KQM | Genshin Impact
echoes-basic-mechanics
56 messages
Ticket ToolMon, April 4, 2022 at 03:53 AM
@๊งเผบ ๐๐๐๐๐ธ๐ฝ๐๐๐โจเผป๊ง - As an author, it is your responsibility to complete the ticket
Guidelines
- Name your ticket with
- When you are ready to submit the ticket, compile everything into one message following the format below. Then type
- The ticket will be scrapped if: No activity >1 week or open for >1 month.
Guidelines
- Name your ticket with
$rename <ticket name>- When you are ready to submit the ticket, compile everything into one message following the format below. Then type
$close or click the button; the ticket will automatically be moved to be reviewed.- The ticket will be scrapped if: No activity >1 week or open for >1 month.
Write-up Format
Theory/Finding/Bug: Title of your submission
Evidence: Explanations with calculations and/or Youtube/Imgur proofs
Significance: Conclusion
Evidence: Explanations with calculations and/or Youtube/Imgur proofs
Significance: Conclusion
Weke | Knave EnjoyerMon, April 4, 2022 at 03:54 AM
first
yolitme0 | crab%Mon, April 4, 2022 at 03:54 AM
yay
๊งเผบ ๐๐๐๐๐ธ๐ฝ๐๐๐โจเผป๊งMon, April 4, 2022 at 04:09 AM
Reminder that this ticket is only for submissions for the Echoes artifact set basic mechanics. Please keep any discussion and conversation in #โtc-discussion relating to this topic.
We are looking for findings about the following list of mechanics. This list is NOT comprehensive. Feel free to add additional findings that you may encounter.
1. does the set work as written (probably not)
2. do proc increases fall off
3. can you actually reach 100% proc chance
4. does hitting normal attacks during the 0.2s cd do something
5. can mihoyo code a set properly ( see hypothesis for question 1)
6. How does echoes work in AoE?
We are looking for findings about the following list of mechanics. This list is NOT comprehensive. Feel free to add additional findings that you may encounter.
1. does the set work as written (probably not)
2. do proc increases fall off
3. can you actually reach 100% proc chance
4. does hitting normal attacks during the 0.2s cd do something
5. can mihoyo code a set properly ( see hypothesis for question 1)
6. How does echoes work in AoE?
itzsomebody | chrisMon, April 4, 2022 at 04:19 AM
When uploading evidence, please make sure to use YouTube/Imgur/related service (not streamable.com) instead of directly uploading to the ticket.
Some other things to look at are:
* Does ping affect Echoes? If so, how?
* What is the duration of how long Echoes stacks last (if any at all)?
* How does Echoes interact with 0-damage attacks (e.g. hitting Abyss mage/Mitachurl shield)?
* Is Echoes affected by hitlag?
* How do the 0.2s and 0.05cds interact with each other (is there overlap)?
Some other things to look at are:
* Does ping affect Echoes? If so, how?
* What is the duration of how long Echoes stacks last (if any at all)?
* How does Echoes interact with 0-damage attacks (e.g. hitting Abyss mage/Mitachurl shield)?
* Is Echoes affected by hitlag?
* How do the 0.2s and 0.05cds interact with each other (is there overlap)?
Finding
Echoes Flat DMG is not limited to single-target (works in AoE)
Evidence
Video: https://youtu.be/LaPF7dqCwhM
Math: https://docs.google.com/spreadsheets/d/14S-_hWzHOtxKHfd2G3x179Wp9MgkGvlPOXIcnGZ5VcA/edit?usp=sharing
Echoes Flat DMG is not limited to single-target (works in AoE)
Evidence
Video: https://youtu.be/LaPF7dqCwhM
Math: https://docs.google.com/spreadsheets/d/14S-_hWzHOtxKHfd2G3x179Wp9MgkGvlPOXIcnGZ5VcA/edit?usp=sharing
Finding
Echoes uses total ATK (as expected) including buffs from Bennett, 4pc Noblesse, and ATK from weapon passives.
Videos:
* Bennett + Echo Tao: https://youtu.be/z7z64VLHLkI
* Homa + Echo Tao: https://youtu.be/giX6K8wMDow
* Homa (< 50% HP) + Echo Tao: https://youtu.be/9gOWwzUQt2s
Math: https://docs.google.com/spreadsheets/d/14S-_hWzHOtxKHfd2G3x179Wp9MgkGvlPOXIcnGZ5VcA/edit?usp=sharing
Echoes uses total ATK (as expected) including buffs from Bennett, 4pc Noblesse, and ATK from weapon passives.
