00:00:34 <Eksell> Unstable branch on CRAWL.XTAHUA.COM updated to: 0.23-a0-730-g0b75920 (34)
00:13:16 <Cheibriados> 03PleasingFungus02 {stenella} 07https://github.com/crawl/crawl/pull/948 * 0.23-a0-715-gf197562: Initial work on C++ cloud generators 10(2 years, 2 months ago, 3 files, 111+ 1-) 13https://github.com/crawl/crawl/commit/f197562f0a0d
00:13:16 <Cheibriados> 03stenella02 07https://github.com/crawl/crawl/pull/948 * 0.23-a0-716-g04f6012: C++ Cloud Generators 10(6 hours ago, 7 files, 179+ 89-) 13https://github.com/crawl/crawl/commit/04f6012f58c1
01:07:00 <alexjurkiewicz> is it possible to compile webtiles to allow wizmode?
01:07:40 <Pinkbeast> alexjurkiewicz: Yes
01:07:54 <Pinkbeast> Er. Sorry. Misread the question; ignore me.
01:16:00 <gammafunk> alexjurkiewicz: most servers have wizmode enabled; they simply run crawl with arguments to disallow it for non-dgl-admins
01:16:30 <gammafunk> so the answer to your question is yes; just build crawl with wizmode and then possibly modify your config to change the crawl startup args
01:16:50 <gammafunk> depending on how you might want it conditionally enabled
01:16:58 <travis-ci> The build was fixed. (master - 0b75920 #10880 : Edgar A. Bering IV): https://travis-ci.org/crawl/crawl/builds/480701189
01:17:10 <ebering> thanks travis
01:17:27 <gammafunk> yeah well you fixed it this time, ebering, but consider yourself on probation
01:17:55 <gammafunk> advil: question, would it be possible to allow spectators to mute people in webtiles chat so that they'd not see that persons messages?
01:18:07 <gammafunk> advil: we're having some trouble with webtiles chat trolls on CAO right now
01:18:38 <gammafunk> bothering admins to ban them is sort of a hassle so it'd be nice if the non-player spectators could mute them
01:28:40 -!- muffindrake1 is now known as muffindrake
01:53:53 -!- amalloy_ is now known as amalloy
02:09:28 <Cheibriados> Transmutation spells can show max duration 13https://crawl.develz.org/mantis/view.php?id=11831 by kitchen_ace
03:00:05 -!- adibis is now known as aditya
03:30:50 <Kelbell> Fork (bcrawl) on crawl.kelbi.org updated to: 0.23-a0-1461-g396401564d
03:33:19 -!- amalloy is now known as amalloy_
04:28:59 -!- Tiobot is now known as Guest72062
05:45:04 <alexjurkiewicz> i wonder if you could use the webtiles "titles" system to allow only accounts with >n hours of play to comment on other player's games
05:46:18 <gammafunk> make tiles chat embedded discord imo
05:46:32 <gammafunk> with me as the admin
05:52:54 <bgiannan> alexjurkiewicz, why would you prevent noobs from asking questions ?
06:06:31 <alexjurkiewicz> to stop trolls ruining noobs experiences
06:55:09 <bgiannan> a ban feature then
06:55:45 <bgiannan> when i started playing if i had been blocked from asking questions i would have gone back playing something else
07:05:43 <gammafunk> requiring n hours of play just wouldn't be a good solution, anyone can register and just run an idle game or a bot
07:18:22 <Eksell> Rhok (L27 DsGl) Crash caused by signal #15: Terminated (Dis:7)
07:44:09 <Cheibriados> 03gammafunk02 07* 0.23-a0-731-gcf8a540: Revamp the Singing Sword again 10(83 minutes ago, 3 files, 18+ 19-) 13https://github.com/crawl/crawl/commit/cf8a540e26e9
07:44:09 <Cheibriados> 03gammafunk02 07* 0.23-a0-732-g5913316: Allow monsters to use the Singing Sword 10(73 minutes ago, 2 files, 1+ 2-) 13https://github.com/crawl/crawl/commit/591331613060
07:44:09 <Cheibriados> 03gammafunk02 07* 0.23-a0-733-ge7febcc: Sometimes place the Singing Sword in a vault 10(60 minutes ago, 1 file, 5+ 0-) 13https://github.com/crawl/crawl/commit/e7febcc0c011
07:47:27 <Kelbell> Unstable branch on crawl.kelbi.org updated to: 0.23-a0-733-ge7febcc0c0 (34)
08:26:44 <NP7> !crash Rhok
08:26:44 <Sequell> 4. Rhok, XL27 DsGl, T:91575 (milestone): https://crawl.xtahua.com/crawl/morgue/Rhok/crash-Rhok-20190117-121712.txt
09:15:29 <advil> what's the CAO trolling situation exactly?
