Register a SA Forums Account here!
JOINING THE SA FORUMS WILL REMOVE THIS BIG AD, THE ANNOYING UNDERLINED ADS, AND STUPID INTERSTITIAL ADS!!!

You can: log in, read the tech support FAQ, or request your lost password. This dumb message (and those ads) will appear on every screen until you register! Get rid of this crap by registering your own SA Forums Account and joining roughly 150,000 Goons, for the one-time price of $9.95! We charge money because it costs us money per month for bills, and since we don't believe in showing ads to our users, we try to make the money back through forum registrations.
 
  • Post
  • Reply
No Wave
Sep 18, 2005

HA! HA! NICE! WHAT A TOOL!
You can deal chip damage to them to get them low then use Diluc/Keqing/Pyronado/geo MC/whatever.

Adbot
ADBOT LOVES YOU

Tae
Oct 24, 2010

Hello? Can you hear me? ...Perhaps if I shout? AAAAAAAAAH!

MarcusSA posted:

Ok who has the best ult/ skill to kill these in 2 seconds?



I did it with big D before but I think I just got lucky.

Whittle down each one, then charge up Lisa's Skill.

Nakar
Sep 2, 2002

Ultima Ratio Regum

dogsicle posted:

i've hit runs where i was only getting a few pieces of a given set like that, as well as others with basically nothing of one set in favor of another. with cups you probably notice bad stats more because they have a wider pool of mainstats to roll but otherwise there's not supposed to be and secret weight to certain outcomes
Based on some research I saw, I think it is weighted in a certain sense. As far as I know the set is a coinflip and each slot is equally weighted, but the research (which I sadly can't locate now) said that the odds for the Sands/Goblet/Circlet mainstats are essentially this:
  • 25% ATK
  • 25% DEF
  • 25% HP
  • 25% "Other"
Where "Other" is Elemental Mastery and whatever else is available in that slot, all at an equal chance. So on a Sands, the least complicated one, it'd be 75% for ATK/DEF/HP and 12.5% each for ER% or EM; the alternative case where every mainstat is fully independent would have a 20% chance for all five. Anecdotally, I don't think Energy Recharge and Elemental Mastery are anywhere close to 40% of all Sands that drop for me, so the notion of the 25% "Other" mainstat table matches my intuitions (though of course I'd need a lot of data that I don't have at present).

The difference between the two drop chances is significant over large sample sizes.
  • For fully equal mainstat chances, you'd expect to roll a great artifact (ATK or ER Sands, any DMG Goblet, and a CR or CD Circlet) about 26.44% of the time, an okay one (any Feather/Flower, EM on anything, ATK% Goblet or Circlet) 53.35% of the time, and a bad one (DEF/HP, Healing Bonus Circlet) 20.21% of the time.
  • With the "Other" odds, a great mainstat artifact drops to a 14.37% chance and a bad one jumps to 31.25% (okay odds stay about the same, mostly because Feathers/Flowers make up 40% of artifacts dropped).
  • The odds of a random mainstat artifact (Sands/Goblet/Circlet) being good would be about 2 in 3 if mainstat chance was equal.
  • If mainstat chance isn't equal, the odds of a random mainstat artifact being good is about 48%.
Since most artifacts are intentionally supposed to be bad, and since I see EM on anything incredibly infrequently, my guess is it's the "Other" table as equal chances would mean that most mainstats would, on average, be something you're happy to see while substats would be what let you down. My experience is you get let down in both places!

Also note that this isn't taking into account the fact that artifact domains have two 5☆ sets, because that should balance out over a very large number of runs and not substantially influence the ratios for actual pieces, but it does mean if you don't want both sets in the domain that you're basically going to have to spend twice as much resin, and it means that half the time that already-rare Crit-related Circlet is going to be the wrong set (which is a problem if you're using an offset Goblet, which statistically is the most likely compromise you'll have to make).

