About the new DPS test dummy
Now that the dps test dummy has been out for a few days and people have been posting a lot of their showcases on different subreddits and youtube I started to lose the feel for what the dps standard is for an average player, or maybe I only now started to concern myself with this because previously we couldn't actually see our teams' dps but now we can.
Anyway I wanted to ask everyone how this has affected your opinion on your own teams and characters? I personally was very disappointed with Mavuika (I knew she was worse without Citlali, whom I don't have, but I could still clear with her so I didn't notice it) when I found out she was much much worse(75K dps) compared to Arlechino(100K dps).
But most importantly I wanted to see, what kind of dps do you guys think is the standard amount of dps an average player has in the game right now and *how high should our expectations for our teams' dps be?*
Anyway I wanted to ask everyone how this has affected your opinion on your own teams and characters? I personally was very disappointed with Mavuika (I knew she was worse without Citlali, whom I don't have, but I could still clear with her so I didn't notice it) when I found out she was much much worse(75K dps) compared to Arlechino(100K dps).
But most importantly I wanted to see, what kind of dps do you guys think is the standard amount of dps an average player has in the game right now and *how high should our expectations for our teams' dps be?*
Commentaires
Although honestly it is good to know that vertically investing in waifu was the right play as people are in an absolute panic about only doing 50K with their "meta" comps, while C6 BAE doubles that solo so I can run Tights Tour of Teyvat with confidence of clearing.
That's 3 and a half pulls per patch.
Also, for some reason replacing Chevreuse on my Arrlechino overload team with C1 Mavuika turned out to out-dps the Chevreuse team. I am very confused about that...
Now if you been clearing easy enough before nothing changes with DPS dummy added. It's just there for players that want to push more.
I'm not surprised DPS levels are low because a lot of players are allegeric to Crit Rate.
Mavuika literally has bigger numbers than Arlechinno in all teams.
Arle does more damage than Mavuika if you're not tryharding.
You should be disappointed with yourself not the character.
- theorycraft only goes so far, what you get in practice probably varies a lot, the vast majority of players doesn't get anywhere near the theorized numbers with their teams
- DPS doesn't really matter as much as people make it sound. If Mavuika could be so meta defining while at the same time being so "not that high" if you aren't tryharding combos and cancels, it doesn't really matter what the ceiling is. Most people won't reach it. A 100k dps character with straightforward gameplay will be more useful for the average player than a 150k dps character that requires a perfect rotation otherwise it dips to 75k.
That's why wheelchairs are good deals. Ineffa + Columbina takes 0 effort to combo and produces amazing damage output, and you can use them alongside whatever onfielder you want to play and not have to worry about how you do your rotation.
with ororon, chevy and iansan. Mavuika is C0 with R1 serpent spine, all supports are c6 I think.
Now the other 150k+ dps without melt team. Her overload team is Ofcourse the option. This typically requires c6 chevruese and c6 iansan for the higher end dmg profile. Luckily they reran together twice since mavuika banner. Now overload donuts vs combos is a pretty big difference unlike vape team. The hardest combo vs 8-9 donuts is around 127% better lmao. But there’s many combos between that and donuts that are significantly better still. The easiest combos is 7cdn1s. That’s 69% better then donuts and easy to remember and will beat arle. 7 charge attack + dash + n1. Her dash on motorcycle is an attack. (Make sure your geared appropriately for overload mavuika). Like vape and melt mavika, there may be gear differences. If you want a more complicated combo for slightly more gains , cdcdcf 2(dcdccf) for 87% more dmg over donuts. That’s charge attack dash charge attack dash charge attack finisher followed by 2 sets of dash charge attack dash charge attack charge attack finisher. But it’s much harder to do Ofcourse.
To be honest, I was expecting Durin, but Albedo has C3+Uraku, and it might swap now that I can get Durin's proper artifacts rather than whatever I had laying around.
<150k: nope
Anyway, 60k DPS is enough to 36* Abyss. You can extrapolate the average player's DPS from there.
Being one or two seconds off on any rotation, or waiting for an enemy to stop stalling makes spreadsheet dps go out the window, and that it makes characters that can get their dps out without needing such strict rotations worth investing into.
Spreadsheets are only valid in specific situations to begin with.
i.e., you can't compare numbers from 2 different theorycrafters because they're probably using different assumptions and situations.
They're only good for A/B comparisons within calcs from the same theorycrafter.
