __declspec(noreturn) 17:17 nor __attribute__((noreturn)) 17:17 first one is easy enough, _msc_ver. Second--are old versions of gcc really common/relevant enough to care about? 17:17 moon-child: they were at the time 17:17 fair enough 17:17 especially on Mac OS X 17:18 ais523, much of the point of optimization is to be able to write code at a higher level 17:18 [hdf-us] [nh370] Luxidream (Bar Orc Fem Cha) reached Mine Town, on T:4263 17:18 for example, modern code tends to have smaller functions than code from 20 years ag 17:18 this depends on the compiler inlining for good performance 17:18 jilles: well, that's technically true, but the optimisations you get at -O1 allow you to do things like leave the "register" off variable declarations 17:18 so, has dnhslex been ascended? 17:19 which is something that is unthinkable to not do with today's compilers 17:19 yes, by Demo 17:19 [hdf-us] [nh370] hyperion67 (Arc Hum Mal Law), 124 points, T:176, killed by a jackal 17:20 ais523, also, not reusing variables for a different purpose 17:20 ok, so -B stays out to compile a minor change, will the compiler itself know what minor change i made? 17:20 that depends on optimization to reuse the space on the machine level 17:20 <@riker> porkman: for running make? 17:20 porkman: it is make that knows which minor changes you made 17:20 porkman: yes, it'll recompile only the files you've changed. if you change files that require other files to be recompiled, it does so automatically 17:21 it looks at the timestamps on the files and tries to figure out what parts of the build need redoing 17:21 <{Demo}> i ascneded dnhslex cause im a real one 17:21 in theory -B should never be necessary if the makefile is set up accurately enough 17:21 but SLEX's was written by a human and almost certainly contains mistakes that mean the -B is required for large changes 17:21 huh, just tested: even on modern gcc, "register" does make a difference at -O0 17:21 without it, the variables will be spilled between every usage 17:21 <@riker> i always use B when i've touched like 4 files or more, just for kicks 17:22 annoyingly the -B is required if you change artilist.h, I guess that could be configured if I knew how :P 17:22 <@riker> demo: what combo? 17:22 <@riker> cha inc barb?/ 17:22 <{Demo}> dwarf ana i think 17:22 <{Demo}> cause carry cap is high af 17:22 <{Demo}> so i could just hold a lot of bullets 17:23 <{Demo}> so early game was breeze 17:23 <@riker> oh that's really stupid 17:23 LarienTelrunya, the trick in avoiding -B is to have the compiler generate the dependencies via options like -MMD 17:23 jilles: that still isn't technically correct when generated files are involved, although it does work 99% of the time 17:24 this is the sort of problem tup was supposed to avoid 17:24 and of course, not all the dependencies in NetHack are related to a C compiler, so those would need to be hard-coded 17:24 moon-child: aimake, too 17:24 but, NetHack's internal structure means that full rebuilds are needed quite frequently 17:24 jilles: huh, I thought by adding them to the .c files? or is that only when a .c depends on a .h? 17:24 because every source file includes almost every header file 17:25 -!- Tangles has quit [Remote host closed the connection] 17:25 in order to reduce the number of rebuilds, you'd have to split up hack.h and extern.h and instead make source files depend only on the headers they actually needed 17:25 at one point I was threatening to automatically generate extern.h as a number of smaller files, automatically split up, but the other NH4 devs wouldn't let me :-( 17:25 didn't DNHslex mostly remove carry caps? 17:25 lol what a question 17:26 [hdf-us] [nh370] Vigorix (Tou Hum Mal Neu) entered the Gnomish Mines, on T:3932 17:26 if it's made by me, the player can carry more than a single item before being burdened :P 17:27 this is also true for nethack products not made by yuo 17:27 you 17:27 it's even true in NetHack products made by me! 17:28 -!- NAOrsa has quit [Quit: Lost terminal] 17:28 not if you pick up two suits of plate mail! 17:29 ~~not with that attitude~~ 17:29 hahaha 17:29 LarienTelrunya: you misspelled "ice box" 17:29 haha 17:29 most iceboxes in vanilla are unliftable due to the weight of the items inside 17:29 (which are often also unliftable so you can't just take them out) 17:30 this is one of the big advantages of #tip 17:30 -!- Tangles has joined #hardfought 17:30 -!- mode/#hardfought [+o Tangles] by ChanServ 17:30 the only unliftable items i have seen in slex are post-nerf ultra heavy balls of some kind 17:30 that shall be my first change 17:30 put balls back down to a manageable weight 17:30 and create 17:30 Porkman's balls of steel, the artifact 17:31 <@riker> what's it do 17:31 [hdf-us] [nh370] hyperion67 (Arc Hum Mal Law) entered the Gnomish Mines, on T:1412 17:32 yay 17:32 the compile works 17:32 the graphics are screwy 17:32 I considered creating a heavy iron ball of holding 17:32 do you have a .exe now? 17:32 ie, water appears as something 17:32 where you put items in and drag them around 17:32 ais523: ha 17:32 ¸ 17:32 and they don't count against your inventory weight because the ball is on the ground 17:32 i don't even know what this character is 17:32 somehow I feel like LarienTelrunya would disapprove, though 17:33 porkman: so it runs! great! from here on you are on your own :P 17:33 [hdf-us] [nh370] efot (Hea Gno Fem Neu) entered a shop, on T:1240 17:33 <@riker> i feel like amy dislikes vanilla on principle at this point 17:33 <@riker> :p 17:33 <3 17:33 not nearly as much as she is going to dislike Pork'em 17:33 Like you have to ask lol 17:33 ais523: heh that idea would probably not be so bad at all, since you're still punished and thus moving at half speed whenever the bag-ball moves 17:33 porkman: ooh, I think your terminal has a codepage that's incompatible with Slash'EM's 17:33 [hdf-us] [nh370] hyperion67 (Arc Hum Mal Law) reached Mine Town, on T:1553 17:34 LarienTelrunya: oh, I thought you'd dislike it because the item /should/ be unhelpful because you should be able to carry that much in main inventory 17:34 [hdf-us] [nh370] Luxidream (Bar Orc Fem Cha), 6134 points, T:5661, quit 17:34 porkman: try exiting/saving the game, running the command "chcp 437" and reloading theg ame 17:34 *game 17:34 [hdf-us] [nh370] hyperion67 (Arc Hum Mal Law) entered a temple, on T:1590 17:34 I don't know for certain that this will work, it's just an educated guess 17:34 [hdf-us] [xnh] bouquet (Cav Gno Fem Neu) wished for "uncursed magic marker named Gutenberg Enchant ARmor Scroll Press", on T:73961 17:35 if it doesn't help, turn off IBMgrahpics in the options, that definitely will work but it'll be a bit ugly 17:35 nice, it worked, thanks 17:35 but what is chcp 437? 17:35 it's back to its fleecy slex graphics 17:35 porkman: it's kind-of a long story 17:36 it's related to how the terminal and game agree on how to name each possible character 17:36 there's a huge range of possible conventions 17:36 437 is a really old-fashioned convention that has some history in roguelikes 17:36 now a fountain is a ¶ though 17:36 [hdf-us] [nh370] hyperion67 (Arc Hum Mal Law) entered a shop, on T:1693 17:36 i can live with that 17:36 the chcp command tells the terminal to use convention ("codepage") number 437 in order to interpret the instructions from Slash'EM 17:37 and Slash'EM is presumably using the same convention to encode the characters, so they agree with each other now 17:37 TAKE THAT DEMOGORGON 17:37 isn't chcp only on unix? 17:37 when they don't match, you have to change the settings at both ends until you find some format they both agree on 17:37 sorry 17:37 mobileuser: no, it's a DOS command 17:37 I don't think it exists on UNIC 17:37 *UNIX 17:37 it still exists in Windows for backwards compatibility 17:37 ahh 17:38 on UNIX the equivalent would be "export LANG=en_US.IBM437" 17:38 but that doesn't reconfigure the terminal, just tells the program what format to output in, so it isn't tge same 17:38 *the 17:38 to reconfigure the terminal itself, you'd have to go into its menus 17:38 what terminal emulators do you guys use? 17:39 anyway this is mostly a solved problem nowadays given that everyone on UNIXes uses UTF-8, but encoding matching was a problem there for a long time 17:39 amateurhour: I have a huge number installed for debugging purposes, but normally default to gnome-terminal (which is just a wrapper around libvte) 17:40 I've been using st for quite a while 17:40 on UNIX it's standard for the program to prefix its output with a few escape codes to tell the terminal what format it's in 17:40 and a few terminals will even honour these sequences 17:40 what do you use to open .lev files? 17:40 but encoding matching is kind-of a minefield, before everyone standardised on one encoding it was a huge pain to get it working 17:40 echo $TERM 17:40 porkman: you don't, they're built from the .des files with the same name 17:41 -!- LarienTelrunya has quit [Quit: Connection closed] 17:41 and you can open a .des file in Notepad++ 17:41 xterm-256color 17:41 <@riker> amateurhour: I just use the default mac terminal lol 17:41 bouquet: $TERM became useless for its intended purpose fairly quickly 17:41 <@riker> you don't need a funky terminal at all 17:41 the basic issue is that almost every terminal claims to be xterm-compatible, but most of them are lying 17:41 that would explain why my pants are on fire 17:42 yeah mine is screen-256color, can't remember why I set it to that 17:42 <@riker> how many of them say xterm-256color 17:42 something about vim colours in tmux 17:42 so $TERM isn't very useful for determining what codes to send the terminal 17:43 amateurhour: either Konsole or mate-terminal or gnome-terminal 17:43 depending on your current DE? :P 17:43 (yeah, I like to try differnt WMs/DEs now and then) 17:43 my output was from gnome-terminal. 17:43 @riker: almost all of them, the main exceptions I know of offhand are screen and rxvt/urxvt 17:44 ok, so 17:44 item properties seem fun 17:44 (this is including both "xterm" and "xterm-256color") 17:44 <@riker> porkman: this is a slex variant right? so amy should have code for those already and all that 17:44 note that xterm also fits in my list of "terminals that tell the truth in $TERM" because it actually /is/ xterm-compatible 17:44 <@riker> i'd hope xterm is xterm-compatible 17:45 maybe at some point xterm should change its $TERM to actually-xterm or something 17:45 so we can determine for sure whether the terminal actually accepts xterm commands 17:46 thanks, ais523. TIL. 17:47 $ echo $TERM 17:47 stterm-256color 17:47 yesyes 17:47 messing with them is fun 17:47 ooh, stterm gets a reward for truthfulness, too 17:47 what's stterm? 