Certain sets also desire certain pieces more, but I didn't bother going to the effort of mathing that out. You know, Gladiator being pretty okay with any DMG booster, Crimson Witch wanting Pyro Goblet over any other DMG booster, Bloodstained not wanting any DMG booster but Physical (generally), Elemental Mastery being significantly more valuable on Venerer and less valuable on Petra, Healing Bonus actually being tolerable on Maiden and maybe Noblesse, etc.

None of this takes substats into account because that would just make me very sad, but the main takeaway is that the first place you should be pursuing great substats is on Feathers and Flowers, because their consistent mainstat means any Feather/Flower is usable if it has great subs, and it's more likely to have good subs because one of the three flat stats is taken out of the pool. Research on substats is pretty unclear and I've heard conflicting things about how they're weighted but my guess is each substat is equally weighted. That might not be true, which would make the substat odds even worse, but they're so bad as it is that you really shouldn't worry about it and just use anything that happens to roll well, but especially any Feathers/Flowers that have four good ones as you're just more likely to see those.

EDIT: As far as how substats are rolled, I have to think it rolls the first slot, chooses the substat, rolls the second with the same pool minus the original (or perhaps rerolls when it hits a sub that it's already assigned?), repeat for all substats (which are probably predetermined and just revealed when leveled enough). It might also do a randomization of the visible order once all substats are assigned, that would make it impossible to tell which got rolled in what order, but I haven't studied the patterns of substats to tell if that might be true.

Nakar fucked around with this message at 21:11 on Dec 18, 2020

No Wave
Sep 18, 2005

HA! HA! NICE! WHAT A TOOL!
This is the infographic I saw once:



EM helmet randomly being super rare is consistent with my experience.

No Wave fucked around with this message at 21:10 on Dec 18, 2020

Cao Ni Ma
May 25, 2010



MarcusSA posted:

Ok who has the best ult/ skill to kill these in 2 seconds?



I did it with big D before but I think I just got lucky.

this is the dumbest loving daily and I wish it never existed

dogsicle
Oct 23, 2012

i'll be chasing 1-2 EM VV hats forever

Nakar
Sep 2, 2002

Ultima Ratio Regum

No Wave posted:

This is the infographic I saw once:



EM helmet randomly being super rare is consistent with my experience.
Same, but I'm curious as to where they got these figures. If they analyzed a lot of drop data then it may well be that each mainstat in each slot is given a preset percentage, although I'll note that the percentages for ATK/DEF/HP are pretty close to 25% everywhere but the Goblet.

Guessing at the weighted odds that infographic suggests would give us:
  • SANDS: 25% each ATK/DEF/HP, 12.5% ER/EM. Essentially the same as the "Other" table.
  • GOBLET: 20-22% each ATK/DEF/HP, 30-33% DMG (~4.29% for any specific type), 1-10% EM. Most likely 22/30/4 but would require way more data to be sure.
  • CIRCLET: 20-22% each ATK/DEF/HP, 33% Healing/CR/CD (~11% for any specific type), 1-7% EM. I'd guess 21/33/4 but again that's a wild guess.
With those numbers plugged in, things are actually a little better than the "Other" table would've produced. You'd get a good mainstat around 18% of the time, an okay one 53% of the time, and a bad one 29% of the time. For random mainstats only, they'd be alright a little more than half the time (50.99%). The main beneficiary of these odds is DMG% Goblets and Crit Circlets, which would jump from ~6.9% of all artifact drops to ~10.4%. On the other hand, Elemental Mastery falls from ~4.4% of all artifact drops to ~4.1%, mainly because the odds of EM Circlets drop but EM Goblets rise; the overwhelming majority of EM drops will be Sands though, essentially 50% more common than EM Goblet/Circlets combined.

Moral of the story: If this is correct, never throw out an EM Circlet or Goblet at 4-5☆, because you have no idea when you'll ever be seeing one again.

Framboise
Sep 21, 2014

To make yourself feel better, you make it so you'll never give in to your forevers and live for always.


Lipstick Apathy

moush posted:

Maybe they change what 4* to include depending on initial reaction to the 5* so they can try to wring the most money out of people if it looks like no one's going to pull.

