Skip to main content

Resources and Efficiency

Main Page:

πŸ“„οΈ Resources and Efficiency

For entries related to resources or resource efficiency.

Resin​

Maximum Original Resin​

By: CQLQRS#7853
Added: 2021-10-18 (v. 2.2)
Last tested: 2021-10-18 (v. 2.2)
Discussion

Finding:
Original Resin can go past the cap of 160, capped at ~2000.

Evidence:

  • Unable to go past 1993 Resin: kimchihotpot#8630
  • Unable to use 11 Fragile Resin (660 Resin) when at 1360 Resin, which would result in 2020 Resin: Tibo#4309
  • Unable to use 1 Fragile Resin when at 2000 Resin: Tibo#4309

Significance:
Better understanding of the Resin system.

Resin to Stardust Exchange Comparison​

By: Mateng#4507
Added: 2021-04-14 (v. 1.4)
Last tested: 2021-04-14 (v. 1.4)

Findings: Blue (XP) leylines are more stardust-efficient than Gold (Mora) leylines. They are 65.33 Stardust/Resin, while Gold leylines and Magic Crystal Ore conversion from the Blacksmiths are both 60 Stardust/Resin.

Evidence: Reddit

I used simple exchange rate math to relate XP materials, Magic Crystal Ores, and Mora to their cost in Resin, and back to their cost in the Stardust Shop. I assumed that the chance of receiving 6 or 7 Adv. Exp and 4 or 5 Hero's Wits was 50%.

Significance: These findings identify that the most efficient use of Stardust (outside of buying the 10 Fates, if the Traveler wishes to pay for them) is to clear the limited amount of Mora, then the Magic Crystal Ores, then the Adventurer's Experience. Additionally, it is more Resin-efficient to be farming more Blue leylines than Gold ones.

120 Original Resin Cap​

By: NZPIEFACE#8439
Added: 2022-11-08 (v. 3.2)
Last tested: 2022-11-08 (v. 3.2)
Discussion

Finding:
120 Original Resin cap still exists for accounts that last logged in before the cap was updated to 160. It only updates after you log in.

Evidence:

Significance:
More information on the Original Resin Cap change.

Artifacts​

Artifact EXP from Vendors Vs Domains​

By: ceeeleste#3568
Added: 2021-06-28 (v. 1.6)
Last tested: 2021-06-28 (v. 1.6)
Discussion

Finding:
Buying 2-star artifacts from vendors is more worth than running a domain for artifact exp (Once a week)

Evidence:
There are 3 artifact vendors, each selling five 2-star artifacts for 2k mora each. Shop refreshes once a week. Wiki Link.

Artifact vendors:
Mora/Resin cost: 15 * 2k = 30k mora (10 resin)
Total artifact exp: 15 * 840 = 12600 exp
Resin worth: 12600/10 = 1260 exp/resin

Domain Run (20 resin):
Drop rates: Wiki Link
Average artifact exp: (1.068 * 3780) + (2.486 * 2520) + (3.546 * 1260) = 14769.72 exp
Best case artifact exp: (2 * 3780) + (3 * 2520) + (4 * 1260) = 20160 exp
2525 Mora: 2525/60000 = ~0.0421 resin
Average resin worth: 14769.72/20.0421 = ~736.9 exp/resin
Best case resin worth: 20160/20.0421 = ~1005.9 exp/resin

Vendor vs Domain:
Average: 1260/736.9 = ~1.7099 (vendor 70.99% more worth)
Best case: 1260/1005.9 = ~1.2526 (vendor 25.26% more worth)

Significance:
Buying out the 3 artifact vendors once a week is more worth than running a domain for artifact exp. Vendor is 70.99% more worth your resin on average, and 25.26% more worth for a best case domain run scenario.

Vendor Artifacts vs Housing Sanctifying Essence​

By: K\ឡឡ#4000
Added: 2021-07-07 (v. 1.6)
Last tested: 2021-07-07 (v. 1.6)
Discussion

Finding:
NPC artifact shops are a better deal than the sanctifying essence in terms of artifact experience.

Evidence:
12600=360
10000=360
For 360 realm currency you can get 10k artifact exp. However, the 2 star artifact experience you get from all three NPC shops is actually equals to more than 10k artifact EXP while costing the same amount of realm currency. The total mora cost for the npc shops is 30k mora for 15 2 star artifacts. 1 two star artifact gives 840 exp and they sell 5 which means 5x840=4200 and 4200x3(amount of NPC artifact shops)=12600.

Significance:
It's better to buy 3 10k mora with 360 realm currency than to buy 1 sanctifying essence for 360 realm currency because you get 2600 more artifact exp from the 30k mora when you buy artis from the npcs.
360 Realm Currency can be converted to 10K artifact EXP.
NPC shops sell 12.6K artifact XP worth of 2* artifacts for 30K Mora.
The mora cost of buying out the NPC shop is equal to 360 realm currency.

Artifact Probability Thesis​

By: tooflesswulf#8456
Added: 2022-05-12 (v. 2.6)
Last tested: 2022-05-12 (v. 2.6)
Discussion

Theory:
Performed a thorough mathematical analysis of the artifact rolling system, deriving equations for:
1. Exact probability that an artifact's substats upgrade beyond some threshold.
2. Approximate probability than an artifact, once upgraded, will yield an increase in damage.
3a. Approximate probability/expected time it takes to improve the overall damage of a build.
3b. Expected damage increase per resin spent for farming an artifact domain.

