00:29:51 <amalloy> !tell greensnark someone at the roguelike celebration said he organizes some kind of postgres convention, and he thought a talk about how Sequell works (the database format, !lg query language, whatever) would be very well received. if you might be interesed in giving one i can send you his contact info
00:29:53 <Sequell> amalloy: OK, I'll let greensnark know.
00:32:32 <Rast> ok this is an interesting depths vault
00:32:48 <Rast> the armour shop has mundane -2 and -1 armours
00:32:57 <Rast> the "jerky" shop has meat rations, lol
00:33:15 <Rast> and the " Miraculous Elixirs" shop has 10+ potions of mutation
00:33:24 <Rast> which in current version are extremely valuable
00:33:39 <Rast> probably not intended that way
00:38:56 <nikheizen> huge stocks of !mut are not really super valuable imo
00:39:01 <nikheizen> unless you plan to play badly
00:39:12 <Rast> you can scum with them
00:39:36 <Rast> and still have plenty left over for bad luck with oofs or whatever
00:39:43 <nikheizen> you can play roulette with them...
00:39:50 <Rast> sure
00:39:57 <Rast> in most games you only have a handful for that
01:15:36 -!- tat is now known as Guest41774
01:19:25 <Gretell> Unstable branch on crawl.develz.org updated to: 0.21-a0-442-g848f44b (34)
01:53:32 -!- amalloy is now known as amalloy_
01:54:33 -!- amalloy_ is now known as amalloy
01:57:30 <Gretell> Windows builds of master branch on crawl.develz.org updated to: 0.21-a0-442-g848f44b
02:52:52 <Gretell> Monster database of master branch on crawl.develz.org updated to: 0.21-a0-442-g848f44b
03:11:35 <Rotatell> Unstable branch on crawl.beRotato.org updated to: 0.21-a0-442-g848f44b (34)
03:32:13 <Cheibriados> Wrong portal message at stairs 13https://crawl.develz.org/mantis/view.php?id=11279 by Yermak
04:32:31 <aidanh> advil: nice work on that font rendering bug!
04:51:15 -!- amalloy is now known as amalloy_
05:47:15 <Cheibriados> Visual glitch with range missiles sticking on the screen 13https://crawl.develz.org/mantis/view.php?id=11280 by Yermak
09:18:32 <TZer0> I'm trying to set up my crawl server from scratch - anyone got any idea why this is happening? https://pastebin.com/iBfmqwHT
09:18:44 <TZer0> Step 6. in this guide: https://crawl.develz.org/wiki/doku.php?id=dcss:server:setting_up_dgamelaunch_and_webtiles
09:21:17 <TZer0> Further debug info: https://pastebin.com/bxPGfbrV
09:40:59 <TZer0> Nevermind, fixed it and updated the documentation
09:42:08 <Ge0ff> a missing package (flex)?
09:42:14 <TZer0> Ge0ff: yeah
09:42:51 <TZer0> I realised what was up when I saw LEX = : in the makefile
09:54:46 <Tux[Qyou]> http://prntscr.com/h9gqq6
10:01:58 <TZer0> Morgues are now up https://underhound.eu/crawl/
10:11:34 <TZer0> |amethyst: can you change logfile-reading so that it no longer reads from port 82? I've set things up here: https://underhound.eu/crawl/meta/
10:11:38 <TZer0> oops.
10:11:42 <TZer0> !tell |amethyst can you change logfile-reading so that it no longer reads from port 82? I've set things up here: https://underhound.eu/crawl/meta/
10:11:43 <Sequell> TZer0: OK, I'll let |amethyst know.
10:11:46 <TZer0> There we go.
10:30:48 <|amethyst> TZer0:  I'll change scoring
10:30:48 <Sequell> |amethyst: You have 1 message. Use !messages to read it.
10:39:28 <|amethyst> TZer0:  okay, scoring daemon is restarted, will take an hour or two to rebuild player pages and finish the first update
10:39:42 <|amethyst> TZer0:  it was on 81 before, not 82, btw
11:41:42 <|amethyst> TZer0:  I assume it's also going to be https://underhound.eu/crawl/morgue and /crawl/ttyrec once they're there?
11:42:28 <|amethyst> TZer0:  also, I just updated the sequell config in git, but I imagine snark still needs to update the running bot?
11:59:38 <SteelNeuron> hey |amethyst
12:00:04 <|amethyst> SteelNeuron: hi
12:00:17 <SteelNeuron> Good to see our time zones line up
12:00:23 <|amethyst> SteelNeuron:  I didn't get a chance to read the backlog, can you fill me in on the discussions so far?