As for the event, it seems like they're already using it as a kind of survey because I got a lot of question about mechanics for characters. The UI is at least better than the Childe/Zhongli one a week or so ago.

Is my sense of time just that absolutely hosed, or was that really just a week ago?

Space Opera
Jun 5, 2011

That rabbit's got a vicious streak a mile wide! It's a killer!

Cao Ni Ma posted:

this is the dumbest loving daily and I wish it never existed

I can usually do it with "chip them down, then pluge attack/noelle's ult" but I think that's the configuration where I have to use QiQi's ult to get them all in one go. It wouldn't be so bad if the harder configs gave you 2-3 more seconds. I almost always skip this particular one and the "destroy the balloon over a water course" because they just arent worth it.

No Wave
Sep 18, 2005

HA! HA! NICE! WHAT A TOOL!
Stone Harbor was before the Zhong Li banner so no, it was about a month ago. It was supposed to fill the dead time before 1.1 released but came out at the beginning of 1.1 due to bugs that were found.

Ashenai
Oct 5, 2005

You taught me language;
and my profit on't
Is, I know how to curse.

Framboise posted:

Is my sense of time just that absolutely hosed, or was that really just a week ago?

It was a month ago, Nov 13-22.

Nakar
Sep 2, 2002

Ultima Ratio Regum
So yeah with those weighted mainstat averages an on-set EM Circlet/Goblet is the rarest possible mainstat roll.

You'd expect to get an on-set DMG% Goblet (of some sort, not necessarily matching the set) around once every 34 runs, an on-set Crit Circlet (Rate or Damage) every 46 runs, an on-set matching DMG% Goblet every 234 runs, and an EM Circlet around once every 250 runs. Yeah, more than five times rarer than a Crit Circlet and less likely to drop on average than the specific type of Goblet you want. That's less than one on average per month of resin dumped into the domain (assuming 160 resin/day is spent); you'd expect 3 EM Sands in the same span.

Gaardean
May 28, 2012

For that training dummy commission, if you pick the second response when you first talk to the NPC (it mentions something about "with friends"), you'll get a torch in the middle of them you can light up with a fire attack and then hit with any lightning attack for an Overload proc to knock them all out. Makes it simply trivial.

Fur20
Nov 14, 2007

すご▞い!
君は働か░い
フ▙▓ズなんだね!

MarcusSA posted:

Ok who has the best ult/ skill to kill these in 2 seconds?



I did it with big D before but I think I just got lucky.

the timer doesn't start until you break one of the targets, so you can just dust them lightly until they have a sliver of health left and use any aoe you like. why the dummies don't have 1hp to begin with i will never understand.

Nakar posted:

So yeah with those weighted mainstat averages an on-set EM Circlet/Goblet is the rarest possible mainstat roll.

thank god, i don't need MORE crappy stats gumming up my already crappy drops. you can get almost as much EM from secondaries as you can from a primary stat, its growth rate as a primary a real bummer.

Fur20 fucked around with this message at 21:59 on Dec 18, 2020

Bussamove
Feb 25, 2006

Klee’s E will also jump across all of them if you throw it from the side.

MarcusSA
Sep 23, 2007

Gaardean posted:

For that training dummy commission, if you pick the second response when you first talk to the NPC (it mentions something about "with friends"), you'll get a torch in the middle of them you can light up with a fire attack and then hit with any lightning attack for an Overload proc to knock them all out. Makes it simply trivial.

Lol that’s a solid tip! I’ll do that next time.

I just took them down a bit and did a plunge attack. They actually do have quite a bit of HP fwiw.

Flair
Apr 5, 2016

MarcusSA posted:

I just took them down a bit and did a plunge attack. They actually do have quite a bit of HP fwiw.

Yeah the HP scales with your world level.