17:47 st 17:47 some variant of st? 17:47 oh 17:48 it's been renamed to avoid naming clashes on my OS 17:48 somewhat goes against the spirit of st really 17:48 amateurhour: thanks for mentioning a terminal that I wasn't running as part of my terminal testsuites 17:48 you're meant to compile changes and config in 17:48 [hdf-us] [nh370] hyperion67 (Arc Hum Mal Law), 2237 points, T:2983, killed by a gnome lord 17:48 well, my OS has source packages that are intended to "just install" if you edit the source 17:48 so having it packaged is still useful 17:49 if i'm going to alter drop chances on things, do i have to keep the total numbers the same, or will it calculate its probabilities as it goes on? 17:49 source based distro like gentoo? 17:49 porkman: they need to add up to 1000 17:49 within one item class 17:49 so if you're increasing some numbers you need to reduce others to compensate, and vice versa 17:49 amateurhour: it's Ubuntu, which is binary-based by default but allows you to install a package source-based if you want to 17:49 (very few people do, but I'm me) 17:50 i see, awesome, thanks 17:50 oh right, fair enough 17:51 st's results on my terminal tests: it's missing many newer features but seems to be well-behaved on the older ones 17:51 it doesn't have scrollback by default, you have to patch it in :) 17:51 they seem to add up to a lot lot more than 1000 in slex though 17:51 <@riker> amy may have changed it to 10k 17:52 <@riker> where's the object parser again? 17:52 right, any advice I give you will be based on vanilla NetHack (or possibly vanilla Slash'EM in cases where I'm aware of differences) 17:52 and thus might not apply to SLEX 17:52 @riker: you mean the wish parser? or something else? 17:52 the wish parser is "readobjnam", I forget which file it's in but it might be doname.c 17:52 <@riker> no, the file that parses objects[], artilist[] into POT_XXX / ART_XXX 17:52 that's makedefs 17:52 <@riker> etc. 17:52 which is in util I think 17:53 <@riker> yah found it 17:53 <@riker> thanks 17:53 <@riker> yeah amy changed the total to 10k porkman 17:55 [hdf-us] [fh] groovejumper (Tou Hum Fem Neu), 9253 points, T:402, killed by a falling rock 17:58 [hdf-us] [nh370] hyperion67 (Arc Hum Mal Law) entered the Gnomish Mines, on T:1824 18:00 [hdf-us] [nh370] Grandolar (Wiz Gno Mal Neu) reached Mine Town, on T:3030 18:05 rhel question: anyone know where I can find (if it exists) src.rpm for nethack? I would like to yoink some of its config for xnethack/splice/etc 18:05 amateurhour: st just scored the highest score on my terminal tester for any terminal without scrollback :-D 18:06 hah, nice 18:06 I would have no qualms about recommending this to someone; the main oddity is that the palette reset code is apparently the terminal reset code rather than having a separate code of its own 18:06 is alacritty on your list? 18:06 I don't have it installed, so no 18:06 [hdf-us] [fh] groovejumper (Tou Hum Fem Neu), 31996 points, T:1745, killed by a rabid rat 18:06 also the tester isn't finished yet 18:07 it's meant to be gpu accelerated or something, idk 18:07 looking at st's stderr output, it seems to have correctly parsed everything it didn't understand, too, just didn't know what it means 18:07 <@riker> ais523: what does your terminal tester check? 18:07 [hdf-us] [nh370] Grandolar (Wiz Gno Mal Neu) entered a temple, on T:3193 18:08 @riker: it primarily looks at things that I care about for NetHack4, so understanding of encodings, colors, cursor motion commands 18:08 but I'm hoping to expand it to all the terminal codes eventually 18:08 <@riker> oh neat 18:08 <@riker> is it online like jettyplay? 18:08 I have two terminal testers 18:09 a very basic one that's online but I haven't linked everywhere, except in IRC every now and then 18:09 and a more complex one I'm working on that's unfinished and doesn't generate the output in useful form 18:09 alacritty claims to be the fastest terminal emulator, but I've not used it 18:09 -!- Umbire_ has quit [Remote host closed the connection] 18:09 last I heard it's missing some things 18:12 [hdf-us] [nh370] Grandolar (Wiz Gno Mal Neu), 2330 points, T:3265, killed by a white unicorn 18:13 terminal emulator speed can be hard to measure objectively 18:13 as it depends a lot on what's being printed 18:14 looks good in benchmarks at least 18:14 I think it was discovered a while back that some terminal emulator went quadratic when printing very long strings of letters, but not very long strings of punctuation marks 18:14 likewise, I'd expect terminals to have very different results for rendering and for parsing control codes 18:15 and there are some operations, such as retroactive palette sets, whose performance is likely to vary wildly on terminals due to different internal implementations of them and even different meanings for the codes 18:18 [hdf-us] [evil] dingotron (Con Hob Fem Cha) stole 7 zorkmids worth of merchandise from Chibougamau's general store, on T:21295 18:18 I don't think any programs actually sit there spamming retroactive palette sets, but on some terminals, such as Linux fbcon, it'd allow you to get truecolor out of a terminal that only supports 16 colors 18:18 so it's a technique that would be worth considering if it weren't so insane 18:21 time to sleep, can barely keep my eyes open 18:21 seeya :) 18:21 gnite amateurhour 18:21 where would i find definitions for the various random functions? 18:21 say, rn2 18:22 porkman: src/rnd.c contains the implementations of the functions themselves 18:22 if(rn2(2).... 18:22 and also has comments saying what they're for 18:22 but a very brief summary is that rn2(x), which is the most commonly used, returns a random number from 0 inclusive up to x exclusive 18:22 and if it's used as the test in a conditional, it therefore has a 1 in x chance of being false, and is true the rest of the time 18:23 "if (rn2(3))" thus means "2/3 of the time" 18:23 and "if (!rn2(3))" means "1/3 of the time" 18:23 I think I see the negated version (with !) more often 18:23 <@riker> rnd is from 1 to x, rn2 is from 0 to (x-1), rnl is luck-biased rn2 (high luck makes it more likely to return 0, and rnz is funky 18:24 if you're interested in rne, rnl or rnz I recommend checking the wiki 18:24 so the ! is a negation 18:24 <@riker> oh yeah rne is xp biased 18:24 porkman: right 18:24 <@riker> it's negation for anything, not just random functions 18:25 -!- amateurhour has quit [Ping timeout: 256 seconds] 18:26 @riker: have you added any extra uses of rnz? or have you decided that it's best kept well away from? :-D 18:26 <@riker> dnethack adds some more things 18:26 <@riker> and it's annoying as hell 18:26 it strikes me as the sort of function that might actually be useful in slex when you want a very unpredictable result 18:26 <@riker> (it's cooldowns for dragon breath, among other things) 18:26 <@riker> *other cooldowns 18:26 ooh, rnz on monsters, I hadn't thought of that 18:27 <@riker> well, it's not for monster dragon breath actually 18:27 <@riker> it's for player half-dragon race breath 18:27 <@riker> er, not that actually, for DSM breath 18:27 so if (rn2(10) foo) means, 2/10 of the time? 18:27 <@riker> the foo should be on the outside 18:27 <@riker> and no, tha'ts a 9/10 chance? 18:27 porkman: if(rn2(10)) foo would mean 9/10 of the time 18:27 <@riker> rn2(10) would return 0 1/10 of the time, which would be false 18:27 the 2 is misleading, the function doesn't have anything to do with 2, it's just to distinguish the function from the other random functions 18:27 oh, right 18:27 <@riker> random-number-to 18:28 so it returns false on a 0 and true otherwise 18:28 <@riker> yes porkman 18:28 porkman: that's a general rule of C, 0 (and NULL, which is 0 in disguise) are the only false values 18:28 so (!rn2(10)) would be 10% 18:28 <@riker> ais: dnethack lets you breath dragonbreath if you're wearing matching dragon scale(s| mail) and a a dragon scale shield 18:28 yep 18:28 <@riker> it's one of my favorite flavor-based mechanics, even if it's not that good 18:28 ("rn2(100)) 1% and so on 18:28 what happens if the rnz cooldown hasn't run out yet? 18:28 porkman: assuming that's a typo for !, yes 18:29 <@riker> it pulls up in #monster if i'ts there 18:29 <@riker> if it's not - it's just not there 18:29 excellent, i get it, thanks <3 18:29 <@riker> which leads to annoying checking repeatedly 18:29 also, huh, you're using a UK keyboard layout 18:29 I normally assume that unknown people are American, but there's a European bias towards variant devs for some reason 18:29 <@riker> you're UK, right? 18:29 yes 18:30 I'm not sure where French and German layouts put their double-quote 18:30 but with US layouts it's nowhere near ! 18:30 so you can get some information from the typos people make 18:31 <@riker> but yeah rnz is annoying, since it's hard to rely on (which is probably intentional for prayer at least) 18:31 -!- nabru has quit [Ping timeout: 256 seconds] 18:31 the best way to think of rnz is "the output is /normally/ low, but is unreliable and could be very high" 18:32 to comment on a .c fileit's still /*? 18:32 Ais523, it's right next to it 18:32 <@riker> yeah, i'm familiar with the distribution (showed the std dev equations to a math major friend, he was horrified) 18:32 <@riker> porkman: // for a single line comment 18:32 !"£$%^&*() 18:32 is 1234567890 with a shift 18:32 <@riker> /* starts a multiline comment tha'ts ended by */ 18:33 porkman: the £ is a huge giveaway of a UK layout :-) 18:33 <@riker> porkman: in the US our 1-9 go !@#$%^&*() 18:33 <@riker> double quote is where your @ is I think, next to right enter 18:33 porkman: it's worth mentioning that // comments are disallowed by NetHack's style guide, but Amy probably doesn't care what your SLEXy comments look like 18:33 <@riker> oh, are they really? 