12:00:47 <SteelNeuron> There hasn't been a lot in the way of discussion, I made a writeup for a possible direction for WJC
12:00:53 <SteelNeuron> Gammafunk has uploaded it to the wiki
12:01:26 <SteelNeuron> But we were mostly waiting for your take on it, since I wrote it trying to go in line with what you were considering for the god
12:01:34 <SteelNeuron> hopefully we can make some variation of it work
12:01:57 <|amethyst> where at?
12:02:45 <|amethyst> I don't see it on recent changes
12:03:17 <|amethyst> oh, I'm looking at the wrong wiki
12:05:23 <Henzell> Unstable branch on crawl.akrasiac.org updated to: 0.21-a0-442-g848f44b (34)
12:05:56 <SteelNeuron> sorry, let me fetch that link
12:06:22 <SteelNeuron> https://github.com/crawl/crawl/wiki/Wu-Jian-Council-Changes
12:12:42 <|amethyst> SteelNeuron:  I assume pin just prevents movement, not attack?
12:12:53 <SteelNeuron> exactly
12:13:04 <|amethyst> I like these, though
12:13:17 <|amethyst> I think 120% on lunge is small
12:16:08 <|amethyst> hm, I'm not sure how whirlwind lets you get one tile of distance
12:16:45 <Cheibriados> Ru sacrifice mutations duplicate when extended by transient mutations. 13https://crawl.develz.org/mantis/view.php?id=11281 by NormalPerson7
12:16:45 <SteelNeuron> I think I've been quite bad at explaining it :)
12:16:45 <|amethyst> you do the whirlwind move and pin a monster, but that means you ended adjacent to it
12:16:45 <SteelNeuron> the duration of the pin effect would start right after the monster has moved
12:16:50 <SteelNeuron> so in some way, it lasts two "turns"
12:17:04 <SteelNeuron> I just wanted to express it that way to avoid indefinite stacking
12:17:14 <SteelNeuron> if it just gave, say, 20 auts of pin, then it would stack up and that's not the intention
12:17:25 <SteelNeuron> but rather that there are 10 auts left of pin time after the monster had its counterattack
12:17:41 <|amethyst> it doesn't have to stack up
12:18:00 <|amethyst> you can say 20 aut of pin, and that it caps at 20
12:18:11 <|amethyst> so just "set the pin duration to 20" instead of adding
12:18:51 <SteelNeuron> gotcha
12:18:52 <SteelNeuron> that works
12:19:04 <|amethyst> also, I'd rather not do "measured by your movement speed" there... other than Chei, there's not much that gets better when you are slower
12:22:45 <|amethyst> re wall jump, maybe it makes sense to make it a delay so it actually counts as two moves
12:23:34 <|amethyst> that would also allow some flexibility in deciding where the two turns are spent
12:23:53 <|amethyst> e.g. you can spend the first turn at your source location and the second turn at the target
12:24:40 <|amethyst> 20-aut actions are weird because you perform the whole thing at once, then (normal-speed) monsters get two opportunities to advance/swing
12:26:27 <|amethyst> also, as long as score is based on "turns" instead of AUT, you don't want speedrunners/highscorers using walljump repeatedly to increase score (like breadswinging, except it moves you) :)
12:27:31 <|amethyst> OTOH, maybe the ability would be too unattractive if you took hits before you got the benefit
12:31:36 <SteelNeuron> Well
12:31:37 <|amethyst> On lunge, I'm not sure where the 120% came from... repeated lunge/pin combo would need to be  lunge 120%, move adjacent 80%, back up (monster pinned), repeat, so that's 200% of attacks in 3 moves
12:31:55 <SteelNeuron> Sorry, I was only half online
12:31:57 <SteelNeuron> now I can pay attention
12:31:58 <SteelNeuron> :)
12:32:25 <SteelNeuron> so about wall jump, the front loading is meant to be the tradeoff for the ability
12:32:39 <SteelNeuron> I'm not saying it's the right solution, but I was thinking of giving it a relatively niche use
12:33:03 <|amethyst> when you say "front loading"?
12:33:03 <SteelNeuron> niche being that it's good for alpha striking a monster or group of monsters in exchange for the risk of more burst damage back
12:33:06 <SteelNeuron> exactly
12:33:11 <|amethyst> ah
12:33:18 <SteelNeuron> if you allow the flexibility of staging it, then the risk is gone and it becomes mandatory
12:33:25 <|amethyst> so front-loading outgoing damage, back-loading incoming damage
12:33:29 <SteelNeuron> exactly
12:33:33 <|amethyst> what's that trading off against?