I'd say that around 50k is the "average good team"
I was disappointed with the DPS on single target I could do with Mavuika too... but she demolishes anything in AoE and multiwave, so there's that.
Not everything is just raw damage, but HOW that damage gets to the enemy. Consider AoE, Atk frequency, elemental Aplication, damage profile (is it frontloaded, backloaded, or evenly distributed on the rotation?), kit synergies, survivability, consistency, comfort (sometimes a character is great, but is so tiring or boring to play that you as the player start to underperform).
And then there's ping. I did 20% less dps on my phone in an area with bad connection where my ping was hitting 400-700ms, it is a huge deal.
trying to min-max for max damage has always been an unnecessary personal pursuit by the few, not the majority of players.
The average DPS is 20k, many people don't build teams, it's probably lower. expectation is around 70k.
But it's cool.
People would have more fun if they cared more about meta.
Looking these numbers is like watching news. Cool if you know. But your life is most likely just going to be fine if you don't.
The thing is: in practice, there are layers of complexity that can push away these numbers from the synthetic benchmark.
Enemy positioning, down time, gimmicks will majorly affect your performance more than you'd think it'd be. Heck, even damage profile matters quite a bit given some circumstances. I found my C2R1 Mualani is exceptional on any single target without gimmicks. Her 1st shark chomp easily score ~1.2m for which executed in ~1 second. And most mobs just disappear. The standard 7s, 15s, 21s DPS window in benchmark doesn't get to see the whole play.
I generally don't pull for DPS characters that often and would rather invest in the ones I have (dps cons, support cons, sigs), so I'm pretty happy that they're all still relevant.
Across 4 rotations each:
- C1R1 Wriothesley Melt (8 cost) - 157k
- C0R1 Neuvillette Hyper (12 cost) - 158k
- C0R1 Varesa Overload (3 cost) - 143k
- C0R1 Nefer Hyper (5 cost) - 177k
- C6R5 Gaming Melt (5 cost) - 160k
- C0R0 Mavuika Melt (8 cost) - 225k (disgusting)
Key supports:
- C2R2 Furina (lost to her sig twice, go figure)
- C2R1 Xilonen
- C2R0 Citlali
- C1R0 Kazuha
- C0R1 Xianyun
I often TC my own team builds to find out best rotations and comps, but the test dummy had proven that calcs have limits. My current Wrio Melt team was calced at 139k and in reality it's 157k. That's a huge jump and I'm happy with that!
It has skewed my views a little tho. Makes me feel that anything below 100k is kinda stinky. You'd want 100k to comfortably clear Abyss nowadays, and maybe even higher for Fearless. So I guess that's really the standard now.
If you have Xilonen, give melt with Durin a shot! It'll be much better I think.
Revelatory for showing me faults in my teams and mechanics. My hypercarry, hyperinvested Noelle was barely hitting 100K – energy at fault, as well as being single target. Some unexpected, viable teams I hadn't considered: full wheelchair team (no DPS), was my highest at 130k; Ningguang suddenly at 115K @ 120 secs! Ecstatic. I am a 1.0 try-hard f2p for perspective.
Oldschool Keqing aggravate with Kazuha/Fischl/Nahida barely reached 45k DPS while using Albedo instead of Kazuha actually gets me up to around 52k DPS. Using Bineffa wheelchair with Sucrose gets me to around 70-80k lol.
A better way used to be to fight an enemy with known HP and see how long it takes while managing the fights mechanics. This limits rotations and give you real world performance. A test dummy doesn't fight back and stands perfectly still and has no mechanics you need to play around.
Plus, most damage showcases on it cherry pick the section or do a single burst for a high DPS when what should really matter is a DPS for the first X seconds it would take to do Y damage where that number is tied to beating a specific boss.
I think it was definitely needed overall, but what I really would have liked is an in-game plugin that tracks damage in real fights. If they can make a DPS tester that's just real fights with a log at the end, that would be perfection.
Spreadsheets have never been a good way to evaluate a character.
I'd say the good news is that players are finally starting to understand this.
There are simply variables you can't calculate.
I'm not going to be full clearing Dire SO anytime soon, but I'm satisfied with fearless, and I like being able to play around with diffetent weapons and artifacts to get different results.
The feature I appreciate most is actually the combat log. I usually have a hard time getting a clear understanding of how much and how often each character is proccing damage, and the combat log makes it easy to see.