18:33 most modern compilers understand // comments just fine, it's only old compilers that struggle with them 18:33 <@riker> whoops 18:34 there's some sentiment that we shouldn't be making things any harder for old compilers than necessary, though 18:34 I can just put /* --- */ 18:34 <@riker> meh my comments aren't making it into vanilla so it's fine 18:34 <@riker> porkman yes 18:34 and, well, // comments were invented over 20 years ago, and even if compilers were slow to adopt them, they've had plenty of time to catch up by now :-D 18:34 <@riker> the --- is ignored and you can put anything that's not */ 18:35 <@riker> spaces are ignore as well, actually 18:35 [hdf-us] [nh370] Luxidream (Wiz Orc Fem Cha) entered a shop, on T:1028 18:35 <@riker> ais523: oh that reminds me, any clue why the function declarations in cmd.c all have /**/ after them? 18:35 [hdf-us] [nh370] Luxidream (Wiz Orc Fem Cha), 451 points, T:1200, killed by a system shock 18:36 @riker: those definitions are a workaround to some complier bug or other; it may be to make them look sufficiently different from normal `extern` declarations that automated tools can tell them apart, or it might just be a reminder by the dev who wrote them that they aren't "real" declarations 18:36 <@riker> they're in an ifdef DUMB I believe 18:37 <@riker> so likely it was some old compiler then 18:37 [hdf-us] [nh370] Luxidream (Pri Elf Fem Cha) entered a shop, on T:550 18:37 there's something of a meta-convention in C that doing something obviously pointless for no reason is a clue that "yes, I really mean this, even though it looks wrong" for some very close part of the code 18:37 <@riker> hm that's reasonable 18:37 e.g. if (a=b) looks wrong, it should normally be if (a==b) to do a comparison without an assignment 18:38 so if you really want the assignment you write if ((a=b)) to make clear that whatever nonsense is inside is intentional 18:38 <@riker> ^ clang's got a warning for that 18:38 [hdf-us] [nh370] Luxidream (Pri Elf Fem Cha) entered the Gnomish Mines, on T:782 18:38 because you wouldn't have added the second pair of parentheses by mistake 18:38 <@riker> src/bones.c:106:16: warning: equality comparison result unused [-Wunused-comparison] 18:38 both clang and gcc have a warning for that, and both will turn off the warning if you double-parenthesize! 18:38 <@riker> bleh that one is a bug lol 18:39 <@riker> and I just realized that's the wrong warning, but I know i've seen it 18:39 that's the opposite mistake, isn't it? using == when you meant = 18:39 <@riker> src/restore.c:903:19: warning: using the result of an assignment as a condition without parentheses [-Wparentheses] 18:39 <@riker> yeah 18:39 -!- Chris_ANG has quit [Ping timeout: 265 seconds] 18:39 [hdf-us] [evil] dingotron (Con Hob Fem Cha) killed Warden Arianna, on T:21655 18:39 (porkman: the difference is that a==b means "a currently equals b", and a=b means "change a to be the same as b"; these are very different operations but the spelling is almost the same, so it's easy to confuse them) 18:40 <@riker> i hate clang sometimes 18:40 the problem being that you have two very fundamental operations that both want to be called =, but you need to distinguish somehow, so every language has to compromise somewhere 18:40 <@riker> generally = is called assignment and == is equality, I believe 18:40 maybe it would be best to change them both, and have := for assignment and == for equality 18:40 <@riker> ^ 18:40 [hdf-us] [nh370] Luxidream (Pri Elf Fem Cha) reached Mine Town, on T:2039 18:41 <@riker> a couple languages do that I know 18:41 <@riker> r also uses <-, which is funny 18:41 it stole that from INTERCAL :-P 18:41 [hdf-us] [nh370] Kevkat (Hea Gno Mal Neu) killed the invisible Cyclops, on T:28595 18:41 which picked it to be arbitrarily different 18:42 probably the other languages that used it were unaware of the inspiration 18:42 <@riker> didn't you write a intercal derivative at once 18:42 <@riker> *at one time 18:42 [hdf-us] [nh370] Kevkat (Hea Gno Mal Neu) acquired the Bell of Opening, on T:28597 18:42 @riker: I helped/help to maintain the main compiler 18:42 <@riker> ah maintaining a interpreter 18:42 I guess that's a derivative in some sense 18:42 it's a compiler, not an interpreter 18:42 <@riker> I just knew your name associated with it 18:42 <@riker> > He also maintains a modern Intercal interpreter. 18:43 <@riker> your esolangs page (not your userpage, the one about you) is wrong 18:43 hmm, someone out there doesn't know the difference between compilers and interpreters :-D 18:43 <@riker> or didn't know what intercal was 18:43 I guess it's technically an interpreter if you run the debugger 18:43 <@riker> and defaulted to interpreter, since most esolangs are probably interpreted 18:43 yes, most esolangs are interpreted 18:43 INTERCAL interpreters exist too, but C-INTERCAL is a compiler 18:44 also, whowever wrote that doesn't know how to spell INTERCAL, either :-D 18:44 (the name is allcaps) 18:44 <@riker> lol 18:47 (it's an acronym for Compiler Language With No Pronounceable Acronym) 18:47 [hdf-us] [nh370] Luxidream (Pri Elf Fem Cha), 4357 points, T:4504, killed by a killer bee 18:48 <@riker> how's that lead to INTERCAL then? 18:48 <@riker> where'd that name come from? 18:48 and acronyms are apparently written in allcaps even if the letters that make them up don't match the words they stand for 18:48 @riker: well the whole language is a joke, so there probably isn't a good reason 18:48 <@riker> fair 18:48 the stated etymology of the name is obviously ridiculous 18:48 [hdf-us] [nh370] Luxidream (Pri Elf Fem Cha) entered a shop, on T:344 18:48 but it's all we have to go on 18:49 [hdf-us] [nh370] Luxidream (Pri Elf Fem Cha) entered the Gnomish Mines, on T:904 18:50 -!- nabru has joined #hardfought 18:50 the wikipedia page on INTERCAL is very funny 18:50 thanks, I think I'm responsible for quite a bit of it 18:50 <@riker> mobileuser: the esolangs.org page may have better information, if you're curious 18:50 "if "PLEASE" does not appear often enough, the program is considered insufficiently polite, and the error message says this; if too often, the program could be rejected as excessively polite. " 18:51 although there are still factual inaccuracies I can't correct because I'm not a reliable source, and the reliable sources have incorrect information in them 18:51 [hdf-us] [nh370] Luxidream (Pri Elf Fem Cha), 1832 points, T:1330, killed by a soldier ant 18:51 mobileuser: that was actually an undocumented feature in the original INTERCAL-72 implementation; it still exists in more recent implementations but has been documented now 18:51 the correct ratio of PLEASE to DO is around 1:3 18:52 fantastic!! 18:52 although there's a lot of leniency 18:52 Emacs' intercal-mode will replace DO with PLEASE ¼ of the time, because of this 18:52 and autogenerated INTERCAL code normally puts a PLEASE on every fourth line 18:52 although when I write it by hand, I normally put the PLEASEs in places where they look right (e.g. PLEASE GIVE UP) 18:53 <@riker> hah your wiki page history is amusing ais 18:53 <@riker> including at least one troll & somebody reverting a link to your own work, by you, as vandalism? 18:53 which page is this? there's more than one page you could be referring to 18:53 so, say i want to create a new double damage artifact 18:54 <@riker> user:ais523 18:54 if i look at grayswandir in the source, I can't see where it gets its double damage from 18:54 [hdf-us] [nh370] Luxidream (Ran Orc Fem Cha) entered a shop, on T:1200 18:54 [hdf-us] [evil] dingotron (Con Hob Fem Cha) acquired the Bell of Opening, on T:22050 18:54 <@riker> porkman: PHYS(to-hit, dmg) is the syntax 18:54 <@riker> if dmg = 0, it's double damage 18:54 [hdf-us] [nh370] Luxidream (Ran Orc Fem Cha) entered the Gnomish Mines, on T:1347 18:54 <@riker> (or FIRE, or COLD, etc. kinds of damage) 18:55 how odd 18:55 @riker: on which wiki? Esolang? 18:55 but thanks 18:55 <@riker> wikipedia itself ais 18:55 ah right 18:55 well I used to be an admin there, so attracting trolls was quite likely 18:55 <@riker> porkman: if you don't want extra damage, just don't put any kind of damage bonus, is the idea I think 18:55 <@riker> as it stands, you can't put a to-hit but no damage bonus, but that's a non-issue I think 18:56 so, 5,0 is +5 to hit, double damage 18:56 but 5,10 is +5 to hit, +10 dmg? 18:56 <@riker> +d5 I think but yes 18:56 <@riker> er yeah +5 18:56 <@riker> I think to-hit is a constant but the damage is a rnd call 18:56 <@riker> but yes 18:56 in slex it's d5 and d10, that gets done somewhere else 18:56 the INTERCAL hello world example ends with PLEASE GIVE UP :') 18:56 i think 18:56 <@riker> that would be +5 to-hit, +d10 damage 18:56 @riker: in Slash'EM the damage is not an rnd call, it's flat (this is different from vanilla) 18:56 [hdf-us] [nh370] Luxidream (Ran Orc Fem Cha) reached Mine Town, on T:2014 18:56 <@riker> this is NOT true in slex 18:56 however I believe SLEX changed this back in at least some cases 18:56 <@riker> slex it's specifically changed back for damage 18:57 <@riker> dunno about to-hit 18:57 <{Demo}> slex nerfed it cause slex doesn't like fun 18:57 [hdf-us] [nh370] Luxidream (Ran Orc Fem Cha), 2221 points, T:2068, killed by a soldier ant 18:57 <@riker> demo: you shoudl contribute to porkhack if you have opinions on what woudl make slex more fun :) 18:57 <{Demo}> can't touch that 18:57 you're welcome to! 18:57 <@riker> also hey demo did you address the clusterfuck of compile warnings in notdnethack 18:57 <@riker> or nah 18:57 porkman: be aware that NetHack is a complete minefield of special cases, not only in the gameplay, but also in the source code, and most of them make no sense and are unguessable unless you know about them already 18:57 e.g. 0 in an object's stats often doesn't actually mean 0, it depends on what field you're talking about though 18:58 <{Demo}> no but I should 18:58 <{Demo}> -Wnoall fixes them, right? 18:58 I'd forgotten about the double artifact damag case 18:58 <@riker> demo: well, yes that would probably fix it 18:58 <{Demo}> ;P 18:58 <@riker> ais: I don't see a better solution for that particular special case 18:58 those warnings are there for a reason :-P 18:58 <@riker> possibly -1 but that would probably run into signed / unsigned issues 18:59 <{Demo}> yeah nah the ones I get when I compile are valid 18:59 @riker: SPFX_ flag might make sense 18:59 <{Demo}> like, yes, this should be changed 18:59 [hdf-us] [nh370] Luxidream (Ran Orc Fem Cha) entered the Gnomish Mines, on T:1404 18:59 [hdf-us] [nh370] hyperion67 (Arc Hum Mal Law), 1814 points, T:1903, killed by a crossbow bolt 18:59 <@riker> ais: does 3.6 make any major artilist.h changes, minus the decorative formatting? 