12:33:56 <|amethyst> because that's strictly better than (say) alternating outgoing, incoming, outgoing, incoming
12:33:58 <SteelNeuron> You are losing fine control, in exchange for the possibility of eliminating the threat completely before they attack back
12:34:12 <SteelNeuron> it is better, yes, but if it doesn't work well, two incoming turns of damage may be lethal
12:34:24 <SteelNeuron> and you're probably in the middle of a pack :)
12:34:37 <|amethyst> ah, so the downside is that you don't get to decide in the middle to stop attacking
12:34:40 <SteelNeuron> exactly
12:34:45 <SteelNeuron> it's a single execution that may go poorly
12:34:57 <SteelNeuron> barring Serpent's Lash, which would take the risk away for a piety price
12:35:37 <|amethyst> I still think it's better to count it as two turns, regardless of when the counterattacks happen
12:35:58 <SteelNeuron> yeah that's fair, I don't remember enough of the back end to comment on that
12:36:09 <SteelNeuron> there is a secondary tradeoff here, which is that you need to be careful with your positioning
12:36:17 <SteelNeuron> if you land only one tile away from a monster, you made a mistake
12:36:25 <SteelNeuron> since you'll hit nothing but they will be able to advance and hit you once
12:36:49 <|amethyst> what do you hit with wall jump?
12:36:59 <|amethyst> the proposed version
12:37:06 <SteelNeuron> 1 tile aoe around you
12:37:12 <|amethyst> when you land?
12:37:16 <SteelNeuron> yes
12:37:41 <SteelNeuron> so like the old one, aside from the higher AUT count
12:37:49 <SteelNeuron> and no damage modifiers
12:38:13 <|amethyst> and no special bonus for axes?
12:38:31 <SteelNeuron> so far, axe cleave in WJC has not worked with everything except lunge
12:38:42 <SteelNeuron> I considered making cleave work with lunge enough of an axe incentive, because you can actually hit a monster twice
12:38:51 <SteelNeuron> if you whirlwind against it, while lunging to something else and cleaving
12:39:10 <SteelNeuron> s/everything/anything
12:39:17 <|amethyst> don't all the things let you hit monsters twice because of the attack # compensation?
12:39:22 <SteelNeuron> oh sure
12:39:25 <SteelNeuron> you're right
12:39:30 <|amethyst> but I guess that one is guaranteed twice
12:39:35 <SteelNeuron> I was commenting on a 10 aut move, 10 aut attack situation yep
12:40:07 <|amethyst> I'm not sure how well randomly rounded attack numbers work, but no one's complained about those yet so
12:40:39 <|amethyst> An alternative might be to make lunge and whirlwind take attack speed rather than movement speed, but that's kind of weird
12:41:21 <SteelNeuron> Getting to the rounded attack numbers was a long journey :), before (not sure if you got around to playing these versions) attack damage scaled with the speed ratios
12:41:37 <|amethyst> yeah, scaling damage sounds problematic
12:41:37 <SteelNeuron> that didn't work very well since it really helped slow races with fast weapons get through armor
12:41:44 <SteelNeuron> yep, naga stabbers were fun tho
12:41:51 <SteelNeuron> I've never seen as many exclamation marks
12:42:16 <SteelNeuron> rounded attack numbers are nice because they do not help stabbers that much, since only the first attack gets the stabbing bonus
12:42:24 <SteelNeuron> I'm open to other methods however
12:42:30 <SteelNeuron> but this one seems to work
12:42:53 <SteelNeuron> One thing I considered was making it deterministic by keeping a running modulo
12:43:14 <SteelNeuron> but I didn't like that either since good players would exploit it
12:43:15 <|amethyst> then players have to keep track of that too
12:43:16 <|amethyst> yeah
12:43:23 <SteelNeuron> by "missing" against a rat a couple times and then lunging an orc
12:43:50 <|amethyst> what do you think about my suggestion of scaling the martial moves to attack speed?
12:43:57 <|amethyst> then there's only one "attack"
12:44:11 <SteelNeuron> what aspect of them would you scale?
12:44:11 <|amethyst> (except for wall jump), but it's faster the better you are with your weapon
12:44:16 <|amethyst> the time
12:44:27 <SteelNeuron> I'm not sure how that would look
12:44:32 <SteelNeuron> after all, martial attacks trigger off movement
12:44:39 <SteelNeuron> so you are saying that your move speed would change near enemies?
