02:36 -!- mode/Beholder [+Zi] by Beholder 02:37 -!- tepper.freenode.net changed the topic on #hardfought to: Welcome to Hardfought - https://www.hardfought.org/ - public irc logs available on website ¤ Public NetHack server - ssh nethack@hardfought.org, nethack@eu.hardfought.org or nethack@au.hardfought.org ¤ hdf-us will be down for a couple hours the night of Dec 8th (EST), moving to AWS instance 02:37 -!- mode/#hardfought [+v Beholder] by ChanServ 02:37 -!- mode/Beholder_ [+Zi] by Beholder_ 02:37 -!- card.freenode.net changed the topic on #hardfought to: Welcome to Hardfought - https://www.hardfought.org/ - public irc logs available on website ¤ Public NetHack server - ssh nethack@hardfought.org, nethack@eu.hardfought.org or nethack@au.hardfought.org ¤ hdf-us will be down for a couple hours the night of Dec 8th (EST), moving to AWS instance 02:37 -!- Beholder_ has joined #hardfought 02:37 -!- mode/#hardfought [+v Beholder_] by ChanServ 02:37 -!- mode/#hardfought [+v Beholder_] by ChanServ 02:38 -!- Beholder_ has quit [Remote host closed the connection] 02:38 -!- mode/Beholder_ [+Zi] by Beholder_ 02:38 -!- card.freenode.net changed the topic on #hardfought to: Welcome to Hardfought - https://www.hardfought.org/ - public irc logs available on website ¤ Public NetHack server - ssh nethack@hardfought.org, nethack@eu.hardfought.org or nethack@au.hardfought.org ¤ hdf-us will be down for a couple hours the night of Dec 8th (EST), moving to AWS instance 02:38 -!- Beholder_ has joined #hardfought 02:38 -!- mode/#hardfought [+v Beholder_] by ChanServ 02:38 -!- mode/#hardfought [+v Beholder_] by ChanServ 02:38 !ping 02:38 K2: Pong! 02:38 !ping 02:38 K2: Pong! 02:38 K2: Pong! 02:38 K2: Pong! 02:39 rofl 02:39 rofl 02:39 -!- Beholder has quit [Remote host closed the connection] 02:40 -!- mode/Beholder [+Zi] by Beholder 02:40 -!- card.freenode.net changed the topic on #hardfought to: Welcome to Hardfought - https://www.hardfought.org/ - public irc logs available on website ¤ Public NetHack server - ssh nethack@hardfought.org, nethack@eu.hardfought.org or nethack@au.hardfought.org ¤ hdf-us will be down for a couple hours the night of Dec 8th (EST), moving to AWS instance 02:40 -!- mode/#hardfought [+v Beholder] by ChanServ 02:41 weird 02:41 !ping 02:41 K2: Pong! 02:41 !who 02:41 K2: [hdf-us] No current players | [hdf-eu] No current players | [hdf-au] Mishka [nh343] 02:41 so now i need to work on the web part 02:41 otherwise !who wont work for hdf-us ;) 02:44 <@Luxcalibur> I can ping hardfought but the hterm and website are still rip 02:45 i know 02:45 working on that 02:45 you can ssh in and play just fine though 02:45 that was the most important bit ;) 03:01 gotta be able to get killed by newts! 03:05 "The newt whips you with its long tail. You die." 03:10 !who 03:10 K2: [hdf-us] luxor [slshm] ShivanHunter [nh362] | [hdf-eu] Raisse [nh362] | [hdf-au] Mishka [nh343] 03:13 [hdf-us] [nh343] k2 (Hea Gno Mal Neu), 129 points, T:262, killed by a jackal 03:13 yay 03:14 !lastgame k2 03:14 K2: [hdf-us] https://www.hardfought.org/userdata/k/k2/nh343/dumplog/1537121150.nh343.txt | [hdf-eu] https://eu.hardfought.org/userdata/k/k2/xnethack/dumplog/1535943458.xnh.txt | [hdf-au] https://au.hardfought.org/userdata/k/k2/splicehack/dumplog/1532712486.splice.txt 03:14 now lets get the website up 03:22 [hdf-us] [nh362] ShivanHunter (Wiz Elf Mal Cha) killed the Oracle, on T:13586 03:23 blame the arch-lich! 03:25 [hdf-us] [nh362] ShivanHunter (Wiz Elf Mal Cha) made his first wish - "the blessed greased fixed +3 Master Key of Thievery", on T:13744 03:25 woo fountain wish 03:26 -!- LarienTelrunya has joined #hardfought 03:26 -!- mode/#hardfought [+v LarienTelrunya] by ChanServ 03:31 Something must have gone wrong with my game in the server migration. I had 5 wands of wishing, 10 magic lamps, 24 amulets of life saving and 4 pet archons 03:31 website up 03:31 ShivanHunter yeah lemme fix that... 03:32 need to get hterm up 03:36 -!- raisse has quit [Ping timeout: 240 seconds] 03:41 I had an "uncursed +0 small shield" that showed up in the prompt to identify items. I selected it, and was told it was an "uncursed +0 small shield". What exactly was unidentified about it? 03:43 rustproofing not known? 03:43 <@Luxcalibur> there are some items that spawn fooproof, like angel gear 03:44 Hmm. Might be that 03:44 that seems new but maybe I've just never noticed it before 03:49 revisiting dlvl 1 on a whim, on T:15464, and find a random " of reflection 03:49 lol 04:32 [hdf-us] [nh362] ShivanHunter (Wiz Elf Mal Cha) killed the invisible Dark One, on T:18221 04:32 [hdf-us] [nh362] ShivanHunter (Wiz Elf Mal Cha) acquired the Bell of Opening, on T:18223 04:37 [hdf-us] [nh362] ShivanHunter (Wiz Elf Mal Cha) killed Medusa, on T:18317 05:37 -!- mode/Beholder [+Zi] by Beholder 05:37 -!- asimov.freenode.net changed the topic on #hardfought to: Welcome to Hardfought - https://www.hardfought.org/ - public irc logs available on website ¤ Public NetHack server - ssh nethack@hardfought.org, nethack@eu.hardfought.org or nethack@au.hardfought.org ¤ hdf-us will be down for a couple hours the night of Dec 8th (EST), moving to AWS instance 05:37 -!- mode/#hardfought [+v Beholder] by ChanServ 05:49 hterm works - https://www.hardfought.org/hterm/ - but some permalinks on website broke during migration. will fix after i've had some sleep. gnite 05:58 permalinks fixed. gnite ;) 06:04 -!- rebatela has joined #hardfought 06:08 -!- rebatela has quit [Read error: Connection reset by peer] 06:31 -!- K2_ has joined #hardfought 06:31 -!- K2 has quit [Quit: Rico, smoke! *poof*] 06:32 -!- K2 has joined #hardfought 06:32 -!- mode/#hardfought [+o K2] by ChanServ 06:36 -!- K2 has quit [Disconnected by services] 06:36 -!- K2 has joined #hardfought 06:36 -!- mode/#hardfought [+o K2] by ChanServ 06:37 -!- K2 has quit [Client Quit] 06:38 -!- K2_ is now known as K2 06:38 -!- mode/#hardfought [+o K2] by ChanServ 06:42 [hdf-us] [nh362] ShivanHunter (Wiz Elf Mal Cha) wished for "holy marker", on T:19187 06:51 [hdf-us] [nh362] ShivanHunter (Wiz Elf Mal Cha) wished for "holy marker", on T:19280 06:51 (0:33) RNG pls 06:51 [hdf-us] [nh362] ShivanHunter (Wiz Elf Mal Cha) wished for "holy marker", on T:19281 06:52 [hdf-us] [nh362] ShivanHunter (Wiz Elf Mal Cha) wished for "uncursed marker", on T:19282 07:03 [hdf-us] [nh362] ShivanHunter (Wiz Elf Mal Cha) polymorphed his first object, on T:20788 07:12 [hdf-us] [nh362] ShivanHunter (Wiz Elf Mal Cha) entered Gehennom, on T:21521 07:23 lol I have a pet titan that found a c corpse and is stoning everything. 07:29 -!- raisse has joined #hardfought 07:44 -!- stenno has quit [Ping timeout: 268 seconds] 07:48 -!- lonelyBoy has joined #hardfought 07:49 PuTTY is giving me a warning of a potential security breach? 07:50 <@Tmasher> Server moved 08:04 -!- workimer has joined #hardfought 08:04 -!- mode/#hardfought [+v workimer] by ChanServ 08:06 -!- stenno has joined #hardfought 08:08 -!- remirol has quit [Ping timeout: 246 seconds] 08:18 [hdf-us] [nh362] Tmasher (Cav Hum Fem Law), 2621 points, T:3120, killed by an Uruk-hai 08:47 -!- Tangles changed the topic on #hardfought to: Welcome to Hardfought - https://www.hardfought.org/ - public irc logs available on website ¤ Public NetHack 08:47 ugh no 08:48 public nethack?! but what about my private nethack? 08:48 Welcome to Hardfought - https://www.hardfought.org/ - public irc logs available on server - ssh nethack@hardfought.org, nethack@eu.hardfought.org or nethack@au.hardfought.org ¤ hdf-us 08:48 -!- ProzacElf has quit [Ping timeout: 268 seconds] 08:48 tangles: my grunthack curses interface is screwed up 08:49 -!- Tangles changed the topic on #hardfought to: Welcome to Hardfought - https://www.hardfought.org/ - public irc logs available on server - ssh nethack@hardfought.org, nethack@eu.hardfought.org or nethack@au.hardfought.org ¤ hdf-us NEW SSH KEY - f0:5f:fb:da:6d:ac:1c:88:2a:91:ea:24:fe:44:d4:50 08:49 hothraxxa: how so? 08:49 hang on i'll show you 08:49 !whereis hothraxxa 08:49 Tangles: hothraxxa is not playing. 08:50 try now 08:50 EPI: if you do something that would bless an item but it's already blessed, the item becomes cursed instead. 08:51 LarienTelrunya: that makes no flavor sense, whereas most EPIs come about by taking flavor to the fullest 08:51 hothraxxa: your symbol-set is screwed up on my terminal, but that's probably my terminal. Looks like the window borders aren't drawing properly. Is that it? 08:52 looks that way 08:52 aosdict: huh? EPIs are about being evil to the player, which this one fulfills :P 08:52 same settings worked before the move 08:53 Yeah. Something is not right there. 08:54 Does the same for me. 08:54 I'm not going to be able to look at it for a day or two. 08:54 np, i'll play without curses 08:56 LarienTelrunya: Right, but lots of them are about being *too* rational with the game mechanics 08:56 It's affecting curses on all the variants. 08:56 :( 08:56 fuck I wish I knew about this 2 hours ago 08:56 Tangles: oh, if you play the game today there are no curses in it? sign me up :P 08:57 a chest containing a scroll of light, a spellbook of light, and a wand of light 08:57 and a potion of water 08:58 i've always wanted a light chest 08:58 it's probably light water 08:58 Isn't most normal water light water, though 08:58 have to industrially produce heavy water :) 08:58 well, this one is extra light 08:59 not a single deuterium atom to be found! 09:05 -!- stenno has quit [Quit: WeeChat 2.3] 09:09 -!- stenno has joined #hardfought 09:10 -!- stenno has quit [Client Quit] 09:12 -!- stenno has joined #hardfought 09:14 -!- stenno has quit [Client Quit] 09:15 -!- stenno has joined #hardfought 09:15 is this a new installation? 09:16 K2 did some symlinking for $reasons_I_dont_remember 09:17 bhaak: yeah. also the path to the chroot changed so all the install scripts are now wrong. 09:19 oops 09:22 hmm ok I got grunthack rebuilt, but no difference to the curses window drawing. I don't have time to look at this properly now. 09:25 -!- Bugpowder has joined #hardfought 09:25 Tried to log into hardfought at got “The ECDSA host key for hardfought.org has changed, 09:25 and the key for the corresponding IP address 2600:1f18:607c:4fcc:f4ee:8b86:440b:6298 09:25 is unknown. This could either mean that 09:25 DNS SPOOFING is happening or the IP address for the host 09:25 and its host key have changed at the same time." 09:26 Bugpowder: see /topic on this channel. 09:27 cool 09:27 thanks 09:27 hdf-us was moved recently and the ssh key changed. 09:27 so it's the second option and everything is all right 09:27 but curses on all variants is broken on the new server. It probably won't be fixed for a couple of days. 09:29 I have to go to Sydney for a work trip tomorrow and I'm not going to have time to look at it, probably until Tuesday night 09:29 I can’t fiugre out how to run curses anyways 09:31 -!- nooodle has joined #hardfought 09:31 [hdf-us] [nh362] Tmasher (Wiz Elf Mal Cha), 387 points, T:1083, killed by a crossbow bolt 09:33 Anyway I have to pack, and get a couple hours sleep, so I'm off. Good luck everyont 09:33 s/t$/e/ 09:33 p sure curses works fine on Nitro-derivatives 09:34 yeah I'm sure it does, 09:34 I didn't bother to test. 09:35 yeah those look fine. 09:39 [hdf-us] [nh362] Bugpowder (Sam Hum Mal Law) wished for "uncursed magic marker", on T:28841 09:40 have fun! 09:44 SO I'm debating if the easiest way to review the OS/2 port is to cross-compile NetHack from Linux 09:44 With OpenWatcom 09:45 That way it can easy be tested with Travis 09:46 [hdf-us] [nh362] Tmasher (Sam Hum Fem Law) made her first wish - "blessed greased fixed +3 grey dragon scale mail", on T:1681 09:47 -!- raisse has quit [Ping timeout: 250 seconds] 09:56 [hdf-us] [nh362] Bugpowder (Sam Hum Mal Law) wished for "blessed ring of teleport control", on T:29033 09:58 -!- raisse has joined #hardfought 10:03 FIQ, you're on the dev team, right? 10:03 or am I misremembering 10:03 fiq is not on the vanilla devteam 10:03 <[Demo]5> no he's just knowledgeable and experienced about C and nethack 10:04 Ah, I'm just debating on reviving the old nethack ports. SPecifically DOS and OS/2 10:04 as cross-compiled builds from Linux 10:04 Using openwatcom 10:04 Since I provided I can build Linux NetHack with OWC 10:05 and it's a lot more up to date than most of the other compilers 10:05 the ECDSA fingerprint I'm seeing is SHA256:tSr/Wkp1spcddmH1GLjdPU8XurSku7/MUPTUMeErPFM. Is that just a different encoding of the key in the topic? I'm kind of ignorant on SSH keys 10:05 nooodle: server moved 10:06 *nods* i just want to learn how I can verify the f0:5f:... key, not because I don't trust the server I'm seeing, just because I feel like I should know how 10:06 aosdict, I'm surprised you didn't copy the host keys 10:06 me? what would I have to do with it? 10:07 ... I'm failing to remember who does what 10:07 K2/Tangles 10:07 -!- Bugpowder has quit [Quit: Bugpowder] 10:07 K2 is sleeping like a log right now probably 10:07 and you (aosdict) and ais are devteam right? 10:07 [hdf-us] [nh343] strangefella (Tou Hum Mal Neu), 1399 points, T:5123, quit 10:07 or should I give up braining today 10:07 ais is. I'm not. 10:07 bhaak is too 10:08 I guess it's mostly seeing how much the dev team wants to see the ports revived 10:08 the devteam level was very educational. Get to know your devteam! 10:08 I want to do it, but I also know the devteam somewhat wants to modernize the codebase and a non-GCC/MSVC compiler might complicate that 10:09 <[Demo]5> more like get to kill your devteam one member at a time :P 10:10 more like killing one of the devteam is like killing one of the four elephants that holds up the world 10:11 <[Demo]5> thats why u gotta play multiple games 10:11 Ok I figured out the answer to my question, I just had to do ssh -o FingerprintHash=md5 nethack@hardfought.org , so that it showed the md5 in the topic rather than the sha256 I was seeing 10:11 <[Demo]5> just luxi ur way to them and have some creative way to kill them 10:12 -!- Bugpowder has joined #hardfought 10:15 I was thinking of playing xnh but since it hasn't been updated yet it doesn't have the changes I want to test 10:17 I am not a DT member, no 10:18 <[Demo]5> u can still name ur pet fiq in tnnt and kill it though if thats what u really need :P 10:20 aosdict, updated on the server or in the source? 10:21 what does "updated in the source" mean 10:24 aosdict, is the feature in xnethack itself, and just the server needs an update, or is it in 362 and needs porting. 10:25 or rebasing 10:26 <[Demo]5> give someone give me a refresher on 3.4.3 alchemy to make hella potions of full healing? 10:29 NCommander: the changes in question are things I added yesterday, so just waiting on hdf to rebuild 10:30 ++ 10:39 [hdf-us] [slshm] luxor (Val Dop Fem Neu), 186050 points, T:38356, killed by a xorn, while helpless 10:46 [hdf-us] [nh362] Bugpowder (Sam Hum Mal Law) killed Orcus, on T:30448 10:48 <[Demo]5> hmm, i now have 126 potions of full healing 10:48 lol 10:48 be sure to keep 1 in open inventory. would be a shame to die with 126 pots 10:48 [Demo]5 alchemy? 10:48 either that or MASSIVE farming 10:49 <[Demo]5> medium farming, hella alchemy 10:49 <[Demo]5> did like 8 dips, 0 booms 10:49 <[Demo]5> dlvl 41 throne farming for a bit 10:49 <[Demo]5> currently on T:413290907 10:50 <[Demo]5> spent most of that time waiting in soko 10:50 !whereis Demo 10:51 aosdict: [hdf-us] Demo [dnh]: (Ana Inc Fem Cha) T:413290517 Gehennom level: 41 10:51 you went back in time evidently 10:51 -!- Bugpowder has quit [Quit: Bugpowder] 10:53 <[Demo]5> there we go, 1337 hp 10:54 <[Demo]5> hmm i could continue farming for gain level and get enough to ride them all the way up from sanctum 10:55 [hdf-us] [slshm] luxor (Val Dop Fem Neu), 1242 points, T:972, killed by a wererat 10:56 <[Demo]5> I think I wanna do an Astral takeover 11:00 -!- LarienTelrunya_ has joined #hardfought 11:00 -!- mode/#hardfought [+v LarienTelrunya_] by ChanServ 11:00 I've always wanted to do that, but then get nervous that I'll astral splat. 11:00 -!- LarienTelrunya has quit [Ping timeout: 256 seconds] 11:00 -!- LarienTelrunya_ is now known as LarienTelrunya 11:00 * NCommander has astral splat 11:01 That sucked 11:01 hard 11:01 I have done it a few times. It's not so bad when you get used to it ;) 11:03 Did K2 move ssh to a new IP address? I'm getting a key mismatch. 11:03 bouquet: yes, it's now on AWS 11:03 yes, he did 11:03 -!- Bugpowder has joined #hardfought 11:03 Ohhh! 11:04 that happened faster than I expected. 