Skip to main content

Weapons

Bane of ... Series​

Lion's Roar Passive Applies to all forms of DMG inflicted on Electro-Debuffed Enemy​

By: Doug#8888

Evidence:
Keqing with no artifacts on and only a Lion's Roar. I autoattacked a hilichurl, cast stiletto for Electro-debuff, then autoattacked again. Here are the 2 white(physical) numbers recorded.

  • 400dmg - no debuff
  • 469dmg - with Electro aura

Conclusion:
Lion's Roar passive applies to all forms of damage inflicted upon Electro-debuffed monsters. This means that it is not only Electro-based damage being increased, but also Physical damage.

Lion’s Roar Passive Works while the Holder is Off-Field​

By: mikan#2834 and IonFox#8887

Finding:
Lion's Roar passive is still in effect and will boost damage even when equipped character is off-screen

Evidence:
YouTube

  • C2 Xingqiu with Fillet Blade and 631 total attack.
    • 1st wave: 161
    • 2nd wave: 183 with reduced Hydro Res
  • Same Xingqiu with base Lion's Roar and 635 total attack.
    • 1st wave: 186 damage on Electro-afflicted enemy
    • 2nd wave: 211 damage on Electro-afflicted enemy with reduced Hydro Res

Significance:
Equipping support characters with Lion's Roar/Dragon's Bane/Rainslasher that deal damage even when off screen is a viable choice to increase damage output. For example, XingQiu in an Electro (Keqing) or Pyro (Diluc) focused team.

Weapons with Bane Passive Applies to Amplifying Reactions, but not Transformative Ones​

By: Lix-#3417
Added: 2020-12-07 (v. 1.1)
Last tested: 2020-12-07 (v. 1.1)

Findings:
Can Lions Roar, 4TS, dragon's Bane, Raven's Bow increase reaction DMG?

  • The answer is no, for Transformative reactions
  • But yes for Amplifying Reactions (melt and Vaporize)

Evidence:
Test 1: EC

Test 2: Overload

Test 3: Vaporize

Significance:
Lions Roar and other weapons that have similar passive don’t increase transformative reactions damage but do increase Damage for amplifying reactions.

Blackcliff Series​

Blackcliff Sword effect Persists through a Character Swap​

By: Cenpi#3224

Evidence: YouTube

Blackcliff Sword effect doesn’t activate if a Support takes the Kill​

By: Cenpi#3224

Evidence: YouTube

Blackcliff Sword effect doesn’t activate if a CO-OP Member takes the Kill​

By: Cenpi#3224

Evidence: YouTube (refer to link above for original ATK values)

Blackcliff Sword Effect Persists Even if the Holder Switches Off-Field​

By: Lix-#3417
Added: 2020-12-07 (v. 1.1)
Last tested: 2020-12-07 (v. 1.1)

Theory/Findings:

  • Does Blackcliff bonus apply while the wielder is off field?
  • And can you still get the buff if you kill the enemy with your Skills while you're in the off field?
  • The answer is yes, it does apply to any char Skills that remains in the field even if the character already switched

Evidence:

  • Test 1: XQ using blackcliff without inactive buff
  • Test 2: XQ using blackcliff with the active Buff
  • Test 3: XQ using blackcliff killing enemies while in off field, does he still get the buff?
    • No, he does not

Significance:
From the test we can conclude that the buff can be applied to their Skills even if they are not in off field, this also works on Fischl's Oz and Xiangling's Guoba.

  • For example, you can get a kill on Blackcliff Fischl, cast Oz, then swap off and Oz will still receive the Blackcliff buff

The BC holder must be on field to activate the Blackcliff effect.

Favonius Series​

Favonius Weapons do not proc on Shielded Enemies​

By: Sakuno#7950
Added: 2021-02-04 (v. 1.3)
Last tested: 2021-02-04 (v. 1.3)

Evidence:
Testing was done using Diona @ around 40% CRIT Rate spamming normal attacks with an R1 favonius warbow, and a Xingqiu @ around 40% CRIT Rate using normal attacks with an R1 favonius sword. 3 minutes on average I attacked a Hydro abyss mage and a shielded cyro slime, on both characters. All 4 tests resulted in 0 procs. When shields were broken, favonius proc'd as expected (60% chance per CRIT). Shielded Fatui skirmishers on the other hand, does not prevent favonius from working since it does not block DMG completely like the others.

Short demonstration: Streamable link lost to the sands of time

Significance:
Favonius weapons suffer the same drawback as sacrificial and does not provide itself as an alternative weapon choice against shielded enemies. This leaves favonius where it was before, the inferior option against sacrificial unless you have no better.

Favonius Weapons Procs On Multiple Enemies​

By: Nass008#8577
Added: 2021-10-26 (v. 2.2)
Last tested: 2021-10-26 (v. 2.2)
Discussion

Finding:
Favonius weapons passive proc per enemy against multiple targets

Evidence:
Tested with R2 Favonius Sword (70% proc chance) and 100% CRIT Rate
Using Xinyan E (duration 12s) to time passive cooldown

Expected proc chance if the passive proc per enemy : 91%
1 - 0.3^2 = 0.91

Result : 134 attempts, 123 proc, 11 non-proc
91.8% proc chance

Video 1 (21 attempts)
Video 2 (48 attempts)
Video 3 (65 attempts)

Significance:
Favonius Weapons are more effective on character with large AoE abilities/against multiple targets

Favonius Proc Calculator​

By: Mr. Pillow#0065
Added: 2022-02-21 (v. 2.5)
Last tested: 2022-02-21 (v. 2.5)
Discussion

Theory:
Favonius weapon passive activation probability plotter (Desmos graph).

Evidence:
The graph plots the probability of activating the Favonius weapons' Passive against the holder's CRIT Rate, depending on the number of rotations, the number of hits per rotation and the refinement level of the weapon.

I will only summarize the derivation of the equation used, since the problem at hand is elementary. The assumption is that, after a hit, there are two possible ways to not activate the weapon's passive:
#1 Your attack deals CRIT DMG but the passive doesn't activate, with the probability P1 = C × (5-R)/10 and
#2 Your attack doesn't deal CRIT DMG, with the probability 1 - C,

where C and R are the CRIT Rate and Weapon Refinement Level, respectively. After h hits, there's then 2^h different ways to not activate the passive. However, most different ways haven't a unique probability of occurring. This makes it possible to simplify the total probability of not activating the passive to (P1 + P2)^h. This expression is then subtracted from 1 to give the probability of activating the passive, and this is raised to n to get the probability for n rotations. The inverse function is also included in the graph.

Significance:
Offers an easy way to visualize how the probability changes with the CRIT Rate. I initially made it to more easily come up with reasonable CRIT Rate assumptions for the KQMC.

Infusion Series​

Crescent Pike Procs Benefit From All Relevant Damage Bonuses​

By: Xreejan#1180
Added: 2022-07-19 (v. 2.8)
Last tested: 2022-07-19 (v. 2.8)
Discussion

Finding:
Crescent Pike Procs benefit from all relevant damage bonuses

Evidence:
Pike Procs without Geo resonance does 66 damage per proc: Imgur
Pike Procs with Geo resonance does 73 damage per proc: Imgur

Pike Procs with 4LW and 4TS:
Thanks to EatYourVeggies#7493 for providing the clips for 4LW and 4TS

4LW calc: 1391*(1+31.5%+35%)*40%*0.50*0.7=322
4TS calc: 1078*(1+31.5%+35%)*40%*0.50*0.7=249

Significance:
Similar to how Skyward Weapon procs benefit from Physical damage bonus. Crescent Pike Procs benefit from all relevant damage bonuses (phys, 4LW, 4TS, Geo Resonance) since it's based off the user's stats.

Lithic Series​

Lithic Weapons And Childe​

By: @pinneous
Added: 2024-06-06 (v. 4.7)
Last tested: 2024-06-06 (v. 4.7)
Discussion

Finding:
Childe does not contribute a stack to the Lithic Weapon Series while in the party.

Evidence:
YouTube (Footage provided by haafr)

Significance:
Characters that could reasonably use Lithic weapons will not get a stack from Childe while he is in the team. A character like Xiangling will have to account for not getting a stack of the Lithic weapon buff, notably the CRIT Rate buff, while doing calcs/sims/optimizing/building.

Millennial Movement Series​

Millennial Movement Stack Mechanics​

By: rakka#0099
Added: 2021-05-22 (v. 1.5)
Last tested: 2021-05-22 (v. 1.5)
Discussion

Theory/Finding/Bug: Millennial Effect Overlap Behavior

Evidence:
Different weapons with the Millennial Movement effect can stack, however effects of the same buff-type will override each other. IE Elegy at R5 produces a 40% ATK Bonus and a 200 EM bonus for the team however an R2 Song of Broken Pines produces a 25% ATK Bonus and a 15% normal ATK speed bonus. If Elegy procs first and Pines second, the EM bonus from Elegy will stay applied however the 25% ATK Bonus will override the 40% ATK Bonus and lower the overall effect of the shared buffs takes the most recent proc even if it takes the lower value.

Significance:
Millennial Movement passives can stack, however if a buff type overlaps (ATK, movement speed, EM, etc....) then the last effects will override the previous effects while leaving the non-overlap effects alone.
This means that mismatched refinements with overlapping effects can affect the complexity of when Millennial buffs need to proc, so either an Elegy and Pines on the same team should both be R1, R2, etc or it means that the player has to be aware of the order of operations for their swap comps.

Elegy Stack Inconsistencies​

By: Wüolf#3334
Added: 2022-03-09 (v. 2.5)
Last tested: 2022-03-09 (v. 2.5)
Discussion

Findings:
On-field stack gain is consistent across all FPS with damage ticks excluding Amber due to her multiple damage ticks/0.2s and sometimes Sara’s Burst where the different pairs of thunderbolts appear at different times.

Off-field stack gain:

  • Amber: We count stacks gained by the end of the Burst. At 30 FPS she can get 2-3 Elegy stacks, at 60 FPS she could consistently gain 4 stacks by the end of her Burst. Note for some people at lower pings there seems to be inconsistencies at 60 FPS.
  • Venti:
    • Q only: At 30 FPS it will take 7 damage ticks consistently to proc elegy across all pings from our testing. At 60 FPS Venti can proc the Elegy buff at 4 damage ticks, however for some people it isn’t consistent, Venti will proc Elegy at around 4-6 DMG ticks and very rarely 7.
    • E-Q : Using Vye’s consistent videos. At 30fps it takes 4 Q damage ticks after E to get 4 stacks. At 60fps it takes 3 Q damage ticks to get 4 stacks.
  • Sara: After the initial hit the 2 sets of lightning bolts can spawn at different times. Counting the damage ticks and stack count of 1 set of lightning bolts after the initial hit to make it less confusing. At 3 bolt damage ticks she could get 2 more stacks (total 3 stacks) and she could also get 1 more stack(total 2 stacks). At 60 FPS there are similar results inconsistently getting 2 or 3 stacks at 3 damage ticks.