19:00 [hdf-us] [nh370] Luxidream (Ran Orc Fem Cha), 903 points, T:1662, killed by a dwarf 19:00 @riker: 3.6.0 is the version I know least about, because I a) wasn't on the devteam at the time and b) wasn't paying attention 19:00 <{Demo}> a lot of the warnings are in code I don't support 19:00 so any questions about 3.6.0, or changes that would have happened alongside it, are best asked of someone else 19:00 <{Demo}> like curses 19:00 <@riker> well, I meant the change from 3.4.3 to 3.[67] 19:01 <@riker> but yeah i get that 19:01 <@riker> demo: yeah, but they're mostly easy to fix I think 19:01 right, but it's /changes/ I pay attention to 19:01 <@riker> I try to go through the warnings every version or two and fix most 19:01 so if something changed in the 3.4.3 to 3.6.0 transition I probably didn't notice 19:01 <@riker> ais: reasonable 19:01 could you add slow digestion to an artifact? 19:01 <@riker> demo: for example, chris added a new func to apply.c that's an int return type but has exactly 0 return statements anywhere 19:02 porkman: not easily, at least in the vanilla codebase artifacts don't support arbitrary trinsics 19:02 <{Demo}> a lot of them I didn't want to fix to make merging into upstream easier 19:02 SLEX might have changed this, it's the sort of game where it would make sense to do so 19:02 [hdf-us] [nh370] Luxidream (Ran Orc Fem Cha) entered the Gnomish Mines, on T:761 19:02 <{Demo}> incase our fixes conflicted 19:02 <{Demo}> that is no longer the plan 19:02 <@riker> lmao 19:02 <{Demo}> instead I am gonna rebase onto a future dnh version manually in the future in the future one day some day 19:02 NH4 changed this, incidentally 19:02 <{Demo}> coming soon 19:03 {Demo}: which variant do you develop? 19:03 <@riker> notdnethack 19:03 slex manages to add a lot more stuff though 19:03 <@riker> which is forked off like 2017 dnethack 19:03 <3 ndnh 19:03 <@riker> porkman: for a start check artifact.c 19:03 the trinsic code in vanilla is kind-of painful 19:03 <@riker> maybe that would be in there 19:04 i mean, you have artifacts in slex that do...anything 19:04 <@riker> yeah she probably changed it 19:04 set_artifact_intrisic in src/artifact.c looks promising 19:05 <@riker> not in slex 19:05 it's structured based on the spfx field, but the easiest way to make artifacts do anything would be to change that to look at the artifact type tag 19:05 so maybe slex has a more complex version 19:05 <@riker> ah nvm it is, i copied the typo 19:05 <@riker> :p 19:05 I mean, i suggested as a joke to make an artifact victorian underwear that would stop you from refusing seduction attacks, and she managed to code it 19:05 <@riker> bleh masks really wack sometimes 19:06 The King Of Porn 19:06 huh, not there 19:06 [hdf-us] [nh370] Luxidream (Ran Orc Fem Cha) reached Mine Town, on T:2520 19:06 [hdf-us] [nh370] Luxidream (Ran Orc Fem Cha) entered a temple, on T:2520 19:06 [hdf-us] [nh370] Luxidream (Ran Orc Fem Cha) had Stormbringer bestowed upon her by Mars, on T:2524 19:06 <@riker> ok porkman, you have a couple options for adding slow digestoin 19:06 porkman: it's actually easier to implement a special case like that than a trinsic 19:06 all you'd have to do would be to find the code for seduction attacks and add a special case to see if the artifact is equipped 19:06 <@riker> 1. (the hackiest but whatever) just check wherever slow digestion is called for hunger reductions (allmain.c probably) and make it check for the artifatc being worn/wielded/carried/whatever 19:07 <@riker> the downside of that is that it probably wouldn't affect the SLOW_DIGESTION check, so it would reduce hunger but if something else wants to check for slow digestion you'd need to add it there 19:07 @riker: more importantly it wouldn't show up on enlightenment 19:07 -!- noomy has quit [Quit: Bye!] 19:07 <@riker> have you seen slex's enlightenment 19:07 <@riker> but yeah that's true 19:07 that's why, when setting a trinsic, you normally want to actually set the trinsic rather than emulate its effect 19:07 [hdf-us] [nh370] Luxidream (Ran Orc Fem Cha) entered a shop, on T:2755 19:08 and the way vanilla (and probably slex) handles trinsics is /really/ confusing even to an experienced developer 19:08 hmm 19:08 [hdf-us] [nh370] Bliss56 (Val Dwa Fem Law) wished for "uncursed magic marker", on T:70091 19:08 <@riker> 2. slightly easier, add a SPFX flag to artifact.h called like SPFX_SLW_DGST 19:08 <@riker> not sure how full those bitfield things are though 19:08 <@riker> a quick glance looks like you can fit one more in 19:09 the extra SPFX flag would be the correct way to do it in vanilla 19:09 <@riker> then, you can add a check for that in set_artifact_intrinsic 19:09 but I can't imagine that doing that in slex would be correct because you'd run out basically instantly 19:09 unless, that is, Amy decided not to add any trinsics on artifacts other than the ones that already existed in Slash'EM 19:09 <@riker> I think she mostly didn't, and the special cases are non-trinsics 19:10 -!- moony has joined #hardfought 19:10 <@riker> there's also a lot of base item types that do stuff though (like dragon scale mail for basically everything) 19:10 <{Demo}> slex doesn't add a lot, it just looks like it does 19:10 ooh, two of the SPFX flags are marked "Amy" 19:10 <@riker> SPFX_AMY 19:10 and there is room for two more 19:10 <@riker> lowers sanity when carried 19:10 <@riker> > /* artifact never spawns again if it was already generated --Amy */ 19:10 [hdf-us] [nh370] Luxidream (Ran Orc Fem Cha) became literate by engraving "nx", on T:4400 19:10 <@riker> i'm concerned why that's a thing 19:10 porkman: btw, NetHack's license requires you to distinguish between changes you made and changes thaat other people made, which is why you often get variant devs signing their name on changes 19:11 [hdf-us] [nh370] Bliss56 (Val Dwa Fem Law) wished for "uncursed magic marker", on T:70187 19:11 but most of us have version control systems set up for that 19:11 <@riker> ais523: how strictly is that license enforced? 19:11 where people can get the old and new versions and compare them 19:11 @riker: it's kind-of confusing because many of the copyright holders aren't part of the devteam 19:11 <@riker> like, dnethack does that if you count checking git authors, but chris doesn't leave comments in the code normally 19:11 <@riker> i've commented some of the stuff i do, and nero's files don't even have the classic nethack header lol 19:12 I think we've decided that a VCS tracking the changes is reasonable 19:12 for NH4, I put a last-modified date and author on every file 19:12 <@riker> porkman: adding #define SPFX_SLW_DGST 0x40000000L should work there, I think 19:12 so it's impossible for a file to be changed without the change being credited 19:12 what's a VCS tracking? 19:12 <@riker> version control system = vcs 19:12 porkman: VCS = version control system, it's something that records all the changes you make, with reasons and timestamps, etc. 19:12 i see 19:12 <@riker> git is the main one used by most people i'd say 19:12 well, i have been signing changes anyway 19:12 it's very useful if you change your mind about a change, or need to know why you did something 19:13 <@riker> some IDEs have built-in ones, on a smaller scale 19:13 though how would I implement a VCS? 19:13 and almost all developers start using one when they get a little experience because they see how useful it is 19:13 you don't implement a VCS, you download a pre-existing one and use that 19:13 <@riker> LMAO amy signed chris's name here in slex's artifact.c 19:13 git is the most popular, but very hard to learn 19:13 Riker, most slex artifacts can generate an unlimited amount of times 19:13 <@riker> it's not that hard to get a hang off 19:13 I'm not convinced it /should/ have won the VCS wars 19:13 <@riker> porkman: yeah, I think tha'ts pretty funny 19:14 because she has effects that equip really nasty artis on you 19:14 and so if you have been slapped with nastyfoo already, you can still get slapped by it again! 19:14 <@riker> don't want those to run out of artis obviously 19:15 so where would i put that line, in artifact.h? 19:15 <@riker> anyway porkman, once the SPFX_ for slow digestion is added to include/artifact.h, you can use it in src/artifact.c and include/artilist.h 19:15 <@riker> see the top where it has other spfxs? 19:15 aha 19:15 <@riker> add it belong the onlyone 19:15 <@riker> *below 19:15 [hdf-us] [nh370] Luxidream (Ran Orc Fem Cha), 8082 points, T:6611, killed by a queen bee 19:15 <@riker> then go into src/artifact.c, probably control-f for SPFX_HPHDAM 19:15 porkman: technically it doesn't matter, #define directives can exist anywhere before they're used, but for code organisation purposes it's best to put it next to the other lines that are doing the same thing 19:16 @riker: is half phys a trinsic? 19:16 <@riker> it's in the function in slex at least 19:16 best to pick something that's a traditional trinsic, like searching or stealth 19:16 <@riker> and I mean it's SPFX_HPHDAM 19:16 <@riker> they're coded the same way? 19:16 ah, OK 19:16 probably it is a trinsic then 19:16 <@riker> it's just that 1/2 phys is the bottom one in the func 19:16 <{Demo}> who is KAA? 19:17 that number is already picked though 19:17 <@riker> what do you mean porkman 19:17 <@riker> what number? 19:17 oh no, it isnt 19:17 40000000L 19:17 <@riker> yeah, 2xxx should be 19:17 does that invoke slow digestion? 19:17 <@riker> no 19:17 <@riker> it does not 19:17 porkman: these are bitfields, so they consist of seven zeros, and one digit that's a 1, 2, 4 or 8 19:17 giving 32 possibilities 19:17 <@riker> that defines a flag for it to be invoked though 19:17 <@riker> go into artifact.c and ctrl-f for HPHDAM 19:18 porkman: so the idea is, you need to be able to explain to the game what the artifact does 19:18 <@riker> that should take you to the middle of a function with a bunch of if statements 19:18 yep 19:18 currently there's no way of expressing that, so adding a "this bitfield bit means slow digestion" concept means you have a way to store that information 19:18 but you also need to tell the game to use that information once it's stored 19:18 <@riker> ok copy the whole if statement for hphdam and duplicate it 19:18 <@riker> so it's just the same if statement, inside of it included 19:18 <@riker> from the first line to the ending brace } (curly bracket) 19:19 then you edit the statement so that it applies to slow digestion rather than half phys 19:19 <@riker> and change the SPFX_HPHDAM in the check to be SPFX_SLW_DGST (or whatever you named the spfx flag there) 19:19 <@riker> and then the EHalf_physical_whatever into ESlow_digestion 19:19 https://cdn.discordapp.com/attachments/400474774440247317/697948905253502976/Screen_Shot_2020-04-09_at_4.19.42_PM.png 19:20 -!