Evidence:
See ArtiSubMath-v10.pdf or the latest version on Overleaf.
There is also python demo code available at artifact-math-demo-py.zip.

Significance:

  1. One form of the long-awaited artifact-substat-to-resin relation has been found.
  2. The theory can answer common questions such as:
    • Which artifact should I upgrade? Should I fodder this artifact or keep it?
    • Should I continue farming artifacts or should I focus on levels/talents?

Realm Currency​

Realm Currency Exchange Rate Analysis​

By: Mateng#4507
Added: 2021-06-06 (v. 1.5)
Last tested: 2021-06-06 (v. 1.5)
Discussion

Theory
Realm Currency (RC) is a new kind of currency that will release in 1.5 allowing Travelers to exchange it for various goods including Mora, EXP, Ores, and Transient Resin (60 Resin that must be used quickly). This exchange rate analysis will examine its value alongside Resin and Stardust.

Evidence

Assumptions:

  • The Traveler uses the Transient Resin and consumes 60 Original Resin
  • Stardust/Resin values are seen here: Resin Stardust Comparison

Abbreviations:

  • RC = Realm Currency
  • MEO = Magic Enhancement Ore

RC conversions based at 1200 RC:

4 Wits / 1200 RC       (1 Wit = 120 RC)
100000 mora / 1200 RC  (1 mora = .012 RC)
20 MEOs / 1200 RC      (1 MEO = 60 RC)
60 resin / 1200 RC

Stardust conversions based at 60 Resin:

18.38 Wits / 60 Resin   (20 Resin = 6.125 Wits)
180000 Mora / 60 Resin  (20 Resin = 60000 Mora)
36 MEOs / 60 Resin      (20 Resin = 12 MEOs)

Significance

Stardust/RC Values:

1 Wit  = .0153 Stardust/RC (18.38/1200)
1 Mora = .015 Stardust/RC (180000/1200/10000)
1 MEO  = .03 Stardust/RC (36/1200)

RC/Resin Values:

1 Wit  = .9995 RC/Resin (.0153 * 65.33)
1 Mora = .9 RC/Resin (.015 * 60)
1 MEO  = 1.8 RC/Resin (.03 * 60)

Based at 20 Resin:

6.125 Wit  = .05 RC/Resin (.9995/20)
60000 Mora = .045 RC/Resin (.9/20)
12 MEOs    = .09 RC/Resin (1.8/20)

Analysis
The most Resin-efficient purchase order at the Realm Depot is to claim the Magic Enhancement Ores every week, then the Wits, and then the Mora (assuming that the Traveler has already claimed the Transient Resin if they wish to).

For Travelers who wish to maximize Stardust and minimize RC: target MEOs -> EXP -> Mora at Paimon's Bargains in the Shop. Inverse this priority (Mora -> EXP -> MEOs) if Travelers wish to maximize RC and minimize Stardust.

Realm Currency Extended Analysis​

By: chowr#0522
Added: 2022-03-16 (v. 2.5)
Last tested: 2022-03-16 (v. 2.5)
Discussion

Abstract:
I was looking at the library submissions for Realm Currency and found that I had completely different conclusions in my personal findings than what had been submitted (specifically the submission above).

My 2 main criticisms of the previous analysis is that it does not include artifact XP, and it does not take into account the "weekly budget" that a player receives from their teapot. Additionally, I wanted to analyze the actual value of Transient Resin in comparison to each shop item.

In my finding, I have come to the conclusion that Transient Resin is by far the most efficient item in the shop, by a substantial amount. In order for any item in the shop to achieve at or above the same Resin Value as Transient Resin (equivalent to 60 Resin), the other options must be bought either almost or completely out.

Regarding the other shop items, the three "main" resources: XP books, Mystic Enhancement Ores, and Mora, are roughly equivalent to each other (though XP books are worth slightly less). On the other hand, the Artifact XP Tinctures are notably more efficient than the main three, as buying them out costs less while still reaching over 60 Resin Value, as well as having two of them available.

Finding:
Transient Resin and Artifact XP are the most resin efficient resources available for purchase in the Realm Shop.

RC/Resin for each resource is as follows:

  • XP Book: 0.0272 Resin/RC (Resin per RC)
  • MEO: 0.0278 Resin/RC
  • Mora: 0.0278 Resin/RC
  • Artifact XP (Small): 0.0376 Resin/RC
  • Artifact XP (Large): 0.0376 Resin/RC
  • Transient Resin: 0.05 Resin/RC

Order in efficiency is as follows: Transient Resin > Artifact XP > Mora = MEO > (very slightly) XP Books.

Evidence:
*All calculations are done assuming resin is being spent to obtain resources.

Calculations: Google Sheets
Extended version: Pastebin

Significance:
Buying out both of the Artifact XP options allows the player to essentially receive a full day's worth of resin from their Realm Shop, should they purchase all of the artifact XP tinctures, as well as the Transient Resin. It is entirely possible to do this, as the total cost comes out to 4800 RC, and the highest possible expenditure each week is 5040 RC.

However, opting for a lower budget of buying out only one of the artifact options in order to save coins is also recommended.

There are other possible permutations of what to buy in the shop, but these are my recommendations if a person is looking to buy out the items as efficiently as possible.

Closing Thoughts:
There are extra factors to consider about why artifacts are worth more to buy. Primarily because there are often times that 5* artifacts will not be used as fodder.

I also want to object to the idea that purchasing 360 RC of Mora to buy out the 2-star artifacts in all vendors is better than buying a 360 RC artifact tincture, made in the following library submission: "Vendor Artifacts vs Housing Sanctifying Essence".