11:04 yes it did 11:04 Did your homelab catch fire? 11:04 as they sometimes do. 11:05 related: someone posted their energy consumption for their homelab on reddit recently. Very eye opening when you consider how many electrons are sacrificed to AC/DC. 11:06 hey, we're not in the business of destroying matter here! 11:06 #SaveTheLeptons 11:06 they go to the big particle accelerator in the sky 11:07 anywho, I check the ecdsa and sha fingerprints onthe website before I asked here. I think you need to update those. 11:07 yeah I bet that's how K2 will justify the expense of AWS to his wife. think of the electricity savings! 11:07 if you check server info in dgl on hdf-us, i've updated the info there with the new RSA/ECDSA public keys (md5) 11:08 also, like, the topic :P 11:08 yeah i saw that, yay Tangles 11:09 ah, well, I leave irssi running in tmux for backlogs so I don't often see it. 11:09 K2: so to ssh into the new hdf-us, we need to first trust the new key of hdf-us? ;) 11:09 [hdf-us] [nh362] Bugpowder (Sam Hum Mal Law) killed the invisible Baalzebub, on T:31016 11:10 to ssh in to see the dgl server info I men 11:10 mean 11:10 yeah 11:12 i have to leave again here in a couple min 11:12 bleh 11:13 bouquet: I'm rusty on this, given the fingerprint in dgl, how do I verify that against the known_hosts ecdsa key? 11:14 ask a trusted source? THat's why I asked here. 11:14 I deleted the saved key from known_hosts and reconnected. 11:15 we can trust K2 11:15 I know there's some way to compare my known_hosts entry against the SSH fingerprint listed in dgl, but I don't know what 11:15 assuming the dgl screen is a trusted source 11:16 i get it, that info should be elsewhere 11:16 like on webpage 11:16 <[Demo]5> yikes 11:17 [hdf-us] [nh362] Bugpowder (Sam Hum Mal Law) killed Juiblex, on T:31186 11:17 <[Demo]5> why am i so bad at video game """ethics""" 11:17 the fingerprint I'm getting out of known_hosts is 2600:1f18:607c:4fcc:f4ee:8b86:440b:6298 which doesn't match what I see in dgl 11:17 ok there 11:17 exit dgl and reconnect 11:17 Maybe K2 is being held captive and forced to verify ssh fingerprints? 11:18 any changes to the dgl config requires exiting/reconencting to see changes 11:18 updated key info - https://www.hardfought.org/nethack/ 11:18 <[Demo]5> oh well 11:18 gotta go, be back later 11:18 why is mine formatted like an ipv6 address and not byte:byte:byte? 11:19 [hdf-us] [dnh] Demo (Ana Inc Fem Cha) wished for "Blessed +2 fixed greased stormhelm", on T:413291052 11:19 what i show is in md5 format 11:19 ah there we go 11:19 are you sure you are checking /etc/ssh/known_hosts and not /etc/known_hosts ? 11:20 [hdf-us] [dnh] Demo (Ana Inc Fem Cha) wished for "Blessed +7 fixed greased stormhelm", on T:413291042 11:20 had to ssh-keygen -l -E md5 11:21 I feel like a LPB on this newfangled AWS stuff. 11:21 -!- K2 changed the topic on #hardfought to: Welcome to Hardfought - https://www.hardfought.org/ - public irc logs available on server - ssh nethack@hardfought.org, nethack@eu.hardfought.org or nethack@au.hardfought.org ¤ hdf-us SSH keys have changed due to server migration, see https://www.hardfought.org/nethack for new public key info (md5 format) 11:22 T:413291042?!? 11:22 <[Demo]5> yeah im pro 11:23 Bugpowder: it's Demo, he does stuff like this 11:25 [hdf-us] [dnh] Demo (Ana Inc Fem Cha) wished for "Blessed +2 fixed greased helping hand", on T:413291049 11:27 [hdf-us] [slshm] luxor (Val Dop Fem Neu), 146 points, T:688, killed by a large mimic 11:29 I just removed the old key and said yes to the new one, much easier than replacing it in the file 11:31 -!- AntiGulp has joined #hardfought 11:31 -!- mode/#hardfought [+v AntiGulp] by ChanServ 11:31 hi 11:31 Hello AntiGulp, Welcome to #hardfought 11:34 -!- Bugpowder has quit [Quit: Bugpowder] 11:34 [hdf-us] [dnh] Demo (Ana Inc Fem Cha) wished for "Blessed +2 fixed greased helping hand", on T:413291057 11:36 Turn 413291057? 11:38 <[Demo]5> yeah 11:38 <[Demo]5> im just a really slow player :P 11:39 413 MILLION turns? 11:39 how long has that game been going on for? 11:39 <[Demo]5> since like, the summer? idk 11:39 <[Demo]5> kinda got bored and shelved it 11:39 why on earth did you get bored.. 11:39 My slowest ascension is under 40,000… and that took like 12 hours… how??? 11:39 [hdf-us] [nh362] bouquet (Arc Gno Fem Neu), 1490 points, T:2438, poisoned by a killer bee 11:40 my fastest ascension is about 39K and it took 2 days (all my ascensions take 2-3 days) 11:40 <[Demo]5> well it would be a lot higher but it was using too much cpu and making k2 sad so i had to stop 11:40 my slowest just over a million I think 11:41 er, just over 100.000, order of magnitude error 11:41 <[Demo]5> well im just 4130x slower? :P 11:44 <[Demo]5> nah i achieved stasis 11:46 [hdf-us] [slshm] luxor (Val Dop Fem Neu), 1338 points, T:3797, killed by a gnome, while helpless 11:47 [hdf-us] [slshm] luxor (Val Dop Fem Neu), 102 points, T:166, killed by a water demon 11:50 [hdf-us] [dnh] Demo (Ana Inc Fem Cha) wished for "uncursed magic marker", on T:413291063 11:52 [hdf-us] [dnh] Demo (Ana Inc Fem Cha) wished for "uncursed magic marker", on T:413291065 11:54 <[Demo]5> eh thats enough for now 11:54 <[Demo]5> not really sure what the optimal strategy is right now 11:56 My longest (real life) game was five years 11:56 Had a save on alt.org that I picked started in 2009, and became my first ascension 11:57 [Demo]5: is this an extinctionist run? At that many turns I would imagine there's nothing left to kill. 11:58 i have 13 wins over 30k turns, one of which is over 40k 11:59 [hdf-us] [nh362] CropperS (Wiz Elf Mal Cha) completed Sokoban, on T:16850 12:00 I'm sort of a 50k'ish turn person. I have one run that was blessed with a very helpful RNG which took 25k. Highest was about 80k. 12:00 I'd like to try extinctionist some day. If we could have multiple games going (like 4k/fourk/fiq) then I'd commit a vanilla game to it. 12:02 <[Demo]5> no 12:02 <[Demo]5> most of those turns were spent waiting in soko 12:06 [hdf-us] [slshm] luxor (Fla Orc Mal Cha), 478 points, T:1828, killed by Ms. Rouffiac, the shopkeeper 12:08 whee, ice box, cockatrice corpse, wands and scrolls of create monster. How many meatballs I can make now! 12:09 pity I don't need teleport control, but it's just as well, so tedious 12:12 -!- spontiff has joined #hardfought 12:14 [hdf-us] [slshm] luxor (Fla Dop Fem Neu), 24 points, T:9, killed by touching The Staff of Aesculapius 12:14 rip 12:16 wait how 12:16 without a wish announcement 12:17 <[Demo]5> probably doesn't announce them 12:17 not sure if slash'em livelogging is set up 12:18 -!- Chris__ANG has quit [Remote host closed the connection] 12:20 [hdf-us] [nh362] oh6 (Rog Hum Mal Cha) made his first wish - "blessed fixed greased +3 gray dragon scale mail", on T:20 12:21 Gosh, the new server's throwing quite a welcoming party. 12:22 <[Demo]5> haha 12:22 an altar and a magic lamp right next to the level 1 stairs up 12:22 blessed magic lamp I should add 12:24 -!- ais523 has joined #hardfought 12:24 -!- mode/#hardfought [+v ais523] by ChanServ 12:24 <[Demo]5> jesus 12:24 -!- AntiGulp has quit [Quit: explodes into confetti] 12:25 <@Luxcalibur> !tell K2 hterm is still busted 12:25 Will do, @Luxcalibur! 12:26 [hdf-us] [nh362] Tmasher (Sam Hum Fem Law), 11063 points, T:5935, killed by a leocrotta 12:36 izchak has 5 oil lamps 12:36 but I have a magic lamp! 12:36 I'd be tempted to cash it for the wish and buy all the oil lamps 12:37 "Genie, give me enough gold to buy all these oil lamps! I love oil lamps!" 12:38 now Izchak has 6 oil lamps, I just sold him a death-drop one 12:38 and only 3 assorted candles 12:38 (i.e. one is wax. two are tallow, none of them stack) 12:41 YANI: unidentified items that look the same stack with each other, the game determines their actual identity (BCU, enchantment, etc.) lazily when you do something that would identify them 12:42 hi ais523! I was wondering; if you make something sufficiently unlikely, can it happen that there is no random seed that makes it happen? 12:42 -!- Bugpowder has joined #hardfought 12:42 Of course this is already true to a degree with BUC-id but this incentivises Hypothetically Optimal Man to avoid identifying stuff to fit more stacks in inventory. 12:43 say, if I had 5 consecutive !rn2(1000000) calls, could it be that no 32-bit random number series would exist that passes all those? 12:43 LarienTelrunya: this actually varies by platform, but the answer is yes in almost every case 12:43 It seems evident that if there are more than 2^32 outcomes, some of them cannot be reached by a given 32-bit seed 12:44 there are some platforms that mix observed random behaviour from real-world inputs into their RNGs by default, NetHack would be capable of producing any sequence of random numbers on one of those platforms if it was using that RNG 12:44 but it's rare 12:44 yeah some variants have CSPRNGs apparently, but when I tried to implement one for slex, it was unbearably laggy to play 12:45 Pinkbeast: Final Fantasy 8 is a good example, it has an 8-bit RNG seed (!) and some of its random tables have more than 256 entries, because the devs apparently didn't understand randomness 12:45 they were taking multiple consecutive numbers from the RNG in the hope that that would help (it doesn't) 12:45 <[Demo]5> what the hell 12:45 ais523: I speculate one dur-brain wrote the RNG and none of the others knew, or at any rate, I hope that was the case. 12:46 LarienTelrunya: look at the one UnNetHack uses. 12:46 the RNG isn't enen unbiased, some 8-bit results are produced from more than one seed, meaning some aren't produced at all 12:50 bhaak: uhh I didn't understand that implementation either 12:51 Why did they write their own RNG anyway? 12:51 NH4's is easy to understand and seems to be efficient enough but it may be overkill 12:52 Pinkbeast: presumably the Playstation doesn't have a hardware RNG, and it may not have one in its standard library 12:52 -!- raisse has quit [Ping timeout: 268 seconds] 12:53 It seems extraordinary that a games console wouldn't, but I guess you're right 12:53 many games just ended up using the example RNG in the C standard, which is common enough that you can often find a game's random number generation routine simply by searching for instances of the constant 1103515245 12:53 IIRC the 3DS was the first widely released games console with a hardware RNG 12:53 so fairly recent 12:54 presumably Nintendo got sick of people reverse-engineering their games' random number generation and using it to exploit them 12:54 <@Zachary> I know FF12 on the PS2 had a really bad RNG that was completely predicatble. 12:54 it's nowhere near as bad as FF8's, but yes 12:54 <@Zachary> But didnt the PS2 have a really bad RNG built into the hardware? 12:54 it's just a stock 32-bit RNG 12:55 people have got really good at reversing those, especially when (like FF12's) it's linear-congruential 12:55 LarienTelrunya: what do you need to understand? set a seed, pop out random numbers. neither you nor I are skilled enough to verify that those RNG implementations are truly cryptographically secure. 12:55 ais523: I had a question I was going to ask you about RNG effects. There are a lot of algorithms in nethack that work via the principle of "generate a random coord/array index up to 500 times or until the random value is suitable", so I was looking into reservoir sampling (iterate through the array choosing the nth suitable element with probability 1/n). Is either approach biased via calling rn2 a lot 12:56 in succession? 12:56 bhaak: no I meant code-wise, I'm not sure what all needs to be put into slex to make it work correctly 12:57 aosdict: we learned a lot about this sort of RNG effect in the TAS, calling it "jitter" 12:58 basically what happens is that the algorithm skips to the next valid possibility, so it's unbiased, but effectively reduces the size of the RNG seed (for that check and the near future); there are fewer valid possibilities 12:59 [hdf-us] [nh362] Bugpowder (Sam Hum Mal Law) destroyed Vlad the Impaler, on T:32431 12:59 err, in theory it could be biased because some valid possibilities will have more seeds before them than others 12:59 [hdf-us] [nh362] Bugpowder (Sam Hum Mal Law) acquired the Candelabrum of Invocation, on T:32433 12:59 but the resulting seed space is still sufficiently large that the process mostly averages out 12:59 LarienTelrunya: have you looked at the commit where I added it? I don't think you need to do much different. 12:59 it's like the "random article" button on Wikipedia 13:00 bhaak: uhh if I did, it was long ago; do you have a link? 13:00 basically, each article has a random number attached, the "random article" button picks a random number and then visits the article with the next-higher number 13:00 -!- lonelyBoy has quit [Ping timeout: 256 seconds] 13:00 so some articles are more likely to be picked than others because they have larger gaps before them 13:01 ais523: specifically what I was interested in was a rndmonst() change I implemented recently that greatly simplifies the code by removing the need for a eligible monster cache, but now it has to generate up to, theoretically, NUMMONS rn2 results instead of just one. 13:02 the numbers are small enough that you shouldn't expect detectable bias 13:02 unless you're slex, possibly 13:03 I suppose one could still have it just call rn2 once, but that would require two passes over monst[] instead of one, to count up the frequencies of eligible monsters first. 13:04 in general, when you're writing a game, even fairly odd RNG behaviour is hard for players to notice 13:04 -!- Bugpowder has quit [Quit: Bugpowder] 13:05 anyway rndmonst isn't the main offender I wanted to address, it's things like somexy which are worse offenders. 13:05 Because those have actual, built in bias. 13:07 note that you'll get modulo bias whenever the number of possibilities is not a power of 2 13:07 so if you're really concerned about bias, you'll need to fix that too 13:07 (NH4's RNG has a modulo bias fix in it, not because it really needs one, just because I thought I might as well) 13:10 Well, I was referring to the fallback cases that result from the usual nethack approach. For instance, if somexy tries enough times and fails, it'll just iterate over the room and take the first valid space, so there's a bias for that space 13:11 now I'm wondering if that effect is smaller or larger than the modulo bias 13:13 I'm wondering if this was originally for performance reasons? because the chances are good that you would find a suitable spot in that room on the first go, rather than repeatedly picking random ineligible spots. both of the other algorithms I've considered require a linear loop over the spaces, so it's less performant in the average case but still a small amount 13:16 almost certainly; you nearly always find a good spot on the first few attempts 13:17 and the bias only becomes noticeable if the probability of finding a good spot is very low 13:17 o/ 13:17 in terms of code elegance, though... nethack's algorithm is pretty bad. 13:17 well, it's very simple, and so in a way it's more elegant than the unbiased algorithm! 13:18 -!- raisse has joined #hardfought 13:18 simple != simplistic 13:18 hey ais523 13:18 not really... it has two loops, one as a fallback, and an arbitrary magic number 13:18 hi 13:18 Hello ais523, Welcome to #hardfought 13:19 ais523, how is life 13:19 nowadays, of course, iterating over a room is virtually nothing in terms of performance cost 13:19 when an algorithm is correct, it can be encapsulated and treated as a robust, self-contained component, and that increases overall simplicity 13:19 now I'm wondering if I should automatically respond "life, don't talk to me about life" or give a serious answer :-D 13:19 ais523, :) 13:20 ais523, I'm dealing with the fact my laptop's optical drive is dead so my efforts to build NetHack on OS/2 are somewhat stymied ATM 13:20 Unless I want to write and feed 39 floppies 13:20 ais523, and yet I have an urge to port nethack somewhere exotic ... ish 13:20 NCommander: is the problem writing the optical disks or reading them? 13:20 -!- introsp3ctive has joined #hardfought 13:21 * Pinkbeast blinks, no network on the laptop? 13:21 as in, is the laptop running OS/2, or is this the machine on which you're producing disks to feed to OS/2? 13:21 ais523, reading. 13:21 ugh, there may not be many workarounds then 13:21 (also, is NetHack really up to 39 floppies of disk space for the source?) 