Evidence:
Amber:

Venti:

Sara:

Significance:
Server sided registration on Elegy stack gain means that unstable ping can lead to different amount of damage ticks to proc 4 stacks. The longer time required to gain 4 stacks (if possible) needs to be taken into account for rotations with heavy and fast quick-swapping, to counteract the long time, Amber should stay on the field for 1 second or use E after using her Q before switching to 4-Piece VV supports in Hu Tao teams. Venti should always E+Q as it brings the time to gain 4 stacks down massively. Sara should use her C2 and Charged Attacks before using her Q to make sure she can guarantee the 4 Elegy stacks before switching to other characters.

Freedom-Sworn Interaction With Weapon Infuse​

By: borderline#3100
Added: 2022-04-08 (v. 2.6)
Last tested: 2022-04-08 (v. 2.6)
Discussion

Finding:
Freedom-Sworn does not gain a stack when a weapon gauge infusion takes place.

Evidence:
In this video Keqing equipped with Freedom-Sworn triggers an Overloaded Reaction and gains a visible stack. When infusing her weapon with Electro after having a Cryo Infusion, she does not gain the Freedom-Sworn buff, as seen by the visible stack remaining and her attack not increasing.

Significance:
This is not a method to trigger Freedom-Sworn's passive.

Royal Weapon Series​

The Effective CRIT Rate of Royal Weapons (& Simulator)​

By: Cola#6861

Evidence:

Tool used to run simulations for Royal weapon line CRIT hits.

Link

  • Ran simulations (using google sheet's randomize function) over 10,000 runs, each for 5%, 15% and 50% CRIT Rate.
  • Recorded the average CRIT occurrence for each set of 10,000 runs, for 10 times.
  • Averaged the results from 10 trials and came up with the average effective CRIT Rate.
  • The same was done for refinement rank 1, 3 and 5.

Data

Result

Significance

  • Avoid CRIT Rate circlet and minimize CRIT Rate rolls on your substats.
  • The royal weapon line behaves, strangely, I know.
  • If you are SOL on artifacts however, the Royal weapons can make for a decent choice.

How Royal LongSword’s Passive Works​

By: Pleinair#8321
Added: 2020-11-22 (v. 1.1)
Last tested: 2020-11-22 (v. 1.1)

Evidence: YouTube

Significance:
The sword is a rare case of actually working while off field. While Sacrificial sword is one of Xingqiu BiS this weapon would be more optimized for a dps build over the Harbinger of Dawn due to the overall higher raw stats it provides

By: IonFox#8887

Further elaborating on @Pleinair's submission, Royal Longsword works exactly as written even when XQ is off field.

More evidence from Pleinair: YouTube

Significance:
As with above conclusion, there might be a use case where Royal Longsword offers a higher DPS ceiling than sacrificial or becomes a valid F2P alternative if players can mitigate the recharge loss through substats or a Hydro DPS.

The video also reveals that there might be a tiny ICD on Royal Sword's ability, or that multiple Hydro swords in a single wave count as 1 hit or a single source of damage, since only 1 stack of CRIT was granted. The second seems likely given how each wave only applies one instance of Hydro now. However it has also been observed that swords can CRIT independently.

This also raises the question of other royal weapon use cases with off screen DPS, such as Royal Bow and Fischl's Oz.

EDIT: XQ's rain sword orbitals do not count as a hit when they apply Hydro, not CRIT boost there.

Royal Series Dynamic Stacking on Snapshot DoT Skills​

By: sunnyd#5916
Added: 2021-04-10 (v. 1.4)
Last tested: 2021-04-10 (v. 1.4)

Finding:
Lisa's Q snapshots CRIT stacks from Royal Grimoire but the damage from the Q ticks will dynamically change the number of stacks

Evidence: YouTube

Significance:
The passive of royal weapons continues to stack even when a snapshotting Skill deals damage over time. First, cast the Skill at low CRIT, and have the character gain CRIT from royal weapon's passive while the Skill maintains its low CRIT Rate because it snapshots. This allows stacks to be gained without an increased chance of losing them. Which means, the increased CRIT Rate can then be capitalized on by casting high damage abilities during that window of increased CRIT Rate.

Royal Series AoE Abilities and Mechanics​

By: Bobrokrot#0111
Added: 2021-06-25 (v. 1.6)
Last tested: 2021-06-25 (v. 1.6)
Discussion

Finding: AoE abilities do "snapshot" CRIT Rate.

Evidence:

  • Video 1
  • Video 2
    The tests were conducted as follows: gain 5 stacks of a Royal weapon, exit combat and then raise CRIT Rate up to >100%. Then use an AoE ability and counter the number of non-CRIT hits. If CRIT didn't "snapshot" and instead was calculated one by one, then after the first CRIT the CRIT Rate would drop to ~60% due to losing 5 stacks of a R1 Royal weapon and the 2nd instance of damage would have a reasonable chance to not CRIT. In the above tests, not a single event of non-CRIT has been observed which allows us with a reasonable amount of confidence claim that AoE abilities do "snapshot" CRIT Rate.

Significance:
For Royal weapons it means that they can potentially have an increased efficiency given a right rotation. Although you can't gain multiple stacks with an AoE ability, what you can do is take advantage of the stacks you already gained through single-target attacks. Damage dealt to all targets is guaranteed to have an increased constant CRIT Rate, although it requires further analysis whether it actually yields better efficient CRIT Rate or not for different rotations and combinations of base CRIT Rate and Royal refinements. One particular character that could take advantage of this that comes to mind is Yanfei since her typical attack pattern consists of 2-3 single-target attacks followed by an AoE attack (which coincidentally cares about scoring CRIT hits).

Finding:
Mechanics of Royal weapon stacks:

After gaining a stack you cannot gain another one for a small period of time (cooldown).

For an AoE ability, damage instances dealt by the ability, including damage dealt by reactions Overload, Swirl and Superconduct caused by the ability, are put in a list. Then the game applies the following logic to all elements of the list one by one:

  • if no_crit then
  • add CRIT_stack
  • add CRIT_stack_CD
  • if no_crit and CRIT_stack_CD=true then
  • do nothing
  • else remove CRIT_stack

Stacks can be gained when the character equipped with the Royal weapon is off-field.

Evidence:

  1. Sucrose's quick N1-N2 gaining 1 stack instead of 2 and multiple evidence of an AoE ability gaining no more than 1 stack when it non-CRITs multiple times. YouTube video
  2. After a single-target ability CRITs and causes a Swirl, the number of stacks is reset to 1 instead of 0 implying that damage from a Swirl is taken into account and is considered as non-CRIT. YouTube video
  3. Superconduct and Swirl chain reactions YouTube video
  4. Overload and Electro-Charged YouTube video

Significance:
Royal weapons have an increased efficiency in situations when the holder triggers Overload, Swirl or Superconduct due to the fact that they can potentially always have at least 1 stack.

Mathematical Analysis of Royal Weapon Effective CRIT Rate​

By: NZPIEFACE#8439
Added: 2021-10-28 (v. 2.2)
Last tested: 2021-10-28 (v. 2.2)
Discussion

Finding:
A mathematical formula to find the effective CRIT Rate of Royal weapons is not yet in the library, there is only a table and a simulation.

Use the following equation to analytically calculate for the final CRIT Rate after using Royal Weapons:

a = 1 - base_cr
b = 1 - base_cr - cr_from_royal
c = 1 - base_cr - cr_from_royal * 2
d = 1 - base_cr - cr_from_royal * 3
e = 1 - base_cr - cr_from_royal * 4
f = 1 - base_cr - cr_from_royal * 5
effective_cr = 1/(1 + a + a*b + a*b*c + a*b*c*d + a*b*c*d*e/(1-f))

Evidence:
Math and explanation in here: Google Doc
There is a python function in the doc to calculate effective CRIT Rate from character CRIT Rate and weapon refinement level. It's a useful tool if anyone ever wants to get this to work with fringe snapshot builds.

Significance:
It was found that the CRIT Rate that the weapons’ passive had decreased the higher the CRIT Rate of the character was. The passive of the weapons performs poorly in comparison to other CRIT value stat sticks (Blackcliff series that is also from the shop, Harbinger of Dawn, etc.), and is only really comparable at extremely low values of investment into CRIT Rate. While there is an argument that these would have value in the hands of new players who cannot invest into CRIT Rate that much, it would also be a terrible long term investment for them as the weapon would naturally become worse and worse as they play.

Sacrificial Series​

Average ER Gain from Sacrificial Weapons​

By: Raladin#3752
Added: 2022-02-02 (v. 2.4)
Last tested: 2022-02-02 (v. 2.4)

Finding: Average ER gain from Sacrificial Weapons has decreasing returns when increasing refinement.

Evidence:
Math and Explanation
Graph 1
Graph 2

Significance: This mathematically shows why units with low Skill cooldowns are not great pairings with sacrificial weapons, and also show which units are ideal for these types of weapons.

Sacrificial Weapons do not proc on Shielded Enemies​

By: Exuma#1917
Added: 2021-02-04 (v. 1.3)
Last tested: 2021-02-04 (v. 1.3)

Evidence:
Testing was done with Diona hold E with R1 sacrificial bow and XQ E with R1 sacrificial sword. Both units used 50 E Skills on shielded Pyro abyss mages and shielded Geo slimes with 0 procs for both cases. When shields were broken, sac weapon procs were as expected. (40% chance per hit on R1)

Short Demonstration: Streamable video lost to the sands of time

Significance:
This has many applications in combat, especially in abyss where timed trials will be affected by proc frequency and subsequently Burst uptime. Sacrificial weapons are among the best support options so understanding their mechanics are necessary.

Sacrificial Passive With Multiple E Stacks​

By: mol#3280
Added: 2021-05-06 (v. 1.5)
Last tested: 2021-05-06 (v. 1.5)
Discussion

Finding:
When a character with multiple charges of their Elemental Skill procs the sacrificial passive, the Elemental Skill cooldown shown on the timer is the cooldown that will be reset.

Evidence:
I cast one Sucrose E, waited until the first cooldown was down to 5s, and cast the second E. A reset occurred, and the time of the second cooldown was ~15s (minus 10% due to Anemo resonance in video). So, whatever timer is currently showing is the one which will be reset.

The same occurs for Ganyu C2 and Klee.

Significance:
Any serial-cooldown, multi-cast Skill user may waste sacrificial procs by causing a reset when their first cooldown is near 0s. Since the timer on stacked Skills begins with the first Skill cast out of the stack, and that Skill is the one which is reset (not the longer, still waiting cooldowns), resets can be completely wasted on characters with multiple stacks. For example, with Sucrose it is possible to reset a Skill with 0.1s of cooldown remaining, then have a stack of 15s go onto cooldown afterward. To get the most out of the cooldown benefit of sacrificial weapons, characters with multiple stacks might consider using stacks as quickly as possible.

Sacrificial Weapons Interactions​

By: gengar#8426
Added: 2021-08-19 (v. 2.0)
Last tested: 2021-08-19 (v. 2.0)
Discussion

Finding:
I documented every Sacrificial Weapons' interactions with characters that have Elemental Skills that hit multiple times. Check the Google Sheet in the Evidence for all the details.