Videos:
* Bennett + Echo Tao: https://youtu.be/z7z64VLHLkI
* Homa + Echo Tao: https://youtu.be/giX6K8wMDow
* Homa (< 50% HP) + Echo Tao: https://youtu.be/9gOWwzUQt2s
Math: https://docs.google.com/spreadsheets/d/14S-_hWzHOtxKHfd2G3x179Wp9MgkGvlPOXIcnGZ5VcA/edit?usp=sharing
Finding
Echoes effect can be shared with subsequent attacks as long as they fit in the 0.05 second interval
Evidence
Video: https://youtu.be/JGRxO2f-4Ks
Math: https://docs.google.com/spreadsheets/d/14S-_hWzHOtxKHfd2G3x179Wp9MgkGvlPOXIcnGZ5VcA/edit?usp=sharing
Echoes effect can be shared with subsequent attacks as long as they fit in the 0.05 second interval
Evidence
Video: https://youtu.be/JGRxO2f-4Ks
Math: https://docs.google.com/spreadsheets/d/14S-_hWzHOtxKHfd2G3x179Wp9MgkGvlPOXIcnGZ5VcA/edit?usp=sharing
Finding
Echoes will stack even when holder's Normal Attacks deal 0 damage
Evidence
Testing by first hitting a Mitachurl's shield for 0 damage and then the Mitachurl to deal NA damage results in a sampled Echoes Flat DMG bonus average trigger chance that is abnormally high if the set did not stack on 0-damage Normal Attacks.
Video: https://youtu.be/x_B1wD1tSNM
Math: https://docs.google.com/spreadsheets/d/14S-_hWzHOtxKHfd2G3x179Wp9MgkGvlPOXIcnGZ5VcA/edit?usp=sharing
Echoes will stack even when holder's Normal Attacks deal 0 damage
Evidence
Testing by first hitting a Mitachurl's shield for 0 damage and then the Mitachurl to deal NA damage results in a sampled Echoes Flat DMG bonus average trigger chance that is abnormally high if the set did not stack on 0-damage Normal Attacks.
Video: https://youtu.be/x_B1wD1tSNM
Math: https://docs.google.com/spreadsheets/d/14S-_hWzHOtxKHfd2G3x179Wp9MgkGvlPOXIcnGZ5VcA/edit?usp=sharing
mystathiMon, April 4, 2022 at 07:40 AM
Finding:
Something weird is going on. We aren't confident as to what is.
Evidence:
In the following youtube video, at 14s in, Childe attempts a string of N4 Jump. The N3 of this string triggers Echos. The N4 of this string does not trigger Echos. For the next three strings of N4 Jump, Childe does not trigger Echos, totally in 13 hits between Echos procs.
The video does not show Childe's artifacts, but this video does have Echos procs, and the spreadsheet provided has a DMG table and counts of which hits proc Echos (1 being proc, 0 being no proc). Provided on the spreadsheet is framecounts for Childe's N4.
Others in CM are reporting similar things, and I have more footage I have partially gone through that have similar issues. I do have more footage I haven't gone through, and it's in a Google Drive linked on the spreadsheet. If necessary it can be posted to YouTube.
We also are further looking into more strings and potential causes.
Video: https://youtu.be/VDEnbWo4GBI
Spreadsheet: https://docs.google.com/spreadsheets/d/1deBV5unvhWWbNgLBUSUI4sriiAdYZkNXEHIBTSnaKlg/edit?usp=sharing
Significance:
Not only can Childe not crit, he also can't proc Echos!
All joking aside, there is something wrong witht the Echos proc chance here. This could change the way Echos is looked at, as it messes with how well the set works on Childe, and other characters may have similar issues. I know ZLM have reported weird stuff too.
This is updated from a ticket I made on this topic, with more information that has been gathered since then
Something weird is going on. We aren't confident as to what is.
Evidence:
In the following youtube video, at 14s in, Childe attempts a string of N4 Jump. The N3 of this string triggers Echos. The N4 of this string does not trigger Echos. For the next three strings of N4 Jump, Childe does not trigger Echos, totally in 13 hits between Echos procs.