13:21 Drive's laser seems to be dying. I ordered a replacement 13:21 ais523, it's installing OS/2 13:21 !tell k2 grunthack crashed, now says i have a game in progress 13:21 I'm on it, hothraxxa. 13:21 NCommander: oh, I see 13:21 This is an old ThinkPad 380D. I could use a VM. 13:22 ais523, or cross-compile from Windows/Linux; I proved NetHack will build with OpenWatcom with minor code tweaks 13:22 there probably aren't that many computers left that are capable of running OS/2 but still have working optical drives 13:22 ais523, that would let the OS/2 and DOS ports be compiled from Windows/Linux 13:22 ais523, there's modern versions of OS/2. eCS, and AcraOS 13:22 I just want a vintage copy because vintage machine. 13:23 huh, I assumed OS/2 died ages ago 13:23 ais523, I mean, if I really want to hurt the devteam, I could get the classic Mac OS port building again. 13:23 ais523, https://www.arcanoae.com/ 13:23 someone ported a different one of my programs to Mac Classic a while back 13:23 just for the sake of it, I think 13:24 you want vintage? i'll show you vintage. https://youtu.be/uFQ3sajIdaM 13:24 ais523, which ports are currently scheduled for deathrow 13:24 I always liked the fact NetHack ran on almost anything 13:24 not on the 1401 13:24 for the record, i worked with one of these when it was still current 13:26 NCommander: I can't remember, and am not 100% sure the devteam is internally consistent on this 13:26 but "most of them" is probably a good example 13:26 obviously, Windows, UNIX/Linux, Mac OS X aren't 13:26 ais523, well I know there was an argument on modernizing the C source code 13:26 and I think maybe VMS also has a maintainer? 13:26 *probably a good rule 13:26 ais523: I thought it was just the ports which couldn't compile ANSI C 13:26 which was not that many of them? 13:26 it's the ports where nobody's maintaining them 13:27 we also want to move onto ANSI C (or at least, most of us do) 13:27 because if a port's maintainable, it almost certainly has a C89 compiler 13:27 Well, if I modify the OS/2 port to use openwatcom, problem solved 13:27 (it's probably harder to find K&R C compilers than C89 compilers nowadays!) 13:27 And we can cross build it and a DOS port from a modern system 13:28 how different is OS/2 from DOS in terms of the binaries? I know they were meant to be compatible, but don't know how closely that relationship holds 13:28 DOS binaries will run on OS/2 13:28 As long as it's DOS 5 compat 13:28 basically like Windows 95 13:28 -!- shadowcat22 has joined #hardfought 13:29 maybe we should be producing a joint "DOS / OS/2" binary, then? 13:29 how does DPMI come into this? because that's necessary with the current DOS ports 13:30 being able to use 32-bit binaries is so helpful, as almost every platform is at least 32-bit nowadays 13:30 ais523, you can actually create OS/2/DOS binaries as one package 13:30 Which use OS/2 specific functionality when on OS/2 13:30 (it's part of the LE standard, and Watcom can create these type of FAT binaries) 13:30 I *tried* to get the DOS port going, but it's bitrotted 13:31 And DPGPP is ... ugh 13:31 the problem with 16-bit binaries is the 640K limit 13:31 DJGPP is a very easy way around that 13:31 Well, OpenWatcom has DOS/4WG 13:31 (also, the fact that people have been very slapdash about int vs. short vs. long) 13:31 Yeah 13:31 (and even more slapdash about near vs. far!) 13:31 I learned that the hard way, OWCC bitches on mismatch 13:32 Well ... bonus, DOS/4G supports 32-bit 286 protected mode 13:32 so we'd only loose support for XT class machines 13:33 ais523, unless the dev team wants to dump DOS support entirely, my recommendation is to port NetHack to be cross compiled from Linux for DOS with OpenWatcom and it's extender. 13:33 [Demo]5: I don 13:33 't endorse killing fiqs 13:34 And that buys about 90% OS/2 support 13:34 I think at least one variant uses cross-compilation for Windows (Un?) 13:34 ... hell, it actually buys us Win16 port 13:34 ais523, WinCE does that 13:34 Since no native compilation 13:34 ais523: @ YANI, that's something I have considered for candles 13:34 FIQ: doing it for everything would reduce the number of inventory slots required, which would be helpful interface-wise 13:34 ... fuck it, I'm curious how hard it would be to get nethack running under Windows 3.1 13:35 although it'd be very weird if you identified a stack of candles and now it didn't fit in your inventory any more 13:35 ais523: I mean, I've considered merging candles in general even with different lengths 13:35 an old computer I have (which still worked as of several years ago, the last time I tried to turn it on) had a compiler on it that could produce Windows 3.1 binaries 13:36 FIQ: I think NH4 is more aggressive about that than vanilla but it may not be enough 13:36 and having applying them allow you to light/snuff out each of them on its own 13:36 Much of this would cease to be an issue did the 52-item limit also go away 13:36 so you have a stack of candles you can do something with 13:37 "a - a stack of candles (4 lit)" something like that 13:37 the reason I haven't done something like this is that nobody actually uses candles 13:37 because they don't really last, and you don't get that many of them even with the increased amount 13:37 ais523, well openwatcom had Wat386, which allowed 32-bit binaries on WIndows 3.1 13:37 Similar to Win32s, but doesn't require an unmaintained compielr 13:38 ais523: do you understand what I mean? 13:38 Pinkbeast: no, it'd still be an issue; in general the more inventory lines/letters are in use, the harder it is to produce a good interface for them 13:38 FIQ: yes, interface-wise 13:38 ais523, I guess my question is, IMHO, beside my own personal enterainment, is there a desire for the dev team to keep the old ports alive. 13:38 I'm not sure how you'd implement this in turns of duration, unless you're just adding together all the light turns 13:39 in which case, what's to stop you double-lighting the same candle? 13:39 NCommander: within the devteam or outside it? 13:39 both 13:39 I think many of us view it as being something that's nice but not necessary 13:39 ais523: I know what you mean but it does avoid all the cases where Hypothetically Optimal Man holds off on (buc-)IDing to save stacks, or where "we can't do X because it might make the inventory overflow and that's not worth it"... gah, but now HOM will carry 256 named rocks as curse protection 13:39 ais523: oh, I'd still have them all seperate 13:39 [hdf-us] [slshm] luxor (Fla Dop Fem Neu), 6344 points, T:3959, killed by a watchman, while helpless 13:39 not as individual objects, but an object extension (oextra) 13:40 or possibly a container 13:40 in that case, how do you specify whether you light the shorter or longer ones? (you could change how the candelabrum works, then there'd never be a reason to light the shorter ones) 13:40 Or make the candelabrum come with candles. 13:40 . o O (you can't carry more than one lit candle) 13:40 (just like Sting now comes with batteries) 13:40 ais523: I would make applying a stack of candles give a list of them all t hat you can light/snuff individually 13:41 *that 13:41 I had one game where I had to bloody wish for candles 13:41 hmm, this is quite similar to an old UI proposal of mine 13:41 It sucked 13:41 NCommander: this is what the candlegnomes in NH4 are for 13:41 (letting you get 7 candles before leaving the Mines) 13:41 in 3.4.3, I often skip the Mines and just wish for candles 13:42 in 3.6 there are candles in Vlad's, so you don't even have to pay a wish if you want to skip the Mines 13:42 I approve of this trend of using -gnome as a suffix 13:42 Deathgnomes for gnome with the /oD? 13:42 * NCommander had that happen to me finally 13:42 basically, "apply a - a stack of 52 candles (2 lit)" -> "Snuff out/light up which candles? a - a blessed wax candle (0:205), ..." 13:42 something like that 13:43 NCommander: specifically a gnome, and specifically a wand of death? 13:43 Yeah 13:43 the general idea is common, but that specific combination is rare 13:43 For when the RNG just flat out hates you 13:43 FIQ: I thought that sort of specific inventory juggling was unpopular and one of the things that drives people away from NetHack 13:43 although it isn't any worse than in vanilla 13:43 ais523: The main issue with this idea that has made me not bother, besides that nobody actually uses candles, is that it would make candles behave somewhat like gold 13:44 I actually use candles in the mines if I have more than seven 13:44 And no lamp 13:44 people will put so much effort into trying to optimise the decision when it's very unlikely to matter 13:44 in that it's treated as an object in some cases, a stack of non-objects otherwise 13:44 NCommander: Me too (but like you I only want to light one at once...) 13:44 <@Luxcalibur> it would be nice if you could multishot daggers without worrying about only ones with specific names or enchantments stacking 13:45 ais523: The idea is that it would *remove* inventory juggling... 13:45 because all candles would attach themselves as a single item 13:45 irregardless of BUC or length 13:45 @Luxcalibur The better way, IMO, is not to block things merging based on their various known flags. 13:46 FIQ: no, say you want to light a candle 13:46 <@Luxcalibur> right, NH4 does this 13:46 now you have this over-precise list of candles 13:46 and if you're the sort of person who optimises everyone, you make a Reddit post or whatever asking what the best algorithm for lighting candles is 13:46 you want to light a candle? apply, candle list, candle 13:46 <@Luxcalibur> but is there really an in-game reason for a rogue not to be able to throw a -1 dagger after a +0 dagger? 13:46 I don't get the problem 13:46 so you have an inventory-juggling-esque subproblem about which of your many near-identical candles to use 13:47 can you explain how exactly you need/want to optimize this? 13:47 @Luxcalibur flavor-wise no, the technical reason is limitations of the quiver system 13:47 Candles behave like gold; if you've ever had 7 candles, you can use the Candelabrum on the invocation spot, with some fluff text about your candle stubs sputtering to life 13:47 FIQ: Which candle do you want to light? a - a tallow candle (0:100) b - a tallow candle (0:80) c - a wax candle (0:80) d - a wax candle (0:150) 13:47 let's say you have 8 candles of duration 100, 98, 60, 40, 35, 30, 25, 20, 15 13:48 what is it that you want to optimize here? 13:48 FIQ: the fact that it's not obvious just makes the problem /worse/ 13:48 not really a way to specify "after I throw all daggers from this +1 stack, I want to throw this +0 elven dagger, and then the +0 orcish dagger, and then the -1 rusty dagger" 13:48 like, surely there's going to be one candle that's better to light than the others 13:48 and some people will be upset by not knowing which one it is 13:48 ais523: no, because I don't get why you want to optimize anything at all 13:48 why is one better? 13:48 or worse? 13:48 aosdict: Circular around inventory 13:48 high ascension chance if you light that one 13:49 you no take candle 13:49 Pinkbeast: ugh then you have to get into adjusting to define ordering? 13:49 the difference in your asc rate with lighting specific candles is going to be pretty small, but presumably it's there 13:49 [hdf-us] [nh362] dgoddard (Mon Hum Mal Cha) acquired the luckstone from Mines' End, on T:4766 13:49 at least in fh, blessed candles have a light radius of 3 13:49 aosdict: If you care and don't just want to throw whatever comes to hand. You're no worse off than you were now. 13:49 so it is never a good idea to light more than one except in corner cases 13:49 (astral jump, for instance) 13:50 Pinkbeast: it's also not clear whether you should start throwing, say, silver daggers, or arrows 13:50 can you not light 49 uncursed candles for light radius 4? :-D 13:50 or your sword 13:50 (that should really drop off quadratically, not exponentially) 13:50 IIRC I changed the formula in fh 13:50 aosdict: If you throw a foo, you throw other foos 13:50 Come to think of it, this might be a limitation of the *throwing* code. 13:50 1, 2, 4, 8, ... instead of 1, 7, 49, 343, ... 13:50 so it isn't quite as terrible an idea as in vanilla 13:50 but it's still dubious 13:50 anyway 13:50 ais523: As with many things, I have a patch for that if you want it :) 13:51 what I am trying to say is 13:51 you don' 13:51 t need to optimize anything at all 13:51 since you're always just lighting a single candle 13:51 so you do that 13:51 then it goes poof 13:51 then you light another one 13:51 rinse and repeat 13:51 very simple, no strats needed 13:51 FIQ: actually there's a lot of strategy in that 13:51 Second idea; all candles stack. You can only light one candle from a stack; you have no control on what candle is lit (secretly the game lights the shortest). 13:51 because you don't want to spend a turn to light the next candle during combat 13:52 so you need to predict a number of turns in the future at which you won't be in combat 13:52 -!- greqrg has joined #hardfought 13:52 Pinkbeast: then you could just merge the duration of them all 13:52 Only light one candle> and again you can't carry >1 lit candle 13:52 IIRC thats how PSMD handle wand charges 13:52 FIQ: Sure,t hat works. 13:52 Like corpse age merging. 13:52 that's.. not the same thing 13:52 you could have a candle with duration 400 13:52 and another with 400 13:52 merge into 800 duration 13:52 simple and behaves as you said 13:52 FIQ: How do I track if I still have "7 candles"? 13:53 only problem is candelabrum 13:53 how would you get 7 for the candelabrum 13:53 <@Luxcalibur> you can already do this in vanilla by adjusting a single candle to a different letter, so FIQ's system just makes this easier? 13:53 you would have to pick them up and attach them one at a time 13:53 ais523: that's not something you can know on a turn-to-turn basis for anything beyond, say, 10 turns 13:54 FIQ: If the max age for a candle is 400, you could merge one with 300 + one with 200 into "2 candles" with 100 turns remaining 13:54 FIQ: That's why I thought you meant a corpse age style merge, to track the number of candles. (I'm indifferent between this and the game secretly tracking a bunch of individual candles) 13:54 FIQ: that just makes the agonising harder 13:54 then once those 100 turns are up, they go out and obj->quan-- 13:54 aosdict: but what if you have one with 2, another with 52 turns left 13:54 aosdict: Shurely 300 + 200 makes 250? 13:55 Pinkbeast: 300 + 200 = 400 (full candle) + 100 13:55 aosdict: Oh, that's another idea 13:55 rather than 250 + 250 13:55 -!- ais523 has quit [Quit: food] 13:55 if you distributed the ages evenly it would be like your character was cutting off bits of wax and sticking them onto other candles 13:56 who knows, maybe he/she does 13:56 aosdict: It would but no-one would notice or care, just like being near a fresh corpse makes an old one less manky 13:58 [hdf-us] [nh362] shadowcat (Val Hum Fem Law) changed form for the first time, becoming a silver dragon, on T:4589 13:58 ... I mean, this is a _far_ more important thing for corpses but we put up with it. 13:58 [hdf-us] [nh362] hypnotist (Cav Gno Fem Neu) polymorphed her first object, on T:32089 13:59 My averaging of duration also makes it harder not to have 7 candles left than "make as many full candles as possible and one stub" 14:01 -!- spontiff has quit [Ping timeout: 256 seconds] 14:04 [hdf-us] [nh362] dgoddard (Mon Hum Mal Cha) hit with a wielded weapon for the first time, on T:7473 14:04 [hdf-us] [nh362] shadowcat (Val Hum Fem Law) rejected atheism with a prayer, on T:5391 14:10 [hdf-us] [nh362] shadowcat (Val Hum Fem Law), 9029 points, T:5896, killed by a burning scroll 14:12 -!- Chris__ANG has joined #hardfought 14:12 -!- mode/#hardfought [+v Chris__ANG] by ChanServ 14:13 [hdf-us] [nh362] dgoddard (Mon Hum Mal Cha), 19693 points, T:9320, killed by a pyrolisk, while frozen by a monster's gaze 14:13 LarienTelrunya: https://github.com/UnNetHack/UnNetHack/commit/3161230ec and https://github.com/UnNetHack/UnNetHack/commit/aff125955 14:14 bhaak: does that get randomness from some outside source? 14:14 -!- shadowcat22 has quit [Ping timeout: 256 seconds] 14:14 [hdf-us] [slshm] luxor (Val Dop Fem Neu), 4520 points, T:3184, killed by a komodo dragon 14:14 Tangles: I suppose terminfo is broken. would fit the evidence that nh4 derivatives aren't affected. dynahack should then be broken too I guess 14:15 LarienTelrunya: only for the initial seed. because of the dungeon layout reproducability, it must be deterministic after seeding 14:16 <@Luxcalibur> dynahack is completely busted lol 14:19 -!