09:16:08 <advil> in principle it would be possible to do something like that, but I guess the first question would be why the player isn't willing or able to mute the trolls
09:20:38 <advil> !crashlog
09:20:51 <Sequell> 20546. Rhok, XL27 DsGl, T:91575 (milestone): https://crawl.xtahua.com/crawl/morgue/Rhok/crash-Rhok-20190117-121712.txt
09:23:10 <advil> odd...SIGTERM during a deflate call
09:30:34 <NP7> he declined to update to the latest version of trunk just before this crash happens, in case that matters
09:30:43 <NP7> s/happens/happened
09:33:08 <advil> NP7: was it a one-time thing? i.e. was he able to play again without updating?
09:34:02 <NP7> yes, he tried again without updating and it worked I believe
09:35:34 <advil> ok...in that case it may have been some server-specific temporary issue
09:36:02 <advil> haven't seen any crashes exactly like this, though I do feel like we've been seeing SIGTERM (not triggered externally) more than I remember
09:36:15 <advil> (which could be because it's reported more clearly now)
09:46:07 <Cheibriados> New branch created: pull/949 (1 commit) 13https://github.com/crawl/crawl/pull/949
09:46:07 <Cheibriados> 03argonaut02 07https://github.com/crawl/crawl/pull/949 * 0.23-a0-715-g4c26300: The unrand trident "Rift" with distortion and extended reach. 10(22 hours ago, 6 files, 70+ 34-) 13https://github.com/crawl/crawl/commit/4c26300488d1
10:38:48 -!- amalloy_ is now known as amalloy
12:06:12 <Henzell> Unstable branch on crawl.akrasiac.org updated to: 0.23-a0-733-ge7febcc (34)
12:54:33 <switch__> ebering: (and other devs) you can test what the desolation clouds look like on cpo, if you want. the way i've been doing it is entering wizmode, shift+p portal_desolation_entry_ruins , then entering it normally
12:57:33 <Pinkbeast> Not to be Capt Obvious but you can also & ~ 9 to go there directly
13:00:29 <switch__> i didn't know that, so that's super the opposite of captain obvious
13:00:43 <switch__> unless i'm just really dumb in which case you're still captain obvious
13:19:31 <amalloy> switch__: you know about &?, right? it's a pretty good list to read over
13:20:56 <switch__> i've looked at it, but for some reason i have to read it line-by-line, i can't just glance at it
13:21:05 <switch__> which i don't think i've ever actually done
13:21:39 <amalloy> switch__: also, something that is not totally obvious: x? also lists wizmode-only targeting-mode commands
13:24:10 <switch__> wtf
13:25:06 <amalloy> this is mentioned in &? as well, but as i said it's not obvious
13:25:19 <switch__> i wonder why &? is so hard to read
13:26:03 <switch__> maybe it isn't and i'm crazy
13:26:30 <switch__> it's probably that they aren't categorized super well
13:27:05 <amalloy> i have trouble finding stuff there sometimes too. i don't always know what category i'm looking for, and also the double-column layout makes it easy to accidentally skim past stuff
13:27:40 <switch__> yeah, that's what i mean. i can't like, glance past a line, if i do, i miss it completely
13:28:38 <switch__> i think it's that each one has completely different sentences and you can't autocomplete them out of your peripheral vision
13:28:56 <Pinkbeast> BTW, is there a reason the recently-added bones debugging commands are hidden behind DEBUG_BONES, given that all the ghost_dprfs happen in wizard mode anyway?
13:41:11 <switch__> why isn't it possible to run a program inside a container and have that container spit out the execution flow of the program
13:41:13 <switch__> does this really not exist
13:43:26 <NP7> Is there a way to run util/checkwhite without it changing all LFs to CRLFs in all files? It's really annoying to run checkwhite to check I've not missed any whitespace errors and then just get a list of every single file, then also when I change something very minor, have Crawl completely recompile, because checkwhite did something that afaik is completely unimportant
13:44:31 <switch__> yeah, -r i think
13:45:47 <NP7> thanks! :)
13:46:01 <switch__> did that work?
13:46:41 <NP7> well I ran `source/util/checkwhite -r` and got nothing back, so I presume so
13:46:55 <switch__> dope
15:10:19 <advil> !tell NP7 I'm guessing you might be misunderstanding, when checkwhite prints "CR: file" it's *removing* a CR, not adding; the repo uses unix-style line endings. Possibly you have your editor set to use windows-style line endings and it's adding them in for checkwhite to get annoyed at? (git should autoremove them in any case, so checkwhite is a bit redundant)
15:10:20 <Sequell> Maximum message length is 340 characters, but you had 351. Eschew verbosity, Gladys!
15:10:43 <advil> !tell NP7 I'm guessing you might be misunderstanding, when checkwhite prints "CR: file" it's *removing* a CR, not adding; the repo uses unix-style line endings. Possibly you have your editor set to use windows-style line endings and it's adding them in for checkwhite to get annoyed at?