- Chris_ANG has joined #hardfought 19:20 -!- mode/#hardfought [+v Chris_ANG] by ChanServ 19:20 <@riker> i hate worn masks lol 19:20 [hdf-us] [nh370] efot (Hea Gno Fem Neu) entered the Gnomish Mines, on T:2098 19:20 <@riker> i still need to get around to fixing evilhack's funky offhand masks 19:20 E = extrinsic, right? and H = intrinsic, on properties 19:20 <@riker> yes 19:20 I forget how it works because I rewrote it for NH4 19:20 <@riker> ais tell me about it, dnethack overhauls artifact.c almost entirely lol 19:20 and the extrinsic field works by having a separate bit for each item that might convey the property 19:21 <@riker> that sounds like a much better idea 19:21 aha 19:21 <@riker> if I remember i'll tell that to nero, he enjoys rewriting the nethack code from scratch 19:21 so now slow digestion is an intrinsic 19:21 <@riker> it was always an intrinsic 19:21 that can be given by an artifact 19:21 <@riker> now it can be given by artifacts 19:21 <@riker> RIP uhitm/mhitu/mhitm, long live xhity 19:22 porkman: that's the idea, you've changed the artifact code to be able to understand slow digestion artifacts 19:22 and now that the feature exists, you can use it to create a new artifact that has that property 19:22 <@riker> you should probably recompile now 19:22 <@riker> just run make install I think 19:22 <@riker> just to make sure you don't have any typos anywhere 19:22 right, the reason to recompile is to make sure that your new code makes sense 19:22 the compiler will probably complain if you've made a typo 19:22 <@riker> it should 19:23 in the early days of computers, it was usual to spend all day, by hand, double-checking a program to make sure there were no mistakes in it 19:23 <@riker> bleh ais do you know off the top of your head how important doapply's return value is 19:23 but modern compilers can do the same thing 19:23 @riker: at a guess it controls whether the application should take time or not 19:23 that's usually the case for do* functions 19:23 <@riker> thanks 19:23 e.g. if the user tries to apply something and needs an argument, then cancels out 19:24 that probably shouldn't take up in-game time 19:24 but obviously, if they go through with the command, it should 19:24 <@riker> > light_cocktail(obj); 19:24 <@riker> that's a fun function name 19:24 there are two separate ways to track that in vanilla because of course there are, but the return value from do* functions is probably more commonly used 19:24 (the other is flags.move) 19:24 [hdf-us] [nh370] eraserhead97 (Cav Hum Fem Neu) acquired the Mines' End luckstone, on T:8687 19:25 <@riker> ais: that's probably going to be refactored for this code soon anyway lol 19:25 IIRC, the relationship is that setting flags.move to 0 means that the turn will not take up time no matter what, whereas returning 0 means that this command didn't take time, but allows for the possibility that something else that's happening might 19:25 <@riker> dnh abuses the movement code heavily 19:26 in general it's a bad idea to have two different ways to do the same thing because it's really confusing to work with, but that doesn't seem to have stopped NetHack, so… 19:26 also, flags is a bizarre place to put that, given that it's normally used for handling saved options (like autopickup) 19:26 <@riker> my policy is that if it's been relatively unchanged since the 80s, it's a black box for me 19:27 unfortunately, sometimes you have to understand this code to be able to interact with it 19:27 at the very least, a comprehensive understanding of what nomul() does is typically important 19:27 Next, i gotta find where Amy kept the list of descriptions for artifacts 19:27 <@riker> fair 19:28 -!- mobileuser has quit [Quit: Connection closed] 19:28 <@riker> porkman: do you have an example description? 19:28 <@riker> since you can search the code if you have a snippet of one, to see where that's from 19:28 porkman: easiest way to do that is to look up the description of something in-game, then search the source code for the same string 19:28 <@riker> ^ 19:28 this is the reason we ask people for the exact error message when something goes wrong, it's so we can search for it in the source code 19:29 (if the reported error message is even one letter off, it's much harder to search for) 19:29 although NH4's impossible/panic dialog displays the source code location itself so that we don't have to do the search 19:29 <@riker> i've also had fun times remembering to exclude things like "You" from the search 19:29 <@riker> or "The" 19:30 <@riker> just searching for the most distinct subsection, due to string formatting & funky pline functions 19:30 [hdf-us] [nh370] Frozty (Val Dwa Fem Law) killed Lord Surtur, on T:22481 19:30 I'm pretty good at leaving off "You hear" and friends 19:30 the %s placeholders still often catch me out though 19:31 <{Demo}> ceiling and floor descriptors 19:31 <{Demo}> limbs 19:31 <{Demo}> err, body parts 19:31 <@riker> why is there a special type "aligntyp" that's just schar 19:31 <@riker> ??? 19:31 for example, "Hallucination resistance when wielded" 19:31 [hdf-us] [nh370] efot (Hea Gno Fem Neu) reached Mine Town, on T:3202 19:31 though on the source, if i just search github, it won'tshow up 19:32 <@riker> src/invent.c porkman 19:32 @riker: probably to make the code in other parts more readable, if you see an aligntyp you know it refers to an alignment, rather than just that it's a signed character 19:32 it could also be for futureproofing if the game ever has more than 127 alignments, but that seems unlikely even in SLEX 19:32 Riker, what'd you use to find it? 19:32 <@riker> considering invent.c is 18k lines, possibly github just says no 19:32 (that said, amy has changed a number of types to make them larger) 19:32 <@riker> mine is ag but effectively just grep 19:32 [hdf-us] [nh370] Frozty (Val Dwa Fem Law) acquired the Bell of Opening, on T:22497 19:32 <@riker> (I think the package is the-silver-searcher, it's a bit easier to use than grep and it's a bit quicker) 19:33 <@riker> but it's just a command line search tool, for files inside folders 19:33 @riker: try ripgrep 19:33 <@riker> grepping for "when wielded" -> invent.c has them all MANUALLY defined 19:33 the command line search tool wars were very productive 19:33 <@riker> for EVERY SINGLE ARTIFACT 19:33 <@riker> which is horrifying 19:33 @riker: haha 19:33 I hope Amy at least autogenerated that 19:33 but who knows, maybe she typed it out 19:33 <@riker> it's literally 3k lines 19:33 <@riker> of artifacts only 19:33 <@riker> not even items 19:34 at least it'd be a neat mini-spoiler 19:34 https://cdn.discordapp.com/attachments/400474774440247317/697952520756002867/Screen_Shot_2020-04-09_at_4.34.04_PM.png 19:34 <@riker> what the hell 19:34 <@riker> that's over 10k lines of that function 19:34 gah, I'm trying to remember what font that is 19:34 I know I recognise it 19:34 so, notepad++ highlights the same word in lines next to it 19:34 when you select a word 19:34 is there a way to make it change them all in one shot? 19:35 almost certainly, but I don't know what it is 19:35 [hdf-us] [nh370] nabru (Val Dwa Fem Law), 1654 points, T:2219, killed by a winter wolf cub 19:35 @riker: wow, that's not even what the itemactions function is for, either 19:36 that worries me even more than a function that's 10596 lines long 19:37 She said she typed it out 19:37 many months ago, I decided I would stop paying attention to slex development; it was a good decision, I didn't have to worry or think about all the insane development practices 19:38 now, out of kindness, I turn up to help a new variant developer, but I may have to give up because I'm learning far too much to be healthy about slex at the same time 19:39 <@riker> ais: want the answer (what font)? 19:39 @riker: yes 19:39 <@riker> inconsolata 19:39 oh right, makes sense 19:40 I was close but wouldn't have guessed it 19:40 <@riker> based off consolas i believe 19:40 "that looks kind-of like consolas but…" is about as close as I got 19:40 but then went off in the wrong direction 19:40 -!- eLtMosen1 has quit [Ping timeout: 260 seconds] 19:40 <@riker> ^ it's close, it's a 2006-ish variant I think 19:41 <@riker> at least 2006 sounds familiar + inconsolata 19:41 <@riker> anyway porkman to add a description, copy the "case ART_XXX: " and the following line, duplicate it 19:41 <@riker> change the ART_XXX to be the name of your artifact + rewrite the description 19:42 [hdf-us] [nh370] sawbones (Hea Gno Mal Neu), 6442 points, T:7343, killed by a blast of frost 19:43 [hdf-us] [nh370] qt (Arc Hum Fem Neu) entered the Gnomish Mines, on T:1599 19:43 <@riker> why does nethack insist on K&R function declarations? 19:43 [hdf-us] [nh370] qt (Arc Hum Fem Neu), 408 points, T:1613, killed by a gnome lord 19:44 <@riker> + definitions 19:44 I don't know that it _insists_ on it, but it was written before ANSI. 19:44 <@riker> i feel like it would've been updated if it wasn't for an internal style guide saying don't do it 19:46 at least one devteam member prefers NetHack to be compilable on a K&R compiler 19:46 <@riker> i guess i'll just leave all the annoying type promotion warnings alone then :/ 19:47 note that many other devteam members think that this restriction is likely not helpful nowadays 19:48 I personally think that any system that doesn't have at least a C89 compiler is unlikely to have enough memory to run NetHack 19:48 <@riker> how many devteam members are there currently? 