- introsp3ctive has quit [Ping timeout: 252 seconds] 14:19 ha, prediction power skillz 14:26 -!- ProzacElf has joined #hardfought 14:29 @Luxcalibur i know 14:29 K2: Message from @Luxcalibur at 2018-12-09 12:25 EST: hterm is still busted 14:29 K2: Message from hothraxxa at 2018-12-09 13:21 EST: grunthack crashed, now says i have a game in progress 14:29 i'm on it 14:29 looks like i need to re-compile most of not all the variants on th enew box to address some quirks 14:31 was up til 7am working on stuff, crahsed, woke up 3 hrs later and got pulled into a bunch of stuff i didnt want to do 14:31 home finally ;) 14:31 So I was on NHQDB just now and I heard there's green slime changes? 14:31 What's happening? 14:32 patr addressed a report that genociding green slime while turning to slime should cure the sliming. 14:33 Well that's kinda weird but it doesn't sounds like something people would be mad about 14:33 he did not do that, but instead made it so that lifesaving and turning into slime polymorphs you into a slime non-permanently instead, and additionally made it so that if you had genocided green slime as a response to turning into one, you would instadie. 14:33 Ah 14:34 this was unpopular less because of the actual change and more because it's a trend of the devteam taking a report that suggests better behavior and doing something totally different that harms the player, and also because it's inconsistent 14:36 the new template for this is "Report suggested X. I went another direction: [EPI related to X]." 14:36 for your snowcloning use 14:39 BTW, does hdf have the new xnethack yet? 14:39 K2: it's possible that installing a ncurses 5 version could fix it, too. 14:39 PavelB: don't think so 14:40 at least, K2 hasn't said "xNetHack updated" since saying he probably needs to rebuild all the variants 14:40 o shit my script for connecting to hdf went and died 14:40 yeah ssh keys changed 14:40 Ah 14:40 Um, how do I fix this? 14:41 I can't ssh in anymore 14:41 <[Demo]5> delete the known host entry 14:41 ncurses 5 is already installed bhaak 14:41 dynahack plays the the display isnt right 14:41 but the 14:42 [Demo]5: You mean the known_hosts file in .ssh? 14:42 yes 14:43 A line in that file, yes. ssh-ing manually will probably tell you a rune to excise the pertinent line. 14:43 * K2 sighs 14:44 K2: is ncurses 6 also installed? is it picking this up instead of 5? 14:44 chrootlibs.sh doesnt copy anything new, all the required libs are present and in the right locales 14:44 no 6 isnt present 14:44 libncurses.so.5 and libncursesw.so.5 14:45 hmm, then I'm at a loss. my prediction was based on the assumption of ncurses 6 14:45 I just realized the connection between "killed by an exploding rune" spellbook death and D&D Exploding Runes. 14:45 i've dbl checked, the libs between old hdf-us and new hdf-us are identical 14:45 in name 14:46 but going from ubuntu server 16.04 to 18.04, its prob newer versions of same files 14:46 hoping a recompile will do the trick 14:48 Is a morgue a graveyard? 14:48 yes 14:49 IRL morgues don't have graves in the floor... but IRL graveyards don't have dead bodies just lying around aboveground either. 14:49 Eh, maybe the sexton just knocked off for lunch 14:52 What does "Apres moi, le deluge." even mean 14:52 look it up 14:55 after me, the deluge 14:55 very powerful mantra... 14:56 quoting georg schramm: "we ARE the deluge" 14:56 It turns out we were the deluge all along 14:59 -!- captain42 has quit [Remote host closed the connection] 15:05 aosdict: If I find a glass longsword, can I dip for a glass excalibur? 15:06 Um, my game just froze. 15:06 K2: Antichthon reported that he had another slex crash yesterday *sigh* from looking at the ttyrecs, it seems to have been a segfault. Can you check whether there's a core dump? ;) 15:07 prob not 15:07 i asked him to save his game so i could start server migration, he kept playing 15:07 so i had to manually kill his game 15:07 K2: My xnh game on hdf-us just froze. I'm at a >> message prompt but nothing I press makes it do anything. Any idea what's up? 15:07 <@rikersan> pavel: did you try dipping the glass longsword? was that what started it? 15:08 Or what can I do 15:08 curses mode? 15:08 Yeah 15:08 OHHH I know what it is 15:08 yeah 15:08 I still have a few msgtype:alert 15:08 K2: oh, so it might not actually have been a segfault after all, I see 15:08 LarienTelrunya correct 15:08 what version of slex is on master branch? 15:08 And I have it set to msgtype:alert for when my brain is eaten by a mindflayer, and that's what just happened 15:09 <@rikersan> try tab maybe 15:09 no aosdict broke msgtype:alert recently 15:09 <@rikersan> ah ok 15:09 K2: 2.4.2; will probably savebreak again shortly but it's a relatively stable version (fixes a bunch of 2.3.8 crashes), so you can decide whether you want to put it up :) 15:09 soon 15:09 K2: Couldn't terminate one of your stale xnh040 processes gracefully. Force its termination? [yn] 15:10 [hdf-us] [nh362] jmcunx (Wiz Elf Mal Cha), 421 points, T:1444, killed by an invisible gnome lord 15:11 PavelB: I put it back. 15:11 Well, my game froze anyway 15:11 Do I do the force termination thing though 15:12 -!- Antichthon` has joined #hardfought 15:12 anyway artifacts can only be one specific material, dipping a glass longsword would work but it would turn into iron 15:13 @rikersan every mention of glass longswords makes me want to add callandor but I don't want more sword artis :P 15:13 I'm gonna force termination 15:13 well that did nothing 15:13 or you could wait for K2 15:14 Well seeing as how I still get the game in progress warning I don't think it mattered 15:14 Yeah now I just need my game recovered 15:17 !tell hothraxxa had to recover your grunthack game from an older save, day old tops 15:17 No worries, K2, I've got this! 15:19 <@rikersan> aosdict: it's a two-handed sword I believe, not a longsword just fyi 15:19 <@rikersan> (I believe the end of aMoL mentions moridin wielding it with two hands at least) 15:22 well it's "the sword that is not a sword". who even knows 15:23 I don't actually own any of the books so I don't have a reference 15:26 oh. my. gawd. 15:26 hothraxxa: Message from K2 at 2018-12-09 15:17 EST: had to recover your grunthack game from an older save, day old tops 15:27 k2: you have no idea how much i'd done after this restore. huge polypile that involved collecting everything in the castle and had quite satisfying results :( 15:28 this... will be tedious 15:28 sorry :( options were limited to one 15:29 i know, just letting off steam 15:29 maybe the outcomes will be better 15:31 oh shit it's much worse than i thought 15:32 K2 can I get a game recover too 15:33 i already did but dont load it yet 15:33 i'll say when (soon) 15:33 oh ok 15:35 ok i have no idea how far back i've been set but it's a looong way 15:35 hothraxxa :( 15:36 if i can figure out where i am i'll be ok 15:36 i mean, at what point this is 15:36 curses still busted? 15:37 i recompiled grunthack on the new server 15:37 so hoping it isnt 15:37 havent tested yet 15:37 PavelB go for it 15:38 ^-^ 15:38 this is latest xnethack by the way that's why I asked you to wait 15:38 ^ aosdict 15:39 Oh neat it regenerated the level without the mindflayer 15:39 there seem to be ttyrecs up to my crash. i should be able to sort this out 15:39 ...fuck I don't have a musical instrument 15:39 -!- raisse has quit [Ping timeout: 240 seconds] 15:39 hothraxxa curses seems to work 15:39 for gh 15:40 hmm 15:40 ty. that will be a plus 15:40 dark grey is now dark blue 15:40 bhaak: Thanks I'll test dynahack and see if that holds. 15:40 -!- tacco\unfoog has joined #hardfought 15:40 K2: so xnh on server is fully up to date? 15:40 hdf-us yes 15:40 i'll get to eu and au soon 15:40 cool thanks 15:41 k2: curses no work. look at my game 15:41 What do I do here? I've searched all the walls for secret doors. (The ~ are water) https://usercontent.irccloud-cdn.com/file/KexRJgnw/image.png 15:42 there's one behind those pools in the bottom right 15:42 you will need some way to cross water 15:42 Dangit. Up I go. 15:42 or dig down I guess 15:43 ...up I go. 15:43 <@rikersan> pavel: you do know that is medusa right 15:43 haven't touched that level, it's pure vanilla 15:43 @rikersan I've gotten to medusa like a whole two times in my life ok 15:43 <@rikersan> fair 15:43 <@rikersan> you can probably check the wiki for total ?oEarth needed, but cold/ww is gonna be much easier 15:44 <@rikersan> that was merely your reminder to bring blindness/reflection 15:44 dang with luxidream moving from yellow to orange I'm now confusing him with rikersan 15:44 I have a TP scroll but I can see a couple ways that ends poorly for me 15:44 ikr 15:44 Also it's red 15:44 ree 15:45 pavelb: you can dig down then do the popup 15:45 popup? 15:45 <@ChristmIceBox> dig down, find the up-stair and voila 15:45 yes. find the up stairs 15:45 <@ChristmIceBox> face-to-face with the snake lady 15:45 of course might be the castle 15:46 but if i don't have a way across water that's what i usually do 15:46 hoping for levitation boots in perseus 15:46 You know I should probably go back up for a bunch of reasons 15:46 I need a digging tool, a source of blindness, finish sokoban (fucked up earlier and trapped myself) 15:47 <@ChristmIceBox> haha, yeah that'd be good 15:48 hothraxxa i see it 15:48 same i am seeing with xnethack and my game in curses mode 15:48 <@rikersan> aosdict: how are the names gen'd? hashing? 15:48 Maybe I should just wait a bit for curses to get sorted out 15:49 <@riker> does this work better 15:49 @riker: Message from aosdict at 2018-04-27 20:12 EST: "The gnome wields 7 daggers!" :P 15:49 @riker what names 15:49 <@riker> *name colors 15:49 @riker yes you're cyan now 15:49 @riker now figure out what that !tell was in response to :P 15:49 <@riker> is this better for y'all 15:49 -!- ais523 has joined #hardfought 15:49 -!- mode/#hardfought [+v ais523] by ChanServ 15:49 ALL - I wouldnt use curses mode for the time being. working on it right now 15:50 <@riker> it was in regards to buggy offhands attacks aosdict 15:50 <@riker> the dnh bug still exists funnily enough, I left chris a !tell and disappeared and you left one to me 15:50 You should do a MOTD about the migration+curses 15:51 <@riker> aosdict: you underestimate the power of discord search :p 15:51 anyway you would have to ask mtf about name color generation 15:52 Is rld open source 15:52 [hdf-us] [xnh] aosdict (Bar Orc Fem Cha), 74 points, T:78, killed by a sewer rat 15:53 [hdf-us] [nh362] Leggman (Arc Hum Mal Law), 152 points, T:557, killed by a gnome 16:02 haha I'm forgetting that I implemented auto wand ID for obvious messages 16:06 YANI: untrapping a chest also silently reveals its locked state 16:06 Do people commonly #untrap chests 16:07 It's fairly inconvenient to type an extended command and traps are rather rare 16:07 rare but possibly deadly since I haven't made them depth dependent yet 16:08 Never. Max Luck means you need terrible luck to have an issue with a chest trap. 16:08 Pinkbeast: early game tho 16:08 i use u to untrap, and they might be rare but the first time your rings and wands get blown away by a trap you'll reconsider 16:09 aosdict: I never touch chests until I can unlock them; I think there is an interval where I'm opening chests but don't have max Luck and am running a risk, but since it's never caught me out badly enough to leave a scar I'll probably keep doing it. 16:09 *and* you need to specify a direction each time, so you can just do `33u` even if you *do* bind it to u 16:09 also in this grunt game, i have max luck and 18 dex and still had a trap go off 16:10 I want to say s should be able to find door/chest traps in addition to terrain traps 16:10 And then you can't move northeast # yes, I know 16:10 I think Hypothetically Optimal Man untraps chests 16:11 u.y u.y u.y done. what's so hard? 16:11 hothraxxa: numpad? 16:11 For *every chest and door in the game*? 16:11 yes 16:11 Yes, that's hard 16:11 well not every door 16:11 Pinkbeast: I don't, I think a hypothetically optimal player ignores them 16:11 in grunthack, every door :D 16:12 also i only do doors in xnh and grunt. their door traps are nasty 16:12 the risks are higher than the likely reward 16:12 and in grunt, once i have fooproofed everything i stop doing doors 16:12 i think i do that in xnh as well, can't recall 16:13 xnh can only fire blast you if dlvl >= 15 16:13 yeah what about buckets of water on the door, hmm? 16:14 that used to be worse 16:14 not sure if you fixed that 16:14 I did 16:14 it only wets um, d3 items? and doesn't hit containers 16:14 ais523: Altogether, or just "until max Luck"? 16:14 so i still can lose a scroll or spellbook 16:15 Pinkbeast: by luck-max time it's not worth the risk of backtracking for chests you've already skipped; it /might/ be worth opening chests you encounter from then on 16:15 it is not uncommon for me to ascend at +3 or +4 Luck, though 16:15 that is probably non-optimal but I haven't noticed much of a difference 16:15 ais523: I'm a little confused since to me the risk of backtracking would seem to be ~zero 16:16 the more time you spend playing the game the more chance you have of either a) making mistakes, or b) stumbling into an absurd sequence of random numbers that puts you in an absolutely terrible sitaution 16:17 That's certainly true but once a character is well established the risk of either on a trip back to D:1 to open a chest is surely phenomenally small. 16:17 so is the reward from the chest contents 16:18 I dunno it's _that_ small; could well be a potion that'll become FH and I guess I expect the average chest to have something at least nominally useful. 16:18 I mean, sure, that 40th potion of HW doesn't do much, to be fair. 16:20 Hypothetically Optimal Man is a particular kind of optimal, the kind the vanilla DCSS devteam fear; in NetHack, he'll get 9 points of protection from a priest by repeatedly extracting two bits for an ale 16:23 Why do the DCSS team fear Optimal Man? 16:24 * PavelB hasn't played DCSS and doesn't know anything much about it 16:24 potion that'll become fiqhack 16:24 they assume that all their players play like that and will quickly get bored if they don't make that sort of play interesting 16:24 -!- AlbertB has joined #hardfought 16:24 they are probably correct, because such players tend to give up on NetHack quickly and DCSS has been where they traditionally go 16:25 ais523: optimal man untraps chest when traps stop being harmful 16:25 ais523: That is almost completely wrong. 16:26 They take active (some might say excessive) action to prevent you playing like that. 16:26 Pinkbeast: no, they assume that all their players will /try/ to play like that, i.e. do optimal things no matter the cost in boredom 16:26 Pinkbeast: no, that's not what is happening 16:26 so they try to adjust the game so that playing optimally is not boring 16:26 their assumption is that players try for optimal play, always 16:26 the way to do that is by removing pretty much anything that exists in the game that could give you a small advantage 16:26 thus they ensure that optimal play doesn't lead to task they view as tedious 16:27 *tasks 16:27 unfortunately, this has, over time, lead them to remove pretty much the entire game 16:27 and as a result remove a lot of flavour 16:27 <[Demo]5> 45 scrolls of create monster 16:27 <[Demo]5> what should i polymorph them into? 16:27 (I'm exaggerating, but not as much as it might seem) 16:27 Even as someone kind of dissatisfied with the current line of development, that's wildly hyperbolic. 16:27 <[Demo]5> oh? enchant armor? ok 16:27 Pinkbeast: I am being hyperbolic? How? 16:28 This is their entire philosophy 16:28 PavelB: unusually the game has a design philosophy written down in the manual, here https://github.com/crawl/crawl/blob/master/crawl-ref/docs/crawl_manual.rst#n-philosophy-pas-de-faq 16:28 FIQ: ais523 is being hyperbolic; it has obviously not led them to remove "pretty much the entire game", or anything like it. 16:28 Remove things that allow players advantage by playing "strangely"/tediously 16:28 Even if the feature is liked by a large group of players 16:28 This is my issue with DCSS 16:29 Well, that's certainly true to a degree, yes. 16:29 <@Luxcalibur> Is this why DCSS is shorter than nethack? I see absurdly high turncounts in dcss 16:30 FIQ: However, if something that allows players advantage by playing tediously catches on in a big way (pudding farming), it starts to be detrimental to the overall metagame 16:30 <@Luxcalibur> in little more than a few hours 16:30 Lux: There's much wider scope to play longer. 