The video does not show Childe's artifacts, but this video does have Echos procs, and the spreadsheet provided has a DMG table and counts of which hits proc Echos (1 being proc, 0 being no proc). Provided on the spreadsheet is framecounts for Childe's N4.
Others in CM are reporting similar things, and I have more footage I have partially gone through that have similar issues. I do have more footage I haven't gone through, and it's in a Google Drive linked on the spreadsheet. If necessary it can be posted to YouTube.
We also are further looking into more strings and potential causes.
Video: https://youtu.be/VDEnbWo4GBI
Spreadsheet: https://docs.google.com/spreadsheets/d/1deBV5unvhWWbNgLBUSUI4sriiAdYZkNXEHIBTSnaKlg/edit?usp=sharing
Significance:
Not only can Childe not crit, he also can't proc Echos!
All joking aside, there is something wrong witht the Echos proc chance here. This could change the way Echos is looked at, as it messes with how well the set works on Childe, and other characters may have similar issues. I know ZLM have reported weird stuff too.
This is updated from a ticket I made on this topic, with more information that has been gathered since then
โฟ.๏ฝก.:* ๐๐ถ๐๐พ | ๐ฆ๐๐ฑ๐ธ๐ธ๐ธ๐ธ๐ป๐ป๐ป *.:๏ฝก.โฟMon, April 4, 2022 at 07:59 AM
@The Golden Man Here
The Golden ManMon, April 4, 2022 at 08:17 AM
Finding:
How Zhongli's spearkick hits proc Echoes differs depending on the tester. Cause unknown, but the only obvious possible explanation that me and valorha could come up with is ping. I have about 50-60 ms on average, while valorha has 200+ ms on average.
Evidence:
I analyzed 106 Zhongli combos versus a Ruin Guard outside Mondstadt. In all of my combos, the first and second spearkick hits were linked: they either procced Echoes together or not at all. When the first two procced together, the third hit sometimes procced, but the fourth hit never procced. When the first two did not proc, the third and fourth hit were both able to proc and seemed to do so independently.
valorha over on Zhongli mains similarly tested 136 Zhongli combos versus a Ruin Guard and the Childe weekly boss. For valorha, all of the spearkick hits were linked: all four either procced Echoes together or not at all. There was one exception at combo 91, where the first three spearkick hits procced but the last did not.
My Video: https://youtu.be/mo7YwzUcvc8
My Spreadsheet (with help from Zerodyme): https://docs.google.com/spreadsheets/d/1wIGJeNZZqUBb7HZcp6vh4Wad1fIOkGTA5vemBJDKRTI/edit?usp=sharing
valorha's Video #1: https://www.youtube.com/watch?v=dE-1SAl1E-g
valorha's Video #2: https://youtu.be/gkC8XjL0fmM
valorha's Spreadsheet: https://docs.google.com/spreadsheets/d/10yMcbn1AkAHiyaPcwqddvX3Int-atCX5wJVtROjuvDs/edit?usp=sharing
How Zhongli's spearkick hits proc Echoes differs depending on the tester. Cause unknown, but the only obvious possible explanation that me and valorha could come up with is ping. I have about 50-60 ms on average, while valorha has 200+ ms on average.
Evidence:
I analyzed 106 Zhongli combos versus a Ruin Guard outside Mondstadt. In all of my combos, the first and second spearkick hits were linked: they either procced Echoes together or not at all. When the first two procced together, the third hit sometimes procced, but the fourth hit never procced. When the first two did not proc, the third and fourth hit were both able to proc and seemed to do so independently.
valorha over on Zhongli mains similarly tested 136 Zhongli combos versus a Ruin Guard and the Childe weekly boss. For valorha, all of the spearkick hits were linked: all four either procced Echoes together or not at all. There was one exception at combo 91, where the first three spearkick hits procced but the last did not.
My Video: https://youtu.be/mo7YwzUcvc8
My Spreadsheet (with help from Zerodyme): https://docs.google.com/spreadsheets/d/1wIGJeNZZqUBb7HZcp6vh4Wad1fIOkGTA5vemBJDKRTI/edit?usp=sharing
valorha's Video #1: https://www.youtube.com/watch?v=dE-1SAl1E-g
valorha's Video #2: https://youtu.be/gkC8XjL0fmM
valorha's Spreadsheet: https://docs.google.com/spreadsheets/d/10yMcbn1AkAHiyaPcwqddvX3Int-atCX5wJVtROjuvDs/edit?usp=sharing
Finding:
Zhongli's spearkick combos can fail to proc Echoes after 7, 8, or even 9 hits. If this differs between testers, it is likely that whatever explains the difference between spearkick proc patterns (above) also explains this differs between testers.