15:10:44 <Sequell> advil: OK, I'll let np7 know.
15:11:37 <advil> !tell NP7 granted, git should autoconvert to unix line endings in any case, so the checkwhite check is a bit redundant
15:11:37 <Sequell> advil: OK, I'll let np7 know.
17:05:40 <Cheibriados> 03NormalPerson702 {ebering} 07* 0.23-a0-734-gc3daa31: Convert `enum spflag_type` to `enum class spflag` 10(13 days ago, 19 files, 542+ 505-) 13https://github.com/crawl/crawl/commit/c3daa31c0c79
17:05:40 <Cheibriados> 03NormalPerson702 {ebering} 07* 0.23-a0-735-ga6713cf: Convert `enum spschool_flag_type` to `enum class spschool` 10(12 days ago, 35 files, 648+ 645-) 13https://github.com/crawl/crawl/commit/a6713cf8af19
17:05:40 <Cheibriados> 03NormalPerson702 {ebering} 07* 0.23-a0-736-gb067b0d: Convert `enum nothing_happens_when_type` to `enum class nothing_happens` 10(10 days ago, 7 files, 24+ 23-) 13https://github.com/crawl/crawl/commit/b067b0da3149
17:05:40 <Cheibriados> 03NormalPerson702 {ebering} 07* 0.23-a0-737-gc88e511: Add a struct to use for MiscastEffect::special_source 10(2 days ago, 15 files, 89+ 80-) 13https://github.com/crawl/crawl/commit/c88e5110026f
17:05:40 <Cheibriados> 03NormalPerson702 {ebering} 07* 0.23-a0-738-g76aa57b: Convert `enum miscast_source` to an enum class 10(2 days ago, 15 files, 102+ 90-) 13https://github.com/crawl/crawl/commit/76aa57be1008
17:05:40 <Cheibriados> 03NormalPerson702 {ebering} 07* 0.23-a0-739-gc3062a4: Convert `enum recall_t` to an enum class 10(10 days ago, 3 files, 10+ 10-) 13https://github.com/crawl/crawl/commit/c3062a41d611
17:05:40 <Cheibriados> 03NormalPerson702 {ebering} 07* 0.23-a0-740-g2dde14e: Add a note to docs/obsolete/cut_spells.txt 10(5 days ago, 1 file, 9+ 1-) 13https://github.com/crawl/crawl/commit/2dde14e38733
17:15:42 <Cheibriados> 03gammafunk02 07* 0.23-a0-741-g38971ea: Change the Singing Sword's power for monsters (ebering) 10(13 minutes ago, 1 file, 8+ 1-) 13https://github.com/crawl/crawl/commit/38971ea054dd
17:17:45 <floraline> switch__: do you still need me to install your clouds PR?
17:18:17 <switch__> your server is the one in NA, right?
17:18:35 <switch__> it was a little hard for me to test on CPO because there was a baseline lag probably due to distance
17:18:47 <switch__> so, uhh, yes please, i guess
17:19:30 <switch__> if it isn't too much trouble
17:26:29 <ebering> switch__: It's ultimately up to the server admins but
17:26:48 <ebering> I'd prefer if you ran webtiles locally or even used your own local task manager to see if your changes have reduced cpu load from runing clouds
17:27:20 <ebering> testing an issue that can lag the server for all users in production strikes me as a pretty bad plan
17:27:54 <ebering> if alexjurkiewicz and floraline are ok, then it's their servers so go ahead but
17:28:17 <ebering> I'm nervous about such a plan though
17:28:31 <switch__> totally, I would too; it would require me to bother less people. However, I haven't set up webtiles locally, and wouldn't be sure about how well it translates anyway. My testing on my local machine in non-webtiles was a noticable performance increase, but I don't trust my machine for metrics like that, and it's not about performance in an individual game anyway, it's about whether it translates to mass-scale on the service
17:28:37 <switch__> oh, lag isn't a possibility
17:28:52 <switch__> it can't go wrong in that way; i'm just not 100% sure that the performance gain is significant *enough*
17:29:13 <switch__> in which case i will continue digging and improving that rather than just cleaning up
17:29:26 <ebering> ah, I see
17:31:20 <switch__> it's just that, judging by how long it's been since the last person has even *attempted* to look at this lua crap, i feel like i should get this one right by the time i say i'm done with it. but if it's a significant performance increase as-is and i can clean this PR up pretty fast, that can be pushed in time for your next release, and i can keep working from that point.
17:31:47 <switch__> rather than extending this PR until it basically rewrites half the lua
17:32:05 <switch__> which i'm edging closer and closer to feeling comfortable doing
17:32:47 <gammafunk> release does not require any sort of "push" development-ise; this is the sort of thing that can easilly be backported when it's good and ready
17:32:53 <switch__> (i'm sure plenty of the people in this room can easily iterate over my C++ once i'm done and make noticable improvements, but it's pretty clear nobody wants to touch the lua with a 10 foot pole)
17:33:31 <Pinkbeast> switch__: JOOI, I remember you saying the cloud generators in Desolation were odd somehow?