Technically, it is true that, by following this method, it is more efficient to buy out the 2-star artifacts each week with Mora than to buy one tincture. There is a theoretical budget where someone buys all but one large tincture, and purchases Mora 3 times in order to buy out the artifact vendors. I object because there are alternate methods of obtaining Mora that would be more efficient and does not require the purchase of Mora with RC. Utilizing mob kills, daily random events, or even leylines to obtain the mora necessary instead would allow for the purchase of an extra Large Tincture. And if it is worthwhile to buy out the vendors instead of 1 Tincture, I argue it is more worthwhile to buy out the vendors, as well as the tincture. There is only a budget of 5040 Coins per week at the highest level, and so it is not possible to buy Transient Resin, Tinctures, and the 30,000 Mora necessary solely with coins. Therefore, I suggest the artifacts should be bought with Mora from a different source. Though with smaller budgets it certainly could be suggested that the 30,000 mora be purchased instead.

Crafting​

Crafting Refund vs Increase Passives​

By: jdeeyu#9927
Added: 2021-05-22 (v. 1.5)
Last tested: 2021-05-22 (v. 1.5)
Discussion

Finding: Albedo's crafting passive isn't just straight 20% better than Mona's crafting passive (as stated in KQM guide) due to the fact that Mona's passive also allows for more subsequent crafting to be done, therefore more trials, therefore more chances of success.

Evidence: Through simulating a large number of crafting trials, I will post a screenshot of my code below.
Imgur

Significance: Albedo/Eula passive isn't as good as it initially seems when compared to Xingqiu/Mona

Conclusion: From the results of my simulation (feel free to scrutinize my code, I have double checked it 3-4x myself), it seems that Eula/Albedo is passive is actually 10% better than than Xingqiu/Mona instead of the initial 20%. If you take into account the initial talent books for the comparison then the difference is a whopping 1%.

Significance:
Buy the BP if you have the money.

Crafting Refund Vs Increase Passives At 3n+1 Items​

By: @veggie0581, @.whalien
Added: 2023-11-20 (v. 4.2)
Last tested: 2023-10-21 (v. 4.1)
Discussion

Finding:
When you have 3n+2 of a material it is worth using the "25% chance to refund" passive (e.g. that of Xingqiu) repeatedly till you get to a multiple of 3. When you have 3n or 3n+1, you should craft only using the "10% chance to double the product" passive (e.g. that of Layla). In the rare case that you do not plan to craft any more of this product, and you have 3n+1, then you might consider craft using the "25% chance to refund" passive repeatedly till you get to a multiple of 3.

Evidence:
GitHub and Imgur.

  • Image shows the (difference of (expected value of only using 10% passive) minus (expected value of using 25% chance till 3n)) at different values of starting materials. The difference (y-axis) is in terms of equivalent starting/lower-tier materials.

    • Rows represent how much of the low-tier material exists before crafting, categorized into 3n, 3n+1, and 3n+2.
    • Columns represent how the product is considered, with left counting any remaining +1 or +2 product as 1.1x, and right not counting remaining product in any way, only considering the higher tier product.

  • The left column is useful when crafting is expected to continue into infinity, and transition to using the 10% strategy at some point, but realistically it is useful anytime when players expected to craft more in the future. As seen from the graphs, always using "10% double" is better (difference is positive).

  • The right column is useful when crafting is expected to stop. As seen from the graphs, using "25% refund" till a multiple of 3 is reached is better for most values.

Significance:
Optimize crafting output and resin efficiency, prove known strategies with math.

Overworld​

Overworld Farming Statistics​

By: Bobrokrot#0111
Added: 2021-07-13 (v. 1.6)
Last tested: 2021-07-13 (v. 1.6)
Discussion

Finding:

Artifact Drop Spots
There are two types of artifact spots:

  • "No mora" class - Spots from this class never drop Mora and always drop a single piece of artifact with 65% chance it being white artifact and 35% chance for green.
  • "Has mora" class - Spots from this class always drop some amount of mora (200 on average), also the always drop white artifacts (1.5 on average) and never drop green artifacts.

Ore Farming

  • 1.2 ore per average seems consistent for all types of ores (excluding starsilver which there are currently not enough statistics). There are some fluctuations due to low sample size for Iron and White Iron and due to the event of 3 chunks having very low probability. But overall seems consistent. They sometimes drop two chunks, sometimes three and never one.
  • Iron, White and Crystal chunk are all possible drops.

Evidence: Stats Sheet

Significance: Statistics for farming purposes

Crabs Respawn​

By: Mcpie#8672
Added: 2021-07-17 (v. 1.6)
Last tested: 2021-07-17 (v. 1.6)

Finding:
Crabs that you approach and fail to pick up will respawn after leaving their render distances and coming back.

Evidence:
YouTube
YouTube

Significance:
Video

Random Event Cap​

By: derek#9488
Added: 2021-12-22 (v. 2.3)
Last tested: 2021-12-22 (v. 2.3)
Discussion

Finding:
There is a random event cap which is 10 random events a day. After that you get nothing more when you do random events that day.

Evidence:
Doing a random event 11 times, the 11th time there is no reward: YouTube
After the above, other random events also give no reward: YouTube

Significance:
You can do 10 random events for mora, companionship exp, and enhancement ores everyday and its roughly about 15000-16000 mora a day and guaranteed 150 companionship exp a day. The easiest way to do this Is by repeating the warden of konda quest over and over again in that same spot as in the first video, scaring away the pigeons to do it quickly.