12:44:44 <|amethyst> so for a human with mindelay battleaxe, it takes 0.7 to do that move instead of 1.0
12:44:47 <|amethyst> yeah
12:45:00 <SteelNeuron> hm, that's interesting, I never thought of that one
12:45:04 <|amethyst> and for a centaur with untrained battleaxe, it would take 2.0
12:45:08 <SteelNeuron> there are some strange implications however
12:45:19 <|amethyst> yeah, I'm not sure if it would be a good idea
12:45:21 <SteelNeuron> one thing I'd worry about is this:
12:45:28 <SteelNeuron> you are in a pack, wielding a battleaxe, and want to escape
12:45:34 <SteelNeuron> you then swap to your dagger so you move faster
12:45:39 <SteelNeuron> seems strange
12:46:09 <|amethyst> hm
12:47:27 <SteelNeuron> Is there anything particular that you find problematic with scaling number of attacks? I'm not opposed to changing it, but I need to know the angle you're coming from
12:47:54 <|amethyst> more attacks per player action = more message window spam :)
12:47:54 <SteelNeuron> Oh, good point
12:47:54 <|amethyst> and harder to keep track of what happened
12:48:14 <SteelNeuron> There has always been some degree of what I was describing before (about Wall Jump) in WJC in the sense that you give up control over the fight
12:48:26 <SteelNeuron> if you have a mindelay dagger and lunge with it, you're basically tabbing thrice
12:48:33 <|amethyst> Yeah, I can't think of a way to avoid the "swap weapons to move better" problem
12:48:56 <SteelNeuron> but it's the same message window spam than tabbing three times I suppose, just condensed in one action
12:50:46 <SteelNeuron> Oh and something you said I didn't comment on
12:50:48 <SteelNeuron> > On lunge, I'm not sure where the 120% came from... repeated lunge/pin combo would need to be  lunge 120%, move adjacent 80%, back up (monster pinned), repeat, so that's 200% of attacks in 3 moves
12:51:23 <SteelNeuron> Yeah that's right, It would be equivalent to tab in the sense that the monster has also only been able to attack you for two turns
12:51:41 <SteelNeuron> but then you'd have no reason to be doing that instead of tab, unless you wanted to reposition for something else, which is more or less the goal
12:52:10 <SteelNeuron> If you raise Lunge a bit more, you have the issue that this "triangle" motion is better than tabbing and there is no reason not to do it, if you want to play optimally, and we're back to square one
12:53:06 <SteelNeuron> I came to 120% by setting whirlwind's 80% first, which seemed about right for the benefit you get, and then giving Lunge the converse. Any combination like 70% and 130% could work, but we'd need to be careful not to overnerf whirlwind
12:53:19 <|amethyst> but it would need to be 220% to be equivalent to tabbing
12:53:24 <|amethyst> in the absence of other monsters
12:53:28 <SteelNeuron> how so?
12:53:45 <|amethyst> because you have to spend a turn to move away from the monster, and then you're not getting an attack
12:53:48 <|amethyst> oh I see
12:53:56 <|amethyst> because you're not getting attacked either
12:53:58 <Tux[Qyou]> how can I feel a craving for dungeon's cuisine when I'm a mummy?
12:53:59 <SteelNeuron> exactly
12:54:23 <SteelNeuron> so if your base damage is 20, you do 12 first, then 8, (meanwhile taking 20 damage from the monster), walk back for no damage from anyone
12:54:30 <SteelNeuron> sorry, if your base damage is 10*
12:54:36 <|amethyst> Tux[Qyou]:  It's a horrible unsatisfying feeling
12:54:55 <Tux[Qyou]> |amethyst: it's apparently not
12:55:38 <|amethyst> Tux[Qyou]:  I don't mean that's what the message says :)
12:55:57 <|amethyst> Tux[Qyou]:  just that you can have a craving for physiologically impossible things :)
12:59:35 <|amethyst> SteelNeuron:  my concern with 120% is that it doesn't feel like much more than a regular attack
12:59:52 <SteelNeuron> what was the number right now? It's been a while
13:01:31 <|amethyst> 160% if distracted, 130% if not
13:01:37 <SteelNeuron> I see
13:01:46 <SteelNeuron> Well, at 130%, people were getting good use out of it
13:01:53 <SteelNeuron> maybe 130%-70% is the sweet spot
13:02:00 <|amethyst> maybe 120% is okay
13:02:09 <|amethyst> it is still a "free" attack after all
13:02:18 <|amethyst> unless you had a polearm
13:02:20 <SteelNeuron> I just think it's important to keep that symmetry so the optimal path isn't going in triangles