19:48 although would be intrigued to be proven wrong 19:48 <@riker> well, any normal system 19:48 @riker: that's hard to answer, the number of members is much higher than the number of /active/ members 19:48 <@riker> i'm certain i can write a funky operating system with enough memory but no c89 compiler 19:48 there are lots of people who are technically part of the devteam and never do anything 19:48 then there are people who are part of the devteam and /rarely/ do anything, but chip in every now and then (e.g. me) 19:48 and then there are levels of activity above that 19:49 <@riker> I discovered a while ago spidermonkey is enough to run an OS off, minus the lowest level things 19:49 @riker: but would it have a K&R C compiler? 19:49 nowadays you'd target C89 as your first C variant 19:49 <@riker> ais: good question 19:49 or even C99/C11 19:49 Recompiling! 19:49 [hdf-us] [nh370] qt (Arc Hum Fem Neu) averted death, on T:1655 19:50 [hdf-us] [sp] dgoddard (Mon Hum Mal Cha) had a a scale mail bestowed upon him by Huan Ti, on T:3995 19:50 [hdf-us] [nh370] qt (Arc Hum Fem Neu) entered the Gnomish Mines, on T:1673 19:50 [hdf-us] [nh370] qt (Arc Hum Fem Neu), 592 points, T:1675, killed by a wand 19:50 -!- ais523 has quit [Quit: sorry for my connection] 19:50 -!- ais523 has joined #hardfought 19:50 -!- mode/#hardfought [+v ais523] by ChanServ 19:52 [hdf-us] [sp] dgoddard (Mon Hum Mal Cha), 14461 points, T:4240, killed by a watchman 19:54 porkman: once your compile finishes, I recommend using wizmode to wish for your artifact, and enlightenment as the easiest way to see if the slow digestion extrinsic got set 19:54 [hdf-us] [nh370] sawbones (Hea Gno Fem Neu) entered the Gnomish Mines, on T:1183 19:54 it's also worth noting that if your artifact doesn't naturally cause ring/amulet hunger, then it will turn hungering off entirely if equipped, although that might not be a problem 19:54 -!- Umbire has joined #hardfought 19:55 (slow digestion works by eliminating natural hungering, but the ring of slow digestion still causes ring hunger, the net effect is a huge reduction but not elimination of hungering) 19:55 (but if your artifact isn't a ring or amulet then the ring hunger won't apply) 19:59 [hdf-us] [nh370] qt (Arc Hum Fem Law), 1703 points, T:2795, killed by a bugbear 20:01 come think of it, Amy has made a helm of no digestion 20:01 which gives slow dig 20:01 on a hat 20:01 or rather, no dig, because it doesn't consume the ring hunger 20:02 well, presumably your artifact will have better stats 20:02 it's still compiling, no faster than with the -B it seems 20:03 <@riker> once you use -B any more makes will do the whole thing again as well, I believe 20:03 HELM("helm of no digestion", "full helmet", 20:03 no, the problem is that porkman changed a .h file 20:04 [hdf-us] [nh370] qt (Arc Hum Fem Law), 1179 points, T:1193, killed by a water demon 20:04 so she must have defined SLOW_DIGESTION somewhere 20:04 which legitimately forces most of NetHack to recompile because the depenencies are too meshes in with eacah other 20:04 porkman: that's a base item 20:04 those get one trinsic for free 20:04 oooh 20:05 giving them more than one is a complete nightmare (the alchemy smock in vanilla is the only base item with two trinsics and it's a mess of special cases) 20:05 * in vanilla, the alchemy smock… 20:05 is it possible to code a weapon to do +dmg only to a specific monster? 20:06 say, only shopkeepers, but not all @ 20:06 <@riker> yes 20:06 <@riker> that requires some funky code though, but it should be doable 20:06 [hdf-us] [nh370] BLARGO (Val Hum Mal Law) killed Medusa, on T:43901 20:06 <@riker> i believe you'd make it hate @ and then have a special check for isshk in the artifact_hit function, or something like that 20:06 I'm going to create another artifact, a "filthy shopkeepers must die", a non-polearm randart 20:07 [hdf-us] [nh370] eraserhead97 (Cav Hum Fem Neu) entered Sokoban, on T:10509 20:07 chaotic intelligent artifact 20:07 porkman: does your slow digestion artifact work? and what are its stats apart from slow digestion? 20:07 (or is it still compiling? :-D) 20:07 <@riker> sigh, "-Wno-knr-promoted-parameter": there is literally no reason for this to be NOT disabling -Wknr-promoted-parameter, but guess what - it doesn't work 20:07 [hdf-us] [nh370] qt (Arc Hum Fem Law) murdered Smut, her faithful kitten, on T:804 20:07 @riker: maybe it's turned back on later on the command line 20:07 <@riker> for some godforsaken reason i can't turn this warning off without turning off warnings in general 20:07 try moving it to the end 20:08 <@riker> ais: i'm checking the makefile, it should be fine? 20:08 if it's not that, it's probably a bug 20:08 it's a double damage heavy silver ball with slow digestion 20:08 <@riker> it's definitely a bug ais, considering the make output tells me it's compiling with that and i'ts not 20:08 @riker: well, CFLAGS isn't the only thing that determines the command line; although it's the only place you're meant to put warning flags it's possible that someone manually edited warning flags onto the build commands for some reason 20:08 <@riker> cc -Wno-knr-promoted-parameter util/makedefs.o src/alloc.o src/monst.o src/objects.o util/panic.o -o util/makedefs 20:09 OK, that seems fairly comprehensive 20:09 you could try "gcc" rather than just "cc" to make sure you weren't hitting a wrapper script 20:09 but it's likely just a bug at this point 20:09 [hdf-us] [nh370] qt (Arc Hum Fem Law) entered a shop, on T:1308 20:10 <@riker> it's probably a my-computer bug 20:10 <@riker> since my mac has some things installed it really shouldn't 20:10 [hdf-us] [nh370] qt (Arc Hum Fem Law) entered the Gnomish Mines, on T:1469 20:10 what compiler is this, anyway? I can't find that warning in the gcc docs 20:10 nope, i broke it 20:10 <@riker> clang ais 20:10 <@riker> my mac doesn't run gcc, it uninstalls when i reboot the computer 20:10 Configuration incompatibility for file "dungeon". Oops... Hit to continue. 20:10 <@riker> gcc is aliased to run clang for some godforsaken reason 20:10 <@riker> porkman: what did you touhc 20:11 porkman: that's likely caused by a lack of -B 20:11 the level files and the game engine disagree on how many artifacts there are 20:11 and that difference is enough to cause the wrong artifacts to generate 20:11 <@riker> ais: it's listed as a diagnostic in clang's official docs, and it doesn't like trying to disable it for some reason 20:11 Alright, i'll do it again 20:11 ah, OK, clang 20:11 <@riker> also out of curiousity ais: is the i in ais a middle initial? 20:11 I'm less of a clang fan than most people, although I normally assume it'll work in cases like that 20:12 <@riker> I am not a clang fan either :) 20:12 @riker: yes, the whole ais523 username was autogenerated by a computer 20:12 Discordians apparently find this hilarious 20:12 <@riker> oh huh 20:12 [hdf-us] [nh370] Grandolar (Wiz Gno Mal Neu), 213 points, T:772, killed by a boulder 20:12 <@riker> which one? 20:12 <@riker> like, which computer generated it and why? 20:13 <@riker> i'm assuming a uni username or something? 20:13 yes, university username 20:13 <@riker> aha! i found the problem, it was mine 20:13 <@riker> i edited one set of flags but not the other 20:13 <@riker> since there are 2 for some reason 20:13 then why did the make verbose output print your new warning flag? 20:13 <@riker> good question 20:13 <@riker> no clue 20:14 <@riker> i'm willing to ignore that for now since I got what I wanted 20:15 [hdf-us] [nh370] qt (Arc Hum Fem Law), 892 points, T:2190, killed by a rock mole 20:17 [hdf-us] [nh370] qt (Arc Hum Fem Law) entered a shop, on T:378 20:21 <@Suzune Takada> I think "tricky" eating races need ring of slow digestion 20:21 <@Suzune Takada> like ghoul/vampire in spl 20:21 clang's documentation of warnings is lacking compared to gcc's 20:21 unless there's better documentation somehow and I simply haven't found it 20:22 [hdf-us] [dnh] NeroOneTrueKing (Rog Inc Fem Cha) was given Stormbringer, on T:5819 20:22 still recompiling 20:22 <@riker> ais: I found a TODO at one point, so it's not great 20:22 porkman: it does that 20:22 <@riker> todo for the entire "warnings on the command line" section lol 20:23 gcc's list of warnings explains what they're all for 20:23 i need a powerhouse desktop lol 20:23 https://cdn.discordapp.com/attachments/400474774440247317/697964838382600252/Screen_Shot_2020-04-09_at_4.56.01_PM.png 20:23 clang's just tells you what the flag is and what the error message is, but doesn't have extended discussion 20:23 <@riker> it has one of those, just not for enabling/disabling via cli easily 20:23 ah, I'm looking at https://clang.llvm.org/docs/DiagnosticsReference.html 20:23 is there a better page to look at? I'm curious about what sort of warnings clang supports 20:23 the compiler just goes into a mind melt every now and then 20:24 porkman: that is mostly a slex thing I think 20:25 hyperrogue does that too, but that's because the entire game is, from the compiler's point of view, one file 20:25 (it's shipped as lots of separate files but they're combined into one before compiling) 20:25 the all-of-hyperrogue file is probably a comparable size to some of slex's 20:25 what was the chcp you said, 437> 20:26 ? 20:26 that seems to be the one that the slex .bat sets itself to 20:26 <@riker> that's what I found too ais 20:26 porkman: 437 is standard for roguelikes 20:27 there are a lot of arbitrary codepage numbers but that one has been memorized by most roguelike devs 20:27 it emulates an old-fashioned VGA PC, which is where it became standard I think 20:28 (those PCs might have had multiple codepages too, I think, not sure on that; but even if they did have more than one, 437 was the default) 20:29 <@riker> int n = wdie->damn; for (n; n; n--) 20:29 Code page 437 is the character set of the original IBM PC (personal computer). 20:29 hmm 20:29 the compile failed again 20:29 <@riker> why even bother with the n then? not like it's used elsewhere 20:29 the first n does nothing 20:30 <@riker> porkman: what exactly did you change? 20:30 the second and third are useful though 20:30 <@riker> ais: yes, but why not move the first line into the for loop? 20:30 should be "for (; n; n--)" if you want to use n outside the loop 20:30 A few item stats and rarity, made the new artifact 20:30 @riker: in C89 you can't create a variable for just one for loop 20:30 if it's used in the control test 20:30 <@riker> oh, really? 