17:33:43 <ebering> I'm not clear on why you'd need to re-write "half the lua" or even what you mean by that (the lua code, or the lua library?) or what exactly your goal would be
17:33:50 <switch__> sure, i'm not even necessarily talking about an actual release e.g. .23, just that it might be a 10% drop in CPU load or some crazy shit as-is
17:33:59 <switch__> well, you could re-write the marker system in C++
17:34:59 <ebering> yes, but *why* does that need to be done; other than the extensive cloud generators the lua marker performance isn't a major concern
17:35:51 <switch__> Pinkbeast, they're initialized with delay set to 0. PleasingFungus' initial C++ code had a while delay<0 loop, so these are completely unhandled in his c++ code. How they actually behave in the lua code I'm not 100% sure, but they seem to run many dozens of times every tick. I had to simulate this because I don't have a full picture of how those run and I can't force myself to step through multiple-thousand-instruction-long
17:35:51 <switch__> lua loops
17:36:30 <Pinkbeast> switch__: Thanks.
17:36:33 <switch__> oh, it would be in the interest of letting other people work on the code in the future. it's super clear nobody is left that wants anything to do with the lua, and having it all be debuggable in gdb and worked on by basically any community member would be good for the future
17:37:46 <ebering> switch__: have you uh, asked? I spent several weeks this summer documenting clua and learning how it worked in the process
17:37:51 <gammafunk> I think you need to drop this wierd "no one works with lua" idea
17:38:46 <gammafunk> git log of all the files in dat/dlua and all the l-FOO.cc files would show that this is just not the case
17:39:21 <gammafunk> mapmark.cc as well
17:39:29 <gammafunk> cloud generates are just one aspect of things that use lua
17:39:34 <gammafunk> *generators
17:39:41 <switch__> it's possible i'm completely off the mark, but when i was paging through the history it all seemed very dated. i'm sure stuff has been added, but the older code didn't seem to be getting touched very oten
17:40:19 <gammafunk> tons of crawl code is pretty dated; it's a codebase over over 400k lines of C++ and a couple hundred k lines of des and lua, etc
17:40:34 <switch__> yeah totally. maybe i just assumed it because this hadn't been done sooner. it's a complicated problem, and i got lucky by finding the entry point i did, and i still don't understand the majority of it, but it did make me think that if someone could have, someone would have
17:40:36 <gammafunk> it's just not helpful language to throw about since people here have worked on these things :p
17:40:48 <switch__> that's true
17:41:17 <switch__> i'm not trying to criticize, truly. i'm far from the best programmer in this room; i wouldn't even guess i'm close.
17:41:41 <ebering> switch__: you'd do well to disabuse yourself of the idea that "if someone could have, someone would have"
17:41:53 <ebering> can isn't just a measure of ability, understanding, or desire
17:41:59 <ebering> it's also a measure of volunteer time
17:42:26 <ebering> if crawl were a 40hr a week gig for the whole team this and much more would have been done
17:42:54 <switch__> i don't doubt that at all
17:44:29 <Kelbell> Unstable branch on crawl.kelbi.org updated to: 0.23-a0-741-g38971ea054 (34)
17:44:56 <advil> "it's super clear nobody is left that wants anything to do with the lua" ???
17:45:12 <advil> well, I guess I don't need to add to the pushback on that
17:45:25 <switch__> hasn't this specific problem of cloud generators come up in this chat at least dozens of times?
17:46:05 <switch__> it's not like it was a low priority problem (unless it was, in which case, i take back basically everything i've written up until now)
17:47:40 <advil> maybe by "the lua" you mean "the lua code involved in cloud generators"?
17:48:19 <ebering> switch__: I think you'll do much better if you assume that the reason that code cleanup and refactor projects sit uncompleted is one of time available unless an explicit reason to hold off has been mentioned
17:48:57 <ebering> e.g. we haven't merged datafiles yet because it involved getting all the servers on the same page and finding a time when several people can be on hand at the same time post merge in case something unexpected goes wrong
17:49:30 <switch__> advil: yes, 100%
17:51:40 <switch__> ebering, i do basically assume that. obviously anyone on the dev team is capable of looking at any part of the crawl code and figuring it out
17:58:49 -!- ProzacElf__ is now known as ProzacElf
18:14:09 <alexjurkiewicz> ebering: switch__: worst case if an experimental takes a full CPU is no different to someone playing in desolation normally
18:14:21 <switch__> lol
18:14:24 <switch__> does it actually
18:14:45 <alexjurkiewicz> if you wait 100 turns it takes a few seconds
18:15:10 <Pinkbeast> switch__: Pretty much - faster CPU just means the player is done sooner
18:19:28 <floraline> switch__: i'm going to hold off on installing the branch until it's closer to being done, if you still need it by then
18:20:02 <switch__> sure
18:20:22 <floraline> on a related note, it would be nice if i could regenerate the game links without restarting webtiles
18:21:47 <floraline> it looks like it's supposed to render on the fly but it seems to be getting cached on the server side
18:22:12 <Lantell> Unstable branch on underhound.eu updated to: 0.23-a0-741-g38971ea054 (34)
18:23:34 <floraline> oh right, it needs to reload the config file. i wonder if that could be changed
18:24:44 <floraline> alexjurkiewicz: does webtiles-changes allow you to reload your config files without dropping players?