Nakar
Sep 2, 2002

Ultima Ratio Regum
More math: If you're wondering how much more sensible it is to chase good substats on a Feather/Flower instead of the random pieces, consider that to get a 5☆ DPS kit of a Crit Circlet (of some sort), a DMG% Goblet (of the right element but not necessarily from the right set), and an ATK% Sands, you'd need around 120 runs of a domain. That's basically one of each (3 of the ATK% Sands), so you'd be kinda stuck with whatever substats you get. In that time you'd have about 12 each of Flowers and Feathers, which would give you a lot more choice in substats and a lot more chances to roll good ones. The perfect Crit Rate% Circlet with ideal substats could potentially come after hundreds of Feather/Flowers, so don't consider those pieces crappy just because they have a flat mainstat; the overall quality of your best ones is likely to be higher with less of a luck factor involved.

Fur20 posted:

thank god, i don't need MORE crappy stats gumming up my already crappy drops. you can get almost as much EM from secondaries as you can from a primary stat, its growth rate as a primary a real bummer.
While most people don't want EM mainstat, it does help VV sets quite a lot and especially is what Sucrose wants more than anything else (in some circumstances, carries like Diluc can also deal more DPS with an EM over ATK% Sands). A +20 mainstat has 187 EM, while a natural 4-subs artifact that rolls EM five times can have between 96-138 EM. The difference between 138 and 187 is not huge, but the worst case scenario is like half as much and the mainstat growth is predictable. Plus all your substat rolls have to go into the EM, which isn't very high-value compared to Crit stats.

Mostly this hurts Sucrose and to a lesser extent Venti; a VV support Jean probably would rather have ATK% to bolster her healing anyway. Venti gets away just fine as a support with a DPS set though (Crit/Anemo/ATK), so yeah it's mostly anyone trying for a BiS Sucrose VV set with perfect substats is gonna be doing that domain approximately forever.

LorneReams
Jun 27, 2003
I'm bizarre
I'm not at the end game yet, but is anything gated by this insane min-maxing, or is this essentially e-peen and masturbation?

No Wave
Sep 18, 2005

HA! HA! NICE! WHAT A TOOL!
I don't consider finding mainstat artifacts in the set you want to be insane min/maxing. You'll see what I mean eventually, most artifact farming is just a byproduct of trying to find a crit rate/crit dmg helmet. The only things at stake are biweekly primogem rewards from the spiral abyss.

Something like having an EM hat vs an atk hat on sucrose, sure, that's min/maxy and probably not totally necessary. But you're going to want to farm for artifact fodder to upgrade your artifacts anyways so you're going to be doing it somewhere.

Fur20
Nov 14, 2007

すご▞い!
君は働か░い
フ▙▓ズなんだね!

LorneReams posted:

I'm not at the end game yet, but is anything gated by this insane min-maxing, or is this essentially e-peen and masturbation?

abyss 12 is tough even with a lot of optimized equipment and focused builds, but ofc you don't need a perfect set of gear on everyone and you can get away with things like "underleveled and undergeared kaeya/xiang/xing supports" outside of your ~6 mains

Tae
Oct 24, 2010

Hello? Can you hear me? ...Perhaps if I shout? AAAAAAAAAH!

LorneReams posted:

I'm not at the end game yet, but is anything gated by this insane min-maxing, or is this essentially e-peen and masturbation?

Abyss, but the returns are so low that it's not worth the time invested in it.

So basically exactly like Fate Grand Order.

Xun
Apr 25, 2010

LorneReams posted:

I'm not at the end game yet, but is anything gated by this insane min-maxing, or is this essentially e-peen and masturbation?

I'm at AR47 and I haven't ran into anything that's gated except maybe abyss but it could also easily be execution issues.

I just go for good main stats and hope for good substats, if they're poo poo and I roll a possibly better artifact I try to replace them though.

No Wave
Sep 18, 2005

HA! HA! NICE! WHAT A TOOL!

Fur20 posted:

abyss 12 is tough even with a lot of optimized equipment and focused builds, but ofc you don't need a perfect set of gear on everyone and you can get away with things like "underleveled and undergeared kaeya/xiang/xing supports" outside of your ~6 mains
Xiangling is one of the most important characters for players to level to beat abyss 12 (unless they own both Diluc and Klee, but even then...). I don't think she's someone you should neglect.