13:03:10 <|amethyst> hm, triangle polearm strat
13:03:33 <|amethyst> get 180% attacks while taking only one move's worth of attacks
13:03:40 <SteelNeuron> yeah polearms have always been more exploitable with WJC, but then again they always are if you can get distance at will
13:04:20 <|amethyst> anyway, I personally don't care that much if WJC players need to make weird dances to optimise their attacks
13:05:01 <|amethyst> as long as that doesn't make it optimal to take WJC instead of whoever else
13:05:22 <SteelNeuron> well said :)
13:05:28 <SteelNeuron> After some time out of the game, I've changed my view a bit
13:05:44 <SteelNeuron> I used to say WJC was all about movement and that you shouldn't be even thinking about tabbing when picking it
13:06:08 <SteelNeuron> But right now, what I see as the best outcome is having it so tabbing and dancing is roughly equal for most of the time
13:06:20 <SteelNeuron> with punctual situations where you get a better outcome from "dancing"
13:06:46 <|amethyst> hm
13:06:47 <SteelNeuron> it can be subtle, just a whirlwind+lunge combo here, a good wall jump there, and a very cheap Serpent's Lash would be in charge of pushing the power level up
13:07:06 <SteelNeuron> so if you really want to ramp the effectiveness of the martial attacks up, you mix them up with Serpent's Lash
13:07:15 <|amethyst> if lash is cheap I could see getting rid of HoE
13:07:27 <|amethyst> (still thinking of the possibility of moving that to Usk)
13:07:37 <SteelNeuron> Well
13:08:00 <SteelNeuron> I can't comment on whether HoE is a better match for Uskayaw, I don't play Uskayaw much, so I can only talk about it in the context of WJC
13:08:17 <|amethyst> I have to admit I haven't used it
13:08:24 <SteelNeuron> The role that it holds in WJC is forcing you to enter full dance mode for a very significant reward
13:08:40 <SteelNeuron> not only that, but rewarding you for actively hitting as many monsters at once as possible
13:08:48 <SteelNeuron> there isn't really anything else in the WJC kit that rewards this kind of reckless play style
13:09:15 <|amethyst> that's the thing, that feels more like Uskayaw playstyle
13:09:20 <SteelNeuron> could be
13:09:30 <SteelNeuron> I'm not opposed to moving it to uskaway, but I think there should be a replacement
13:09:35 <SteelNeuron> because it does have a role in WJC
13:10:46 <SteelNeuron> I like that it puts you in a situation where you *have* to think ahead in terms of jumps and tricky moves to keep it going
13:10:56 <SteelNeuron> and how it works with Serpent's Lash to help you guarantee you always have a path
13:11:27 <SteelNeuron> An alternative for WJC could be this:
13:12:03 <SteelNeuron> A "combo meter" that counts the number of consecutive different martial attacks. If you reach 5, you get an active that costs a good chunk of piety and deals a lot of damage or similar
13:12:14 <|amethyst> hm
13:12:22 <|amethyst> that's even more like Uskayaw :)
13:12:34 <SteelNeuron> well, there is a difference
13:12:44 <SteelNeuron> Uskayaw is almost entirely AoE based
13:12:54 <SteelNeuron> while this would be more about precision, you wouldn't have to hit multiple enemies anymore
13:13:09 <SteelNeuron> just being able to seamlessly chain wall jump-lunge-whirlwind-lunge-whirlwind for example, which isn't trivial
13:13:19 <SteelNeuron> and then you get access to the "secret technique"
13:13:29 <SteelNeuron> which tbh could take any shape that moves the god away from Uskaway's space
13:13:52 <|amethyst> hm, good point, but unfortunately what about "dance diagonally around a single enemy" :(
13:13:52 <SteelNeuron> then it won't work
13:13:56 <|amethyst> oh, different
13:13:57 <SteelNeuron> as they only count if they are different
13:14:09 <SteelNeuron> consecutively different at least
13:14:31 <|amethyst> that could be kind of neat
13:14:43 <SteelNeuron> for example, you need 2 wall jumps, 2 lunges and 2 whirlwinds without repeating any
13:14:48 <|amethyst> I'll think about it, though you made a good case for keeping HoE on WJC too
13:15:12 <|amethyst> SteelNeuron:  oh, I was thinking "no two repeated in a row"
13:15:18 <|amethyst> SteelNeuron:  which is simpler to keep track of
13:15:20 <SteelNeuron> yeah yeah me too
13:15:29 <SteelNeuron> this last point was just an alternative
13:17:18 <SteelNeuron> Some variation:
13:17:46 <SteelNeuron> Ultimate: Every monster in the vicinity that you have whirlwinded, wall jumped and lunged gets massive damage (at a piety cost)
13:18:08 <SteelNeuron> cost based on the number of monsters
13:18:25 <SteelNeuron> that would be cool and thematic... Internal damage death punch kind of thing
13:21:10 <SteelNeuron> could be 5 piety to use, 2 piety for each affected monster, so it isn't quite linear
13:21:37 <SteelNeuron> and also rewards precision instead of Usk-style indiscriminate AoE
13:23:40 <gammafunk> There's not great need to move HoE to usk if the ability fits on WJC
13:24:02 <gammafunk> I more wanted to do this so we could use the parts of WJC that were more liked at the time if we were going to remove it
13:24:19 <gammafunk> and it would be a good fit there, but it's also fine if it fits the now-existing WJC
13:24:49 <gammafunk> MPA raised some concerns about the actual implementation of HoE on a technical level, but don't remembe what those are
13:25:03 <gammafunk> and that's probably less of a concern compared to getting the god's design nailed down, anyhow
13:25:35 <SteelNeuron> Oh HoE's code is a bit shit, I give you that
13:25:44 <SteelNeuron> thankfully it's quite small
13:26:05 <SteelNeuron> I just wanted a non linear slaying increase/decrease and I probably didn't express it the best way possible ;)
13:26:24 <SteelNeuron> I can clean it up if need be
13:32:36 <gammafunk> also "Uskayaw is almost entirely AOE based" is not really true
13:33:58 <SteelNeuron> Fair enough
13:34:04 <SteelNeuron> More like... dealing a lot of damage quick I guess
13:34:09 <SteelNeuron> which is easier against multiple targets
13:34:38 <SteelNeuron> in any case, the ultimate ability, while necessary I think, isn't the particular bit of WJC that I'm most fussed about
13:35:06 <SteelNeuron> But I do think my proposal could at least be a good attempt to fix the relationship between all three martial moves and Serpent's Lash, so it may be worth a try
13:37:17 <|amethyst> I like the proposal as written, with the caveat that wall jump should somehow count as two player turns (which both solves the serpent's lash issue and prevents silly turncount scumming)
13:37:43 <|amethyst> hopefully you're going to implement most of it so I can sit on my ass and take credit
13:37:51 <|amethyst> I am a *professor* after all, not a grad student
13:38:32 <|amethyst> (well, "assistant professor*" with an asterisk)
13:39:58 <|amethyst> hm, on wall jump ev
13:40:09 <|amethyst> should the duration depend on the number of attacks you made?
13:40:13 <|amethyst> or just the strength?
13:40:49 <|amethyst> One possibility is to have the two decoupled, with a prop to keep track of the size of the bonus
13:42:11 <|amethyst> another would be to gave, say, (20 aut * #attacks) duration (extending if greater, rather than adding to, existing duration) and make the bonus equal to (auts remaining / 10)
13:42:31 <|amethyst> s/to gave/to give/
13:42:50 <|amethyst> but that's quadratic scaling in a sense
13:43:27 <|amethyst> whatever you do, there's still a benefit to wall-jumping periodically (at the very least, when the duration expires), but not constantly
13:46:06 <gammafunk> you're not a real professor to me until you manage to win a VSBe
13:46:33 <gammafunk> yeah, I just got done reading all the backlog
13:46:37 <|amethyst> oh, I heard US News and World Report was terrible at ranking CS programs, is that why?
13:47:26 <gammafunk> because the ranking doesn't properly include DCSS wins of VSBe?
13:47:27 <gammafunk> probably
13:48:53 <gammafunk> re the EV bonus, SteelNeuron had mentioned maybe wall jump doesn't need it, not sure if that's still the case
13:49:31 <gammafunk> oh yeah, this is because serpent's lash might get its cost lowered and become avail earlier
13:49:39 <gammafunk> so your incentive to wall jump would be a bit greater
13:57:09 <gammafunk> it sounds like having attacks be randomized/averaged based on the move speed is working out best overall
14:00:41 <TZer0> |amethyst: fixed.
14:00:56 <|amethyst> TZer0:  morgues and ttyrecs you mean?
14:01:06 <|amethyst> TZer0:  they're at the URLs I mentioned?
14:01:14 <TZer0> Yeah
14:01:17 <TZer0> I broke things when I moved them around
14:01:23 <TZer0> See https://underhound.eu/crawl/meta/
14:01:32 <gammafunk> TZer0: figure out if any data loss happened?
14:01:33 <TZer0> And https://underhound.eu/crawl/ttyrec/ etc.