Bake-Danuki Respawn​

By: Mcpie#8672
Added: 2021-08-08 (v. 2.0)
Last tested: 2021-08-08 (v. 2.0)
Discussion

Finding:
Some bake-danukis that end up with investigation spot respawn. They seem to follow common enemies respawn pattern (12h), however it's still unclear.

Evidence:

Significance:
Some of them respawn artifacts so it might be a reliable way of getting more artifact fodder everyday.

Bush Droprates​

By: gideleon1#2188
Added: 2021-04-11 (v. 1.4)
Last tested: 2021-04-11 (v. 1.4)

Finding: There is an 8% chance to get something from a broken bush.

Out of 1,174 Bushes:

  • 30 Lizards
  • 32 Matsutakes
  • 32 Eggs

It seems like the drop is chosen randomly. Testing was done during the day, because at night, fireflies are added to the pool. They do not replace anything in the loot table, because at night all four items were dropped. There has never been an instance of getting more than one drop out of a bush.

Evidence: Google Docs

I didn't assume anyone would want a two hour video of me just breaking bushes, so I just compiled the results.

Significance: Was curious about the drop rates, and I couldn't find anything about them in the TCL, online, or in the data gathering server, so I figured I'd just do it myself. I wanted exact percentages just for the sake of thoroughness.

Chasm Special Ore​

By: sexyeboy69#2687, Raismin#5493
Added: 2022-09-05 (v. 3.0)
Last tested: 2022-09-05 (v. 3.0)
Discussion

Finding:
The Chasm offers some special ores that can be easily identified by a blue aura around them. These ores drop extra chunks when broken.

Evidence:
Google Sheets

  • On 3 mining trips (unfortunately, some footage is missingβ€”only trips that had footage were recorded in the sheet), I mined 164 Chasm Crystal Ores (CrystalB on the sheet) and 33 normal ores (just crystals).
  • Of these, 128, 35, and 1 dropped 2, 3, and 4 chunks respectively. This evidence means that these special ores are dropping more chunks than what normal ones should. It seems that the drop rates are 0% for 1 chunk (normally 81%), 78% for 2 chunks (normally 18%), 21% for 3 chunks (normally 1%), and 1% for 4 chunks (normally impossible).
  • When these drop rates are shifted by 1 drop then compared to normal drop rates, they are 78%-21%-1% compared to 81%-18%-1%.
  • There are 3 theories that explain this behaviorβ€”but only one is actually possible.
    1. "The drops are normalβ€”I just got really lucky"
      This is impossible because 4 chunks can't drop from a single ore. Hence, the drop table must have been modified for the special Chasm ore.
    2. "Each special ore drops one extra chunk, but each ore can only drop 3 chunks maximum" This is impossible because 4 chunks can drop from a single ore.
    3. "Each special ore drops one extra chunk" This is the most likely theory, but more data should be recorded to be sure.
  • Note that there is not enough data to draw any conclusions about special Noctilucous Jade ore.

Significance:
Special ores from The Chasm don't just look different, they also have different drop rates. This means that interactive Genshin maps and mining routes need to be revisited after taking this in consideration (and the wiki also needs to add this).

Domains​

Pre Daily Reset Freeze Time​

By: Mcpie#8672
Added: 2022-11-21 (v. 3.2)
Last tested: 2022-11-19 (v. 3.2)
Discussion

Finding:
Selecting a Talent/weapon domain before the daily reset will freeze the material type until you leave the domain. This means that if you re-enter the domain using the Resin reward screen UI then you will receive the same material on consecutive runs.

Example: Taishan Mansion
Tuesday: Diligence
Wednesday: Gold
If you enter Taishan Mansion at Tuesday 3:59 AM (1 min before reset), you will receive Diligence books. If you re-enter using rewards screen UI after claiming rewards and it's 4:01 AM, you will still receive Diligence books instead of Gold books.
Additionally, you will not be able to see the Welkin reward screen when you're inside the domain - you have to leave it first. However, you get the Primogems from the Welkin immediately at 4:00 AM.

Evidence:
Talent domain video:
0:00 - Taishan Mansion rewards for current day (indicates that time is 4:00 AM+, so Wednesday)
0:14 - Claiming rewards (Diligence books)
0:27 - Re-entry using domain UI
0:59 - Claiming rewards (Diligence books, even though it's still Wednesday)
1:15 - Leaving domain, Welkin screen appears

Weapon domain video:
0:00 - Display of weapon domain rewards from Wednesday, 3:58 AM (2 minutes prior to daily reset)
0:32 - Start of a domain run (still Wednesday)
1:33 - Domain clear, (still Wednesday), showing Primogem count prior to Welkin (1274)
1:56 - Daily reset, primogems updated (1274 + 90 = 1364)
2:05 - Visible domain rewards have changed to Thursday rewards
2:22 - Claiming rewards (from Wednesday, got Wednesday rewards)
2:38 - Re-entry using domain UI
3:37 - Claiming rewards (it's Thursday, got Wednesday rewards)
5:11 - Claiming rewards (it's Thursday, got Wednesday rewards)
6:11 - Leaving domain, Welkin screen appears

Significance:
If you need to farm certain Talent/weapon material and it is few minutes before daily reset, you can condense all Resin, and enter the domain 1 minute before reset and keep getting rewards for previous day. Another possible usage is during new region releases, where you want to speedrun getting into Talent/weapon domain prior to reset in order to farm Talent/weapon materials for freshly released character without waiting few days.