Evidence:
Google Sheet

Significance:
Helpful for character building, documents which characters are able to proc sacrificial weapons with multi-hit Skills

Sacrificial Weapons Procs On Multiple Enemies​

By: Nass008#8577
Added: 2021-10-26 (v. 2.2)
Last tested: 2021-10-26 (v. 2.2)
Discussion

Finding:
Sacrificial weapons passive proc per enemy against multiple targets

Evidence:
Tested with R3 Sacrificial Sword (60% chance to proc) against two targets
Using Noelle E (cooldown 24s) to time passive cooldown

Expected proc chance if the passive proc per enemy : 84%
1 - 0.4^2 = 0.84

Result : 137 attempts, 114 proc, 23 non-proc
83.2% proc chance

Significance:
Sacrificial Weapons are more effective on character with large AoE Skills/against multiple targets

Sacrificial Weapons Can Proc Even If The Elemental Skill Is Not On Cooldown​

By: f̸̒͂ỏ̶̂o̵͌̚s̶͊̏h̷̤̀ḯ̴̊#9920
Added: 2022-12-18 (v. 3.3)
Last tested: 2022-12-11 (v. 3.3)
Discussion

Finding:
The passive of Sacrificial weapons can proc even if the Elemental Skill is not on cooldown.

Evidence:
Imgur

Significance:
Better understanding of Sacrificial weapon mechanics.

Elemental Skills That Don't Proc Sacrificial Weapons​

By: f̸̒͂ỏ̶̂o̵͌̚s̶͊̏h̷̤̀ḯ̴̊#9920
Added: 2022-12-18 (v. 3.3)
Last tested: 2022-12-11 (v. 3.3)
Discussion

Finding:
There are certain specific Elemental Skills that can't proc the passives of Sacrificial weapons, despite being considered Elemental Skill DMG. This includes Sayu hold E roll, Anemo Traveler hold E Cutting DMG, Mona A1 and Keqing C1.

Evidence:
Sayu Roll - Imgur
Anemo Traveler Cutting DMG - Imgur
Mona A1 - Imgur
Keqing C1 - YouTube

Significance:
Sacrificial weapons are not a reliable indicator of whether or not something is considered Elemental Skill DMG.

Skyward Series​

Skyward Harp​

Skyward Harp’s passive has the ability to CRIT​

By: Cola#6861
Added: 2021-01-31 (v. 1.2)
Last tested: 2021-01-31 (v. 1.2)

Finding:
When shooting with the Skyward Harp, it is possible to achieve a Critical hit on the passive ability.

Evidence:
User Cola Sen was discussing in theory crafting about the Skyward Harp when its ability to CRIT came into the conversation. Afterwards they decided to test out their hypothesis and reported on their findings:

[video] Skyward Harp passive scoring CRIT hits

Significance:
When using the Skyward Harp, it is now apparent that it is slightly better then we originally thought as its own passive gains benefits from the CRIT Rate and CRIT DMG that the Harp provides.

Skyward Blade​

Skyward Blade Description Clarification​

By: Special#1775
Added: 2020-12-06 (v. 1.1)
Last tested: 2020-12-06 (v. 1.1)

Finding:

  • Skyward blade does not increase your damage after Elemental Burst but adds an additional line as bonus damage using the values on the description.
  • Its separate line of damage CRITs individually from the normal attacks.

Current Description: "Increases the DMG of Normal and Charged Attack hits by 20% for 12s."

More Accurate Description: "Normal and Charged Attacks deal an additional 20% of Normal/Charged ATK as DMG for 12s" (this additional damage is dealt separately)

Evidence: YouTube

Significance

  • Less synergy with Electro builds since the bonus damage is a separate instance of white damage
  • Synergises well with Physical builds

Skyward Pride​

Skyward Pride Damage Interaction with 4 Piece Gladiators​

By: Chiefy#2036
Added: 2021-03-30 (v. 1.4)
Last tested: 2021-03-30 (v. 1.4)

Finding:
Skyward Pride's vacuum blades count as raw Physical damage and not as normal or charged attacks for the purpose of 4 set Gladiators.

Evidence: Abyss 7-1 with 3 gladiator pieces vs 4 gladiator pieces

Note: The glad goblet had a bit of DEF, so the video of 4 pc Glad had a bit more ATK than the bottom from Noelle C6 Q, resulting in 25 more vacuum blade damage.

Vacuum Blades can CRIT​

By: Phana#0420
Added: 2021-06-26 (v. 1.6)
Last tested: 2021-06-26 (v. 1.6)
Discussion

Finding:
Skyward Pride's vacuum blades can CRIT.

Evidence: YouTube

Significance: Calculations should account for the CRIT possibility of skyward pride's blades.

Skyward Atlas​

Skyward Atlas effect can CRIT​

By: Phana#0420
Added: 2021-06-26 (v. 1.6)
Last tested: 2021-06-26 (v. 1.6)
Discussion

Finding: Skyward Atlas' weapon proc can CRIT.

Evidence:
YouTube

Significance:
Calculations should account for the CRIT possibility of weapon procs.

Specific Weapon Mechanics​

A Thousand Floating Dreams​

A Thousand Floating Dreams Can Have More Than 3 Stacks​

By: addicted2reading#7262
Added: 2022-12-13 (v. 3.3)
Last tested: 2022-11-05 (v. 3.2)
Discussion

Finding:
A Thousand Floating Dreams can have more than 3 stacks as long as the stacks are not all of the same type. There is a maximum of 3 Elemental DMG Bonus stacks and 3 EM stacks. Trial characters count toward the passive, allowing players to have a total of 4 stacks.

Evidence:
YouTube

Significance:
Better understanding of how the weapon passive works.

Amenoma Kageuchi​

Seed duration​

By: MkO#1192

Gaining a new seed will refresh the duration of previously gained seeds.

Flat Energy Confirmation​

By: Sylfae#1210

Aqua Simulacra​

Aqua Simulacra Vertical Range Limit​

By: Yume#1519
Added: 2022-07-14 (v. 2.8)
Last tested: 2022-07-14 (v. 2.8)
Discussion

Finding:
Aqua Simulacra's passive, The Cleansing Form, seemingly has no vertical range limit.

Evidence:
To determine whether Aqua Simulacra's passive is in effect, the presence of the blue glow on the bow is checked.

Test 1: Testing whether an enemy above the bow user procs the passive. (The enemy is a whopperflower, which is also aggro'd in the end of the video)
Imgur

Test 2: Testing whether an enemy below the bow user procs the passive.
Imgur

As we can see, in both instances, blue glow is present in the bow, which means that the passive is active.

Significance:
Overworld stuff, fluff, more understanding on Aqua Simulacra's passive.

Aquila Favonia​

Aquila Favonia Proc And Shield Interaction​

By: gvo#4200
Added: 2021-10-26 (v. 2.2)
Last tested: 2021-10-26 (v. 2.2)
Discussion

Finding:
There is some inconsistency with shielded sword characters and aquila favonia procs.

Evidence:

  • YouTube: Bennett with Diona's shield
  • YouTube: Xingqiu with Diona's shield

The videos show:

  1. How Aquila does not proc when protected by a shield
  2. How Aquila can proc while in Burst i-frames, without a shield
  3. How Aquila can proc while in Burst i-frames, while protected by a shield

Significance:
This interaction may affect other "triggers on taking DMG" + shield interactions. The only other ability I can think of that could be similarly affected would be Zhongli A2.

Aquila Healing is Affected by Healing Bonus​

By: IAmOnYourSide#5431
Added: 2022-03-09 (v. 2.5)
Last tested: 2022-03-09 (v. 2.5)
Discussion

Finding:
Aquila's healing effects are affected by incoming and outgoing healing bonuses.

Evidence:
No healing bonuses

  • HP restored: 291
  • Bennett with almost no artifacts.

Hydro Resonance

  • HP restored: 379
  • 379/291 = 1.302405498281787
  • Yields 30% from Hydro Resonance as expected.

Hydro Resonance + 2-Piece Maiden

  • HP restored: 422
  • 422/291 = 1.4501718213058419
  • Yields 45% from Hydro Resonance + 2-Piece Maiden as expected.

Significance:
Aquila's healing effects are affected by incoming and outgoing healing bonuses. Prototype Amber and The Black Sword are expected to be the same, but were not tested.

Ballad of the Boundless Blue​

Ballad of the Boundless Blue stacks stay when off field​

By: @emmaus
Added: 2024-05-26 (v. 4.6)
Last tested: 2023-10-14 (v. 4.1)
Discussion

Finding:
Ballad of the Boundless Blue stacks do not reset offield. There's also no visible effect on gaining stacks.

Evidence:
YouTube - Yanfei still does the max 3432 CRIT / 1467 non CRIT damage even after swapping to Layla and back.

Significance: Useful to keep in mind while crafting rotations.

Beacon of the Reed Sea​

Beacon of the Reed Sea's passive does not trigger if shielded​

By: @twce_
Added: 2023-08-13 (v. 3.8)
Last tested: 2023-03-01 (v. 3.5)

Finding:
Beacon of the Reed Sea's DMG passive will not trigger if shielded

Evidence:
YouTube

Significance:
Characters that equip Beacon will want a team with no shielder

Beacon of the Reed Sea's passive affects Beidou's C1 shield HP​

By: @bobrokrot
Added: 2023-08-13 (v. 3.8)
Last tested: 2023-03-01 (v. 3.5)

Finding:
Beacon of the Reed Sea's HP bonus does affect Beidou's C1 shield HP.

Evidence:
YouTube

Significance:
nice defensive utility for Beidou.

Cinnabar Spindle​

Cinnabar Passive Deactivation Bug​

By: TheLocalMilkMan#0813
Added: 2022-03-16 (v. 2.5)
Last tested: 2022-03-16 (v. 2.5)
Discussion

Bug:

  • After hitting a shield (Abyss Mage, Mitachurl Shield, Kairagii Parry, etc) Cinnabar Passive will deactivate indefinitely until party swap/unequip and reequip the weapon.
  • This bug only happens with Albedo due to his specific Skill trigger requiring a hit to deal damage first. It'll require a non-melee hit to trigger, whether it be a projectile, AoE attack, or even burning grass.
  • There is, however, some slight inconsistency. The bug will not trigger 100% of the time, it may trigger even without a shield, and it may even reactivate.

Evidence:

Significance:
This will hurt Albedo's damage until this is fixed. This bug is known and has been known to MHY for several months now.

Cinnabar Spindle's Passive Is Dynamic​

By: Mr. Pillow#0065
Added: 2022-12-17 (v. 3.3)
Last tested: 2022-12-11 (v. 3.3)
Discussion

Finding:
Cinnabar Spindle's passive is dynamic, even for abilities that can snapshot.

Evidence:
In this video, Albedo's Elemental Skill snapshots Gorou's DEF buff and his Transient Blossoms are triggered twice, once when General's War Banner is active and once when it is inactive. The DMG calculations below show that Cinnabar Spindle uses Albedo's current DEF, i.e. its passive is dynamic.

  • General's War Banner active, 3611 DMG dealt: (2.405*(1480.4 + 438.09) + 0.80*(1480.4 + 438.09))*(1 + 0.288)*(0.50666)*(0.9)
  • General's War Banner inactive, 3405 DMG dealt: (2.405*(1480.4 + 438.09) + 0.80*(1480.4))*(1 + 0.288)*(0.50666)*(0.9)

Legend for values used: 2.405 = Transient Blossom Talent%, 0.80 = Cinnabar%, 1480.4 = Albedo's DEF, 438.09 = DEF Increase, 0.288 = DamageBonus, 0.50666 = EnemyDefMult, 0.9 = EnemyResMult.