16:30 @Luxcalibur NetHack and DCSS turncount isn't comparable 16:30 DCSS features a lot of backtracking to get things 16:30 ... and NetHack doesn't? 16:30 and inter-level travel 16:31 <@Luxcalibur> Nethack has bags 16:31 this is what ramps up turncount 16:31 Pinkbeast: NetHack has level teleportation 16:31 in DCSS, you inter-level travel 16:31 <@Luxcalibur> I see that’s a rare thing in roguelikes nowadays 16:31 FIQ: Granted, although I've not always had controlled levelport and I don't find my turncount varies that dramatically. 16:32 Pinkbeast: even the most simple DCSS ascension strat involves travelling back and forth in lair and dungeons 16:32 that is a lot of turns spent 16:32 that you don't do in nethack in that way 16:32 FIQ: I know, having done several; and all my NetHack ascensions have involved repeated returns to stashes / altars, just way fiddlier ones. 16:33 another thing 16:33 DCSS levels are larger 16:33 The big thing is surely that most of DCSS is optional so some games have whopping huge turncounts compared to "just go and win" 16:34 oh yeah, extended is *much* larger than a "lightweight" win 16:34 <@Luxcalibur> Do monsters not spawn in completed levels in dcss 16:34 did 15-rune once, took forever to get everything in pan, but I think I was unlucky 16:34 @Luxcalibur they do 16:34 My DCSS wins start around 65,000 turns - NetHack, 40,000. This seems like it's in the same ballpark to me. 16:34 <@Luxcalibur> Or do you stop to fight while inter-level traveling 16:34 and if you stay long enough, they spawn out of depth things 16:34 but the spawn rate isn 16:34 't nearly as high 16:34 No, monster spawns were removed recently. 16:34 as in NetHack 16:34 oh? that's new then 16:35 <@Luxcalibur> in nethack if there were inter level travel you would stop to fight monsters every few turns 16:36 @Luxcalibur: DCSS has always had the concept of a "cleared level"; most monsters in DCSS spawn with the level, in some versions you got spawns later on too but much more rarely 16:36 Pinkbeast: is it possible that lux is seeing chei games? 16:36 chei is absurdly popular for some reason 16:36 and he'd certainly ramp up turncount 16:36 even in minimal wins 16:36 meanwhile, in NetHack, on non-special levels, the levels are often empty or nearly so when you enter them; the vast majority of monsters you encounter were spawned while you were exploring 16:36 this means that a level can never really be cleared, you can't make it much emptier than it starts 16:36 "Turns" in DCSS land is "player turns taken" - Chei doesn't increase it. 16:36 Pinkbeast: ah, didn't know that 16:37 meanwhile, (very) fast speed would make you move more than once per turn in nethack :P 16:37 Yes, this is super confusing but since NetHack measures "moves" I think I am comparing like with like. (Now I look, my quickest NH win is 46,000 moves) 16:37 and the largest part of the game is presumably spent in Gehennom, where everyone is very fast 16:38 Lux: Very high turncounts are probably extended games; high turncounts per clock time might be bot wins, there being at least one winning bot. 16:38 -!- NeroOneTrueKing has quit [Remote host closed the connection] 16:38 Pinkbeast: higher turns vs clock could also point to autoexplore 16:38 -!- NeroOneTrueKing has joined #hardfought 16:38 *everyone* autoexplores in DCSS 16:39 autoexplore + long range autotravel + cleared levels make the turn counter more or less meaningless 16:39 NetHack vanilla doesn't have autoexplore at all, and it's much less useful in NetHack even where the variant does have it 16:39 the only effect it has is on nutrition, which for most characters is rarely a problem in DCSS 16:39 has DCSS removed food yet? 16:39 I know it has been discussed 16:39 it's got to eventually, surely? 16:39 I wonder if, once you get strong enough to survive Abyss, if you could farm Xom 16:40 since there are places where you can survive indefinitely with certain races 16:40 FIQ: I don't know if you have heard this, but pets in 3.6.2 seem to really prefer to hover 4-5 spaces away from me constantly picking up and dropping items. 16:40 and Xom randomly does things 16:40 aosdict: as opposed to? version-wise 16:40 why does Abyss have such a scary reputation? my experience playing DCSS is that it's one of the easiest places in the game 16:40 I hate many of the recent food changes and didn't take them but I think presently we're in a situation where they won't just come out and remove food but probably should. Hellcrawl removes food. 16:41 FIQ: 3.6.0 I guess 16:41 ais523: because monsters get Banishment before Abyss becomes safe 16:41 I suppose? 16:41 ais523: You can easily land there "too early" and there are a number of ways situations can get out of control. 16:41 and it isn't safe at all 16:41 obviously early D is easier, but by the time I'm considering getting runes, the abyssal rule often seems like one of the simplest 16:41 unless you're playing very old dcss I suppose 16:41 aosdict: at some point, DT added a bug fix by me 16:41 making pets reliably follow you 16:41 ais523: That seems extraordinary. (I've been known to do it 3rd but I'm a weirdo) 16:42 3.6.0 doesn't have this 16:42 3.6.2 does 16:42 up until hellephants were implemented, it was pretty survivable with a starting character who was faster than human speed 16:42 actually killing things was hard, but you could just run around picking up items 16:42 FIQ: yes, this is behavior observed in 3.6.2, they don't want to come to me, they want to hang around a pile of items picking up and dropping them 16:42 hellephants ruined that by having a powerful ranged attack, though 16:42 aosdict: the behaviour only triggers at >=5 spaces away 16:43 because the alternative would lead to major issues in corner cases 16:43 IIRC, at least 16:43 [hdf-us] [nh362] Leggman (Bar Hum Mal Neu), 486 points, T:1488, killed by a dwarf 16:43 I did at least consider adding something like that 16:43 let me see 16:43 This must be ages ago! There's plenty of things can nuke you at a distance in the Abyss now. (Also, faster than human... spriggan (fragile), kitty (double fragile), centaur (super noisy) - other than Ce it's a tricky list) 16:44 aosdict: e4db58bdf3376eddf9ad6416c0664975e7787035 16:45 I don't remember what "issues" refer to, sorry 16:45 But I know there was a reason I added it 16:45 -!- LarienTelrunya has quit [Quit: rn2(0) attempted - returning zero Program in disorder...] 16:45 the 5-tile restriction, that is 16:45 Going back a bit I do seem to be significant faster in ascensions with controlled levelport, yes 16:45 Pinkbeast: I haven't played DCSS for ages, indeed 16:45 it's not really my sort of game 16:46 I keep meaning to try DCSS, but eh 16:46 Same with ADOM 16:46 also I don't like the interface, although most DCSS fans won't believe me when I say I find NetHack's to be better 16:46 at least in NetHack all the menus use the same controls 16:46 ais523: I do like ingame help 16:47 And the general lack of spoiler demand 16:47 NetHack for me has the advantage that I remember pre-menustyle NetHack :-) 16:47 FIQ: it used to be worse for spoilers than NetHack, although I think they've been working on fixing that? 16:47 in NetHack at least you can normally figure out what an enemy does by letting it hit you a few times 16:47 * NCommander remembers playing the original rogue 16:47 Pinkbeast: ais523 stopped playing dcss entirely when they removed stairdancing IIRC 16:47 in DCSS that's not advisable 16:47 if you want a point in time 16:48 FIQ: it could have been about that time, yes, although I personally didn't mind the stairdancing removal 16:48 FIQ: I infer ais523 is a Time Lord because it's still in there 16:48 or attempted removal 16:48 wait, not stairdancing 16:48 I meant the exp thing 16:48 whatever it was called 16:48 I remember complaining about energy randomization because it a) reduced the strategic depth of the game and b) didn't have the desired effect 16:48 Want to know a monster's approximate HD, AC, EV; attack damage and spell set? x-amine it. So less spoiler demand. :-) 16:48 a term used for ancient dcss exp abuse 16:48 FIQ: Victory dancing etc? 16:48 victory dancing? 16:48 I don't think he stopped playing because of the abuse removal itself 16:49 but the result of it 16:49 yes 16:49 that 16:49 I did think removing victory dancing removed a lot of flavour from the game 16:49 * NCommander does sometimes play DoomRL 16:49 it was weird being able to train skills without using them at all 16:49 I think victory dancing should have been modified to be nongrindable somehow, rather than removed 16:49 Its one of the few roguelikes I play beside Nethack 16:49 ais523: right, not saying you *liked* the abuse 16:50 I've won DoomRL a few times but never got up to the level of the good players, partly because there seems to be a lot of mechanics cheese in there 16:50 it's the sort of feature that is really awkward and you don't really want to have 16:50 e.g. exp from a monster could be applied to skills immediately based on how you killed it 16:50 but the result wasn't optimal 16:50 Pinkbeast, it's a game though I can just pick up and go 16:50 It's not like NetHack where I need to think a lot about it 16:50 Also no sokoban helps :) 16:50 ais523: that still has issues 16:50 there are games which does that 16:51 NCommander: Oh, sure, I like it a lot. Also, not to brag, but the combat shotgun pump-and-move mechanic is mine. :-) 16:51 and it leads to very strange play at times 16:51 [hdf-us] [nh362] bouquet (Arc Gno Fem Neu), 909 points, T:2561, killed by a homunculus, while sleeping 16:51 letting monsters hit you constaintly to train up armor or similar 16:51 I've never managed a win in DoomRL 16:51 but I don't play it a LOT 16:51 I've gotten to Hell 16:51 which is weird flavour-wise 16:51 (and also tedious) 16:51 NCommander: Not even on ITYTD? Also, last time I won the end boss was the Cyberdemon, so it's been a while. :-) 16:51 !lastgame bouquet 16:51 bouquet: [hdf-us] https://www.hardfought.org/userdata/b/bouquet/nh361/dumplog/1544391139.nh361.txt 16:52 FIQ: you'd probably need separate offensive and defensive exp pools and to base the defensive ones on what armour you were wearing 16:52 FIQ, ... sounds like Dwarf Fortress 16:52 either that, or just not have defensive experience full stop 16:52 DF annoys me. So much to love, so many bugs 16:52 hmm 16:52 DCSS still has victory dancing theoretically, right? 16:52 The "removal" was adding Manual 16:52 FIQ: I think the other thing there is it might be "realistic" but it makes it painfully hard to switch what you do. The current system where you switch to polearms by carrying a polearm around and thinking about it while you stab monsters with your sword is kind of absurd but works. 16:52 to skill distribution 16:53 Just wondering how the old system worked 16:53 FIQ: There's no "unspent XP" now. 16:54 FIQ: the original system was that killing enemies filled an "XP pool", performing actions drained that pool to give you skills 16:54 so first you get the victory, then you dance 16:54 ais523: ah 16:54 so 16:54 <@Luxcalibur> The problem in polearms in nethack is that enemies have move and attack and will happily run around kiting you with them 16:54 that was changed into a system close to the current "skill autotrain" mode 16:54 manual didn't exist, but on top of that, there was a delay 16:54 <@Luxcalibur> But you get at most one use out of them until the enemy is in melee range and it’s completely useless 16:54 then a manual option was added, but you needed an appropriate item to unlock skills (e.g. a long blade to train long blades) 16:54 I've occassionally leveled polearms to basic to deal with ; 16:55 makes sense 16:55 I think I did that with either my ranger or valk 16:55 @Luxcalibur I considered this a bug and sent DT instructions on how to fix it a while back IIRC 16:55 the move+attack thing, that is 16:56 <@Luxcalibur> Heh, is it fixed in FH? 16:56 -!- AlbertB has quit [Quit: Page closed] 16:56 yes 16:56 no moveáttack ridiculousness 16:56 <@Luxcalibur> I assume the dev team didn’t listen to you 16:56 of course they didn't 16:56 biggest problem with NetHack is a lot of the items/stuff/skills are kinda useless :(, there's only a few things that are really worth using 16:56 YANI: while wielding a polearm, moving towards an enemy that's 3 spaces away (meaning that it's now 2 spaces away) attacks it for free 16:56 this was back when they didn't listen to anyone ever 16:57 they still don't to some extent 16:57 but at least they do it somewhat now 16:57 FIQ: "I went a different direction." 16:57 PavelB: that's not the same thing as not responding at all 16:57 They listened to us sometimes! # note the lack of any "and acted upon it" clause :-) 16:57 I don't even know if it's intentional or not 16:57 Because, as usual, there was no reply 16:57 <@Luxcalibur> An artifact that gives move and attack to the player would be pretty neat 16:57 There never was 16:57 back then 16:58 @Luxcalibur and a huge PITA UI-wise 16:58 unless you mean as an invoke 16:58 for a free move 16:58 (and only move) 16:58 <@Luxcalibur> attack only in the same direction the player moved in? 16:59 hmm 16:59 * NCommander downloads DoomRL 16:59 did DT improve polearm UI 16:59 a Necrodancer rapier would be interesting in NetHack 16:59 (after moving, it automatically attacks in the same direction) 17:00 <@Luxcalibur> ah that’s where I got the idea from 17:00 -!- raisse has joined #hardfought 17:00 <@Luxcalibur> I’m a big necrodancer fan 17:00 <[Demo]5> https://i.imgur.com/t1MmOM3.png yey dnh 17:00 I've played it quite a bit, there are some things I like and some things I dislike 17:00 <@Luxcalibur> it’s how I found out about nethack 17:00 it's balanced very differently from a typical roguelike, even in Bard mode; in fact, it's sufficiently different strategically that I think it might be in a different genre 17:01 <@Luxcalibur> All enemies have a movement quirk that lets you defeat them without taking damage 17:01 the fundamental difference between the games being that in Necrodancer, a character with no useful equipment and no levelling up is assumed to be able to beat any encounter in the game if the player is sufficiently skilled 17:01 <@Luxcalibur> But things become more complicated when they attack in groups 17:01 right, that 17:02 even then, it's normally possible, even with minimal equipment, to scout out a group and approach it from a direction where you can take advantage of all the group's AI weaknesses simultaneously 17:02 that's the whole idea of the game I think 17:03 <@Luxcalibur> It’s been a topic of contention whether or not it’s a roguelike 17:03 hm 17:03 they did actually improve it 17:03 a bit, at least 17:03 nice 17:04 <@Luxcalibur> What did they do FIQ 17:04 I prefer my own implementation since it's even smoother :P 17:04 but it's not nearly as bad as in 3.4.3 17:04 @Luxcalibur the fact that the cursor autotargets a monster when asking where to apply 17:04 is a 3.6.x thing 17:04 <@Luxcalibur> Polearms are still unusable because x still takes a turn though 17:04 @Luxcalibur: I hold the somewhat controversial view that Necrodancer isn't a roguelike, but FTL is 17:05 <@Luxcalibur> that is a controversial view 17:05 even then, though, FTL has some roguelite combat aspects, such as door-dancing 17:05 (On top of this, fh also lets you "auto-attack" with F) 17:05 @Luxcalibur ...all weapons take a turn to attack 17:05 FIQ: what are the controls? F followed by some key? just F on its own? 17:06 ais523: F on its own will attack the last monster you attacked with a polearm 17:06 FIQ: I think @Luxcalibur is talking about `x` swap weapon 17:06 <@Luxcalibur> FIQ: you need to autoswap to your melee weapon 17:06 if there is no monster, it will find one to attack 17:06 <@Luxcalibur> Once they get in melee 17:06 which is free in AceHack and NH4 and probably FIQHack and Fourk, but not vanilla 17:06 oh yeah, true 17:06 <@Luxcalibur> This costs a turn, meaning you might just use a sword 17:06 x is not a free action in vanilla 17:06 in several variants, it's free 17:06 <[Demo]5> which it should be :P 17:06 <@Luxcalibur> might as well* 17:06 on top of that, fh specifically has autoswap 17:07 aosdict: is x free in xnh? 17:07 <[Demo]5> it should be :P 17:08 ais523: IIRC fiqhack went even further and made wielding a free action in all cases 17:08 to reduce UI annoyance by the limitation of only having 2 slots of weapons to swap freely between 17:08 FIQ: w becomes quite similar to Q when you do that 17:08 it's certainly a defensible attitude to take, though 17:09 <@Luxcalibur> A good compromise would be to make it not take a turn with a dex/20, similar to twoweapon 17:09 it also made monster wielding a free action 17:09 for consistency 17:09 <@Luxcalibur> dex/20 chance 17:09 I think there's a split in roguelike designers between people who think that players should have to commit to a weaponn before a fight, and people who are happy for optimal play to consist of having a golf-bag of weapons and picking the perfect one for every combat 17:09 they can wield+attack in 100% of the cases now 17:09 (there are corner cases where they can in vanilla, in fh they always can) 17:09 the corner cases is a bug more than anything I think 17:10 <@Luxcalibur> In the current state of vanilla you use a grand total of about two weapons in many games 17:10 ais523: I don't think the latter is a good mindset 17:10 <@Luxcalibur> the aklys buff has mitigated this somewhat, however 17:10 <@Luxcalibur> It’s quite significant 17:10 The reason I made w as free as I did is mostly to reduce cases where you want to juggle what to have in your free vs off-hand 17:11 FIQ: I'm not sure if there is a good solution to this problem 17:11 for melee vs pole vs missile vs pick 17:11 ALL - to use curses interface and have it look normal, set TERM to 'linux' 17:11 ^ hothraxxa 17:11 in practice, saying "it's free, please don't abuse it" normally works quite well but it won't work with everyone 17:11 its not a fix, but a doable workaround. more to follow hopefully 17:11 <[Demo]5> oops 17:11 K2: oh boy, I remember writing entire articles and indeed entire rendering libraries based on this particular topic 17:11 ais523: well, the point is that it's not really something that exist in NetHack 17:11 I'm reading a DoomRL stragety guide 17:11 <@Luxcalibur> K2: current status of hterm? 