14:01:42 <TZer0> Very little, I'm going to go over things later
14:01:47 <TZer0> May be missing some recordings
14:01:52 <TZer0> Maybe a morgue-file here or there
14:01:59 <gammafunk> cool
14:02:01 <TZer0> Logfiles/scoring are all good
14:02:27 <gammafunk> I guess if a save went missing it would just be "bye bye that game" and not really cause any headaches
14:03:19 <TZer0> Right now I'm trying to figure out passthrough in my VM
14:03:30 <TZer0> Once that is done.. the server should be good to go
14:07:10 <TZer0> gammafunk: I think it is very unlikely that someone actually lost a game - especially one they cared about/remember right now
14:09:21 <gammafunk> ??cue
14:09:21 <Sequell> cue[1/5]: CUE (underhound.eu) is currently offline until the server can be replaced. As of 2017-10-29, a server has been ordered to arrive in 1-5 days, and it will need time to be configured, so there may be a week or more of down time.
14:10:59 <gammafunk> !learn set cue[1] CUE (underhound.eu) is currently offline due to a disk failure, but should be back online in the next week. See status updates here: https://underhound.eu/2017/11/04/serverstatus/
14:10:59 <Sequell> cue[1/5]: CUE (underhound.eu) is currently offline due to a disk failure, but should be back online in the next week. See status updates here: https://underhound.eu/2017/11/04/serverstatus/
14:12:59 <|amethyst> TZer0:  btw, your web site's header still uses :81
14:13:10 <|amethyst> TZer0:  at https://underhound.eu/2017/11/04/serverstatus/
14:13:38 <TZer0> Ah, will fix
14:18:26 <TZer0> |amethyst: how do I debug dgamelaunch again?
14:19:51 <|amethyst> TZer0:  it makes a log file, in the usual config in the chroot root directory, with terse indecipherable messages
14:20:05 <|amethyst> then you look up those messages in the source code to see what the actual problem was :/
14:22:24 <TZer0> I seem to have identified the problem anyway
14:22:32 <TZer0> For some reason my passthrough works fine for root, but not for other users.
14:23:05 <|amethyst> passthrough?
14:23:26 <TZer0> Yeah, what I've done is that I'm running crawl inside of a vm with some directories mapped from the host
14:23:41 <TZer0> Such as morgues
14:24:17 <|amethyst> ah, and the permissions (or the VM-guest shared filesystem implementation) are such that only root can access those?
14:24:30 <TZer0> It is supposed to work for anyone right now
14:24:35 <TZer0> I'm looking at the perms and they're 777
14:25:56 <|amethyst> the shared-FS thing might not actually use perms, or might have some other issue
14:26:05 <|amethyst> does the user that's running the VMs have access?
14:26:20 <|amethyst> (and which VM?)
14:27:04 <TZer0> QEMU/KVM
14:28:41 <|amethyst> ah, my only experience is with virtualbox
14:53:03 <TZer0> I resolved it
14:53:13 <TZer0> But now I'm getting other kinds of errors
14:53:14 <TZer0> So that's fun.
14:53:18 -!- aditya1 is now known as aditya
14:53:42 <TZer0> https://pastebin.com/00Qikzj0
14:53:44 <TZer0> Hmm
14:55:13 <TZer0> Do you think this may be a problem with locks vs. having this specific folder in a share?
15:02:57 <TZer0> holy damn, I'm actually playing in terminal now
15:04:28 <TZer0> still getting the bad channel message though
15:15:01 <geekosaur> that's a version mismatch somewhere
15:15:32 <TZer0> Yeah, that doesn't sound unlikely.
15:16:38 <geekosaur> what version are you playing? I vaguely recall that being a thing in some version
15:17:01 <TZer0> This was latest from like a couple of weeks ago
15:17:20 <TZer0> What kind of mismatch are we talking about? Compile libs. vs current libs on system?
15:18:58 <geekosaur> I mean when prayer was removed, some startup messages weren't adjusted to fit?
15:19:20 <geekosaur> and refer to the pray channel which was removed when prayer was
15:19:24 <geekosaur> I think
15:30:34 <|amethyst> hm
15:31:04 <geekosaur> I do recall this being a thing for a bit, I vaguely recall it having to do with startup messaging
15:31:25 <geekosaur> but I didn't spot it checking the log messages
15:31:35 <|amethyst> I'm not seeing "pray:" anywhere, particularly in dat/descript/
15:32:15 <|amethyst> does /crawl-master/crawl-git-291f19a919/saves/cache.0.21-a0-404-g291f19a/db/ have reasonable-looking permissions?
15:32:28 <|amethyst> SteelNeuron:  anything you want me to work on?