Significance:
Confirmation that Cinnabar Spindle's passive is dynamic, even for abilities that can snapshot.

Cloudforged​

Cloudforged Stacks Refresh​

By: @clevernt
Added: 2024-06-07 (v. 4.7)
Last tested: 2024-06-07 (v. 4.7)
Discussion

Finding:
Gaining a stack of Cloudforged's passive refreshes both stacks.

Evidence:
YouTube
Gained a stack at 0s, and a second stack at ~16s.
at 24s, both stacks are still active, meaning the second stack refreshed the first.

Significance:
Cloudforged's passive can be utilized in short (=<18s) rotations even if the character does not have a way to drain energy outside of Burst.

Deathmatch​

Deathmatch Passive Snapshot​

By: Aftermath#7658
Added: 2022-07-29 (v. 2.8)
Last tested: 2022-07-29 (v. 2.8)
Discussion

Finding:
The ATK or ATK/Def bonus of the Deathmatch snapshots the last effect when the holder goes off-field. When swapping back on-field, the passive update is not immediate.

Evidence:
YouTube

When first equipping the Deathmatch, it will have the ATK boost passive active. Going to a group of enemies and swapping in will activate the split ATK/DEFbonus for 2+ enemies. If you then swap her back off-field and teleport away, the stat screen will show the DEFbonus is still active.

Opening the menu as Yun Jin swaps on-field still displays the DEFbonus at first, so there seems to be some sort of tickrate for the passive. Going back to the group of enemies with Yun Jin off-field, the passive still has the effect for 0-1 enemies.

Significance:
When you can run Yun Jin on deathmatch, the passive snapshot can sometimes maintain her buff stats even when enemies start to die or spawn.

Engulfing Lightning​

4 ESF Interactions​

By: Vye#9195
Added: 2021-09-11 (v. 2.1)
Last tested: 2021-09-11 (v. 2.1)
Discussion 1
Discussion 2

Finding:
4ESF DMG% bonus works with the ER% from Engulfing Lightning's effect but does not snapshot.

Evidence:
Buff Evidence - Vye's stats, shown in video, are calculated for Burst damage here assuming that the ER passive from Engulfing does affect the 4-Piece ESF set: (804+899)×(1+1.098)×(1+0.466+0.582)×(0.9)×100+70(100+70)+(100+85)×2.508=7909(804+899) \times (1+1.098) \times (1+0.466+0.582) \times (0.9) \times \frac{100+70}{(100+70)+(100+85)} \times 2.508 = 7909 Which about matches up with the damage number of 7906 in the video.
No snapshot evidence - In the video, Engulfing Lightning is being used. The ending hits of Xiangling's Burst are lower damage, indicating that the Burst DMG% from Engulfing Lightning's additional ER bonus on Burst cast has worn off.

Significance:
Engulfing + 4ESF have better synergy.

Engulfing Lightning ER% buff and off-field Interaction​

By: TWICE#9958
Added: 2022-11-24 (v. 3.2)
Last tested: 2022-11-24 (v. 3.2)
Discussion

Finding:
Engulfing Lightning's ER% buff stays on the character even after switching.

Evidence:
YouTube
Shenhe has R1 EL equipped, as you can tell from the NAs. 201.1% ER before swapping, keeps the ER% buff (for 231.1% ER total) after swapping

Significance:
In case people have questions of the buff not staying post-switch/off-field for some reason

Fading Twilight​

Fading Twilight Does Not Snapshot​

By: TheFirstRapher#9207
Added: 2022-07-14 (v. 2.8)
Last tested: 2022-07-14 (v. 2.8)
Discussion

Finding:
Fading Twilight doesn't snapshot

Evidence:
Excuse the scuffed clip but you can see my Fischl's CRITs doing 2008 first, then 2072, then 1944.
YouTube

Significance:
Can't keep the buff for the whole duration of a snapshottable Skill.

Hakushin Ring​

Hakushin Ring can buff the holder.​

By: jas#8888
Added: 2022-08-14 (v. 2.8)
Last tested: 2022-08-14 (v. 2.8)
Discussion

Finding:
Hakushin Ring can buff it's holder.

Evidence:

  • Anemo Trigger - Electro Swirl YouTube
  • Electro trigger - Electro Swirl YouTube
  • Electro trigger - Electro-Charged YouTube
  • Electro trigger - Overload YouTube
  • Electro trigger - Superconduct YouTube
  • Geo trigger - Crystallize YouTube
  • Hydro trigger - Electro-Charged YouTube
  • Pyro trigger - Overloaded YouTube

Significance:
Better documentation of Hakushin Ring mechanics.

Crystalize, Swirl and Swirl chain reactions work for the effect.​

By: ProfHugo#6478

Crystalize
Swirl and chain reactions

Gaining a new buff does not refresh duration.​

By: ProfHugo#6478

Video

Hakushin Ring Doesn't Trigger Off Field​

By: Greyhound#7836 and Terrapin#8603
Added: 2021-12-13 (v. 2.3)
Last tested: 2021-12-13 (v. 2.3)
Discussion

Finding:
Hakushin Ring doesn't trigger off field.

Evidence:
YouTube

Significance:
Bad uptime.

Hakushin Ring Does Not Buff Hydro Party Members When Hyperbloom Is Triggered​

By: @oliveuwu.
Added: 2023-07-13 (v. 3.8)
Last tested: 2023-07-12 (v. 3.8)
Discussion

Finding:
Hakushin Ring does not buff Hydro party members when Hyperbloom is triggered.

Evidence:
YouTube

Significance:
Despite Hyperbloom occurring from Hydro + Dendro and then Electro on top, Hakushin only considers the core (Dendro) and Electro when buffing.

Hunter's Path​

Hunters Passive Benefits From Percentage Based EM Buffs​

By: Mr. Pillow#0065
Added: 2022-10-14
Discussion

Finding:
Hunter's passive and Tighnari A4 benefit from percentage-based EM buffs.

Evidence:
In the first video (link: YouTube) and the second video (link: YouTube,) Tighnari hits twice with his Level 1 Aimed Shot and twice with his Elemental Burst, respectively, once before and once after Sucrose has triggered her A4. The attached calculations show that the passive of Hunter's Path and Tighnari A4 (both the CA DMG% and Burst DMG% part) benefit from Sucrose A4.

Significance:
Confirmation that the bugfix introduced with the 3.1 update "Fixes an issue whereby effects gained based on a character's attribute were incorrectly increased" (link: genshin.hoyoverse.com) is actually true. Better knowledge of how percentage-based effects work.

Kagotsurube Isshin​

Cursed Blade Reduction Mechanic​

By: Aevean Leeow#1362
Added: 2022-07-18 (v. 2.8)
Last tested: 2022-07-18 (v. 2.8)
Discussion

Finding:
The cursed blade you use in the Kazuha story quest says the wielder deals 50% less DMG. This effect is additive with DMG bonus, and does not affect transformative damage.

Evidence:
Kazuha a1 deals 2079 DMG, and Swirl does 2020 DMG.

YouTube

A1 calc with multiplicative reduction: 1.088∗0.5∗19090+92+200∗2∗2058∗1.919∗0.9=1923.451.088*0.5* \frac{190}{90+92+200}*2*2058*1.919*0.9=1923.45

A1 calc with additive reduction with the Pyro DMG bonus from a4: (1.088−0.5)∗19090+92+200∗2∗2058∗1.919∗0.9=2079.03(1.088-0.5)*\frac{190}{90+92+200}*2*2058*1.919*0.9=2079.03, same as video

Swirl calc with no reduction applied: 868.1∗2.586∗0.9=2020.42868.1*2.586*0.9=2020.42, same as video

(A1 CRITs here or otherwise it is doomed)

Significance:
Assuming kazuhas level does not approach infinity while the enemy level and/or DEF does, and assuming DEF Shred/reduction does not approach 100%, the DMG reduction on the quest's cursed blade matters less and less because the DEF formula would make non transformative or true damage approach 0 anyways. It will not impact transformative or true damage, which both bypass DEF, so the effect of the DMG reduction is negligible. In such a case Kazuha should probably use an em weapon, but at the very least the DMG reduction on the cursed blade will not be very detrimental in these scenarios.

Kagura's Verity​

Kagura Stack Refresh​

By: ArchedNosi#1484
Added: 2022-02-21 (v. 2.5)
Last tested: 2022-02-21 (v. 2.5)
Discussion

Findings:

  • Kagura's Verity's Stacks' Durations Refresh and act as one
  • Gaining a Stack of Kagura Dance refreshes the total duration to the full 16s.
  • In tandem, if the 16s runs out, all stacks deplete.
  • Additionally, Stacks of Kagura Dance are gained and start their 16s timer at the moment an Elemental Skill is used/inputted, to Skills with a Hold Variant, the passive is gained at the initial start-up of whatever the Hold Skill may be.

Evidence:
YouTube

  • Press E, wait ~10s, Press E, wait ~9s, Press E, wait ~17s, Press E
  • Press E #1 was increased by 1 Stack of Kagura Dance
  • Press E #2 was increased by 2 Stacks of Kagura Dance
  • Press E #3 was increased by 3 Stack of Kagura Dance (despite being longer than the initial 16s)
  • Press E #4 was increased by 1 Stack of Kagura Dance (depleted, also the 16s timer runs out according to as if the buff gets applied upon input, which it does)

Imgur

  • Calculated proof of the numbers shown in the clip.

Significance:

  • Confirms that Kagura's Verity's Stacks are not independent and refresh each other.
  • Similarly, fully depleting the 16s timer of any amount of Stacks erases the Stack count.
  • Additionally, Stacks of Kagura Dance are gained upon input, this applies to both Press and Hold variants to Elemental Skills. In particular for Hold Variants, the 16s timer begins when the Hold Skill is first casted/channeled.

Kagura Stacks Dynamic​

By: ArchedNosi#1484
Added: 2022-02-21 (v. 2.5)
Last tested: 2022-02-21 (v. 2.5)
Discussion

Findings:

  • Kagura's Verity's Stacks' are Dynamic.
  • Stacks of Kagura Dance are dynamic and are not snapshottable, this also implies if an Elemental Skill were to gain a Stack of Kagura Dance, it is also affected by the gained Stack.
  • In addition, this applies to Hold Variants of Elemental Skills, whereby it is also affected by the Stack it were to gain since the Stacks are gained upon initial input.

Evidence:
YouTube

  • Press E, wait ~8s, Press E, wait ~8s, Press E, wait 16s, Press E
  • Press E #1 was increased by 1 Stack of Kagura Dance (1 Stack, NOT 0)
  • Press E #2 was increased by 2 Stacks of Kagura Dance (2 Stacks, NOT 1)
  • Hold E #1 was increased by 3 Stack of Kagura Dance (3 Stacks, NOT 2)

The evidence above concludes that the Skill% buff is dynamic because Lisa’s Press Variant of her Elemental Skill snapshots.
Since the buff is yet still applied to Skills that produce it, it is conclusory that the weapon passive is, in fact, dynamic.

Imgur

  • Calculated proof of the numbers shown in the clip.