20:30 anyway, i'm going to go sleep, i'll figure it out tomorrow 20:30 <@riker> that's really odd 20:30 thanks for the help! 20:30 so the correct layout would be "int n; for (n = wdie->damn; n; n--)" 20:30 <@riker> anyway this is not compiling on c89 anyway editing it 20:30 <@riker> night porkman 20:31 "for (int n=…" is a C99 thing 20:31 <@riker> that did NOT send via discord right 20:32 <@riker> "int n=…" 20:32 <@riker> but yeah I get the idea 20:34 [hdf-us] [nh370] eraserhead97 (Cav Hum Fem Neu) completed Sokoban, on T:12653 20:34 huh, is Discord broken on non-ASCII characters generally? 20:34 let me try a smiley face: ☺ 20:34 did that come through OK? 20:35 appears ok here 20:35 <{Demo}> appears ok here 20:36 nabru: {Demo}: you're replying on IRC, not Discord :-P 20:36 [hdf-us] [nh370] Kevkat (Hea Gno Mal Neu) bribed Asmodeus with 3 zorkmids for safe passage, on T:31590 20:36 <{Demo}> or so it appears to you 20:36 if it was via discord, the message would be relayed by rld 20:37 <{Demo}> https://i.imgur.com/A5aeGYQ.png 20:37 * if it were 20:37 <{Demo}> nope 20:37 <{Demo}> i've had my own private bridge long before rld was a thing 20:37 <@riker> yes it does ais 20:37 …that's the wrong smiley face character 20:37 <@riker> but yeah demo has a really ghetto setup 20:37 -!- tacco\unfoog has quit [] 20:37 <@riker> it becomes :relaxed : 20:38 <@riker> ☺️ 20:38 hmm, the bridge must be translating it 20:38 both ways 20:38 <@riker> rld is wack 20:38 Unicode on one side, emoji code on the other 20:38 what about e-acute? é 20:38 <@riker> demo why tf does my name ahve unicode boxes in your ghetto bridge 20:38 <@riker> fine 20:38 <@riker> é sending one back 20:38 yep, fine back as well 20:38 strange that it failed on … then 20:39 <@riker> i think it's just emojis -> discord emojis 20:39 <@riker> since that's a particular discord emoji, and it probably translates unicode emojis into those 20:39 @riker: the boxes around your name appear to be incorrectly translated IRC color codes 20:39 rld gives people random nick colors 20:39 <@riker> ah 20:39 just like most IRC clients do for IRC users 20:40 <@riker> bleh this makefile has some funky unicode character that's rendering as an error code for me 20:41 <@riker> I think i'ts a carriage return but i'm not 100% 20:41 <@riker> aha it is 20:42 CR should never have been part of files for computers 20:42 <{Demo}> thanks 20:42 <{Demo}> carriage returns piss me off a lot 20:42 <@riker> util/lev_comp.y:1727 20:42 <{Demo}> if i had a scroll of genocide 20:43 <@riker> why does that have so many but none of the other util files 20:43 the only reason CR and LF were split in the first place was that some printers couldn't do a carriage return fast enough, so making the newline character two bytes wide gave them a little extra time 20:43 (to catch up on the CR by stealing some time from the LF) 20:43 <@riker> I understand why but with all due respect please get them out of my 2020 code thanks 20:43 although even that seems dubious because a CR should take well over twice as long as most characters to print 20:44 <@riker> 45548 extra CRs, that's too many 20:44 [hdf-us] [nh370] eraserhead97 (Cav Hum Fem Neu) entered the Bigroom, on T:13485 20:44 ah, Wikipedia explains it a little differently: CR without LF is occasionally useful, and because CR is so slow, splitting CR from LF didn't slow the printers down 20:45 <@riker> as a comparison from dnao, evilhack has 1555 20:45 <@riker> it has the side effect - printing that many CRs in my terminal somehow locked half the lines to 80x24, despite a current terminal size of 190x60 20:45 <@riker> that's pretty funny 20:46 I just checked vanilla: the only CRs are found in .vcxproj files, where they're probably part of the file format 20:46 <{Demo}> alright, CR genocide time 20:46 <@riker> ^ same with evilhack, which is my vanilla baseline for most things 20:46 <@riker> they're all nero's fault demo 20:46 <@riker> xhity has 17k, which is probably 1 per line lol 20:47 <@riker> the only ones that I don't think are his are lev_comp.y (unless he touched that with the room generation changes he did a while back) and gnumakefile 20:48 huh, so I was reading about the history of newlines; apparently the newline was invented in the 1800s and encoded in Morse code as dot-dash-dot-dash 20:49 then the CR/LF nonsense came later 20:50 [hdf-us] [nh370] zoya (Wiz Elf Mal Cha) entered a shop, on T:530 20:50 I guess Morse code operators also spend some time to do a carriage return but it's less than one character long, and/or they have a good enough memory to catch up 20:50 [hdf-us] [dnh] NeroOneTrueKing (Rog Inc Fem Cha) completed Sokoban, on T:10405 20:50 [hdf-us] [nh370] qt (Arc Hum Fem Law), 1450 points, T:2516, killed by a bugbear 20:51 [hdf-us] [xnh] bouquet (Cav Gno Fem Neu) wished for "blessed fireproof +2 speed boots", on T:76309 20:54 [hdf-us] [nh370] qt (Arc Hum Fem Law) entered a shop, on T:711 20:56 getting rid of my iron ball: good. not having a weapon anymore: bad 20:58 [hdf-us] [nh370] nabru (Val Dwa Fem Law) entered a shop, on T:247 20:59 [hdf-us] [nh370] qt (Arc Hum Fem Law) entered the Gnomish Mines, on T:1563 21:00 [hdf-us] [nh370] Vigorix (Tou Hum Mal Neu), 10385 points, T:9545, killed by a wolf 21:02 -!- ais523 has quit [Quit: quit] 21:02 [hdf-us] [nh370] nabru (Val Dwa Fem Law) entered the Gnomish Mines, on T:976 21:03 [hdf-us] [nh370] zoya (Wiz Elf Mal Cha) entered the Gnomish Mines, on T:2620 21:04 <@Pacra> @dk657 what did you do with it 21:04 [hdf-us] [nh370] nabru (Val Dwa Fem Law), 3 points, T:14, killed by a water moccasin called Jairzinho 21:05 [hdf-us] [nh370] zoya (Wiz Elf Mal Cha), 925 points, T:2730, killed by a bolt of cold 21:05 nymph 21:06 [hdf-us] [dnh] NeroOneTrueKing (Rog Inc Fem Cha) killed Sir Garland, on T:12597 21:06 i still don't have a weapon... 21:06 i've been punching things and everything is scary now 21:07 i have a club but i'm afraid it's cursed and it wouldn't take much to be as bad as bare-handed 21:07 <@riker> kick things? 21:07 [hdf-us] [nh370] qt (Arc Hum Fem Law) destroyed qt's ghost, the former Field Worker, on T:1877 21:08 [hdf-us] [nh370] nabru (Val Dwa Mal Law) entered the Gnomish Mines, on T:594 21:09 [hdf-us] [nh370] nabru (Val Dwa Mal Law) reached Mine Town, on T:676 21:09 <@microlance> now im held hostage by a sadistic shopkeeper again, i tested a cloak of invis that was cursed and now he wont let me out. instead he will take great pleasure in watching me starve to death 21:09 rld: i think kicking is a little worse than basic bare-handed 21:09 [hdf-us] [nh370] qt (Arc Hum Fem Law) destroyed hyperion67's ghost, the former Field Worker, on T:2136 21:10 riker i mean 21:10 <@microlance> legalize shopkeeper murder now! 21:12 [hdf-us] [nh370] nabru (Val Dwa Mal Law) destroyed qt's ghost, the former Field Worker, on T:987 21:12 lol random wear testing 21:12 Umbire: Message from LarienTelrunya at 2020-04-09 04:29 EDT: you got it wrong, the real benefit of backwards long jumping is that it lets you get into Moloch's sanctum without having to collect the invocation tools :D 21:12 <@riker> dk: it doesn't have an awful to-hit I think though 21:13 [hdf-us] [dnh] NeroOneTrueKing (Rog Inc Fem Cha), 80098 points, T:13585, killed by a shark 21:14 [hdf-us] [nh370] nabru (Val Dwa Mal Law) destroyed microlance's ghost, the former Rhizotomist, on T:1229 21:14 <@microlance> i got levelteleported into this gigantic antique shop after taking a few steps, so ofc i had to to wear test 21:14 <@riker> remove level teleporters today 21:14 hmm 21:14 <@riker> !lastgame NeroOneTrueKing 21:14 @riker: [hdf-us] https://www.hardfought.org/userdata/N/NeroOneTrueKing/dnethack/dumplog/1586365956.dnh.txt 21:15 <@riker> sharks are surprisingly lethal sometimes 21:15 [hdf-us] [nh370] qt (Arc Hum Fem Law), 2519 points, T:2585, killed by a gnome 21:16 [hdf-us] [nh370] qt (Arc Hum Fem Law) entered a shop, on T:443 21:21 [hdf-us] [nh370] qt (Arc Hum Fem Law), 829 points, T:1236, killed by a werejackal 21:22 [hdf-us] [nh370] nabru (Val Dwa Mal Law), 1716 points, T:2293, killed by a hill orc called Uhai-aibane of Thosoggris 21:22 <@RenanBirck> !lastgame nabru 21:22 @RenanBirck: [hdf-us] https://www.hardfought.org/userdata/n/nabru/nethack/dumplog/1586480676.nh.html 21:23 [hdf-us] [nh370] qt (Arc Hum Fem Law) entered a shop, on T:435 21:27 [hdf-us] [evil] k2 (Val Gia Fem Law), 18104 points, T:12228, killed by a horse 21:27 -!- tonehack has quit [Quit: Leaving] 21:28 [hdf-us] [nh370] qt (Arc Hum Fem Law) entered the Gnomish Mines, on T:1502 21:29 oof, trapped by a large mimic with no weapons and AC 9 21:29 i guess i'll don an unidentified amulet and start reading unidentified scrolls 21:29 [hdf-us] [nh370] BLARGO (Val Hum Mal Law) wished for "magic marker", on T:45302 21:30 [hdf-us] [nh370] qt (Arc Hum Fem Law), 1091 points, T:1813, killed by a giant bat 21:30 ah got away 21:30 thank the lord 21:31 [hdf-us] [nh370] qt (Arc Hum Fem Law) entered the Gnomish Mines, on T:405 21:32 [hdf-us] [nh370] qt (Arc Hum Fem Law), 327 points, T:531, killed by a kitten called Maunz 21:35 thank the gnome lord 21:35 [hdf-us] [nh370] qt (Arc Hum Fem Law), 188 points, T:729, killed by a boulder 21:38 <@Pacra> @riker how did he die to a shark 21:38 <@Pacra> are dnh sharks all great whites 21:39 [hdf-us] [nh370] qt (Arc Hum Fem Law) entered a shop, on T:632 21:40 [hdf-us] [evil] davidkent (Con Hum Fem Cha) hit with a wielded weapon for the first time, on T:5098 21:43 [hdf-us] [nh370] qt (Arc Hum Fem Law) entered the Gnomish Mines, on T:1167 21:44 <@riker> pacra: no but 5d6 isn't inconsiderable 21:44 <@Pacra> true 21:45 <@Pacra> probably thought they had another round of combat in them before disengaging 21:46 <@Pacra> !lastgame krm26 21:46 @Pacra: [hdf-us] https://www.hardfought.org/userdata/k/krm26/evilhack/dumplog/1585835357.evil.html 21:47 <@riker> it may have been suicide as well 21:47 <@riker> i can't tell 21:49 [hdf-us] [nh370] dbanethacker (Val Hum Fem Neu) killed the invisible Lord Surtur, on T:32970 21:49 [hdf-us] [nh370] dbanethacker (Val Hum Fem Neu) acquired the Bell of Opening, on T:32971 21:52 finally found a short sword 21:58 [hdf-us] [nh370] qt (Arc Hum Fem Law), 2314 points, T:2674, killed by a straw golem 21:59 [hdf-us] [evil] davidkent (Con Hum Fem Cha) became literate by engraving "hi", on T:6319 21:59 hi 21:59 Hello K2, Welcome to #hardfought 21:59 <@riker> o/ 21:59 [hdf-us] [nh370] BLARGO (Val Hum Mal Law) wished for "2 blessed scrolls of genocide", on T:45436 21:59 :) 21:59 [hdf-us] [nh370] qt (Val Hum Fem Law) entered a shop, on T:331 22:05 [hdf-us] [nh370] BLARGO (Val Hum Mal Law) performed his first genocide (class L), on T:45438 22:06 [hdf-us] [nh370] qt (Val Hum Fem Law) entered the Gnomish Mines, on T:1617 22:10 [hdf-us] [evil] davidkent (Con Hum Fem Cha), 6053 points, T:7063, killed by a gnome lord 22:10 [hdf-us] [nh370] qt (Val Hum Fem Law) reached Mine Town, on T:2252 22:10 [hdf-us] [nh370] qt (Val Hum Fem Law) entered a temple, on T:2297 22:12 [hdf-us] [nh370] Frozty (Val Dwa Fem Law) entered Gehennom, on T:23029 22:17 [hdf-us] [nh370] qt (Val Hum Fem Law) read a Discworld novel, on T:2932 22:20 [hdf-us] [nh370] qt (Val Hum Fem Law), 3832 points, T:3678, killed by a giant spider 22:20 [hdf-us] [nh370] qt (Arc Hum Fem Law), 54 points, T:88, killed by a little dart 22:24 -!