Food​

Food Costs​

By: Anmol43#1917
Added: 2021-04-20 (v. 1.4)
Last tested: 2021-04-20 (v. 1.4)
Discussion

Finding: Documenting the cost, in mora, of all food items in Genshin Impact.

Evidence: Food Doc

Methodology: The cost of an ingredient is determined by the lowest cost of that ingredient in an NPC shop if the ingredient is sold. If the ingredient is not sold but is available on expeditions, then the ingredient is assigned a cost of 250, since a 20 hour expedition gives 5000 mora or an average of 20 ingredients (average of 10 ingredients collected per expedition for each ingredient), so the tradeoff for each ingredient is 250 Mora. If the ingredient is neither sold nor available on expeditions, the ingredient is assigned a cost of 0. Calculations are also made with reputation discount costs, which are gained for each region at Reputation Level 4. Character special foods are not considered.

Notable Findings:

  • Cheapest 3* instant heal (assuming 20k HP): Mondstadt Hashbrowns, 8.57 Health/Mora
  • Most Mora-efficient heal (assuming 20k HP): Grilled Tiger Fish, 11.23 Health/Mora
  • Cheapest Revive: Teyvat Fried Egg, 180 Mora if bought from restaurant
  • Most Mora-efficient sprinting stamina reduction: Lotus Seed and Bird Egg Soup, 850 Mora

As other effects such as ATK buffs and CRIT Rate buffs cannot be stacked, it is generally advised to use the best possible food in those categories rather than using the most mora-efficient food. However, the cost calculations are included in the spreadsheet.

Significance: players who frequently use food and are looking to save Mora can see which foods are cheapest/most Mora-efficient for their given effects.

Duplicate Special Dish​

By: denmax#9293
Added: 2021-07-11 (v. 1.6)
Last tested: 2021-07-11 (v. 1.6)

Finding:
Characters with extra dish cooking bonus can double their special dishes.

Evidence:
Dishes

Significance:
Just something to keep in mind.

Crafted Food Items Limit​

By: It3mUs3r#4778 and Creonalia#2818
Added: 2021-10-24 (v. 2.2)
Last tested: 2021-10-24 (v. 2.2)
Discussion

Finding:
Crafted items limit is 2000 per unique item. Both automatic cooking and manual cooking yield this result. Which character is cooking also does not change this limit.

Evidence:
2000 of each kind of Fragrant Mashed Potatoes: Imgur
Unable to craft more Suspicious Fragrant Mashed Potatoes: Imgur
Unable to craft more Fragrant Mashed Potatoes: Imgur
Unable to craft more Delicious Fragrant Mashed Potatoes: Imgur

Significance:
Food is not infinite.

Investigation Mechanics​

Daily Investigation Limit​

By: Mcpie#8672
Added: 2021-08-14 (v. 2.0)
Last tested: 2021-08-14 (v. 2.0)
Discussion

Theory:
There exists an investigation daily limit of 100. It is possible to go beyond this number by not leaving the rendered zone after reaching 100. Bake-danukis that give investigation spots count towards this limit as well.

Evidence:
Video shows me logging into genshin (Welkin, daily commissions displaying due to first login, 160 resin so I crafted condensed resin) and running through 100 investigation spots to showcase that if you investigate 100th spot then rerendering the zone WILL NOT display previously accessible investigation spots.
Video
Above video contains timestamps of each spot and a counter on-screen. I also got artifacts from at least 4 bake-danukis that spawn artifact investigation spot in the video and I increased the counter after each one of them.

Video showcasing that it is possible to go beyond the cap. There's on-screen counter present showing how many investigation spots I have already gathered. I managed to get a total of 174 investigation spots.
Video

Significance:
Since patch 2.0, if you're going to do artifact routes then you don't want to investigate every single spot there is due to this limit and the fact that there are over 100 investigation spots containing artifacts as of now.
Additionally, because it's possible to go beyond 100 by not leaving the zone on 100th spot, it's advised to leave places with high amount of investigation spots for last.

Daily Investigation Limit Reset​

By: Mcpie#8672
Added: 2021-08-14 (v. 2.0)
Last tested: 2021-08-14 (v. 2.0)
Discussion

Theory:
Daily investigation limit resets at daily reset.

Evidence:
Video
Video shows my character in 2 artifact spots that I have not picked on the previous day. Once I was in the Liyue cellar spot, daily reset happened - after teleporting far away and back, I was able to pick up investigation spots.

Significance:
Allows for early artifact route run if you still have uninvestigated spots.

Daily Investigation Spots Post Daily Cap​

By: Mcpie#8672
Added: 2021-08-14 (v. 2.0)
Last tested: 2021-08-14 (v. 2.0)
Discussion

Theory:
We already know that the daily investigation cap is 100 and it's possible to go beyond that number by picking already rendered investigation spots.
If you're about to hit the daily cap and go too far from the place you are at right now without teleporting/relogging and then go back, the rendered investigation spot is lost. This would mean that all investigation spots are stored in your memory and they are strictly connected to your render distance.

Evidence:
Timestamped: Video - gathered 99th spot, going north to get 100th which is the cap. There are 2 spots visible here to gather.
Timestamped: Video - gathered 174 spots, no investigation spots present even though there was no teleport/relog happening.
Area covered by black thick border shows the vicinity of where I was past 100th spot, while the DoT displays timestamped above locations.

Significance:
A test whether already rendered investigation spots get removed from your memory if you go too far or not - if it was allowed then it would be possible to walk through the entire inazuma (if you're some freaking absolute minmaxer), render all investigation spots at least once and then pick all of them up because everything is stored in your memory.