Significance:
Kagura's Verity's Skill% is dynamic and cannot be snapshotted. What this also implies is that Elemental Skills that were to produce a Stack of Kagura Dance would also get it’s buff applied to itself. Which further leads to the conclusion that every Elemental Skill used, if Kagura’s Verity is equipped, will have at least 1 Stack of Kagura Dance applied.
The Skill% buff being dynamic also fits in line with other Skill% buffs which are also dynamic (Solar Pearl).

Klee and Kagura Stacks​

This bug has been fixed in version 2.6, see this Vault Entry for more information.

By: solo#3333
Added: 2022-03-09 (v. 2.5)
Last tested: 2022-03-09 (v. 2.5)
Discussion

Finding:
Klee can get 2 stacks of Kagura's Verity from only 1 cast of her Jumpy Dumpty by inputting the Skill twice or more before it goes on cooldown (aka "double tap"), however with several caveats.

  • The first stack is gained approx. 34 frames into the cast, and the second stack is gained approx. 81 frames into the cast.
  • Klee cannot perform any other actions between these frames or the additional stack will not be granted. This includes dashing, jumping, using Burst, Normal Attacks, Charged Attacks, and using her Skill again - all of which she can do before ~81 frames have passed from the first Skill. If Klee uses both charges of Jumpy Dumpty and she allows ~81 frames to pass for at least one of them, she will end up with 3 stacks of the weapon.

Evidence:

  • Frames: Notice how the Skill icon lights up twice before going on cooldown. There is also a visual and audio indicator when Kagura stacks are granted (a layer of the weapon glows and a bell chime plays). However the visual is extremely subtle. To the best of my knowledge, the cast animation is at 230 frames, the first stack is at ~264 frames, and the second stack is at ~311 frames.

  • Proof of stack gaining techniques (credit to pai#3602).

    • First clip: cast E, no double tap, 1 stack is granted.
    • Second clip: cast E, with double tap, 2 stacks are granted, after ~81 frames.
    • Third clip: cast E, with double tap, 2 stacks are granted after ~81 frames. Use the second charge, and the 3rd stack is granted.
    • Fourth clip: cast E, with double tap, but do not wait 81 frames, only 1 stack is granted. Use the second charge with double tap, and wait ~81 frames, the last 2 stacks are finally granted.

  • Other inputs cancel the additional stack, such as:

Significance:
Potentially affects the viability of this weapon for Klee. The caveat of waiting for ~81 frames has dubious impact on a real rotation since most Klee usage involves animation cancels. See this doc for stack counts on every part of a standard Klee rotation vs their delayed version.

Key of Khaj-Nisut​

Key of Khaj-Nisut does not affect Kazuha's A4​

By: CrystalOfSugar#8624
Added: 2022-11-04 (v. 3.2)
Last Tested: 2022-11-04 (v. 3.2)

Finding:
Key of Khaj-Nisut + Kazuha A4 interaction

Evidence:
YouTube
241 EM should provide 9.6% Electro DMG, not 4.6% Electro DMG, like in this case. (4.6% Electro DMG are provided by 115.2 basic Kazuha's EM from ascensions)

Significance:
Elemental Mastery provided by Key of Khaj-Nisut (self-buff and party-buff both) is not used in abilities which use Elemental Mastery for providing other stats, like Kazuha A4.

Key of Khaj-Nisut Stacks Remains After Switching Out​

By: Alpha Flyte#2031
Added: 2022-11-04 (v. 3.2)
Last Tested: 2022-11-04 (v. 3.2)

Finding:
Key of Khaj-Nisut stacks remain even after switching out, and they share their duration, meaning they expire at the same time when the timer of the last stack gained expires.

Evidence:
YouTube

Significance:
No real demerit to switching out the Key user once stacks have been gained.
Key user must periodically refresh stacks to maintain uptime of the buff, since the buff once maxed is either 100% up or 100% down, meaning all 3 stacks must be gained again.

Key of Khaj-Nisut Stacks Cannot be Gained while Off-Field​

By: Alpha Flyte#2031
Added: 2022-11-04 (v. 3.2)
Last Tested: 2022-11-04 (v. 3.2)

Finding:
Key of Khaj-Nisut stacks cannot be gained when the user is off-field, and the duration of the stacks continues counting down even when the user is off-field

Evidence:
YouTube

Significance:
Key users must have some dedicated field time to refresh the stacks of the buff, affecting who may effectively proc the Key's passive, thus affecting rotations and rotation times.

Both Parts Of Key Of Khaj Nisut Passive Apply To The Wielder​

By: @.chronicler
Added: 2025-06-07 (v. 5.5)
Last tested: 2025-03-23 (v. 5.3)
Discussion

Finding:
Both portions of Key's passive increase the wielder's EM

Evidence:
www.youtube.com

273 + (3*70494*0.0012) + (0.002*70494) = 667.8

Significance:
Weapon mechanics

Kitain Cross Spear​

Kitain Spear Energy Drain Delay​

By: Episodde#8962
Added: 2021-08-28 (v. 2.0)
Last tested: 2021-08-28 (v. 2.0)
Discussion

Finding:
Kitain Spear Energy drain do not happen immediately. There is about a 22-24 frame delay (can be longer depend on latency) from the moment your character Elemental Skill hit an enemy to when you character actually lose Energy.

Evidence:
Frame count: YouTube (Frame count started at 2:05)

Rosaria able to perform E->Q with Kitain Spear equipped.
Imgur

Significance:
Some characters actually do not need to change their rotation, because as long as you can cast your Burst within about 23 frames after the Elemental Skill hit an opponent, your character will still be able to use Elemental Burst to spend all your Energy before getting drained by the weapon passive.
Rosaria can hit enemy with her E to get her A1 CRIT Rate buff and still be able to cast Burst with Q afterward to share the CRIT before the Energy is drained.

Kitain Passive Will Proc Even if Carrier Has 0 Energy.​

By: ItzSomebody#0029

Also doesn't dip the unit's Energy into the negatives. Video

Katsuragikiri Nagamasa​

Katsuragikiri Nagamasa Energy Drain Delay​

By: Episodde#8962
Added: 2021-08-28 (v. 2.0)
Last tested: 2021-08-28 (v. 2.0)
Discussion

Finding:
Katsuragikiri Nagamasa Energy drain do not happen immediately. There is about a 22-24 frame delay (can be longer depend on latency) from the moment your character Elemental Skill hit an enemy to when you character actually lose Energy.

Evidence:
Frame count: YouTube (Frame count started at 2:05)

Beidou using dash cancel to perform E -> Q with Nagamasa equipped. Still able to funnel Energy.
Imgur

Significance:
Some characters actually do not need to change their rotation, because as long as you can cast your Burst within about 23 frames after the Elemental Skill hit an opponent, your character will still be able to use Elemental Burst to spend all your Energy before getting drained by the weapon passive.
Beidou can still do E into Q to funnel Energy, as long as you dash cancel the E and then Q immediately.

Lost Prayer to the Sacred Winds​

Lost Prayer Stack Mechanics​

By: Konbo#0004
Added: 2020-12-07 (v. 1.1)
Last tested: 2020-12-07 (v. 1.1)

Findings:

  • Lost Prayer does not reset upon leaving combat, ONLY upon switching to a different character or if the character falls.
  • Stacks will also continue to build outside of combat as long as the character remains on the field.

Evidence:
Switched to a character with Lost Prayer, observed 0% DMG Bonus. Went into combat with a single left click, left combat, observed stacks continuing to build up, observed a 4 stack DMG bonus that remained indefinitely (tested 5+ minutes afk) until I switched off.

YouTube

Significance:
Lost Prayer becomes... even better for a main dps while exploring? The fact that you can carry over stacks over multiple fights will help in open world scenarios.

Mailed Flower​

Mailed Flower’s passive can't be triggered or refreshed from off-field​

By: @asaki_tmk
Added: 2023-03-03 (v. 3.5)
Last tested: 2023-08-13 (v. 3.8)

Finding:
Mailed Flower’s passive (Bonus EM) can not be triggered or refreshed from off-field

Evidence:
YouTube
At 0:08, 310 Swirl without passive, 318 Swirl with passive
At 0:42 consistent 310 Swirls

Significance:
Passive works as expected.

Mailed Flower's passive does not buff the hit or reaction that triggers it​

By: @milktearomance
Added: 2023-03-04 (v. 3.5)
Last tested: 2023-08-13 (v. 3.8)

Finding:
Mailed Flower's passive does not buff the hit or reaction that triggers it

Evidence:
YouTube
Triggered with both Elemental Skill hit and reaction at the same time
First Swirl (unbuffed): 1488
Second Swirl (buffed): 1749

Significance:
Confirmation of the description of the passive which states "Within 8s after the character's Elemental Skill hits an opponent or the character triggers an Elemental Reaction".

Memory of Dust​

Analysis of whether Xingqiu E/Q counts as a “shield” for Memory of Dust and Bolide​

By: ＃＃#0123
Added: 2020-12-07 (v. 1.1)
Last tested: 2020-12-07 (v. 1.1)

Findings:
For Bolide or Memory of Dust purposes, Xingqiu Q/E are not shields. Beidou C1 and Diona E are.

Evidence: YouTube

Significance:
Potential implication for unforged? And you can't get a free shield for memory of dust in your Klee/Xingqiu team.

Mistsplitter Reforged​

Mistsplitter Stacks Upon Burst Usage​

By: gtrain#9987
Added: 2021-09-06 (v. 2.1)
Last tested: 2021-09-06 (v. 2.1)
Discussion

Finding:
Number of Mistsplitter stacks gained on Q is character-dependent, and for some characters it's inconsistent.

The usual number of stacks gained is as follows:
0 stacks: AMC
1 stack: Bennett, Electro MC (usually), Geo MC, Kaeya, Xingqiu (rain sword damage reduction)
2 stacks: Ayaka, Jean, Kazuha, Keqing (initial hit, usually), Qiqi

Non-snapshotted Bursts like Keqing's are important to document, as we must know whether the stacks apply on the initial hit.

Inconsistencies are as follows:
MC in general, but AMC in particular: does very weird things once swapping Elements; it may be possible to manipulate this into getting an extra stack on Q, but requires more testing (see here: YouTube)
Keqing: occasionally gains only 1 stack on the Q initial hit, even more rarely gains only 1 on the entire Burst
Xingqiu gets 1 stack on his DR swords. Casting Q resnaps DMG Bonus on swords first made by E. Casting E during Q does not resnapshot DMG%.

The Burst stack seems to obey the same rules as the 4NO effect, as the only character that usually only gains 0 Mistsplitter stacks also generally does not obtain self-4NO on their Burst (AMC), though this is very inconsistent, and may also be manipulated by changing Elements.

Evidence:
Speadsheet

Significance:
Damage calculations.

Infused plunge attacks do not gain a stack.​

By: MkO#1192

Standard infused plunge
Kazuha's Midare Ranzan

Mistsplitter NA Stack From CA​

By: Trosalio#0420
Added: 2022-01-27 (v. 2.4)
Last tested: 2022-01-27 (v. 2.4)
Discussion

Finding:
A Mistsplitter's Emblem stack that a character obtains when Normal Attack deals Elemental DMG can also be obtained when Charged Attack deals Elemental DMG as well.