18:25:27 <alexjurkiewicz> yes
18:26:46 <floraline> wow, that's convenient
18:41:05 <switch__> https://www.youtube.com/watch?v=gnescBHfei0 comparison video from testing on cpo (thanks alexjurkiewicz <3)
18:42:23 <alexjurkiewicz> tl;dr 🏎 🚀
18:49:58 <gammafunk> floraline: I wouldn't count on that webtiles-changes is pretty broken
18:50:19 <gammafunk> I was the one to add relaoding config without dropping players and I'm not at all confident in that code
18:50:29 <gammafunk> so I wouldn't count on that being very reliable
18:51:16 <alexjurkiewicz> CPO reloads config every time trunk rebuilds or a branch/fork gets added/removed
18:51:38 <gammafunk> yes but this code is just not tested, audited, or anything
18:51:57 <gammafunk> cpo has a very small number of active users, and I'm not sure what code it's even using to use this reload
18:52:09 <gammafunk> as it's not running webtiles-changes exactly, but its own fork essentially
18:52:21 <alexjurkiewicz> not really, it's webtiles-changes
18:52:35 <gammafunk> right; so I was the author of that code, and I can say that I wouldn't trust it
18:53:13 <gammafunk> at a minimum, all that functionality needs an audit and another set of eyes before it were to go in trunk
18:53:20 <gammafunk> probably just a different approach, too
19:04:41 <gammafunk> oh I see, I only did the part where it tries to rollback to its old config if it encounters an error
19:05:55 <gammafunk> what might work would be if parts of webtiles-changes were carefully merged into master, probably specifically things that don't depend on the jsx-react rewrite
19:06:10 <gammafunk> but it's also the sort of thing where you'd have to work with admins to roll that out
19:12:21 <Cheibriados> 03ebering02 07https://github.com/crawl/crawl/pull/944 * 0.23-a0-724-gf9cf2b7: Only list seen branches in Gozag's Extra pane (advil) 10(55 minutes ago, 1 file, 1+ 1-) 13https://github.com/crawl/crawl/commit/f9cf2b74431b
19:12:21 <Cheibriados> 03ebering02 07https://github.com/crawl/crawl/pull/944 * 0.23-a0-725-g2e9402d: Hide some god Extra panes from non-worshippers 10(7 minutes ago, 1 file, 9+ 6-) 13https://github.com/crawl/crawl/commit/2e9402d801d1
19:12:21 <Cheibriados> 03ebering02 07https://github.com/crawl/crawl/pull/944 * 0.23-a0-726-g4ab6b33: Enable god Extra in describe_god_with_join 10(2 minutes ago, 1 file, 2+ 4-) 13https://github.com/crawl/crawl/commit/4ab6b331d2b4
19:43:58 <Cheibriados> 03bjobae02 {ebering} 07* 0.23-a0-742-g1a98732: Add support for an 'Extra' pane in religion screen 10(4 weeks ago, 1 file, 29+ 8-) 13https://github.com/crawl/crawl/commit/1a9873242815
19:43:58 <Cheibriados> 03bjobae02 {ebering} 07* 0.23-a0-743-g509b01f: List named followers for Beogh worshippers 10(4 days ago, 3 files, 52+ 2-) 13https://github.com/crawl/crawl/commit/509b01f02ed1
19:43:58 <Cheibriados> 03bjobae02 {ebering} 07* 0.23-a0-744-gbfe4a4d: Future proof Beogh gift monster props 10(4 days ago, 1 file, 18+ 0-) 13https://github.com/crawl/crawl/commit/bfe4a4df639d
19:43:58 <Cheibriados> 03ebering02 07* 0.23-a0-745-g550063b: God Extra panes for Ash, Gozag, and Hep 10(4 days ago, 1 file, 66+ 60-) 13https://github.com/crawl/crawl/commit/550063bd70e4
19:43:58 <Cheibriados> 03ebering02 07* 0.23-a0-746-g710303d: Add a god Extra pane for Nemelex 10(2 days ago, 4 files, 34+ 14-) 13https://github.com/crawl/crawl/commit/710303dcffce
19:43:58 <Cheibriados> 03ebering02 07* 0.23-a0-747-gf2a872a: Add a god Extra pane for Wu Jian 10(2 days ago, 2 files, 18+ 12-) 13https://github.com/crawl/crawl/commit/f2a872a2867d
19:43:58 <Cheibriados> 03advil02 {ebering} 07* 0.23-a0-748-g5b2007c: Add god extra pane to webtiles 10(22 hours ago, 3 files, 76+ 12-) 13https://github.com/crawl/crawl/commit/5b2007cb45c9
19:43:58 <Cheibriados> 03advil02 {ebering} 07* 0.23-a0-749-g6f75a6c: Code formatting, wording 10(22 hours ago, 2 files, 6+ 5-) 13https://github.com/crawl/crawl/commit/6f75a6ce2cc4
19:43:58 <Cheibriados> 03advil02 {ebering} 07* 0.23-a0-750-g8732018: Remove commented code 10(22 hours ago, 1 file, 0+ 1-) 13https://github.com/crawl/crawl/commit/87320180ed64
19:43:58 <Cheibriados> 03advil02 {ebering} 07* 0.23-a0-751-gc05403b: Code formatting 10(21 hours ago, 1 file, 6+ 0-) 13https://github.com/crawl/crawl/commit/c05403bdbf3d
19:43:58 <Cheibriados> ... and 3 more commits
19:49:53 <alexjurkiewicz> woot
19:51:27 <alexjurkiewicz> looks good on webtiles!