- elenmirie_ has quit [Quit: Going offline, see ya! (www.adiirc.com)] 22:25 [hdf-us] [nh370] qt (Arc Hum Fem Law), 385 points, T:971, killed by a black pudding 22:27 [hdf-us] [nh370] qt (Arc Hum Fem Law), 514 points, T:643, killed by a goblin 22:29 [hdf-us] [nh370] Kevkat (Hea Gno Mal Neu) performed his first genocide (class m), on T:32647 22:31 if I fall into a pit that contains a c corpse will I turn to stone? 22:31 only one way to find out! 22:31 <{Demo}> I think so 22:31 <{Demo}> I'm always careful with ground slam 22:32 <{Demo}> as to avoid that 22:36 [hdf-us] [xnh] bouquet (Cav Gno Fem Neu) killed the Wizard of Yendor on Dlvl:40, on T:79320 22:36 [hdf-us] [xnh] bouquet (Cav Gno Fem Neu) acquired the Book of the Dead, on T:79322 22:42 [hdf-us] [xnh] bouquet (Cav Gno Fem Neu) performed the invocation, on T:80055 22:43 [hdf-us] [nh370] Kevkat (Hea Gno Mal Neu) killed Baalzebub, on T:33025 22:46 [hdf-us] [nh370] Vigorix (Cav Hum Fem Neu) entered a shop, on T:1250 22:50 [hdf-us] [nh370] Vigorix (Cav Hum Fem Neu) entered the Gnomish Mines, on T:1470 22:51 -!- mode/Beholder [+Zi] by Beholder 22:51 -!- cherryh.freenode.net changed the topic on #hardfought to: Welcome to Hardfought - https://www.hardfought.org/ ¤ public irc logs available on website ¤ ssh nethack@hardfought.org, nethack@eu.hardfought.org or nethack@au.hardfought.org ¤ EvilHack 0.5.0 has been released! 22:51 -!- mode/#hardfought [+v Beholder] by ChanServ 22:52 !who 22:52 K2: [hdf-us] eraserhead97 [nh370] Kevkat [nh370] dbanethacker [nh370] Vigorix [nh370] BLARGO [nh370] SuzuneTakada [spl] bouquet [xnh] omghax [xnh] | [hdf-eu] microlance [nh370] | [hdf-au] No current players 22:52 !whereis omghax 22:52 K2: [hdf-us] omghax [xnh]: (Tou Hum Fem Neu) T:1 The Dungeons of Doom level: 1 22:52 [hdf-us] [xnh] omghax (Tou Hum Fem Neu), 0 points, T:6, killed by a falling object 22:53 [hdf-us] [nh370] BLARGO (Val Hum Mal Law) wished for "2 blessed wands of death", on T:46608 22:54 2 blessed wands of death? don't think you're getting that bucko 22:55 <{Demo}> worth a try 22:55 [hdf-us] [nh370] Vigorix (Cav Hum Fem Neu) entered Sokoban, on T:2461 22:56 [hdf-us] [spl] SuzuneTakada (Con Dwa Nbn Cha), 315 points, T:1613, cut down by a dwarf zombie 22:59 anyone else supposed to get 3-6" of snow tonight/tomorrow? 22:59 heh 22:59 <{Demo}> what 22:59 <{Demo}> wild 22:59 sort of 22:59 not uncommon to see snow as late as May here 22:59 <{Demo}> I forgot winter existed already 22:59 where K2? 22:59 vermont, usa 23:00 Helsinki got its record snow cover like a week ago 23:00 for this winter 23:00 (then again, they had 0 permanent snow until late feb, I think) 23:00 when we moved here back in 2009, winter lasted longer 23:00 i remember the mothers day blizzard of 2010 23:00 4+ feet of snow on mothers day 23:00 may something 23:01 Muad no perm snow until february? ugh 23:01 we had that by november last year 23:01 yeah, it was the first winterless year i've ever seen 23:01 :( 23:02 We got a drizzly morning followed by the first thunderstorm of the year, with violent rain and wind, then it all cleared up and went sunny and gusty. Today was weird. 23:02 we had a little bit here but I'm 30km north from Helsinki, and the coast 23:02 ah 23:02 we had a good rainstorm today too, kinda surprised to see we're under a winter weather advisory 23:02 i doubt this snow lasts long 23:03 snow plow drivers will be excited 23:03 guess we'll hold off planting the new garden until next weekend maybe 23:03 heh yeah 23:05 [hdf-us] [spl] SuzuneTakada (Con Dwa Nbn Cha), 807 points, T:766, killed by a gnome lord 23:09 [hdf-us] [xnh] bouquet (Cav Gno Fem Neu) wished for "uncursed magic marker named clever name goes here", on T:82320 23:10 [hdf-us] [xnh] bouquet (Cav Gno Fem Neu) wished for "uncursed magic marker named it's pronounced PEN... ISLAND", on T:82331 23:11 [hdf-us] [nh370] BLARGO (Val Hum Mal Law) entered Gehennom, on T:47235 23:14 [hdf-us] [spl] SuzuneTakada (Con Dwa Nbn Cha), 42 points, T:455, slaughtered by a newt 23:16 [hdf-us] [spl] SuzuneTakada (Con Dwa Nbn Cha), 172 points, T:387, inhumed by a gecko 23:19 [hdf-us] [nh370] Vigorix (Cav Hum Fem Neu), 2280 points, T:4258, killed by a mumak 23:20 [hdf-us] [xnh] bouquet (Cav Gno Fem Neu) entered the Planes, on T:82794 23:20 [hdf-us] [xnh] bouquet (Cav Gno Fem Neu) killed the Wizard of Yendor on the Plane of Earth, on T:82795 23:27 [hdf-us] [spl] SuzuneTakada (Con Dwa Nbn Cha), 436 points, T:1064, murdered by a gnome lord 23:28 [hdf-us] [xnh] bouquet (Cav Gno Fem Neu) genocided class z, on T:83106 23:34 [hdf-us] [xnh] bouquet (Cav Gno Fem Neu) killed Death, on T:83574 23:35 [hdf-us] [xnh] bouquet (Cav Gno Fem Neu) killed Death, on T:83583 23:35 -!- Bliss56 has quit [Remote host closed the connection] 23:35 !who 23:35 K2: [hdf-us] eraserhead97 [nh370] Kevkat [nh370] bouquet [xnh] | [hdf-eu] microlance [nh370] | [hdf-au] omghax [xnh] 23:35 [hdf-us] [xnh] bouquet (Cav Gno Fem Neu) killed Famine, on T:83635 23:36 [hdf-us] [nh370] eraserhead97 (Cav Hum Fem Neu) had Vorpal Blade bestowed upon her by Ishtar, on T:20558 23:36 !whereis omghax 23:36 K2: [hdf-au] omghax [xnh]: (Bar Orc Fem Cha) T:4 The Dungeons of Doom level: 1 23:36 [hdf-us] [xnh] bouquet (Cav Gno Fem Neu) killed Famine, on T:83648 23:37 good ol' omghax 23:39 [hdf-us] [xnh] bouquet (Cav Gno Fem Neu) killed Famine, on T:83737 23:39 [hdf-us] [xnh] bouquet (Cav Gno Fem Neu) killed Pestilence, on T:83760 23:40 [hdf-us] [xnh] bouquet (Cav Gno Fem Neu) killed Death, on T:83770 23:40 [hdf-us] [xnh] bouquet (Cav Gno Fem Neu) killed Famine, on T:83777 23:41 [hdf-us] [xnh] bouquet (Cav Gno Fem Neu), 3801574 points, T:83787, ascended https://www.hardfought.org/userdata/b/bouquet/xnethack/dumplog/1585962423.xnh.html 23:41 aosdict: that covers all the classes for xnh for me! It's been a hoot. 23:41 congratulations to you! 23:41 sweet variant. 23:42 Eminently playable. Entirely non-evil. Very well balanced. 23:42 please don't say "and now I'm leaving it forever for slex" 23:42 !who 23:42 naw, I want to try a pirate in splice. 23:42 K2: [hdf-us] dingotron [evil] dbanethacker [nh370] eraserhead97 [nh370] Kevkat [nh370] | [hdf-eu] microlance [nh370] omghax [xnh] | [hdf-au] No current players 23:42 !whereis omghax 23:42 K2: [hdf-eu] omghax [xnh]: (Ran Elf Fem Cha) T:1 The Dungeons of Doom level: 1 23:43 xNetHack updated (all servers) 23:43 ^ new version 5.1 23:44 I dunno if that's good or bad timing for bouquet 23:45 GG bous 23:45 First 3.6 variant. First 3.7 variant. Now I need to not choke when 3.8-dev comes out ten years from now. 23:45 [hdf-us] [nh370] dbanethacker (Val Hum Fem Neu) killed Medusa, on T:35437 23:45 3.8 will be played on the holodeck 23:46 have fun with those merge conflicts 23:46 holodeck nethack (with like holographic ASCII characters) would be awesome 23:47 actually, reading 5.1 notes I'm reminded that losing see/invis intrinsic from potions is a good design move. Makes extrinsic sources more valuable. 23:48 though, it'd be nice if the see invis potion lasted longer. 23:48 mm 23:48 https://nethack-yanis.github.io/tag/superior_extrinsics_principle 23:48 Does that include invisible stalker corpses? 23:48 it's impossible to get the properties intrinsically now 23:49 stalkers behave like potions in terms of duration 23:50 Mm 23:50 Are there any extrinsic sources? 23:50 are you considering object properties for xnh? 23:50 Apart from amulets of ESP 23:50 the rings are extrinsic 23:50 Ah, right 23:50 and cloak of invisibility 23:50 orb of detection technically 23:51 Is it just those two sources, or any means of becoming invisible? 23:51 that's why I ask about object properties. Could have boots of see invisible or something 23:51 See invisible is intended to be hard to get to the point you need to spend a ring slot on it, because I was dissatisfied with how invisibility basically becomes pointless after the early game 23:51 bouquet: no 23:52 object materials are tricky enough at the moment. 23:52 good plan. It's neat, but doesn't seem to align with xnethack's design philosophy. 23:52 mm 23:52 Andrio keeps saying mm... are you eating bbq ribs or something? 23:52 mm 23:52 no, actually. 23:53 bouquet: Ehh... it *could* align with xnethack's design philosophy, if properties were unwishable. 23:53 Just an alien. 23:53 [hdf-us] [spl] SuzuneTakada (Wiz Inf Nbn Cha), 992 points, T:1698, inhumed by a rothe 23:53 It's more a factor of me being averse to having to balance a system as... random... as object properties would be 23:53 the temporary invis/see invis comes from sporkhack. couple other variants grabbed that also 23:54 oh and also it'd force a fiqhack-like property refactor, because I'd want to do it right, and I don't want to do such a refactor at this time 23:54 what does the alien taste like? 23:54 i think xnethack's temporary effect lasts longer though 23:55 bouquet like chicken 23:55 what else? 23:55 I'm not sure honestly. You'd have to find out yourself. 23:55 CHIKIN 23:55 Don't turn to stone. 23:55 The durations of intrinsic invisibility and see invisible are generally based on how rare the source is 23:56 so throne and fountain random effects give the most, zapping yourself with a wand gives the least 23:56 i actually hung onto those wands of invis my last couple games since they're actually very handy especially in the mines earlygame 23:56 eating the rings gives a good chunk because that's difficult to pull off, and can't be sustained very long 23:57 (in 5.0, eating the rings actually still gave you the intrinsic, but that was an oversight) 23:57 [hdf-us] [nh370] Kevkat (Hea Gno Mal Neu) destroyed Vlad the Impaler, on T:34684 23:57 [hdf-us] [nh370] Kevkat (Hea Gno Mal Neu) acquired the Candelabrum of Invocation, on T:34685