Evidence:

  • Gaining NA stack using Ayaka's CA: YouTube
    • CA DMG with 0 stacks: 458 * 3
    • CA DMG with 1 stack: 491 * 3
  • Gaining NA stack using Keqing's CA: YouTube
    • CA DMG with 0 stacks: 641 + 718
    • CA DMG with 1 stack: 686 + 769

Significance:
You can gain said stack in a case where Normal Attacks cannot reach any enemies, but Charged Attacks can such as using Ayaka's CA to hit enemies in Venti's vortex.

Oathsworn Eye​

Oathsworn Passive Mechanics​

By: Kurt#5846
Added: 2022-03-01 (v. 2.5)
Last tested: 2022-03-01 (v. 2.5)
Discussion

Findings:

  1. Oathsworn Eye passive does not stack
  2. Oathsworn Eye passive duration is refreshed every time the Elemental Skill is used
  3. Oathsworn Eye passive increases the Hydro% from Mona's A4
  4. Oathsworn Eye passive works from off-field

Evidence:

Significance:
Accurate calculations and simulations with this weapon.

Polar Star​

Polar Star ICD​

By: Latiwings#3308, GottaGoFast#7876, Cenpi#3224, NZPIEFACE#8439, Xreejan#1180
Added: 2022-01-13 (v. 2.4)
Last tested: 2022-01-13 (v. 2.4)
Discussion
(Debunked entry: Polar Star has ICD across independent stacks)

Finding:
Polar Star has no ICD for gaining independent stacks.

Evidence:

  • YouTube - This was the evidence to prove that Polar Star won't acquire stacks if the NA and CAs are too close together resulting in a loss of 20% ATK which is 1 stack of Polar Star. However on closer inspection it can be observed at 0:02 [Imgur] that Childe did NOT receive the buff from Bennett's Noblesse Oblige, which is also exactly 20% ATK.
  • YouTube - In this video evidence, contrary to the above, Childe did receive the Noblesse Oblige buff, as seen at 0:05 [Imgur].

In other words, in both video evidences Polar Star had acquired max stacks regardless of attack spacing.

Significance:
Bring justice to Polar Star's name, in that it had no ICD all along.

Portable Power Saw​

Portable Power Saw Mechanics​

By: @milktearomance
Added: 2024-05-26 (v. 4.6)
Last tested: 2023-10-14 (v. 4.1)
Discussion

Finding:

  1. Roused effect does not trigger without Stoic's Symbols.
  2. Eating food can add Stoic's Symbols.
  3. There is little to no cooldown between accumulation of stacks.

Evidence:

  • YouTube - Steps in video:

    • Elemental Skill used without Stoic stacks

    • Food eaten to gain Stoic stacks

    • Elemental Skill used again within 15s (when Roused's cooldown should expire if in effect)

    • Stoic stacks consumed, Sayu's EM buffed

  1. Roused was triggered on the second Skill press, consuming the Stoic's Symbols. If Roused had triggered on the first Skill press, cooldown would have been in effect and would have made it impossible to trigger on the second Skill press. Thus, Roused will not trigger if there are no stacks.
  2. The only healing source was food, and stacks were acquired.
  3. Two stacks were made after consuming one Sweet Madame, since Sweet Madame actually heals twice in the same instance. Thus, there seems to be little to no cooldown between accumulation of stacks.

Significance:
Effective use of this weapon when crafting rotations.

Portable Power Saw Passive procs on overhealing​

By: @shampoobrain
Added: 2024-05-26 (v. 4.6)
Last tested: 2023-10-14 (v. 4.1)
Discussion

Finding:
Portable Power Saw passive procs on overflow healing.

Evidence:
YouTube

Significance:
Don't need to take damage to make use of the weapon fully.

Primordial Jade-Winged Spear​

PJWS On Hit Definition​

By: Vye#9195
Added: 2021-04-15 (v. 1.4)
Last tested: 2021-04-15 (v. 1.4)

Finding:
PJWS passive can be stacked from Elemental Skill and Burst Skill hits on an enemy. The "on hit" portion of the passive does not refer to normal attack hits alone.

Evidence:

Significance:
Triggering and maintaining PJWS stacks in real combat scenarios is a lot easier than initially assumed. Especially for certain units with a high hit rate Elemental Skill or Burst Skill like XL.

PJWS New Stacks Extend Duration​

By: hawtlava#2461
Added: 2021-08-09 (v. 2.0)
Last tested: 2021-08-09 (v. 2.0)
Discussion

Finding: New gained stacks on PJWS will extend the duration of existing stacks, beyond the 6 seconds stipulated.

Evidence:
Video

Significance: XBDIT can stack PJWS for her second pyronado.

Prototype Amber​

Prototype Amber Healing Clarification​

By: Reens#9389
Added: 2021-10-26 (v. 2.2)
Last tested: 2021-10-26 (v. 2.2)
Discussion

Finding:
Prototype Amber's heal (4% rounded down) is based off the Max HP of the character being healed, not the max HP of the character equipped with the catalyst.

Evidence:
Imgur
Sucrose equipped with R1 Prototype Amber - max HP 15082 (4% rounded down is 603 HP):

  • Tartaglia Max HP 21466, Healed 858
  • Beidou Max HP 20162, Healed 806
  • Bennett Max HP 16163 Healed 646

Significance:
Stacking HP on a Prototype Amber wielder will not increase its healing.

Multiple Prototype Amber Heals Don't Stack​

By: Pana <3#0992
Added: 2022-03-01 (v. 2.5)
Last tested: 2022-03-01 (v. 2.5)
Discussion

Finding:
When proccing 2 Prototype Amber Passives, the 2nd one overrides the HP regeneration effect of the 1st one.

Evidence:
In this video, R3 Prototype Amber was equipped on Yae Miko, and R1 Prototype Amber was equipped on Lisa. Healing bonus stat, whether incoming or outgoing, was set to 0 for all 3 characters to not affect the results.

Diona's HP before any Prototype Amber Passive was triggered:

Suppose Prototype Amber HP regeneration stacked, we would expect a total heal of 3 * 5% + 3 * 4% = 27%.

Diona's HP after the HP regeneration from both Prototype Ambers:

From the evidence above: (5140 - 50) / 29942 = 16.9%
The total heal rounds up to only 17%. We can see in the video that PAmber heals totaled only 4 procs instead of 6 total, once after Yae's Burst and thrice after Lisa's Burst. This accounts for 1x5% Max HP regen from Yae and 3 * 4% Max HP regen from Lisa.

Thus the previous statement "Suppose Prototype Amber HP regeneration stacked, we would expect a total heal of 3 * 5% + 3 * 4% = 27%" is false. Prototype Amber HP regeneration does NOT stack, and instead activating a 2nd one overrides the regen effect of the 1st one.

Significance:
It is not advisable to run Prototype Amber on 2 or more characters in each team, since the HP regeneration does not stack. If someone really wants to use multiple Prototype Ambers, it is better to space out the characters' Elemental Bursts once every 6s if they want to maximize healing.

Prototype Starglitter​

Prototype Starglitter Stacks Duration Refresh​

By: jamberry#7142
Added: 2021-07-29 (v. 2.0)
Last tested: 2021-07-29 (v. 2.0)
Discussion

Finding:
Gaining Prototype Starglitter stacks refreshes the duration of existing stacks

Evidence:
CA damage on non CRIT before any Skill uses is 181. First Skill use at 30s, CA damage after first Skill use with one stack on Prototype Starglitter is 196. Second Skill use at 37s, CA damage after first Skill use with two stacks is 210. CA damage is 210 again at 48s, after first stack would have expired if stacks were independent, and 181 again at 53s, after both stacks would have expired if gaining stacks refreshes the duration of existing stacks.
YouTube

Significance:
Affects stack uptime considerations for polearm users with Skill cooldowns less than 12s as it can potentially allow them to keep full uptime on two stacks of Prototype Starglitter's passive.

Rightful Reward​

Rightful Reward mechanics​

By: @starlighthorizon
Added: 2024-05-14 (v. 4.6)
Last tested: 2024-04-25 (v. 4.6)
Discussion

Findings:

  • Rightful reward triggers on overhealing
  • Rightful reward passive can trigger and enter cooldown at full energy

Evidence:

Rust​

Rust Damage Decrease​

By: Amadurim#8925
Added: 2021-04-05 (v. 1.4)
Last tested: 2021-04-05 (v. 1.4)

Finding:
Rust DMG reduction on charge shots is additive and is in the same category as DMG% when calculated.

Evidence:
Using level 50 Amber on Lv87 Hilichurl, with +20 Physical goblet

Calculations:

1062 = Amber total attack
1.583 = +20 Physical DMG bonus cup
0.439 = Amber T1 aimed shot

===

1062 * 1.583 * 0.439 = 738.023
DEF + RES mod: (50+100)/((50+87)+200) * 0.9 = 0.400

Amber Damage w/Prototype Crescent = 738.023 * 0.400 = 295.209

===

Rust Passive:

If multiplicative: 295.209 * 0.9 = 265.688
If additive: 1062 * (1.583 - 0.1) * 0.439 * 0.400 = 276.560

Truncating decimals aside, the DMG number of "If additive" is the same as in-game numbers, hence the DMG reduction of Rust is additive and falls under DMG% in calculations. The same result is also seen in Slingshot.

Significance:
The DMG reduction of Rust passive is pretty negligible.

Scion of the Blazing Sun​

Scion of the Blazing Sun works in AoE​

By: @bobrokrot
Added: 2024-05-14 (v. 4.6)
Last tested: 2024-04-25 (v. 4.6)
Discussion

Finding: Scion of the Blazing Sun does work in AoE. The physical damage is AoE, and the effect is applied to all damaged enemies.

Evidence: https://youtu.be/u_NWfeK6-xc

Serpent Spine​

Rate of Loss of Serpent Spine Stacks​

By: Reydriar#8127
Added: 2020-12-07 (v. 1.1)
Last tested: 2020-12-07 (v. 1.1)

Finding: Losing Serpent Spine stacks has an ICD depending on the stack count.

  • 5-4: 1.01-1.02s (after you lost your 5th stack you can‘t lose the 4th stack for the named duration)
  • 4-3: 1.01-1.02s
  • 3-2: 1.17-1.18s
  • 2-1: 1.17-1.18s

Evidence: YouTube

Significance:
Getting hit by multi hit attacks will still only remove one stack thus increasing your overall uptime on stacks. 1 and 2 stacks have a better protection than higher stacks due to an increased ICD.

  • Relevant examples in the overworld only: environmental damage
  • Relevant examples: Ruin guard missiles or the Cryo gun attack from the Cryo Fatui (YouTube)

Serpent Spine Damage Taken affects Shields​

By: Greyhound#7836
Added: 2021-06-26 (v. 1.6)
Last tested: 2021-06-26 (v. 1.6)
Discussion

Finding:
The damage taken increase from Serpent Spine's passive does affect the damage shields take.

Evidence: YouTube

Explanation: The crystalize shield is created by a level 50 character with 0 EM, thus creating a ~786 health shield which has 2.5x effectiveness against its own Element, meaning that it will block ~1965 Pyro damage. Both flower stamens are dealing the same damage, and give the same results, with Beidou equipped with SS (with 5 stacks) taking 1045 damage per tick, and with fav sword taking 908 damage per tick. Beidou with SS 5 stack's shield lasted for 2 ticks allowed Beidou to take 123 damage additional damage, while the shield with favonius greatsword equipped lasted for 3 ticks and allowed Beidou to take 759 additional damage.