Ultiville
Jan 14, 2005

The law protects no one unless it binds everyone, binds no one unless it protects everyone.

LorneReams posted:

I'm not at the end game yet, but is anything gated by this insane min-maxing, or is this essentially e-peen and masturbation?

To be clear about what "Abyss" means, a small amount of challenge dungeon rooms with biweekly rewards are locked behind it, but no exploration regions, no story quests, nothing like that. You get a small mechanical bonus for engaging with all of it but you can still explore everything and see all the story and stuff no problem without it.

dogsicle
Oct 23, 2012

with common mainstat like ATK% i think you can afford to be pickier, but i wouldn't expect most people sit and farm for in-set crit/EM/ER/ele% pieces and that's fine. just getting the baseline of +16 4-5* with good mainstats in relevant set(s) is going to be fine.

there's not a whole lot worth using resin on at AR45+ so just idly running domains for a chance at those better artifact outcomes isn't really hurting you though, and it's the most number go up thing

Nakar
Sep 2, 2002

Ultima Ratio Regum
A decent team with +16 artifacts with the right mainstats can clear Abyss 12 fairly easily. No other content in the game requires being any stronger right now, it's just fun to flex and annihilate things.

But the difference substat optimization makes is pretty huge in the theoretical case of perfect artifacts. Tartaglia with 4pc Depth of Heart in the right mainstats (CR%/Hydro%/ATK%) and no substats at all could do something like 11k DPS with a 60% Crit Rate. Tartaglia with the same artifacts (except CR% switched to CD% on the Circlet) but all substats perfectly assigned would do 39k DPS with a 100% Crit Rate. To be clear, 11k DPS without even figuring in reactions and attacks from other characters is already quite high and very few things will stand up to that, even lv100 enemies, but a fully-optimized character would be over three times as powerful. Same weapon, same level, same mainstats, just perfect subs vs. no/poor subs, triple the power.

There's a reason they don't want it to be easy.

EDIT: Just for fun, I tested what would happen if you had substats all in the right places but rolled the minimum for each possible increase and had the increases spread more or less evenly across all substats. That makes Tartaglia about 67% stronger than no substats at all. The godrolled artifacts make him 252% stronger. The former is far, far more realistic than the latter, so good substats do matter but not to the extent it might appear from me looking at the absolute best-case scenario (which will never happen; you might get one perfect piece ever the whole time you play the game in its current state, and God knows what set it'll be).

Nakar fucked around with this message at 22:55 on Dec 18, 2020

N-N-N-NINE BREAKER
Jul 12, 2014

Ultiville posted:

To be clear about what "Abyss" means, a small amount of challenge dungeon rooms with biweekly rewards are locked behind it, but no exploration regions, no story quests, nothing like that. You get a small mechanical bonus for engaging with all of it but you can still explore everything and see all the story and stuff no problem without it.

Until noobs get stonewalled by the story tartaglia fight, at least. Unless they changed the scaling on that already?

Personally I didn't mind the difficulty though. And maybe it's not a bad idea to slow people's progression by throwing really hard fights in that get nerfed later. Lots of phone games/mmos do that

Ashenai
Oct 5, 2005

You taught me language;
and my profit on't
Is, I know how to curse.

LorneReams posted:

I'm not at the end game yet, but is anything gated by this insane min-maxing, or is this essentially e-peen and masturbation?

There's a hard endgame dungeon (Spiral Abyss) that gives rewards, but I wouldn't call it "content" and it's extremely optional. People grind to beat it because once you're done with the story and exploration, what else are you going to do? It's just there to give the minmaxers something to do.

Everything else can be beaten without any grinding.

Tae
Oct 24, 2010

Hello? Can you hear me? ...Perhaps if I shout? AAAAAAAAAH!

N-N-N-NINE BREAKER posted:

Until noobs get stonewalled by the story tartaglia fight, at least. Unless they changed the scaling on that already?