The above video displays that investigation spots are stored in your memory, and if you go too far then they are removed as it should be, because memory leaks are a horrible thing.

Multiple Artifact Clusters After 100​

By: makeway4pK#3901
Added: 2022-03-16 (v. 2.5)
Last tested: 2022-03-16 (v. 2.5)
Discussion

Findings:

  • 'Co-op helps': Investigates (Artifacts, Weapons, Iron, Energy, Veggies) don't despawn after reaching daily cap as long as a watcher (host or Co-op teammate) is near them [1].
  • 'Order matters': Which spots are included in a cluster can change depending on which spot is being watched when 100th investigate is done [2][5].
  • 'Wide clusters': Clusters can extend pretty far [3].
  • 'Merged!': Tatarasuna and the nearby spots #7, #8 & BK6 can be merged into a single cluster by placing watchers at certain locations. Same goes for Koseki and Seiraimaru (Kosekimaru?) [4].
  • 'Hypothesis': Seems like there are pre-defined 'groups' of these spots that can be held in memory as long as a watcher is in a certain range of at least one or few of its spots. these groups can still despawn if they are left unwatched anytime after reaching the cap.

Notes:

  • Cluster is an area/set of spots that can be watched by a single watcher after reaching the daily investigation cap.
  • Spot numbers are referenced from the maps created in previous tickets:

Evidence:
[1] Basic confirmation
[2] 4 clusters tested, including a 22-spotted one in Seiraimaru (Ship + #6 + 2x BKs)
[3] 11-spotted Hiraumi, Koseki + #6 + 2x BKs and BK10 included in Tatara cluster
[4] Post-cap run for 184 total spots
[5] Kosekimaru watchpoint, read description

Significance:
Allows collecting artifacts from more than (the previous max of) 129 spots in a day. Assuming you have friends with moderate patience stats (max 20 mins afk). Based on current knowledge, theoretical max is: 5\+99\+32\+30\+11\+7=184.

Trees​

Tree Chopping Memory Mechanics​

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

Finding:
The game can only remember the 10 most recently-struck trees at a time. After the 11th tree is hit, the 1st tree's state is wiped from the memory and can be harvested again. However, if a previously harvested tree is struck again BEFORE this, it will be moved back into the sequence as the most recently updated item. Therefore, a loop in the following order where each number represents a specific tree will generate no wood on the second partial loop: 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11. This is because trees 2-11 have become the recently updated set in memory, and only 1 has been dropped and reverted to "harvestable" state. In a typical 11 tree loop, hitting tree 1 again would provide wood, then remove the most recent last tree (tree 2 in this case) from memory. Partial harvests also count the same as full harvest or re-harvest.

Evidence:
Partial harvest evidence & 11 tree loop Hitting tree 2 before 1 on second loop proves minimum loop is 11, only the 11th tree is reset, and updating mechanics.

Significance:
There is no quick method to reset tree resources beyond hitting an 11th tree and removing the 10th-previously-harvested tree from the memory. Aside from logging out of the game, these mechanics are key to quick farming of a targeted type of tree. Fast traveling has no effect, nor does entering a domain, nor does waiting several minutes. To keep farming trees efficiently, the loop must avoid all 10 trees already in memory. Adding a buffer of an additional 1 or 2 trees into the loop can help prevent accidentally updating depleted trees in memory and wasting time. Additionally, if your loop contains mixed tree types, you can exploit the partial harvest finding to farm one type faster by only striking the less-desired trees one time each before moving on.

Temporary Skill Targets: Enabling Unholy Harvesting Methods​

By: mol#3280, wiremash#0433, Aluminum#5462
Added: 2021-06-06 (v. 1.5)
Last tested: 2021-06-06 (v. 1.5)
Discussion

Findings:
Attacks that shake untouched trees, including enemy attacks, can allow them to be harvested using a small number of Elemental Skills in a short window. We explored this to find several things.

  • Shaking a tree marks it as a target for a wider variety of Skills. Some attacks, like Mona's E, can only shake already-marked trees, and do nothing.
  • All Physical damage sources & an odd assortment of character Skills (Jean's E, Kaeya's Q) can shake/mark
  • Marks last a very short time
  • Some Skills and Overloads can harvest. Characters like Bennett and Xingqiu can harvest using E Skills, while Razor can harvest with E only if he uses Overload.

Evidence:
Summary post with multiple videos here YouTube

Significance:
Attacks are not equal in terms player-environment interaction. Most Skills can never harvest nor shake tress; some can shake trees, others can harvest directly, while others can harvest indirectly via Overloading burning grass. Most importantly for combat, damage sources appear to be treated differently depending on the character, attack, and reaction source. Overloading via E to harvest a marked tree works for Razor and Beidou, but not Xiangling or Fischl's Elemental arrow. Overload seems to work on burning grass but not self-overloads. Dealing phys damage with Xinyan harvests using a Normal Attack, but only shakes during her Burst. More broadly, the game appears to be conserving resources by limiting the player's kit unless certain enabling actions are taken. After enabling these attacks, the game shuts them down after a short window. Enemies appear to be permanently enabled for all attacks, but the player must fight enemies within a specific area, otherwise they will teleport back and reset.

Tree Mark Duration (Misc Tree Mark Updates)​

By: an odd medium#6288
Added: 2023-01-24 (v. 3.4)
Last tested: 2023-01-14 (v. 3.3)
Discussion

Extension to the above ticket

Finding:
... β€’ Marks last a very short time ... "

Finding:
Tree marks last for 5 seconds and cannot be extended with subsequent hits during this duration.