19:54:45 <Kelbell> Unstable branch on crawl.kelbi.org updated to: 0.23-a0-754-gf2270e3573 (34)
19:59:07 -!- amalloy is now known as amalloy_
20:03:59 <alexjurkiewicz> I wrote a proposal for "level feeling", keen for any feedback on the idea: https://crawl.develz.org/tavern/viewtopic.php?f=8&t=26016
20:10:38 <gammafunk> one fast piece of feedback: making all monsters on a level skeletons or zombies is a bad example; it's a huge player buff that makes monsters systematically *less* interesting and we'd probably even like to remove derived undead working that way
20:13:23 <gammafunk> I personally think that ideas like this probably work best when thinking of new branch-wide  mechanics for e.g. lair branches and other post-lair content that we could use
20:13:26 <gammafunk> an example:
20:13:57 <gammafunk> https://crawl.develz.org/wiki/doku.php?id=dcss:brainstorm:dungeon:branch:propose:desert_mikee&s[]=desert
20:14:13 <gammafunk> some brainstorming mikee and I did about a desert branch to replace swamp
20:15:56 <gammafunk> making branches work more distinctly aside from the idea of "here's a small set of thematic monsters that are largely similar to other threats you've fought"
20:16:09 <gammafunk> it's the kind of thing that could go hand in hand with some shortening of levels required to finish the game
20:16:18 -!- amalloy_ is now known as amalloy
20:16:49 <gammafunk> and would give a sense of each progressive branch you go through to reach the orb being something special enough to warrant its existence
20:47:18 <Kelbell> Unstable branch on crawl.kelbi.org updated to: 0.23-a0-755-gdb21e031a1 (34)
20:47:29 <Cheibriados> 03ebering02 07* 0.23-a0-755-gdb21e03: Reduce silence radius from Klown pies 10(19 minutes ago, 1 file, 4+ 4-) 13https://github.com/crawl/crawl/commit/db21e031a17d
21:24:43 <Cheibriados> 03mgdelmonte02 07https://github.com/crawl/crawl/pull/939 * 0.23-a0-685-g51f2a98: added guard against both zero-range and non-zapping spells 10(2 minutes ago, 1 file, 7+ 3-) 13https://github.com/crawl/crawl/commit/51f2a98d97dd
22:01:36 <Cheibriados> 03mgdelmonte02 {ebering} 07* 0.23-a0-756-g517b432: Create a CLua function for spell targeter access 10(7 days ago, 1 file, 69+ 0-) 13https://github.com/crawl/crawl/commit/517b43256b2f
22:17:09 <Kelbell> Unstable branch on crawl.kelbi.org updated to: 0.23-a0-756-g517b43256b (34)
22:22:37 <alexjurkiewicz> strange incentives sounds like a good thing :P
22:23:55 <ebering> not the good strange
22:25:51 <gammafun1> ally commands no longer make noise in the first place
22:25:59 <gammafun1> so it might not make sense that they don't work under silence
22:27:01 <gammafun1> but that's getting into interpretations regarding realism and the like
22:27:54 <gammafun1> my concerns are mostly complexity/spoilers, and maybe just the effect being meaningful enough
22:28:41 <gammafun1> they are mostly no-ops for a lot of situations and will be pretty irrelevant much of the time; it's a bit hard to strike the right balance with a kind of grab bag of effects like this
22:33:04 <ebering> ally commands work while silenced
22:34:40 <gammafun1> well a complaint of silence on allies being too dangerous for the player was brought up at one point
22:35:05 <gammafun1> if you can control them during silence that's even less relevant; but even if you couldn't allies just offer a ton of advantages so that doesn't concern me
22:35:38 <ebering> it's more the behavior of monster silence: a silenced ally is radius 5 silence for all but the last 30 aut of its duration
22:35:47 <gammafun1> actual weird behaviour and exploits are certainly more of a thing, but again it's a bit hard to see how this would be some meaningful strategy
22:35:50 <gammafun1> yeah, they're not symmetric
22:36:02 <gammafun1> I don't know what the previous duration's radius was over the course of the duration
22:36:07 <gammafun1> for players, I mean
22:36:28 <ebering> mostly in the 2 range, a high roll could hit 3
22:36:30 <ebering> I think
22:36:31 <ebering> no
22:36:33 <ebering> that's not correct
22:37:08 <gammafun1> I had also thought I used spellpower to apply silence but looking at the diff in your commit I see I just some the "set silence duration with a random component in turns" code
22:37:18 <gammafun1> *I just copied