Significance:
Explanation of how shields interaction with negative damage reduction

Slingshot​

Slingshot Passive Mechanics​

By: jamberry#7142 and ItzSomebody#0029
Added: 2022-08-03 (v. 2.8)
Last tested: 2022-08-03 (v. 2.8)

Finding:
Slingshot has a medium zone where neither buff nor debuff applies. Slingshot buff applies until roughly the same distance as it takes for Amos to move from 3 stack to 4 stack passive for charged shots.

Slingshot appears to dynamically track flight time. On release, arrows have a mixin which gives them a DMG% buff. After .3s, this mixin is removed, and then a mixin giving negative DMG% is added. This allows for the potential for some arrows to hit within a very narrow window where they have neither buff nor debuff.

Evidence:

Significance:
Definitely just Yoimiya things.

Song of Stillness​

Song of Stillness Mechanics​

By: @hungrydave
Added: 2024-05-14 (v. 4.6)
Last tested: 2024-04-25 (v. 4.6)
Discussion

Findings:

  • Song of Stillness buff is snapshottable.
  • The buff procs on overheals.
  • The buff does not appear on the stat screen.

Evidence:

Staff of the Scarlet Sands​

Staff of the Scarlet Sands does not stack off-field​

By: _Maxi#1776
Added: 2022-11-04 (v. 3.2)
Last tested: 2022-11-04 (v. 3.2)

Finding:
Stacks for Staff of the Scarlet Sands cannot be gained from off field but are retained while off-field.

Evidence:
YouTube

Significance:
Characters need more field time to get optimal use of the weapon. Characters that do not snapshot can still make use of stacks while off field.

Staff of the Scarlet Sands Stacks Snapshot​

By: nico#7748
Added: 2022-11-26 (v. 3.2)
Last tested: 2022-11-26 (v. 3.2)
Discussion

Finding:
The Dream of the Scarlet Sands stacks (which increase ATK based on EM) from the Staff of the Scarlet Sands (SotSS) snapshot the character's current EM when obtaining each stack.

Evidence:
YouTube

1283 (total ATK without SotSS passive)+203 (EM before diona C6)×.52 (static SotSS buff)+203 (EM before diona C6)×.28×3 (SotSS stack buff)= 15601283 \ \text{(total ATK without SotSS passive)} + 203 \ \text{(EM before diona C6)} \times .52 \ \text{(static SotSS buff)} + 203 \ \text{(EM before diona C6)} \times .28 \times 3 \ \text{(SotSS stack buff)} = ~1560

1283 (total ATK without SotSS passive)+403 (EM after diona C6)×.52 (static SotSS buff)+203 (EM before diona C6)×.28×3 (SotSS stack buff)= 16631283 \ \text{(total ATK without SotSS passive)} + 403 \ \text{(EM after diona C6)} \times .52 \ \text{(static SotSS buff)} + \mathbf{203} \ \textbf{(EM before diona C6)} \times .28 \times 3 \ \text{(SotSS stack buff)} = ~1663

Significance:
Pre-stacking SotSS will not give you the full value of the Dream of the Scarlet Sands effect if EM buffs are obtained after the stacks.

Staff Of The Scarlet Sands Can Gain Multiple Stacks From One Attack​

By: @xochiyoh
Added: 2024-05-20 (v. 4.6)
Last tested: 2024-05-07 (v. 4.6)
Discussion

Finding:
Staff of the Scarlet Sands can gain multiple stacks from one attack if it hits multiple enemies.

Evidence:
www.youtube.com

Significance:
SoSS can stack faster in AoE situations.

Talking Stick​

Talking Stick mechanics​

By: @bobrokrot
Added: 2024-05-14 (v. 4.6)
Last tested: 2024-04-25 (v. 4.6)
Discussion

Findings:

  • Beidou E in Bennett burst triggers both abilities.
  • Swapping into a character with Talking stick while Diona shield + Xinayn shield are both active triggers both abilities.
  • Suffering an elemental attack from the enemy while in Bennet field does not trigger the 2nd ability.
  • Suffering an elemental attack normally does trigger the ability.

Evidence: https://youtu.be/fKdwoPKplbY

The Bell​

Bell’s %Increased Damage Passive Works With Any Shield​

By: adri#1824

Finding:
The Bell's passive %Inc Damage works with any shield, and not just with the proc from the passive itself like the description suggests.

Testing: Attacked enemies

  • without any shield
  • with the Crystallize shield
  • breastplate
  • and the shield from bell's passive

Evidence: Imgur

Significance:
Although there's better claymores, this makes The Bell significantly better than the passive description lets on. Makes for an ok Noelle weapon if prototypes are being used on another character.

Tidal Shadow​

Tidal Shadow Counts Overheal​

By: @clevernt
Added: 2024-07-08 (v. 4.7)
Last tested: 2024-07-08 (v. 4.7)
Discussion

Finding:
Tidal Shadow passive procs with overheal
Evidence:
YouTube
Significance:
don't need to get hit or furina to proc it

Tome of the Eternal Flow​

Tome of the Eternal Flow doesn't gain stacks when off field​

By: @f99shi
Added: 2024-05-26 (v. 4.6)
Last tested: 2023-10-14 (v. 4.1)
Discussion

Finding:
Tome of the Eternal Flow doesnt gain stacks when off field.

Evidence:
Imgur

Significance:
Useful to keep in mind when crafting rotations.

Luxurious Sea-Lord​

Luxurious Sea-Lord interactions​

By: CQLQRS#7853
Added: 2021-10-09 (v. 2.1)
Last tested: 2021-10-09 (v. 2.1)
Discussion

Finding:
Fishmore aka Luxurious Sea Lord has a passive that summons a titanic tuna to deal damage once every 15 seconds when the character's Burst hits a target. Like other similar passives such as the ones in Prototype Archaic and Star Tombed Starsilver, it only procs when the wielder is on field.

Evidence:
CBeidou on field part 1: Imgur
Beidou on field part 2: Imgur
Beidou off field: Imgur
Eula A4 (by ZyronX#2720): Imgur
Eula off field (by ZyronX#2720): Imgur
Sayu on field (by ZyronX#2720): Imgur
Sayu off field (by ZyronX#2720): Imgur
AoE finding part 1: YouTube
AoE finding part 2: YouTube

Significance:
The passive procs as intended similar to the other mentioned previously. However, ZyronX has confirmed that the passive has an AoE that can affect multiple targets with a single proc. It should be easier to do this if an Anemo cc such as venti Burst was used.

The Flute​

The Flute’s AoE can CRIT​

By: Crome#2644

Finding: Yes, it can CRIT

Evidence:
Imgur

The Alley Flash​

Alley Flash's Disable Timer can be Reset by Damage​

By: kol#1593
Added: 2021-06-26 (v. 1.6)
Last tested: 2021-06-26 (v. 1.6)
Discussion

Finding: Taking damage while The Alley Flash's DMG% increase passive is disabled will reset the disable timer.

Evidence: YouTube

Bennett takes hit 1 at 124 frames (2.083 seconds) Bennett takes hit 2 at 261 frames (4.35 seconds) The Alley Flash's passive symbol fully materializes at 561 frames (9.35 seconds)

Significance: Every time you take damage with Alley Flash equipped, you reset the timer and thus extend the downtime of the passive.

The Widsith​

Widsith Abyss Second Half Proc​

By: Keteru#3020
Added: 2022-03-16 (v. 2.5)
Last tested: 2022-03-16 (v. 2.5)
Discussion

Finding:
The Widsth’s Passive activates upon entering the second half of an Abyss chamber with its holder being on the first slot.

Evidence:
YouTube video
Yae Miko holding The Widsith in slot 1 is buffed for less than 10 seconds, seemingly triggering during the transition screen when proceeding changing from first half to second half, eating into the buff uptime.

Significance:
Changes Abyss team slotting and potentially optimal team rotation when using The Widsith.

Wandering Evenstar​

Evenstar Buff Duration and CD Reset​

By: Casdela#5121
Added: 2022-11-09 (v. 3.2)

Finding:
Xiphos and Evenstar give their respective buffs 64 frames after being equipped, all timers are reset when removed or when entering/exiting domains (including challenge again, abyss try again), allowing the buff to re-apply 64 frames after re-equipping or entering/exiting domains regardless of the previous timer.

Evidence:

  • Buff is applied 64 frames after the weapon is equipped: YouTube
    • Open and check Sucrose's stats after 63 frames (1.05s) shows that the buff wasn't applied.
    • After 64 frames (1.067s) checking Sucrose stats shows that the buff was applied.
  • Buff timer resets when entering a domain: YouTube

Significance:
Mostly fluff, just to confirm that the Tulaytullah Weapon Series (which Xiphos and Evenstar belong in) have 100% buff up-time regardless of the situation, except for the 1.067s delay at the start. This info has no practical application other than to test buff timings.

Wolf-Fang​

Wolf-Fang passive does not stack off-field​

By: @scooty727
Added: 2024-05-14 (v. 4.6)
Last tested: 2024-04-25 (v. 4.6)
Discussion

Finding: Wolf-fang crit rate does not stack off-field

Evidence: Xingqiu's burst misses multiple crits throughout the video which shouldn't be possible at 93% crit rate unless wolf-fang is not stacking off-field https://www.youtube.com/watch?v=ukdNFcvBnI4

Wolf's Gravestone​

Wolf's Gravestone Procs Don't Stack​

By: JonahFarc#7056
Added: 2021-05-11 (v. 1.5)
Last tested: 2021-05-11 (v. 1.5)
Discussion

Finding:
WGS procs do not stack and the most recent proc overwrites the existing proc, even if it has duration left over

Evidence:
YouTube

Significance:
Multiple WGS can kinda suck if they're not equal refine

Wolf's Gravestone Passive Does Not Affect the Attack That Procs It​

By: Goden#2155
Added: 2021-12-16 (v. 2.3)
Last tested: 2021-12-16 (v. 2.3)
Discussion

Finding:
The hit that procs Wolf's Gravestone's passive does not get the buff bonus applied to itself. Only subsequent hits are affected by the buff.

Evidence:
Beidou's N1 is shown to CRIT prior to the buff for 1859. On the hit that procs WGS' buff, she also deals 1859 damage. One final hit is done in the clip to confirm the buff is active for subsequent hits. YouTube

Significance:
Setup your WGS buff before a big nuke or you lose out on potential damage.

Xiphos' Moonlight​

Xiphos and Raiden A4 Interactions​

By: Casdela#5121
Added: 2022-11-4 (v. 3.2)
Last tested: 2022-11-4 (v. 3.2)

Finding:
Xiphos does not affect both parts of Raiden's Ascension 4 passive and does not affect Emblem's Burst DMG% Bonus, which is based on Energy Recharge%.

Evidence:
YouTube
YouTube

With Xiphos, Raiden has 196.7% ER, without it Raiden has 188.3% ER. Both videos have Raiden's Electro bonus at 35.3% and Keqing's Burst is filled to exactly the same point, meaning Raiden's Ascension 4 Passive isn't affected by Xiphos ER bonus.

The final hit of Raiden's 5-hit string is 2240 in both videos, meaning Emblem's Burst DMG% Bonus isn't affected by Xiphos ER bonus.