15:40:36 <SteelNeuron> Sorry, I got distracted
15:40:37 <SteelNeuron> again
15:40:49 <SteelNeuron> On the topic of EV, I'm leaning more towards it not needing the bonus, it seems more elegant that way
15:41:05 <SteelNeuron> even without the bonus, there are still situations where you'd want to use it even without Serpent's Lash
15:41:28 <SteelNeuron> On the topic of two-turn Wall Jump, I'm not fussed either way but I'm not sure how I'd implement that
15:41:42 <SteelNeuron> I don't think I have the back-end knowledge to do that
15:41:53 <SteelNeuron> while just increasing the AUT count for the action is comparatively easier
15:42:00 <SteelNeuron> lastly, on actually getting work done to implement it
15:42:43 <SteelNeuron> this is a bit more touchy because the primary reason why I stopped working on DCSS altogether is that I realize the unhealthy amount of time I was spending on it (WJC alone was 400+ hours of work, spectating and testing) and realized I didn't really have the time
15:42:52 <SteelNeuron> This doesn't mean I wont' work on these changes though, I'm willing to do that
15:43:36 <SteelNeuron> But, it does mean that in order for me to work them, I need at least some guarantee that they will be tested in trunk. Last time we had a conversation like this (back then, it was about some changes that Lasty and I had been discussing/proposing) I put the hours, opened the PR, and it was blocked on the ensuing PR discussion
15:43:48 <SteelNeuron> without ever being tested
15:44:28 <SteelNeuron> So I'd rather make sure that all the discussing is done before writing it, because I really can't justify the amount of work that went into fleshing out every variation that did not make it to trunk
15:44:29 <SteelNeuron> hope it's not an unreasonable expectation
15:47:06 <SteelNeuron> (oh and btw, if increasing the turn count by +2 instead of +1 when using the wall jump ability counts, that should be easy, I'm just not sure how to segment it into two actual, complete turns)
15:47:08 <gammafunk> yeah, there is at least one dev (|amethyst) who's behind the current direction and I think elliptic and I (the other two who have some time right now to look at things) are willing to compromise on the design of the god if some improvement can be made
15:47:33 <gammafunk> elliptic recommended removal to me (and I agreed) when it didn't seem like anyone had time/interest to work on things more
15:47:56 <gammafunk> I can help out with feedback and code as well, so certainly getting an improved WJC tested in trunk shouldn't be a problem
15:48:39 <gammafunk> It would be good if, when you think you've got the major changes in a near-final state, elliptic took a look at them (I haven't heard anything really problematic with proposals so far)
15:48:58 <SteelNeuron> cool, that sounds good to me
15:49:14 <SteelNeuron> And just to be clear I'm not asking for any sort of guarantee that they'll be well received after the trunk testing
15:49:21 <SteelNeuron> only testing can proof it
15:49:40 <SteelNeuron> I'm just making sure that all the pre-trunk discussion is closed before writing the code
15:49:40 <gammafunk> I think the main thrust of elliptic's complaint was how, even if you allow a god that requires these movement decisions, the two movement abilities didn't introduce very good decisions
15:50:22 <gammafunk> making whirlwind attractive to use and making it have some predictability in terms of the player planning their moves should help
15:51:15 <gammafunk> and yeah, testing is great, but where the magic happens is mostly in the design/planning; it's easy to get into a cycle of "just change it, then see what players say", but that can be a way of just trying to push the hard work onto players
15:52:41 <gammafunk> obviously people do release entire games around "just change it, let players yell at you", so I suppose it has its place
15:53:01 <gammafunk> but we like to maintain this pleasant fiction of "we make good designs in DCSS" :)
15:53:36 <SteelNeuron> :) well then, I'll put aside some hours to work on it
15:53:41 <SteelNeuron> as soon as I readjust to the DCSS codebase
15:57:37 <|amethyst> TBH, I was happy with the version before the wall-jump targetting restrictions
15:57:58 <|amethyst> but I guess we should make gf and (curve)ic happy
15:58:25 <|amethyst> I'm not hard to please
15:59:22 <|amethyst> FR: bring back player Twisted Resurrection, Evaporate, Ely piety-for-pacification, and meleebug, so I can win again
16:01:50 <|amethyst> (also, buff Trog; surely e should guarantee a win for VS)
16:37:46 <gammafunk> |amethyst: just think how good the game will be if a chess grandmaster is happy with it
16:37:57 <gammafunk> and if a twitch celebrity is happy with it
18:34:46 <Rotatell> fartboss (L22 GnWz) ASSERT(in_non_diamond_int(r.start)) in 'ray.cc' at line 217 failed. (Shoals:1)
20:01:20 -!- amalloy_ is now known as amalloy
21:05:31 -!- amalloy is now known as amalloy_
22:04:26 -!- amalloy_ is now known as amalloy
23:01:53 -!- amalloy is now known as amalloy_