Story Childe got nerfed a week later.

Bloody Emissary
Mar 31, 2014

Powawa~n
Rolling for Zhongli was 100% worth it. Tall character sprint speed, insta-mining, height-boosting pillars...I'm not using him much in combat yet since I'm low on artifact EXP and he's still below level 60, but the exploration utility he offers is great. My team of Chongyun/Xingqiu/Bennet can handle all the fighting while grandpa sits back and enjoys his post-retirement babysitting duty.

Ashenai posted:

New event, should be available from your Notices, unless it's not available in the US yet?



Beta? Hmm, I wonder if they're planning on expanding it or putting in a permanent version later on.

Ultiville
Jan 14, 2005

The law protects no one unless it binds everyone, binds no one unless it protects everyone.

N-N-N-NINE BREAKER posted:

Until noobs get stonewalled by the story tartaglia fight, at least. Unless they changed the scaling on that already?

Personally I didn't mind the difficulty though. And maybe it's not a bad idea to slow people's progression by throwing really hard fights in that get nerfed later. Lots of phone games/mmos do that

I didn't have much trouble with him, but also didn't face him immediately on release, so maybe he was nerfed?

LorneReams
Jun 27, 2003
I'm bizarre
There is no timer, food is plentiful, and pausing to eat takes no time or has no lag time. How is anything actually difficult? I'm like the most passive player, and the only things that are troublesome are timed poo poo, like kill them all in 30 seconds.

awesmoe
Nov 30, 2005

Pillbug

LorneReams posted:

There is no timer, food is plentiful, and pausing to eat takes no time or has no lag time. How is anything actually difficult? I'm like the most passive player, and the only things that are troublesome are timed poo poo, like kill them all in 30 seconds.

well, in the abyss, you cant eat and there are timers

Mailer
Nov 4, 2009

Have you accepted The Void as your lord and savior?
This game is a lot more fun gearing up supports and preparing my two teams for the eventual abyss clears than it was spending six weeks just doing requirements and hoarding gems for the eventual Bennet drop. :v:

Also gently caress the Noblesse Oblige domain so hard.

aparmenideanmonad
Jan 28, 2004
Balls to you and your way of mortal opinions - you don't exist anyway!
Fun Shoe

Mailer posted:

Also gently caress the Noblesse Oblige domain so hard.
You may know this already but it's significantly easier if you tag the start pillar and run back to the entrance. The hydro mage will just sit and wait for you to kill the other two.

(I'm assuming you're at max domain level)

LorneReams
Jun 27, 2003
I'm bizarre

awesmoe posted:

well, in the abyss, you cant eat and there are timers

I meant people having so much trouble with that event fight that they needed to nerf it.

gandlethorpe
Aug 16, 2008

:gowron::m10:

aparmenideanmonad posted:

You may know this already but it's significantly easier if you tag the start pillar and run back to the entrance. The hydro mage will just sit and wait for you to kill the other two.

You can also lure all 3 into the corner closest to the Hydro mage, then use a swirl to break all their shields!

No Wave
Sep 18, 2005

HA! HA! NICE! WHAT A TOOL!

LorneReams posted:

I meant people having so much trouble with that event fight that they needed to nerf it.
There are people playing this game on their phone using characters with +0 artifacts. Story content should be easy as heck and i'm quite surprised (pleased) that the weekly childe fight is as hard as it is.

Adbot
ADBOT LOVES YOU

King of Bleh
Mar 3, 2007

A kingdom of rats.

LorneReams posted:

I meant people having so much trouble with that event fight that they needed to nerf it.


Childe has multiple OHKO attacks and generally hits very fast and hard if you don't know how to dodge his attacks, and revival foods DO have a lengthy cooldown.

He's quite manageable once you learn his patterns, but so far that specific story fight has been the first and only actual difficulty spike outside of the Abyss (which is primarily just Big Numbers gently caress You anyway, and not truly difficult).

  • 1
  • 2
  • 3
  • 4
  • 5
  • Post
  • Reply