Significance:
Expected results.

Xiphos Buff Duration and CD Reset​

By: Casdela#5121
Added: 2022-11-09 (v. 3.2)

Finding:
Xiphos and Evenstar gives buff 64 frames after equipped, all timers are reset when removed or when enter/exit domains (including challenge again, abyss try again), allowing the buff to reapply 64 frames after re-equipping or enter/exit domains regardless of the previous timer.

Evidence:
YouTube Open and check Kazuha's stats after 56 frames (0.933s) shows that the buff wasn't applied. However on 3rd attempt after 64 frames (1.067s) checking Kazuha stats shows that the buff was applied.

YouTube

Using Noblesse as control test, enter/exit domain removes all effects but doesn't reset Burst and Skill cooldown. However, Xiphos timer reset after removed so it reapplies almost instantly.

Significance:
Mostly fluff, just to confirm that the Tulaytullah Weapon Series that Xiphos and Evenstar belongs in have 100% buff up-time regardless of the situations, except for the 1.067s delay at the start. This info has no practical application other than to test buff timings.

Xiphos Buff Snapshotting​

By: Casdela#5121
Added: 2022-11-09 (v. 3.2)

Finding:
Xiphos buff snapshots once every 10s.

Evidence:
YouTube
9.667s after Xiphos buff first starts, the buff doesn't update to match the new EM value, meaning Xiphos buff is not dynamic, but is snapshots instead. Elegy buff starts 530f (8.833s) after Xiphos buff is first applied, lasts for 746f (12.667s), and ends 612f (11.5s) after Xiphos snapshots Elegy bonus. Which means Xiphos can snapshot Elegy twice if it has a 10s snapshot interval.
The Xiphos buff, that snapshot Elegy's 12.667s EM buff, lasts for at least 18.733s, but no more than 1124f+120f=1244f=20.733s. Which matches the expected time frame of the 10s snapshot interval.

Significance:
Confirms that the Tulaytullah Weapon Series buff re-snapshots every 10s.
On paper it sounds good, but it's unreliable to snapshot any buff that lasts less than 10s, because there is absolutely no indication whatsoever of when the buff re-snapshots.

General Mechanics​

Event Weapons Cannot be Foddered​

By: CQLQRS#7853
Added: 2021-10-16 (v. 2.2)
Last tested: 2021-10-16 (v. 2.2)
Discussion

Finding:
Event weapons, locked or not, will not appear in the enhance menu of other weapons.

Evidence:

Significance:
Prevents accidental foddering of once in a lifetime event weapons.

Weapon Stats for Inactive Characters​

By: Reens#9389
Added: 2021-10-16 (v. 2.2)
Last tested: 2021-10-16 (v. 2.2)
Discussion

Finding: Certain weapon passives do not apply their effect on character stat pages for out-of-party characters. This only applies to weapons that either: 1) have a conditional stat buff or 2) provide one stat in relation to another.

Evidence:

Significance: Characters using these weapons must be in your active party when their stats are viewed to get active numbers.

Attack Speed Buffs on Claymores​

By: EternalDream#5503
Added: 2021-04-02 (v. 1.4)
Last tested: 2021-04-02 (v. 1.4)

Finding: Attack Speed buffs do not dynamically affect Claymore Charged Attacks

Evidence: Side by side comparisons of a C0 vs C1 Xinyan

Significance:
If an Attack Speed buff is active, such as Beidou's A4 after max parry, the charged attack will gain the speed increase for its entire length even if it wears off during the attack. Additionally, as the only character that can currently dynamically receive a speed buff mid-Charged Attack is Xinyan with C1, this means she MUST trigger the buff before-hand, as it will have no effect if triggered during the spin itself.

Charged Attack Speed Buff Update​

By: tamamo#4303
Added: 2022-03-09 (v. 2.5)
Last tested: 2022-03-09 (v. 2.5)
Discussion

Finding:
Charged Attack speed buffs do not dynamically increase claymore Charged Attack speed, but dynamically decrease Charged Attack speed if duration ends during the attack.

Evidence:
YouTube video

Significance:
Better understanding of Charged Attack speed buffs (with the exception of Itto).

Catalyst Effective Attack Comparisons​

By: Maygi#4987

Findings:
Catalyst effective attack comparisons - see spreadsheet below.

Evidence: Imgur

Methodology
Effective attack was calculated with a simple piecewise function that takes into account CRIT Rate, CRIT DMG, attack, and damage % (Elemental damage). Passives that cannot be quantified due to how they can differ in effectiveness with different playstyles are NOT included in the effective attack calculation.

The spreadsheet has three main columns - for low, medium, and high stats. Each one simulates the every possible substat combination with a total of 10, 20, or 30 offensive lines (maximum possible rolls is 38, but nobody will ever get such lines - right?). Royal weapon CRIT Rate is simulated on each possible combination of stats. The base stats used were that of a level 90 Klee using 2 piece Gladiator and 2 piece Witch set.

Significance:
Memory of dust can actually be strong, even without a shield doubling its effect. Widsith is around ~20% weaker than 5* catalysts in general outside its Burst window, and ~10% stronger during the Burst window at R1 and ~30% stronger at R5.

Refinement Costs​

By: JonahFarc#7056
Added: 2021-04-18 (v. 1.4)
Last tested: 2021-04-18 (v. 1.4)
Discussion

Finding:
Refinement costs scales with the refinement rank of the current weapon

Evidence:

Significance:
If refining a weapon from R1 to R5, it's cheaper to constantly use the R1 weapon to refine. Also, if you're refining your levelled weapon multiple times, make the other refines first (EX: For an R3 weapon, combine others to R2 before refining with the R3)

Weapon Stacks On Hit​

By: CQLQRS#7853 and Dunia#2481
Added: 2021-11-05 (v. 2.2)
Last tested: 2021-11-05 (v. 2.2)
Discussion

Finding:
Weapon passives with the description "on hit" or "scoring hits on opponents" can trigger on Normal Attack, Charged Attack, Elemental Skill, and Elemental Burst.

Evidence:

The evidence for Summit Shaper is missing, but it should behave the same way as the other weapons in the same series.

Significance:
Documentation of how the ambiguous description of such weapon passives work.

Lost Prayer and Serpent Spine Stacks​

By: Xardas#2531
Added: 2022-01-04 (v. 2.3)
Last tested: 2022-01-04 (v. 2.3)
Discussion

Findings:
Unpausing the game after equipping a character with Lost Prayer to the Sacred Winds (LP) or Serpent Spine (SS) starts a timer. This timer starts ticking and continues ticking whether the character equipping LP or SS is on the field or not, and every 4 seconds, it checks if that character is on the field. If they are, the game gives them a stack of their weapon's passive.

The timer:

  1. Pauses while your inventory, wish screen or characters screen are open (during single player)
  2. Resets to 0 when you enter any domain, or Spiral Abyss chamber
  3. Runs as normal when you reach 4 stacks of LP or SS, but will no longer grant stacks
  4. (For SS only) Resets to 0 when you lose (or would lose, if already at zero stacks) a stack of SS

Evidence:
Timer starts when you unpause, even if the character with LP or SS is currently not on the field: YouTube

  • YouTube
  • YouTube
  • YouTube
  • YouTube
    NOTE: I could not test the exact times on this, but the timer still resets when you take damage, I just don't know how often the timer can reset itself if you take multiple instances of damage in a short amount of time

Significance:

  • With this knowledge, Claymore and Catalyst users no longer need to stay on field for 4 or more seconds to gain their stacks of SS and their first stack of LP respectively.
  • With proper rotation timing your favourite Catalyst users can always start their DPS windows with one stack of LP right away, adding 8-16 ElementalDMG% depending on LP refinements.
  • After taking damage, your Serpent Spine users can easily regain their lost stack by being on the field exactly 4 seconds after they took damage and lost their stack.
  • For casual play, this means the first LP stack is gained, on average, after 2 seconds of being on the field, and not 4.

Weapon Stacking And Ping​

By: NZPIEFACE#8439
Added: 2022-01-14 (v. 2.4)
Last tested: 2022-01-14 (v. 2.4)
Discussion

Finding:
Weapons that gain stacks on hit are affected by ping. You will only gain stacks when the hit registers on the server, and the server will process hits at the same time, without caring about when the client actually made the hits.

Freedom-Sworn, which gains its buffs from reactions happen, has normal behaviour.

Polar Star behaves differently as it has individual stacks for each type, and will gain all the stacks at the same time when the hit is registered.

Evidence:

Evidence from Kolibri#7675:

Only one stack was gained despite the trigger condition being met multiple times.

Also from Kolibri#7675:

  • FS (Not affected by ping.)

It gained stacks as normal.

Evidence from Latiwings#3308:

It gained all the stacks once connection to the server was restored, meaning that while gaining stacks is affected by ping, the number of stacks gained is not.

Significance:
This isn't an issue for anyone with a stable connection, if someone's connection to the servers are rather spotty, this can cause them to lose stacks they would otherwise have.
Freedom-Sworn doesn't change behaviour.

Widsith and TTDS Work Off-Field​

By: Charisma#5567
Added: 2022-02-21 (v. 2.5)
Last tested: 2022-02-21 (v. 2.5)
Discussion

Findings:
The buff from the Passive of The Widsith and Thrilling Tales of Dragon Slayers remain active even after the character swaps out after obtaining the buff.

Evidence:

  • Video: The ATK% buff that the Widsith gave Yae Miko remained after swapping.
  • Video: The ATK% buff that TTDS gave Kazuha remained after swapping.

Significance:
With this knowledge, characters are free to swap out and still retain Widsith/TTDS buff from off field, which can affect a lot of rotations.

Weapons and Heals​

By: Daibangden#8410
Added: 2022-06-06 (v. 2.7)
Last tested: 2022-06-06 (v. 2.7)
Discussion

Findings:
Healing passive from weapons are affected by Healing Bonus and Incoming Healing Bonus of the character who wields them.

Evidence:
YouTube

  • Weapons tested: Traveler's Handy Sword, White Iron Greatsword, Recurve Bow, Otherworldly Story.
  • Each weapon was tested without any bonus, with 35.9% Healing Bonus (+20 Healing Bonus Circlet), and with 35.9% Healing Bonus + 30% Incoming Healing Bonus (+20 Healing Bonus Circlet + Hydro Resonance).

Significance:
Confirmation of how healing passive works on weapons.

Weapon Stacks and Zero Damage Attacks​

By: Kolibri#7675, Xreejan#1180, Phana#0420
Added: 2022-07-28 (v. 2.8)
Last tested: 2022-07-28 (v. 2.8)
Discussion

Finding:
Some character abilities contain zero-damage attacks which can grant stacks for weapons such as Primordial Jade Winged-Spear, Summit Shaper, The Unforged, Vortex Vanquisher and Memory of Dust. They can also trigger Wolf's Gravestone's passive.
These attacks cannot trigger Fillet Blade's passive even though its description says "On hit" like Wolf's Gravestone's passive.

Evidence:

Significance:
Characters can benefit from the stack granted by a zero-damage attack.
Examples:

  • Raiden can gain a stack before her Burst deals damage. This is because of the zero-damage attack that happens towards the start of her Burst animation.
  • Hu Tao can gain a stack by simply activating her E next to an enemy.