22:37:19 <ebering> 10+random(21) was the duration, the distance is isqrt(max(0,min(3*(dur-5/4),5))
22:37:30 <ebering> brackets are wrong there
22:37:34 <ebering> (dur-5)/4
22:37:57 <gammafun1> so average was 20 turns
22:39:47 <gammafun1> it's probably ok, in the end; it's not as if we couldn't add one more effect that's a little more impactful if we feel the need
22:39:51 <gammafun1> or replace one or something
22:40:29 <gammafun1> silence that's sort of paralysis duration won't be all that nasty but it will certainly be relevant sometimes
22:41:24 <gammafun1> alexjurkiewicz: to prove that ebering is wrong to nerf silence pies, just streak 10 spellcasters
22:41:34 <gammafun1> if you win that easilly it will require buffs
22:44:08 <alexjurkiewicz> https://www.reddit.com/r/dcss/comments/ah449z/the_race_tiers_are_all_over_the_place/ I submit another piece of evidence that arranging species by anything other than "newbie difficulty" is a bad idea
22:45:15 <gammafun1> lol
22:45:32 <gammafun1> that one guy: "The ingame race difficulty and especially recommended backgrounds are 100% misleading."
22:45:36 <gammafun1> don't ever change, reddit
22:46:05 <gammafun1> this is going to be someone who thinks that tm of chei is amazing
22:46:29 <gammafun1> I think revamping that listing is something to consider, but I'm not really sure what the best criteria is
22:46:42 <gammafun1> "player perception of strength" is not a great listing
22:47:01 <gammafun1> players are pretty bad at assessing how strong a thing is, even players who have won a number of times
22:47:34 <gammafun1> Lasty wanted it to be about some combination of complexity and difficulty, as a kind of "here try this first" thing
22:48:05 <gammafun1> so it makes sense to not recommend players start with Ce, which is not exactly most people's pick for a fun species
22:48:47 <gammafun1> but it's pretty difficult to nail down that hybrid approach in a way that makes people happy
22:51:03 <gammafun1> it may be that what we need are just some tweaks to the current species listing, and maybe some way players can get to an explanation of what those listings mean from that menu
22:51:32 <gammafun1> some kind of brief FAQ that they could look at from that selection menu
22:52:20 -!- sedition_ is now known as sedition
22:53:09 <alexjurkiewicz> I think "ease of getting to lair for a <10 game newbie" would be fantastic
22:53:53 <alexjurkiewicz> like once you know the species, the arrangement doesn't matter. You know about all the species anyway. The arrangement should focus 100% on helping newbies pick something they'll be successful with
22:54:49 <gammafun1> it really wouldn't because we'd have to list DD prominently
22:54:49 <alexjurkiewicz> I think the problem with current arrangement is the metric ("complexity") isn't actionable for anyone, really
22:55:23 <alexjurkiewicz> i bet if you gave DD to a <10 game newbie they wouldn't find the ability menu. So it should go at the very end :P
22:55:31 <gammafun1> I bet that's not true
22:55:38 <gammafun1> There are certainly a handful of species that are unabigiuosly easy for "noobs", but this is just very difficult to make subjective
22:55:46 <gammafun1> especially past the first tier
22:56:01 <gammafun1> and we'd be recommending some stuff that's not exactly the most thrilling thing to start the game with
22:56:07 <Pinkbeast> alexjurkiewicz: When I was a <10 game newbie I did, FWIW
22:56:17 <Pinkbeast> ... of course I only had to find it when the wand was dry
22:56:55 <gammafun1> telling them to play DD, Ce, Tr on their first games is less good then telling them to play Mi, HO, Gr, etc
22:57:35 <gammafun1> if "easy to get to lair" if you don't know what you're doing were synonmous with "a fun introduction to the game", I could more agree
22:58:48 <alexjurkiewicz> now that we're talking about this, I remember that someone proposed the idea of having preset combos (eg MiFi, CeHu, DeFE) and then an option to go advanced and fully customise your character
22:58:58 <alexjurkiewicz> I didn't really like it at the time
23:00:01 <floraline> i'm happy to see talk of newbies, my number one concern is newbie accessibility
23:00:43 <floraline> that first time they boot up the game is so important
23:01:04 <gammafun1> it's also worth pointing out for people obsessed with the newbie experience: the tutorials could always use work!