17:11 FIQ: because most weapons in NetHack suck :-D 17:12 weapons don't really have strengths and weaknesses in NetHack 17:12 [hdf-us] [nh362] Leggman (Arc Hum Mal Law), 1023 points, T:2056, killed by a dwarf 17:12 they're either strictly better or strictly worse 17:12 in most cases 17:12 FIQ: when I wrote the weapon descriptions, I found out that that was actually false 17:12 <@Luxcalibur> FIQ: the 3.6 aklys is really good 17:12 there are exceptions 17:12 there's meant to be a big split between good vs. small and good vs. large 17:12 tridents IIRC have like 1d3/1d12 17:12 but that's an outlier 17:12 it's just that the gain from micromanaging that is really small, especially when you're talking about a difference between d6 and d8 damage and your weapon is +7 17:13 @Luxcalibur what about it 17:13 <@Luxcalibur> It’s a game changer for several roles that lacked a good ranged attack prior to 3.6 17:13 K2: http://nethack4.org/blog/portable-terminal-codes.html 17:14 I do not want to mess with my $TERM to get curses, thank you in advance 17:14 should I start working on porting curses interface to libuncursed again? 17:14 @Luxcalibur what about it -> what did they do 17:14 <[Demo]5> some weird tether mechanic 17:15 <@Luxcalibur> They now can be thrown and return like mjolnir 17:15 oh that 17:15 I think I stole that actually 17:15 <[Demo]5> it also draws some weird line 17:15 FIQ: it appears nh4 and variants are unaffected 17:15 not from vanilla, but from dnethack 17:15 I guess vanilla did too 17:15 nice 17:15 you'd just need to set TERM to linux for 3.x and dynahack 17:15 K2: naturally, since NH4 uses libuncursed 17:15 hence my question 17:15 <[Demo]5> in dnh you can use it as a polearm FIQ 17:15 and also, dont be an asshole 17:16 you not following the convo in the hf dev channel? 17:16 K2: forcing people to set $TERM to something that isn't even standard (almost all terminals use "xterm" or possibly "xterm-256color") isn't good 17:16 I am not 17:16 FIQ: it appears nh4 and variants are unaffected ← this problem came up at /dev/null/nethack one year, so I made sure when writing NH4 that it'd be immune; fixing other games like that is harder though 17:16 should I? 17:16 FIQ we're not forcing anyone to do anything 17:16 K2: I didn't mean to be a meanie 17:17 there's an issue using xterm* as TERM on the new hdf-us server and curses 17:17 but it sounded like you said "guys, you need to do this or no curses for you" 17:17 "and I am happy with this" 17:17 i said it was a doable workaround 17:17 which I took a disliking to 17:17 <[Demo]5> mine is screen lol 17:17 if it's just a work-around, I wont complain 17:17 my guess is that the xterm terminal definition is wrong 17:17 FIQ: in xnethack x swaps main and offhand weapons 17:17 since it's obviously a stopgap 17:17 next time read everything i type ;) 17:17 I still want to port curses to libuncursed at some point 17:18 K2: I did -- in here 17:18 we were missing context 17:18 aosdict: but is it free 17:18 or does it cost a turn 17:18 [17:11] <@K2> its not a fix, but a doable workaround. more to follow hopefully 17:18 K2: hmm 17:19 so backstory 17:19 sorry! 17:19 ais523: Other than "because it's how D&D did it", is there any good reason for why weapons need different damages for small vs large? IMO it'd be simpler to have a unified damage amount and differentiate weapons in other ways (e.g. pierce/slash/whack) 17:19 old hdf-us (and current hdf-eu and hdf-au) run on ubuntu 16.04 17:19 new hdf-us is ubuntu 18.04 17:19 FIQ: it costs an action 17:19 aosdict: I see 17:19 aosdict: I guess it's a fairly intuitive way to differentiate weapons 17:19 there are also pierce/slash/whack differentiations in the game but they're rare 17:19 ais523: a much better way that nethack doesn't actually do 17:19 I think either of these could plausibly be made more common 17:20 is pierce vs slash vs whack 17:20 ais523: it's fairly unintuitive 17:20 FIQ: I think it's best to do both, most likely? 17:20 <@Luxcalibur> Don’t be Chris 17:20 I think the good idea here is to have 6 categories 17:20 <@Luxcalibur> And make skeletons resist bashing damage 17:20 pierce/slash/whack + small/large 17:20 and have weapons based on that 17:21 there's a large assortment of weapons that nobody uses 17:21 if one were to remake the existing weapons based on the above 17:21 I wanted to reuse the sdam and ldam fields for a more general 1dX + Y (or XdY, or 1dX + 1dY, perhaps) 17:21 maybe it could make more weapons viable 17:21 <[Demo]5> dnh has that differentiation 17:21 FIQ: yes 17:21 FIQ: It also encourages golf bag strategy. 17:21 aosdict: indeed, that's the obvious disadvantage 17:21 aosdict: well, weapon skill requirements help with that 17:21 one fix is skill 17:21 <@Luxcalibur> Yes demo, but some of the distinctions like the one I said above are silly 17:22 but skill has very little impact in NetHack past earlygame 17:22 weapon-wise 17:22 <@Luxcalibur> longswords are still the best weapon 17:22 <[Demo]5> blamming weapons vs whamming weapons 17:22 <[Demo]5> vs slicing weapons vs knifing weapons 17:22 @Luxcalibur um 17:22 if anything, they'd resist piercing 17:22 resisting blunt damage makes no sense 17:23 in d&d aren't skeletons *vulnerable* to blunt? 17:23 <@Luxcalibur> ask Chris 17:23 (and by extension liches) 17:23 <@Luxcalibur> I don’t make dnh 17:23 what aosdict said 17:23 <[Demo]5> chris is awol rn lol 17:23 <@Luxcalibur> I thought so too 17:23 <[Demo]5> havent seen him in like a week now 17:23 blunt should have a *bigger* impact than slash, and definitely bigger than pierce 17:23 <@Luxcalibur> It’s finals week 17:23 [hdf-us] [nh362] Leggman (Val Hum Fem Law), 122 points, T:656, killed by a small mimic 17:23 skeletons are vulnerable to blunt in NetHack, yes 17:23 err in D&D 17:23 hmmm 17:23 <@Luxcalibur> So give him another few days? 17:23 if I were to make this system 17:24 it should probably be "blunt is best against hard things, slashing is best against soft things, piercing is best against fleshy things" 17:24 <[Demo]5> for some reason dnh zombies resist... piercing i think? 17:24 <[Demo]5> its stupid and makes me sad 17:25 not stupid 17:25 it makes sense 17:25 <[Demo]5> cause chris already made them spawn in large groups and do double damage 17:25 <[Demo]5> well no imagine running into a large group of lecrottas 17:25 I'd probably make it 0.5x/1x/1.5x depending on if a monster is weak and/or strong to a certain weapon type 17:25 <[Demo]5> that are lecrotta zombies 17:25 <[Demo]5> and move a bit slower 17:25 -!- raisse has quit [Ping timeout: 245 seconds] 17:25 FIQ: gonna YANIfy that 17:25 <[Demo]5> but hit twice as hard and you cant do shit dmg to 17:25 there is no need to make monsters impervious to certain weapon types 17:25 <[Demo]5> and they r cold resistant 17:25 <[Demo]5> like what 17:25 except for stuff like shades 17:25 <@Luxcalibur> Is there zombie ward 17:25 because that *forces* golfbag-style play 17:25 and is a terrible idea 17:26 <@Luxcalibur> or is it the same as the creature letter 17:26 @Luxcalibur yes, whatever repels Z 17:26 I think 17:27 <[Demo]5> i dont know if it actually works 17:27 <[Demo]5> if chris remembered to fix it after moving zombies to Z 17:27 there is justifications for making liches immune to blunt weapons 17:27 not regular skeletons though 17:27 <[Demo]5> liches have MG_VBLUNT oh well 17:27 blunt weapons don't damage the bones 17:28 <[Demo]5> wait thats bonus dmg 17:28 <[Demo]5> they r vulnerable to blunt, hmm 17:29 <@Luxcalibur> I mean you’re not fighting any liches besides the alhoons, and have fun taking those in melee 17:29 <@Luxcalibur> amnesia is a good mechanic 17:29 <[Demo]5> alhoons arent vulnerable to blunt 17:30 <[Demo]5> infact they are resistant to piercing 17:30 <@Luxcalibur> so you sword them 17:31 <[Demo]5> spell them 17:32 the main feature of liches 17:32 is the phylactery 17:32 if there is no phylactery 17:32 why even call them liches 17:34 <[Demo]5> alhoon = lich + ceremorphosis 17:38 <[Demo]5> ok im banned from playing my special dnh game for the rest of the night cause ill accidentally break it 17:38 FIQ: how would phylacteries be implemented? 17:38 special type of amulet? 17:38 <[Demo]5> would work best 17:38 in Brogue, phylacteries are a monster 17:38 <[Demo]5> weird 17:39 that spawn when a lich dies and turn into alich after a while 17:41 aosdict: that's the idea 17:41 spawn an amulet randomly on the level when you also spawn a lich 17:41 perhaps make liches not randomly generated as well 17:41 <[Demo]5> id fight a lich with my phasers 17:41 -!- Lavos has joined #hardfought 17:41 Nethack needs a soundtrack 17:41 (or only make level-creation liches contain phylacteries) 17:41 <[Demo]5> what if you had some like, mage @ that could become a lich? 17:41 YANI: ^ 17:42 (for the phylactery stuff) 17:42 EPI: phylacteries make a hostile takeover on whoever steps on it if the lich is dormant ("dead"), including players which is an instadeath 17:44 <[Demo]5> or u just poly into a lich 17:44 <[Demo]5> lol 17:44 [hdf-us] [nh362] Leggman (Rog Hum Mal Cha), 610 points, T:1985, killed by a rothe 17:44 <[Demo]5> death is dumb 17:44 <[Demo]5> LEPI 17:45 <[Demo]5> less evil patch idea 17:54 -!- Lavos has quit [Remote host closed the connection] 17:55 -!- Lavos has joined #hardfought 17:56 -!- Lavos has quit [Remote host closed the connection] 17:56 -!- Lavos has joined #hardfought 17:57 -!- Lavos has quit [Remote host closed the connection] 17:58 -!- Lavos has joined #hardfought 17:59 -!- Lavos has quit [Remote host closed the connection] 17:59 -!- Lavos has joined #hardfought 18:00 [hdf-us] [nh362] archery (Ran Orc Fem Cha) wished for "blessed wand of polymorph", on T:43633 18:00 [hdf-us] [nh362] archery (Ran Orc Fem Cha) polymorphed her first object, on T:43634 18:00 -!- Lavos has quit [Remote host closed the connection] 18:01 -!- Lavos has joined #hardfought 18:02 -!- Lavos has quit [Remote host closed the connection] 18:02 -!- Lavos has joined #hardfought 18:03 -!- Lavos has quit [Remote host closed the connection] 18:04 -!- Lavos has joined #hardfought 18:04 NCommander: Nethack Legacy does 18:05 -!- Lavos has quit [Remote host closed the connection] 18:05 -!- Lavos has joined #hardfought 18:05 <@Luxcalibur> if they have a built-in speedrun timer I'd consider playing it 18:06 <@Luxcalibur> does the crossed out text show up on irc 18:06 <@Luxcalibur> anyway, I wonder how hard it would be to actually put in a realtime timer on the status line 18:06 -!- Lavos has quit [Remote host closed the connection] 18:07 -!- Lavos has joined #hardfought 18:07 nope no crossout on irc 18:07 <@Luxcalibur> rip 18:07 how much demand is there for an in-game realtime timer? 18:08 this is the sort of thing I'd like to implement (although the status bar is very short on room at the moment!) 18:08 -!- Lavos has quit [Remote host closed the connection] 18:08 ais523: animated? 18:08 one problem is that you pretty much have to rewrite the game's UI loop 18:08 -!- Lavos has joined #hardfought 18:08 -!- Lavos has quit [Remote host closed the connection] 18:08 in order to be able to update the screen /while/ waiting for input 18:08 animated is probably too much 18:09 incidentally, a while back someone on Reddit suggested a version of NetHack where nutrition runs down in realtime, not gametime 18:09 just show its value at each redraw 18:09 I think that might be fun too 18:09 but in that case the timer would pretty much have to be animated, it wouldn't be as fun otherwise 18:09 <[Demo]5> that would suck 18:09 (the timer should also run far too fast for most people to keep up with it) 18:09 <[Demo]5> i would never play that 18:09 <[Demo]5> veto 18:09 [Demo]5: this would never be an "official" / "main" game mode, just a bit of fun 18:10 <[Demo]5> yeah :P 18:10 ais523: you would basically be cutting off at some arbitrary value of speedrunability 18:10 <[Demo]5> except for the part where it wouldnt be fun :P 18:10 <[Demo]5> just stressful 18:10 [Demo]5: Think about it. Could you complete a game of nethack without eating IRL and still not have starved to death by the end? 18:10 <[Demo]5> yeah i could 18:10 <[Demo]5> i could probably ascend in 5 hrs? 18:10 <[Demo]5> and i can definitely not eat for 5 hrs 18:10 PavelB: I could, I regularly go ten hours without eating and most of my NetHack completions are faster than that 18:10 <[Demo]5> idk ive never done it 18:10 I could ascend without eating but I would be hungry afterwards 18:11 <[Demo]5> ^^ 18:11 I daresay you could complete a game without drinking, although you might be a wee bit thirsty by the end 18:11 <[Demo]5> i havent ascended in months cause i get chars halfway through gehennom and lose itnerest 18:11 <[Demo]5> dude man i couldnt complete a weekend without drinking, none the less a game 18:12 luckily games are faster than weekends for many people 18:12 Earlier I said something about showing gametime in the c-x menu and in whereis, and Tangles said he might do that 18:13 adding things that the game tracks to ^X is pretty much trivial 18:13 that's half of the reason why so much stuff ended up there in AceHack :-D 18:14 You have the power to port some of these things to vanilla 18:14 yes, but with many features, it's important to get them right first time 18:15 I'd personally prefer an animated realtime timer on the status bar (or maybe message area) if we do one at all, for example 18:15 although as an option 18:16 Howabout how many turns ago you prayed 18:16 Can we get that 18:16 You could probably take the xnh commit for it 18:17 I'd guess that there'd be people who would be upset by that information being added 18:17 either due to spoiling unspoiled players on how the game works or by removing the incentive to track it manually 18:18 that doesn't mean it's a bad idea, but I only have so much effort for trying to push these changes through 18:18 the prayer confirmation prompt might be a good place for it, actually 18:18 "You last prayed 100 turns ago. Really pray?" 18:19 that way the sort of traditionalist who plays !prayconfirm won't see it, and it's not really spoiling things any more than the existence of prayconfirm already does 18:19 except, hmm 18:19 I think the game doesn't directly track the turncount of the previous prayer 18:20 ais523: removing the incentive to track it manually? does anyone *like* the incentive to track it manually? 