Evidence:
I analyzed 106 Zhongli combos versus a Ruin Guard. There are 7 recorded hits which fail to proc in a row between combo 81 (beginning at 4:48 in my video) and 82.
valorha analyzed over 136 combos versus a Ruin Guard and the Childe weekly boss, and got 8 recorded hits versus the Ruin Guard which fail to proc in a row between combo 58 (beginning at 2:27 of their first video) and 59. There were 9 recorded hits versus Childe which fail to proc in a row between combo 88 (beginning at 3:59 of their first video) and 89.
My Video: https://youtu.be/mo7YwzUcvc8
My Spreadsheet (with help from Zerodyme): https://docs.google.com/spreadsheets/d/1wIGJeNZZqUBb7HZcp6vh4Wad1fIOkGTA5vemBJDKRTI/edit?usp=sharing
valorha's Video #1: https://www.youtube.com/watch?v=dE-1SAl1E-g
valorha's Video #2: https://youtu.be/gkC8XjL0fmM
valorha's Spreadsheet: https://docs.google.com/spreadsheets/d/10yMcbn1AkAHiyaPcwqddvX3Int-atCX5wJVtROjuvDs/edit?usp=sharing
Zhongli's spearkick combos can fail to proc Echoes after 7, 8, or even 9 hits. If this differs between testers, it is likely that whatever explains the difference between spearkick proc patterns (above) also explains this differs between testers.
Evidence:
I analyzed 106 Zhongli combos versus a Ruin Guard. There are 7 recorded hits which fail to proc in a row between combo 81 (beginning at 4:48 in my video) and 82.
valorha analyzed over 136 combos versus a Ruin Guard and the Childe weekly boss, and got 8 recorded hits versus the Ruin Guard which fail to proc in a row between combo 58 (beginning at 2:27 of their first video) and 59. There were 9 recorded hits versus Childe which fail to proc in a row between combo 88 (beginning at 3:59 of their first video) and 89.
My Video: https://youtu.be/mo7YwzUcvc8
My Spreadsheet (with help from Zerodyme): https://docs.google.com/spreadsheets/d/1wIGJeNZZqUBb7HZcp6vh4Wad1fIOkGTA5vemBJDKRTI/edit?usp=sharing
valorha's Video #1: https://www.youtube.com/watch?v=dE-1SAl1E-g
valorha's Video #2: https://youtu.be/gkC8XjL0fmM
valorha's Spreadsheet: https://docs.google.com/spreadsheets/d/10yMcbn1AkAHiyaPcwqddvX3Int-atCX5wJVtROjuvDs/edit?usp=sharing
risukeMon, April 4, 2022 at 08:43 AM
Finding:
Echoes does not proc buff off field, nor does it gain stacks off field. This aligns with other artifacts that unless specifically said, will not have its effect unless on field.
Evidence:
Every Riptide Flash while Childe is not on the field does not gain Echoes' buff proc. When swapped to Childe, Echoes was able to gain stacks and proc buff.
https://youtu.be/obRuJFqgUrU
Echoes does not proc buff off field, nor does it gain stacks off field. This aligns with other artifacts that unless specifically said, will not have its effect unless on field.
Evidence:
Every Riptide Flash while Childe is not on the field does not gain Echoes' buff proc. When swapped to Childe, Echoes was able to gain stacks and proc buff.
https://youtu.be/obRuJFqgUrU
The Golden ManMon, April 4, 2022 at 09:09 AM
Finding
How Yoimiya's N1 hits proc Echoes differs depending on range. Cause unknown, but the most obvious explanation is that getting closer causes the second hit to happen within the 0.05 buff procced by the first.
However, I could not get the second N1 hit to proc without the first proccing, which should theoretically be possible if the first N1 does not proc the buff given the above explanation.
Evidence
Two videos show my Yoimiya doing N1 versus a Ruin Guard, one at range and one as close as possible. When at range, each N1 hits proc independently: they can both proc, neither proc, or either one can proc. When as close as possible to the Ruin Guard, the N1 hits become linked: they either proc together or not at all.