23:01:19 <gammafun1> that is where a newbie will start and those kind of just never get updated
23:01:30 <floraline> that is a great suggestion. maybe i should spend some time looking at those
23:02:00 <alexjurkiewicz> hints mode as well
23:02:09 <alexjurkiewicz> but a lot of people skip tutorials
23:02:42 <gammafun1> well, that they might, be we can't really design for every contingency in the end
23:02:47 <alexjurkiewicz> I guess ideally a newbie would start DCSS and the first menu option would be "start game (ingame tutorial: enabled)" and that would be that
23:02:48 <gammafun1> combo recommendations might be a nice part of a hybrid approach
23:03:25 <gammafun1> yeah, maybe it could default to a tutorial unless they disable it for their first game somehow
23:04:03 <alexjurkiewicz> dcss does save the prf file, so you can save "has disabled tutorial by default" for accounts
23:04:10 <gammafun1> maybe we could treat the selections for race/combo according to recommendations
23:04:55 <gammafun1> one thing I like about the current class scheme is the logical grouping that's nice after you're no longer a newbie
23:05:39 <gammafun1> but that's not really required; we could sort of list both things by a kind of recommendation strategy
23:05:43 <floraline> what if you could set the grouping with an option
23:05:51 <floraline> like different styles
23:05:57 <alexjurkiewicz> what do you mean by treat according to recommendations?
23:06:28 <gammafun1> well, I suppose it's sort of what Lasty was probably going for; "here are some stronger, fun things to start out with"
23:06:43 <gammafun1> Gnolls, Minotaur, Gargoyle, ...
23:07:22 <gammafun1> so we don't have to list e.g. DD, Ce in that; maybe Troll would be ok for it since that's less annoying than the former two
23:07:53 <gammafun1> but yeah I'm not sure how we have three tiers, what language we have for those three tiers (especially past the first tier)
23:08:20 <alexjurkiewicz> you know if you give a list of combos i guess there would also be a couple of harder combos at the bottom. MuMo challenge might become canonical!
23:08:47 <gammafun1> well, that would be cute, but the recommended combos should really have that "fun" factor
23:08:52 <gammafun1> MuMo is just not fun
23:08:54 <gammafun1> MuCK is fun
23:08:56 <gammafun1> MuMo isn't
23:09:41 <alexjurkiewicz> MuCK sounds like a pretty good final combo
23:09:43 <gammafun1> it's sort of a wide-ranging project I guess, overhauling the "selecting my character as a newbie" experience
23:10:11 <gammafun1> also we probably should have them pick class first
23:10:43 <gammafun1> species is much more of a difficulty selector, but players don't really know this, and classes determine what you start doing immediately
23:11:12 <gammafun1> they're also kind of better balanced across each other; there's more variation in species difficulty
23:12:36 <gammafun1> might be worth at least making a wiki page with a list of ideas
23:18:42 <alexjurkiewicz> https://imgur.com/a/wSPEZyZ something like this would be pretty MVP, and easy to just insert into the existing flow (no need to modify the existing screens)
23:19:18 <alexjurkiewicz> well, I guess you need a "simple character" go back button on the existing screens
23:31:47 <gammafun1> well you wouldn't put muck on a screen like that
23:31:49 <gammafun1> you'd just have something hard yet sensible
23:32:07 <gammafun1> this should be "hard for a new player" not "attempt to be the hardest combo in the game"
23:33:12 <gammafun1> !I thought you wanted a larger list of combos with some tiers or something
23:34:11 <gammafun1> I'm not sure, that feels like it might be too simple a listing, and again I'd prefer to try to be more universal in terms of things players enjoy
23:34:43 <alexjurkiewicz> oh definitely the list is just a placeholder
23:34:52 <gammafun1> maybe something with some basic tiers, melee, ranged, "hybrid", and "spellcaster"
23:35:02 <alexjurkiewicz> you might want to split the combos up -- yeah exactly
23:35:06 <alexjurkiewicz> really
23:35:29 <alexjurkiewicz> just wanted to illustrate the idea of a page in the "new game workflow" where there's an escape hatch to the old system for regular players
23:35:48 <alexjurkiewicz> but clear guidance / handrails for new players
23:45:45 <floraline> does equip_bar = true not work in webtiles?
23:46:42 <alexjurkiewicz> it does not
23:48:41 <gammafun1> it's only for cool people aka console players
23:55:27 <floraline> is there a way to login automatically in console using environment variables or something?
23:57:03 <geekosaur> no. there are various expect scripts about