18:20 meaning that we can't implement this until 3.7 because it'd mean adding a field to the save data 18:20 aosdict: there's at least one game which had the absence of auto-mapping as a selling point, so presumably yes 18:20 https://github.com/copperwater/xNetHack/commit/c19501ded1b6ffef8ae8aef4309cd70a97088bb4 18:20 PavelB: of note, it was savebreaking. 18:21 How does the game currently track how long ago you last prayed? 18:21 <[Demo]5> u.u??????? 18:21 ...the commit you linked? 18:21 blesscnt 18:21 No, vanilla 18:21 <[Demo]5> close enough 18:21 then yes blesscnt 18:21 <[Demo]5> lol 18:21 It needs to know somehow that it's not safe to pray 18:21 ublesscnt is prayer /timeout/ 18:22 but you can't deduce the previous pray time from it 18:22 blesscnt is how many turns remaining until it's safe to pray, and is how enlightenment can tell you whether it is or not 18:22 Ah 18:22 because a) the timeout is randomized somewhat and b) in-game actions (e.g. wishing) can affect it 18:22 incidentally, the comment by ublesscnt implies that it was originally intended for something else, but apparently morphed into prayer timeout over time? 18:23 <[Demo]5> why does wishing affect it? 18:23 <[Demo]5> also rnz sucks 18:23 wishing has a minor prayer timeout penalty IIRC, presumably because the deities think you shouldn't need to pray if you have wishes 18:24 oh hey I finally have a direct commit in vanilla :D 18:24 <[Demo]5> stupid gods 18:24 That comment dates back to at least 1.3d, FWIW 18:24 the "rnz sucks" comment? 18:25 presumably the comment about ublesscnt 18:25 nethack-1.3d/you.h: int ublessed,ublesscnt; /* blessing/duration from #pray */ 18:25 oh I missed a message 18:25 ublessed appears to be AC from intrinsic protection 18:26 ais523: isn't the "blessing" part referring to ublessed? 18:26 and the "duration from #pray" referring to ublesscnt? 18:26 aosdict: I see "blessing" as referring to ublessed, "duration" as referring to ublesscnt, and "from #pray" as referring to both 18:27 if they were unrelated they wouldn't be grouped like that 18:27 Well, you can get intrinsic protection from praying. 18:27 I find it interesting that ublessed is just a regular int, and not an xchar or something. 18:27 And I think in 1.3d you can only get it from prayer. 18:27 <[Demo]5> i will pray to god for jesus if he is happy for me and if not i will get smited and he will send his leprechauns 18:27 <[Demo]5> thanks nethack 18:28 that is certainly a sentence 18:29 aosdict: is it possible for intrisic protection to be negative (e.g. eating negatively enchanted =oP)? 18:29 <[Demo]5> testing my new ai integration into my irc to discord forwarder to autorespond for me so i never have to think again 18:29 if so it'd have to be at least an schar 18:29 [hdf-us] [nh362] hypnotist (Cav Gno Fem Neu) performed her first genocide (class R), on T:34574 18:32 ais523: I just ate a -5 ring of protection and now have 15 AC, so... yes 18:34 @message LarienTelrunya new race/class idea: starts with intrinsics like negative protection, negative increase damage, negative increase accuracy 18:34 !tell LarienTelrunya new race/class idea: starts with intrinsics like negative protection, negative increase damage, negative increase accuracy 18:34 I'm on it, ais523. 18:34 <[Demo]5> thats slex 18:34 <[Demo]5> it has that 18:34 <[Demo]5> i think 18:34 <[Demo]5> oh ur talking to slex 18:34 <[Demo]5> just play kurwa it probably is that 18:35 slex needs more than three alignments really 18:35 <[Demo]5> agree 18:35 <[Demo]5> needs an elf alignment 18:35 <[Demo]5> and an elf gender 18:35 <[Demo]5> so i can play as elf the elf elf elf elf 18:38 Order of the Stick establishes precedent for an "elf" gender. 18:39 <[Demo]5> ive seen nuts chicks with elf ears before 18:39 <[Demo]5> it might be a gender 18:40 <[Demo]5> cant disprove 18:44 last oots issue I read had a cute dwarf baby in it 18:44 probably missed 100 issues since 18:46 [hdf-us] [nh362] archery (Ran Orc Fem Cha) bribed Asmodeus with 910 zorkmids for safe passage, on T:44987 18:47 [hdf-us] [nh362] archery (Ran Orc Fem Cha) genocided class h, on T:45048 18:49 FIQ: You probably want to catch up, it's been interesting 18:50 [Demo]5: NetHack used to have an elf role 18:50 xNetHack updated (all servers) 18:52 Didn't you already update xnh earlier? 18:52 Oh, all servers 18:59 [hdf-us] [nh362] Leggman (Arc Hum Mal Law), 208 points, T:356, killed by a gnome 19:02 <@Luxcalibur> ais523 backscroll: how many minutes to go from 900 nutrition to not hungry? 19:02 15 19:02 <@Luxcalibur> 1 nutrition per second? 19:02 yes 19:03 I think it should go faster, at least later in the game 19:03 early game might be a problem with particularly fast speeds, though 19:03 make it a function off turncount 19:03 <@Luxcalibur> if it took 15 minutes to get to hungry from a food ration, speedruns would need far less food 19:03 discourage high-TC :) 19:04 either it'll have to run faster as the game goes on or the player will have to start with a nutrition buffer or rations in inventory 19:04 <@Luxcalibur> how do you account for fighting hunger/casting hunger 19:04 it doubles when you get the Amulet, IIRC 19:04 <@Luxcalibur> and ring hunger 19:04 <@Luxcalibur> does that just not exist too 19:04 perhaps it should double in Gehennom too 19:04 ais523: no it doesn't 19:04 it goes up by x1.05 19:05 presumably, rings increase nutrition drain rate, fighting/casting hunger just remove nutrition points directly 19:05 [hdf-us] [nh362] Leggman (Bar Hum Mal Neu), 239 points, T:573, killed by a small mimic 19:05 FIQ: haha, seriously? 19:05 only hunger, regeneration and conflict drastically increases it (x1.5) 19:05 yes 19:05 that seems like something which could be made harsher without negatively affecting the game 19:05 it doubles luck timeout 19:05 that might be what you're thinking of 19:05 also Pw costs 19:05 and prayer timeout 19:05 obviously, people use luckstones at that point 19:05 no, pw cost is 2-3x 19:05 prayer timeout is tied to Rodney 19:05 [hdf-us] [nh362] eraserhead97 (Rog Orc Mal Cha), 619 points, T:759, killed by a dwarf 19:05 <@Luxcalibur> actually force players to bring food? 19:06 tbh, doubling it wouldn't have a significant effect on standard strategy 19:06 <@Luxcalibur> I think a lot of people forget to eat corpses on the ascension run 19:06 you need something like two rations for the asc run, more if you get Famined 19:06 doubling that to four does not make a noticeable difference 19:06 <@Luxcalibur> there was a recent reddit post where someone starved on water starting the run with 12 c-rations 19:06 oh, Water 19:07 I usually don't eat corpses at all once I get to Gehennom 19:07 that explains a lot 19:07 I find it too annoying UI-wise 19:07 and the endless K/C-ration supply is enough generally 19:07 FIQ: I have almost starved in Gehennom before now but that's because I was being me and thus not carrying rations 19:07 coupled with =oSD 19:07 oh, =oSD is cheating :-P 19:07 <@Luxcalibur> wish there was a way to eat while ignoring distractions 19:07 ais523: I usually don't have anything else I want to use 19:07 <@Luxcalibur> taking off armor in vanilla for some reason can't be interrupted by attack 19:07 <@Luxcalibur> s 19:08 <@Luxcalibur> but eating can 19:08 but I think you can get away with it even without SD if you don't cast 19:08 <@Luxcalibur> ? 19:08 FIQ: I normally don't either, but then I normally don't have the =oSD in the first place 19:08 so I normally run into Gehennom either with random resistance rings or with a ring finger or two empty 19:08 especially as I rarely have a full set of intrinsics by the time I reach Gehennom 19:08 Lux: If you're halfway out of your armour and something attacks you, you can't really do much until you finish disrobing (well, ish, but it'll do as a rationale). Lunch, you can leave halfway. 19:08 [hdf-us] [nh362] archery (Ran Orc Fem Cha) killed Juiblex, on T:46593 19:09 (I'll typically have the /important/ ones but there are lots of minor resistances that I won't have bothered to get yet) 19:09 @Luxcalibur: actually, whether taking off armour is interruptible depends on the commands you use; A is interruptible, T isn't 19:09 <@Luxcalibur> oh, interesting 19:09 Naturally 19:09 A is also faster (i.e. fewer turns used) in the vast majority of circumstances 19:09 Consistency is evil 19:09 although T can be faster in corner cases 19:09 oh wow I forgot about that oddity 19:10 <@Luxcalibur> ...why exactly does this happen again? 19:10 in NH4, T is basically just a wrapper around A, so this doesn't happen 19:10 Different codepaths 19:10 they don't share code 19:10 3.6.x is even worse 19:10 different codepaths for P/R vs W/T 19:10 but both are usable in all cases 19:10 Seems like protection, free action, warning, increase damage, shape changers, telecontrol would all be better than slow digestion. Maybe even regeneration... 19:10 leading to slightly different behaviour for them 19:10 and individual bugs 19:11 why would you even do something like that 19:11 Pinkbeast: I'll normally wear a =oRegen if I find and identify one (which doesn't happen often) 19:11 <@Luxcalibur> that's something I won't ever understand about nethack, the P/R W/T split 19:11 I don't get it 19:11 <@Luxcalibur> who decided this was necessary? 19:11 ais523: I guess it's better than SD but it doesn't do much at that juncture. 19:11 @Luxcalibur: that split in 3.4.3 makes sense from a code-internal point of view, P and W have very little to do with each other in terms of what you need to do codewise to implement them; it doesn't make much sense from a UI point of view though 19:12 in NH4, PR vs WT is still a split 19:12 Lux: No-one. In the mists of time someone decided to write commands that dealt with one kind of object 19:12 but only to show different items as "recommended" 19:12 some people like the split because they like the fact that P is single-action, W is longer 19:12 but that is the only difference 19:12 otherwise it's all the same codepath 19:13 hm 19:13 so WT is not an occupation in NH3? 19:13 given what @Luxcalibur said earlier 19:15 no, it's helplessness + aftermv 19:15 aftermv? 19:15 code that runs when helplessness ends 19:16 ah 19:16 in at least 3.4.3, you can actually screw this state up and move around while the game thinks you're in the middle of equipping something 19:16 <[Demo]5> sounds like aftershave 19:16 and then cancel a later helplessness when the item you were "trying to equip" gets stolen 19:16 but it's fragile, equipment storage is lost fairly easily 19:16 but lifesave terminates it, no? 19:17 I think it might lifesaving that /causes/ that state to happen 19:17 there was a reason it wasn't usable in the speedrun 19:17 *might be 19:17 it wasn't usable in the TAS, the reason was different though I think 19:17 could just be that you can't do it as an air elemental 19:17 probably can in 3.6.0 19:18 IIRC there was a bug with WT being slower than PR for blindtool 19:18 in 3.6.0 19:18 which air elementals can use 19:19 <[Demo]5> isnt there a bug where likes_lava monsters can walk in lava without their shit burning? 19:19 hmm 19:19 <[Demo]5> or is that not a bug> 19:19 <[Demo]5> or is that fixed since 3.4.3? 19:19 did NH4 fix the free curse checking of weapons for people who can twoweapon? 19:19 in NH4+friends, you can use A to put a cursed weapon in your offhand 19:20 without it welding to you 19:20 twoweapon then lets you know that it's cursed 19:20 without it getting stuck 19:20 I don't think I fixed that 19:20 it's not that major an exploit 19:21 right, it slips 19:21 that's what it did 19:21 you can test for cursedness of daggers by throwing them 19:21 meh 19:22 maybe even non-throwing-designed weapons 19:22 was juist checking the behaviour 19:22 dagger was the first weapon that came to mind 19:22 K2: shouldn't it tell gamemode? 19:22 for livelogging 19:22 the above can be misinterpreted as a T:1 wish in a normal game 19:23 which is notable 19:23 it isn't in explore mode 19:23 ais523: no 19:23 it doesn't work on that 19:23 but dagger is kind of both, so I don't know there 19:26 [hdf-us] [nh362] hypnotist (Cav Gno Fem Neu) entered Gehennom, on T:36694 19:27 FIQ: Luxidream was exploiting that a while back 19:27 spamming the channel with explore mode wishes as performance art 19:27 heh 19:28 reminds me of the wish rickroll in nhqdb 19:29 <@Luxcalibur> someone in ootc asked me to test fruitname wishes :p 19:29 <[Demo]5> i can spam the channel with dnh wishes from thrones that are cheated abusing autorecover 19:29 oh, fruitnames 19:29 right, because livelog doesn't normally log the wish if it doesn't parse correctly? 19:30 aka the reason for 90% of pluralize 19:31 [Demo]5: dnh *still* has that problem? 19:31 <[Demo]5> as long as there is a reliable crash bug 19:31 <[Demo]5> and who would know a reliable crash bug but the ndnh developer? 19:31 I suggest NH4 save system, or at the very least per-turn checkpointing (rather than per-level) 19:31 <@Luxcalibur> yeah. I found you couldn't wish for your fruitname if its the name of an artifact, and if BUC and other modifiers are part of the name, then it also affects the fruit 19:32 <[Demo]5> wishing sucks 19:32 <[Demo]5> wish parsing 19:32 @Luxcalibur kind of reasonable unless the alternative is no match 19:32 the wish parser is at least fun to mess around wiith 19:32 fruitname should never be able to override other things 19:32 also, you probably don't want people to be able to stick artifact fruits into each others' games via bones to mess up artiwishes 19:32 if you're going to grief, suck it up and get the actual artifact 19:33 where did the term grief come from anyway 19:33 <@Luxcalibur> why would the fruit be counted as an artifact? 19:33 I used to believe it originated in Minecraft 19:33 clearly not 19:34 <[Demo]5> haha 19:34 (grief as in destroy for others, not as in the mental state) 19:34 <[Demo]5> yeah 19:34 @Luxicalibur: I mean you create a fruit called "Magicbane" or whatever so that people trying to wish for it, but who have your fruit in their bones, get the fruit instead 19:35 FIQ: I first saw the term in connection with Second Life 19:35 there was a particular flying penis attack that made the mainstream news 19:35 lovely 19:35 <[Demo]5> i think its just common from minecraft cause it was so easy there 19:35 second life is one of those weird games where you can actually gain real life money from it directly 19:35 <@Luxcalibur> does finding a bones file automatically make the deceased person's fruitname wishable? 19:35 like, there was some very important and formal occasion in-game, and a griefer who thought that people were taking it too seriously sent a horde of flying penises in 19:35 at least it used to be 19:35 that's all I know about it 19:36 @Luxcalibur: yes, unless the fruitname buffer is full already 19:36 even if there isn't any fruit in the level? 19:36 oh, there probably has to be fruit on the level 19:36 otherwise the fruitname likely wouldn't even be saved 19:36 fh abuses something with fruits as a version string 19:36 in bones 19:36 because flags isn't saved in bones 19:37 ah right 19:37 I use count 19:37 <@Luxcalibur> I assume this goes for any other fruitname too, like the "dwarven roots" you find in 3.6.2 orctown 19:37 which for whatever reason was saved as a 2-byte string 19:37 when it caps at 255 19:38 @Luxcalibur: yes, it's tied to the fruitnames in the bones so it even works recursively, if someone gets an "outside" fruit from someone else's bones and then dies and someone gets the double-bones fruit 19:38 [hdf-us] [nh362] Leggman (Bar Hum Mal Neu), 787 points, T:2566, killed by a rothe, while frozen by a monster's gaze 19:39 fourk does that too 19:39 use fruitname to add flavour fruit 19:39 For holidays mainly, and for scurrier role, yes. 19:39 scurriers get some unique fruitnames 19:39 Scurriers get chestnuts. 19:39 [hdf-us] [nh362] archery (Ran Orc Fem Cha) bribed Baalzebub with 2588 zorkmids for safe passage, on T:48498 19:39 Which are just specially named fruit. 19:40 And on certain holidays you can find things like candy corn, etc. 20:00 -!- shadowcat22 has joined #hardfought 20:11 [hdf-us] [nh362] eraserhead97 (Cav Dwa Fem Law), 311 points, T:1060, killed by a boulder 20:12 [hdf-us] [nh362] shadowcat (Val Hum Fem Law), 1169 points, T:1632, killed by a hill orc 20:20 [hdf-us] [nh362] eraserhead97 (Mon Hum Fem Neu), 455 points, T:545, killed by a hobbit 20:24 [hdf-us] [nh362] shadowcat (Bar Hum Fem Cha) chose an elven dagger to be named "Sting", on T:2351 20:40 -!- hpardis has joined #hardfought 20:43 [hdf-us] [nh362] shadowcat (Bar Hum Fem Cha) averted death, on T:4576 20:45 -!