Evidence:
docs.google.com - Tree mark duration (in frames), roughly 300 frames at 60fps.

Significance:
Gotta chop fast

Tree Marks and Duration are Client-Side (Misc Tree Mark Updates)​

By: an odd medium#6288
Added: 2023-01-24 (v. 3.4)
Last tested: 2023-01-14 (v. 3.3)
Discussion

Finding:
Marking trees and the duration of the mark are both client-side.

Evidence:
YouTube

Significance:
Demonstrates that the act of marking trees and mark duration are not dependent on ping.

Missed Attacks Still Mark Trees For Unholy Harvesting​

By: Aluminum#5462
Added: 2021-12-12 (v. 2.3)
Last tested: 2021-12-12 (v. 2.3)
Discussion

Finding:
Missed attacks on trees due to a bug still mark the tree for unholy harvesting methods.

Evidence:
Xingqiu N1 (miss) into Xingqiu Skill: YouTube
Qiqi N1 (miss) into Tartaglia Skill and Attacks: YouTube
Version 2.3 Bennett N1 (whiff) into Tartaglia Skill and Attacks: Imgur

  • In Version 2.3, miHoYo issued a supposed fix for the whiffed attacks on trees, as noted in their patch notes. However, this issue can still sometimes occur post-patch, enabling this trick.

Significance:
I am the Lorax. I speak for the trees.

Lumberjack Tier List​

Contributors: Aluminum#5462, Green sabre#0540, Kourinn#6001, mol#3280, Steph#3614, JenjenJL#6582, Greyhound#7836
Added: 2021-06-06 (v. 1.5)
Last tested: 2021-06-06 (v. 1.5)
Discussion

Link to Tierlist

Summary of Results, most efficient units for tree chopping:

  • Xingqiu (only if you manually time their attack string, spam clicking won't work)
  • Keqing
  • Rosaria
  • Xiao

Wood Daily Drop Limit​

By: Creonalia#2818
Added: 2021-05-21 (v. 1.5)
Last tested: 2021-05-21 (v. 1.5)
Discussion

Finding: You can only collect 2,000 of any specific wood per day. This limit resets at server reset and cannot be reset by relogging/teleporting.

Evidence:

  • Inventory before and after farming wood over 3 days: YouTube
  • Evidence that player cannot collect more bamboo, but can collect other wood: YouTube
  • Same for pine wood: YouTube
  • Hitting cap and testing teleport and logout/login (and showing that neither reset the cap): YouTube
  • Collecting pine wood after server reset (marked by new daily commission quest at 0:12): YouTube

Note that although I did need to teleport to get wood after reset, teleporting alone does not reset the cap, as shown above.

Significance: Useful for players to know if they're farming a massive amount of wood.

Capped Wood Types Register In Chopping Memory​

By: Homura#9806 and mol#3280
Added: 2022-02-20 (v. 2.5)
Last tested: 2022-02-20 (v. 2.5)
Discussion

Finding:
Even after farming one tree type to the cap, it will still register in the 11 tree chopping memory.

Evidence:

YouTube

In the video, thoma hits 4 birch trees that do not drop wood, this is either caused by a.) being in memory as a tree 1-11 or b.) the daily wood cap was hit for birch specifically. Next, 5 fir trees and 2 cuihua trees were cut, all dropping wood. Going back to the birch trees resulted no new wood drops, and since those are 11 trees hit in total it is impossible for all of the birch trees to be "in memory", so the only reason they wouldn't drop wood is that birch daily limit has been reached. 2 additional fir trees were cut and dropped wood, so capped wood types do count in the tree chopping memory mechanics.

Significance:
Since capped trees do register in memory, this allows for hybrid routes (such as the one in the video), especially very unbalanced hybrid routes (like 1 cuihua to 10 cedar) to function even after you have farmed one wood type to completion.

Elemental Infusion Hampers Wood Cutting​

By: lapidibus#8146 and δανιΡλ#6040
Added: 2022-05-07 (v. 2.6)
Last tested: 2022-05-07 (v. 2.6)
Discussion

Finding:
Melee attacks infused with elements usually cannot chop wood from trees.

Evidence:
Chongyun and Ayaka infusions prevent wood chopping - YouTube
Ayato can chop wood during his Skill's infusion if activated immediately after an uninfused normal attack hit (see Unholy Harvesting Methods for an explanation) - YouTube

Significance:
Characters that involuntarily infuse elements (e.g. Ayaka) are sub-par at chopping wood.

Battle Pass​

BP Resin Value​

By: Peekays#2722
Added: 2021-09-14 (v. 2.1)
Last tested: 2021-09-14 (v. 2.1)
Discussion

Finding:
The $10 BP gives resources worth approximately 1800 resin = 45 condensed resin = 30 fragile resins in the form of exp books, mora, and talent books.
This does not include the weapon enhancement ores, fates, primogems, and bp exclusive weapon.