- Antichthon` has quit [Ping timeout: 246 seconds] 20:49 [hdf-us] [nh362] shadowcat (Bar Hum Fem Cha), 8015 points, T:4986, killed by a mumak 20:50 !whereis k2 20:50 K2: [hdf-us] k2 [nh343]: (Hea Gno Mal Neu) T:1 The Dungeons of Doom level: 1 21:03 [hdf-us] [nh362] CropperS (Wiz Elf Mal Cha) had Magicbane bestowed upon him by Anhur, on T:21474 21:05 -!- tacco\unfoog has quit [] 21:11 -!- Bartlem has joined #hardfought 21:16 ALL - you dont have to switch TERM setting to linux to use curses anymore. can use xterm or xterm-256color again without it screwing up the display 21:17 I wonder what the default it 21:17 I didn't know ssh even sent environment data 21:17 depends on what you're using 21:18 OpenSSH? 21:18 [hdf-us] [nh362] Luxidream (Val Dwa Fem Law), 5206 points, T:4095, quit 21:19 [hdf-us] [nh362] Luxidream (Val Dwa Fem Law), 47 points, T:147, killed by a sewer rat 21:19 [hdf-us] [nh362] shadowcat (Bar Hum Fem Cha) chose an elven dagger to be named "Sting", on T:3730 21:19 depends on what terminal you're using 21:20 How does Hardfought know about that 21:20 within the chroot, terminfo folder and contents are mirrored from whats on the server 21:21 whole list of different ones 21:21 I was just wondering how the server would know about the terminal I'm using 21:21 it doesnt, but each terminal would probably have its own default TERM value 21:21 I presume your ssh client voluntarily tells it 21:22 Ohh, I get it. No, what I was wondering was what ncurses assumes it should do if it can't read TERM properly 21:24 [hdf-us] [nh362] CropperS (Wiz Elf Mal Cha) acquired the luckstone from Mines' End, on T:22859 21:28 [hdf-us] [nh362] Luxidream (Val Dwa Fem Law) had Excalibur thrown at her by some watery tart, on T:3532 21:30 [hdf-us] [nh362] hypnotist (Cav Gno Fem Neu) killed the invisible Juiblex, on T:38266 21:33 [hdf-us] [nh362] Luxidream (Val Dwa Fem Law) became literate by reading the fortune inside a cookie, on T:4979 21:35 -!- Bugpowder has joined #hardfought 21:35 [hdf-us] [nh362] Luxidream (Val Dwa Fem Law) completed Sokoban, on T:5860 21:36 [hdf-us] [nh362] CropperS (Wiz Elf Mal Cha) had Fire Brand bestowed upon him by Anhur, on T:23581 21:37 [hdf-us] [nh362] shadowcat (Bar Hum Fem Cha) averted death, on T:6083 21:39 [hdf-us] [nh362] CropperS (Wiz Elf Mal Cha) had Werebane bestowed upon him by Anhur, on T:23934 21:39 -!- Antichthon has joined #hardfought 21:41 [hdf-us] [nh362] shadowcat (Bar Hum Fem Cha), 15645 points, T:6595, killed by an air elemental 21:41 -!- Bugpowder has quit [Quit: Bugpowder] 21:42 -!- MisteryMyra has joined #hardfought 21:42 -!- Bugpowder has joined #hardfought 21:42 -!- Antichthon` has joined #hardfought 21:43 Is the hterm working on the US server for anyone else? 21:43 [hdf-us] [nh362] hypnotist (Cav Gno Fem Neu) killed Baalzebub, on T:39094 21:44 Bartlem: The hosting arrangement for the US server changed earlier today, so if you're having an issue you haven't had before, it could be related to that. 21:44 ah 21:44 (Or was that yesterday at this point?) 21:44 It works on the other servers 21:44 I thought K2 said it was working. But 21:45 -!- MiseryMyra has quit [Ping timeout: 245 seconds] 21:45 hterm is still down on hdf-us 21:45 Ah. 21:45 its being... difficult 21:45 Permissions issues? 21:45 error log shows http 404 error which is stupid 21:46 Hmm. That is odd. 21:46 Unless you forgot an alias in the Apache config or something. 21:46 yeah i've played this game with apache2 proxy settings before 21:46 clearly it should have 418ed instead of 404ing 21:46 new server has newer version of apache2 so something may be a bit off 21:46 Apache 1.x was *so* much easier to configure for a single-site, no-virtual-hosts setup. 21:47 -!- Antichthon has quit [Ping timeout: 268 seconds] 21:47 jonadab the settings are exact from the old hdf-us server 21:47 I understand why 2.0 made the changes it did, but. 21:47 and its also verbatim from hdf-eu and au 21:47 Huh. 21:47 exactly 21:47 Bartlem i'm on it, its on the to-do 21:47 [hdf-us] [nh362] shadowcat (Bar Hum Fem Cha), 240 points, T:617, killed by a bat 21:49 Any recommendation of another terminal to use that supports xterm on windows? 21:50 putty is a good choice 21:50 stick around, i'll let the channel know when hdf-us hterm is back up and running 21:52 [hdf-us] [nh362] Luxidream (Val Dwa Fem Law) performed her first genocide (class L), on T:9482 21:52 [hdf-us] [nh362] CropperS (Wiz Elf Mal Cha) had Dragonbane bestowed upon him by Anhur, on T:27721 21:52 [hdf-us] [nh362] Luxidream (Val Dwa Fem Law) polymorphed her first object, on T:9493 21:54 -!- MisteryMyra has quit [Read error: Connection reset by peer] 21:55 -!- MisteryMyra has joined #hardfought 21:56 -!- Bugpowder has quit [Quit: Bugpowder] 21:57 -!- Bugpowder has joined #hardfought 21:57 Bartlem: WSL 21:57 (Windows Subsystem for Linux) 21:59 [hdf-us] [nh362] shadowcat (Bar Hum Fem Cha), 1459 points, T:2074, killed by an Uruk-hai 21:59 [hdf-us] [nh362] hypnotist (Cav Gno Fem Neu) destroyed Vlad the Impaler, on T:40340 21:59 [hdf-us] [nh362] hypnotist (Cav Gno Fem Neu) acquired the Candelabrum of Invocation, on T:40341 22:00 Bartlem: There's also MobaXterm. Or, if you are really ambitious, you can always Cygwin it up. 22:00 But Cygwin is kind of overkill for this specific use case. 22:02 Cygwin is such a pain to configure it's easier to setup actual Linux (WSL) 22:02 -!- Antichthon has joined #hardfought 22:03 All of these sound a lot more effort than downloading PuTTY 22:04 WSL takes longer for the actual download but I wouldn't say it's more effort 22:05 It has basically no configuration beyond username+password 22:06 -!- Antichthon` has quit [Ping timeout: 246 seconds] 22:06 I started on the WSL install. Putty and curses interface don't seem to work together for me 22:06 https://docs.microsoft.com/en-us/windows/wsl/install-win10 I'm 4 lines in when I have to reboot 22:07 -!- Antichthon has quit [Ping timeout: 250 seconds] 22:07 -!- Bartlem has quit [Quit: Page closed] 22:09 [hdf-us] [nh362] Luxidream (Val Dwa Fem Law) made her first wish - "2 blessed greased scrolls of charging", on T:9635 22:09 [hdf-us] [nh362] Luxidream (Val Dwa Fem Law) wished for "blessed greased fireproof bag of holding", on T:9635 22:10 [hdf-us] [nh362] Luxidream (Val Dwa Fem Law) wished for "blessed greased ring of levitation", on T:9648 22:11 -!- MisteryMyra is now known as MiseryMyra 22:12 [hdf-us] [nh362] Luxidream (Val Dwa Fem Law) entered Gehennom, on T:10112 22:14 -!- Bartlem has joined #hardfought 22:17 Pinkbeast: PuTTY is a terrible terminal, though. 22:17 ... it is? 22:17 Well, I mean, it's relative, I suppose. I'm sure there are worse ones. 22:17 It's not wincon. 22:17 [hdf-us] [nh362] Luxidream (Val Dwa Fem Law) genocided master mind flayer, on T:10620 22:17 But it ain't Konsole either. 22:18 [hdf-us] [nh362] Luxidream (Val Dwa Fem Law) killed Medusa, on T:10763 22:19 Does anyone still use just /usr/bin/xterm 22:19 PavelB: Cygwin can do some things that WSL cannot do. (Most prominently, it can run Windows-native programs; this means it can be used to build software that has a multi-stage build process and generated code, such as NetHack for instance.) 22:19 /dev/false 22:19 PavelB: Yes. 22:19 PavelB: A few people. But mostly no. 22:20 jonadab: Build it on WSL and then run it on WSL :3 22:20 [hdf-us] [nh362] hypnotist (Cav Gno Fem Neu) killed Orcus, on T:41955 22:20 My Clion is configured to use WSL 22:20 -!- Bartlem_ has joined #hardfought 22:20 -!- Bartlem has quit [Ping timeout: 256 seconds] 22:21 PavelB: I had to give up on WSL as a build environment for Fourk. Ended up using Strawberry, instead. I have one user who got Cygwin to work. 22:21 Granted, WSL has some advantages in other areas. 22:21 Not least, its packaging system is *much* better than Cygwin's. 22:22 xterm has always seemed perfectly adequate to me. I remember ancient arguments about all the bloat of the 4014 emulation you never use but these days with Web browsers weighing in in the gigabytes that seems like a bit of a joke. 22:22 [hdf-us] [nh362] hypnotist (Cav Gno Fem Neu) killed the invisible Asmodeus, on T:42121 22:22 Does xterm even support terminal profiles? 22:22 -!- Antichthon has joined #hardfought 22:23 -!- hpardis_ has joined #hardfought 22:23 -!- Chris__ANG has quit [Remote host closed the connection] 22:23 I doubt it. *shurg* 22:23 -!- Chris__ANG has joined #hardfought 22:23 -!- mode/#hardfought [+v Chris__ANG] by ChanServ 22:23 [hdf-us] [nh362] Luxidream (Val Dwa Fem Law) killed Lord Surtur, on T:11486 22:23 Because that's kind of an extremely important and really basic feature. 22:23 [hdf-us] [nh362] Luxidream (Val Dwa Fem Law) acquired the Bell of Opening, on T:11488 22:23 It is? 22:23 It is. 22:24 If you use a lot of terminal windows for wildly different purposes, you NEED them to be easily visually distinguishable. 22:24 I do? I wonder how I manage without. 22:24 Dunno. 22:24 I couldn't. 22:25 In the one case where I do I just have a bit of the Debian menu system to invoke it with the right arguments, which seems to suffice. 22:28 [hdf-us] [nh362] Luxidream (Val Dwa Fem Law) genocided class ;, on T:12001 22:28 [hdf-us] [nh362] Luxidream (Val Dwa Fem Law) genocided class R, on T:12001 22:32 -!- Bugpowder has quit [Quit: Bugpowder] 22:32 -!- elenmirie_ has joined #hardfought 22:33 [hdf-us] [nh362] Luxidream (Val Dwa Fem Law) bribed Asmodeus with 49 zorkmids for safe passage, on T:12731 22:34 -!- Bugpowder has joined #hardfought 22:34 -!- Bugpowder has quit [Client Quit] 22:34 [hdf-us] [nh362] Luxidream (Val Dwa Fem Law) killed Juiblex, on T:12870 22:35 [hdf-us] [nh362] Luxidream (Val Dwa Fem Law) killed the invisible Baalzebub, on T:13113 22:35 -!- elenmirie has quit [Ping timeout: 246 seconds] 22:36 jonadab: Not that xterm supports this, but konsole lets you set a semirandom background for each tab without using profiles 22:36 -!- hpardis has quit [Quit: Leaving] 22:36 [hdf-us] [nh362] shadowcat (Bar Hum Fem Cha), 2021 points, T:3241, killed by a lizard, while frozen by a monster's gaze 22:37 semirandom background> truly what one person sees as a useful feature in a terminal client may be less useful to another 22:37 [hdf-us] [nh362] Luxidream (Val Dwa Fem Law) destroyed Vlad the Impaler, on T:13302 22:37 [hdf-us] [nh362] Luxidream (Val Dwa Fem Law) acquired the Candelabrum of Invocation, on T:13304 22:39 Setting different profiles for nethack and actual work is certaintly helpful for me 22:39 [hdf-us] [nh362] Luxidream (Val Dwa Fem Law) killed Orcus, on T:13539 22:40 That indeed is the one case where I do have terminals appear differently; large text and white on black for roguelikes. 22:41 [hdf-us] [nh362] Luxidream (Val Dwa Fem Law) averted death, on T:13803 22:41 Instead of WSL i setup a linux vm in Hyper-v - works great 22:41 Oh right, you use black on white normally 22:42 Mm, I actually have smaller text for nethack, so I can fit more stuff onscreen 22:42 <[Demo]5> haha hyper-v and works great in the same setence 22:42 -!- hpardis_ has left #hardfought 22:44 for a ssh client 22:45 [hdf-us] [nh362] Bartlem (Wiz Hum Mal Neu), 455 points, T:1125, killed by a gnome 22:45 <[Demo]5> cygwin is bash is an ok ssh client 22:45 <[Demo]5> on windows 22:46 [hdf-us] [nh362] Luxidream (Val Dwa Fem Law) killed the invisible Wizard of Yendor, on T:14627 22:46 [hdf-us] [nh362] Luxidream (Val Dwa Fem Law) acquired the Book of the Dead, on T:14630 22:46 You could probably get away with making a docker for connecting to hardfought 22:47 <[Demo]5> why the fuck would you do that 22:47 [hdf-us] [nh362] Luxidream (Val Dwa Fem Law) performed the invocation, on T:14698 22:47 It's easy, works with any terminal, and reusable by others 22:47 [Demo]5: My theory is that the conversation changed to "what's the most absurdly heavyweight way of getting an ssh client you can think of?" and no-one told us. 22:48 <[Demo]5> i guess so 22:48 No think about it 22:48 If you already had docker installed 22:48 <[Demo]5> no 22:48 <[Demo]5> no no no no no 22:48 [hdf-us] [nh362] Luxidream (Val Dwa Fem Law) killed the high priestess of Moloch, on T:14797 22:48 [hdf-us] [nh362] Luxidream (Val Dwa Fem Law) acquired the Amulet of Yendor, on T:14807 22:48 [hdf-us] [nh362] Bartlem (Wiz Hum Mal Neu), 104 points, T:390, killed by a small mimic 22:48 Like if you don't already have it than yeah it's not a great idea but if you do it's super simple 22:49 <[Demo]5> how big is the docker image for that and how much of your resources does it consume to do that? 22:49 fkn massive 22:49 <[Demo]5> answer: an absurd amount for the task you are trying to compelte 22:49 but really easy to use 22:49 [hdf-us] [nh362] Luxidream (Val Dwa Fem Cha) wished for "blessed greased amulet of life saving", on T:14874 22:49 <[Demo]5> at that point just use hterm 22:49 [hdf-us] [nh362] shadowcat (Bar Hum Fem Cha) chose an elven dagger to be named "Sting", on T:2432 22:50 Pinkbeast: pretty sure a linux vm is more absurdly heavyweight than docker? 22:50 aosdict: I certainly wasn't trying to give these a relative score 22:51 [hdf-us] [nh362] Luxidream (Val Dwa Fem Cha) killed the invisible Wizard of Yendor, on T:15145 22:51 [hdf-us] [nh362] dreadpirateA (Sam Hum Mal Law) changed form for the first time, becoming a gnome lord, on T:6877 22:51 <[Demo]5> both are absurdly heavyweight for the task 22:52 aosdict: Docker uses hyper-v so internally anyway 22:52 [Demo]5: The reason we're having this discussion is hterm is down 22:53 <[Demo]5> run ur own on a vps 22:53 [hdf-us] [nh362] shadowcat (Bar Hum Fem Cha) rejected atheism with a prayer, on T:3090 22:54 [hdf-us] [nh362] shadowcat (Bar Hum Fem Cha) changed form for the first time, becoming a wererat, on T:3160 22:54 Why a VPS? You might as well shill out for metal. 22:56 -!- hpardis has joined #hardfought 22:56 -!- greqrg has quit [Quit: Leaving] 22:57 [hdf-us] [nh362] shadowcat (Bar Hum Fem Cha), 3102 points, T:3482, killed by a gold golem 22:58 [hdf-us] [nh362] Luxidream (Val Dwa Fem Cha) killed the Wizard of Yendor, on T:17049 22:58 [hdf-us] [nh362] Bartlem (Wiz Hum Mal Neu), 198 points, T:790, killed by a bear trap 22:59 [hdf-us] [nh362] Luxidream (Val Dwa Fem Cha) entered the Planes, on T:17089 22:59 [hdf-us] [nh362] Luxidream (Val Dwa Fem Cha) killed the Wizard of Yendor, on T:17090 23:03 [hdf-us] [nh362] Luxidream (Val Dwa Fem Cha) killed Famine, on T:17428 23:03 [hdf-us] [nh362] Luxidream (Val Dwa Fem Cha), 466223 points, T:17456, ascended https://www.hardfought.org/userdata/L/Luxidream/nh361/dumplog/1544408342.nh361.txt 23:05 -!- ais523 has quit [Quit: quit] 23:05 <@Luxcalibur> yooo 23:05 <@Luxcalibur> sub 1:30 23:06 <@Luxcalibur> I'd like to thank the new server for halving my ping 23:06 [hdf-us] [nh362] shadowcat (Bar Hum Fem Cha), 431 points, T:1473, killed by a little dog 23:06 @Luxidream did you save at some point? 23:06 <@Luxcalibur> yeah I did 23:07 hm does the dumplog not contain actual wall clock time counting saves? 23:07 *not counting saves 23:07 <@Luxcalibur> 21:19:02, ended 2018-12-09 23:03:54. 23:07 <@Luxcalibur> I looked up realtime in the xlogfile 23:08 right, so I saw the dumplog and said "well that's not sub 1:30, maybe he saved" 23:09 at some point you should speedrun on a local copy so ping is 0. we won't be able to verify the result but it would clearly be the fastest ever 23:09 My ping went from ~80ms to ~70ms 23:13 <[Demo]5> ping is ~13 lol 23:13 <@Luxcalibur> my ping is 40 23:13 <@Luxcalibur> used to be 80 or so 23:19 -!- Grasshopper has joined #hardfought 23:21 [hdf-us] [nh362] dreadpirateA (Sam Hum Mal Law) acquired the luckstone from Mines' End, on T:8818 23:22 I'm not too concerned with ping - I play on hdf-eu every other game or so because the work firewall doesn't block it for some reason 23:30 -!- Bartlem_ has quit [Ping timeout: 256 seconds] 23:44 [hdf-us] [nh362] eraserhead97 (Tou Hum Mal Neu), 4059 points, T:5784, killed by a giant ant 23:51 [hdf-us] [nh362] shadowcat (Bar Hum Fem Cha), 5436 points, T:4447, killed by Ms. Manlobbi; the shopkeeper 23:51 -!- shadowcat22 has quit [] 23:55 -!- hpardis has quit [Quit: ZNC 1.6.6+deb1ubuntu0.1 - http://znc.in] 23:56 <@Luxcalibur> !tell K2 scoreboards are busted 23:56 No worries, @Luxcalibur, I've got this! 23:57 -!- hpardis has joined #hardfought 23:58 -!- hpard_ has joined #hardfought 23:58 -!- hpard_ has quit [Remote host closed the connection] 23:58 -!- hpard_ has joined #hardfought 23:58 -!- hpard_ has quit [Remote host closed the connection]