Evidence:
Spreadsheet

Also average talent domain rewards sheet (credit to carrier/Data Gathering): Spreadsheet

Weekly BP Cap During Launch Week​

By: Mcpie#8672
Added: 2022-01-10 (v. 2.4)
Last tested: 2022-01-10 (v. 2.4)
Discussion

Finding: Reaching BP Weekly EXP limit during launch week requires sacrifices.
Possible sacrifices are:

  • Spend a total of 500,000 Mora.
    • It's easy to spend it if you're planning to level anything.
  • Purchase 2 items from the Teapot Traveling Salesman in another player's Serenitea Pot.
    • Requires sacrificing possible resin gains from purchasing Mora/fodder/EXP for Realm Currency - it's 80 Realm Currency in order to reach it.
  • Use a total of 1200 Original Resin.
    • Only NA is capable of doing this and not even by a huge margin - you can use up to 1237 resin assuming perfect server maintenance end and x2 Transient Resin.
    • If you're not a hardcore player, you need to either make a refill with Primogems or utilize Fragile Resins.

Evidence: Sheet including options and calcs per server region: Google Doc

Note that 500,000 Mora and Purchase 2 item from the Teapot Traveling Salesman have same value (450), which is the reason as to why there is no extra sheet for salesman / salesman + resin as same values are included in 500,000 Mora sheets.

Significance: People who wanted to reach BP Weekly EXP limit during launch week have to make additional sacrifices in order to do so.

Inventory​

Destroy Function Mechanics​

By: xf3#3123
Added: 2021-04-08 (v. 1.4)
Last tested: 2021-04-08 (v. 1.4)

Finding:

The destroy button in the inventory is the only way to mass delete/destroy items in the game other than the Parametric Transformer. However, the function itself is limited and the resources gained from destroying items are not worth compared to using them as fodder.

Evidence:
Imgur

By destroying artifacts, you either can get XP or Mora, it is the same for each. By destroying weapons, you obtain a number of 1* enhancement ores based on the rarity of the weapon (3 for 3 stars, 2 for 2 stars, 1 for 1 star).

Other Observations:

  • 5* or 4* artifacts and weapons cannot be destroyed
  • Refined 3* weapons cannot be destroyed
  • Character development materials, Food (except apples and sunsettias), gadgets, quest, and precious items cannot be destroyed
  • Material items that can be destroyed (in addition to apples and sunsettia) do not give anything when destroyed
  • Some material items can be destroyed and some cannot. There does not seem to be consistent reasoning for why

Significance:

Destroying weapons is not efficient and causes you to lose weapon XP. Destroying artifacts although has no "real downside," the rewards given are lackluster and are better used as fodder. There is no reason to destroy items that give no rewards. Those items are better used for the Parametric Transformer if they need to be disposed of. The only upside is to mass clear inventory space, but due to the size of the inventory and the loss of XP, it does not warrant using the feature.

Capped Enhancement Ores​

By: Kgbeast#6738
Added: 2021-10-09 (v. 2.1)
Last tested: 2021-10-09 (v. 2.1)
Discussion

Findings/Evidence:
Consequences of having capped (9999) Mystic Enhancement Ore:

  • Abyss rewards are sent to the mail and cannot be collected until there is enough room for ore to no longer exceed the 9999 mystic enhancement ore cap
    Video

  • Mystic enhancement ore from the Battle Pass cannot be collected until there is enough room for ore to no longer exceed the 9999 mystic enhancement ore cap

  • Completing a random event will not yield any ore rewards but will still give companionship EXP and mora
    Video
    Video

  • Mystic enhancement ore cannot be crafted at the blacksmith with either amethyst chunks or crystal ore
    Video

  • Completing a daily commission CAN and will give mystic enhancement ore, exceeding the 9999 cap (previous methods still do not work past 9999)
    Video

Significance:
This will never happen to you though.

Crystal Chunk Inventory Cap​

By: Cypher130n#6831
Added: 2022-03-29 (v. 2.5)
Last tested: 2022-03-29 (v. 2.5)
Discussion

Findings:

  • The natural cap of Crystal Chunks able to be held in a player's inventory is 9999.
  • This can be overcapped by claiming Crystal Chunks as a reward from Expeditions at a maximum of 10006 (9998+8).

Evidence:
Natural cap
Overcap

Significance:
Don't become an Ore Archon like me.

Berry Max​

By: makeway4pK#3901
Added: 2022-03-22 (v. 2.5)
Last tested: 2022-03-22 (v. 2.5)
Discussion

Finding:
Inventory cap for berries is 9999 much like crystal chunks.

Evidence:
YouTube

Significance:
Knowledge... and maybe a hint that all materials might be capped at 9999.

Events​

Reunion Blessing Is Consumed Before Ley Line Overflow​

By: jas#6562
Added: 2023-01-13 (v. 3.3)
Last tested: 2023-01-13 (v. 3.3)
Discussion

Finding:
The Reunion Blessing bonus from the Stellar Reunion event is prioritized over the Ley Line Overflow bonus if both events are active at the same time.

Evidence:
YouTube

Significance:
This won't be a particularly common occurrence, but if you want to use the Reunion Blessing bonus for something other than Ley Line Outcrops while Ley Line Overflow is active, spending Resin on another activity before doing Ley Line Outcrops is recommended.

It might be assumed that this works similarly for the Overflowing Mastery event. In theory, it's likely that Reunion Blessing is also consumed first, but testing is still necessary to verify that claim.

Reunion Blessing Is Consumed Before Overflowing Mastery​

By: jas#6562, spacespace#9403
Added: 2023-02-22 (v. 3.4)
Last tested: 2023-02-21 (v. 3.4)
Discussion

Finding:
The Reunion Blessing bonus from Stellar Reunion is consumed before the Overflowing Mastery bonus if both events are active at the same time.

Evidence:
www.youtube.com

Significance:
If you have both events active and you wish to use Reunion Blessing for something other than talent domains while Overflowing Mastery is active, you should spend resin on that other activity before doing talent domains.