Close Video: https://youtu.be/2Fe58lDrjL8
Ranged Video: https://youtu.be/67Ta1R5ZSbo
How Yoimiya's N1 hits proc Echoes differs depending on range. Cause unknown, but the most obvious explanation is that getting closer causes the second hit to happen within the 0.05 buff procced by the first.
However, I could not get the second N1 hit to proc without the first proccing, which should theoretically be possible if the first N1 does not proc the buff given the above explanation.
Evidence
Two videos show my Yoimiya doing N1 versus a Ruin Guard, one at range and one as close as possible. When at range, each N1 hits proc independently: they can both proc, neither proc, or either one can proc. When as close as possible to the Ruin Guard, the N1 hits become linked: they either proc together or not at all.
Close Video: https://youtu.be/2Fe58lDrjL8
Ranged Video: https://youtu.be/67Ta1R5ZSbo
Finding
How Childe's N1 and N2 hits proc Echoes differs depending on the tester. Cause unknown, but it is likely that whatever explains the difference here also explains the similar difference observed between testers for Zhongli's spearkicks (above). The only obvious possible explanation that me and valorha could come up with for Zhongli is ping.
Evidence
I recorded a video of my Childe doing NA combos, and Myst recorded some videos of his Childe doing NA combos. In my video, Childe's N1 and N2 procced exclusively: either one could proc or neither could proc, but both could not proc in the same combo. In Myst's videos, Childe's N1 and N2 were linked: either they both procced together or not at all.
My Video: https://youtu.be/EckC4Lzoxyg
Myst's Videos: https://drive.google.com/drive/folders/1CBTwHI1wBcQH5mlV_oF1YgTwKufSZR3U?usp=sharing
Myst's Spreadsheet: https://docs.google.com/spreadsheets/d/1deBV5unvhWWbNgLBUSUI4sriiAdYZkNXEHIBTSnaKlg/edit?usp=sharing
How Childe's N1 and N2 hits proc Echoes differs depending on the tester. Cause unknown, but it is likely that whatever explains the difference here also explains the similar difference observed between testers for Zhongli's spearkicks (above). The only obvious possible explanation that me and valorha could come up with for Zhongli is ping.
Evidence
I recorded a video of my Childe doing NA combos, and Myst recorded some videos of his Childe doing NA combos. In my video, Childe's N1 and N2 procced exclusively: either one could proc or neither could proc, but both could not proc in the same combo. In Myst's videos, Childe's N1 and N2 were linked: either they both procced together or not at all.
My Video: https://youtu.be/EckC4Lzoxyg
Myst's Videos: https://drive.google.com/drive/folders/1CBTwHI1wBcQH5mlV_oF1YgTwKufSZR3U?usp=sharing
Myst's Spreadsheet: https://docs.google.com/spreadsheets/d/1deBV5unvhWWbNgLBUSUI4sriiAdYZkNXEHIBTSnaKlg/edit?usp=sharing
pray_primeMon, April 4, 2022 at 01:46 PM
Does echoes proc apply to other stuff coded as normal attack (the way yunjin works), like fischl c1?
weirdfoolMon, April 4, 2022 at 02:53 PM
It applies to Childe Riptide Riptide Flash at the very least, which is triggered from his Charge Shots
Finding
Kazuha's 3-hit n5 is "linked"
Kaz's 2-hit n3 is not "linked"
Evidence
I analysed 107 full N5 strings against a downed ruin guard over 2 sessions
n5:
- the first 2 hits display the same proc (i.e NN or PP) in 105/107 trials
- the 3rd hit only procced in 6/107 trials, and those 6 were all NNP overall for n5. If you assume even the minimum 36% proc chance for n5-3, the chance to get only 6 procs in 107 trials is in the region of 6x10-14
n3:
- every combination has been observed multiple times (NN, NP, PN, PP)
Spreadsheet: https://docs.google.com/spreadsheets/d/1_ZFiEDXpQFTtjHBgjyOXdG-PQW1j8zy8w7w3GK9iaIs/edit?usp=sharing
Videos: https://www.youtube.com/watch?v=UdntmwQl2rk
https://www.youtube.com/watch?v=62pGzyDR67U (might want to mute this one, forgot to strip the audio)
Kazuha's 3-hit n5 is "linked"
Kaz's 2-hit n3 is not "linked"
Evidence
I analysed 107 full N5 strings against a downed ruin guard over 2 sessions
n5:
- the first 2 hits display the same proc (i.e NN or PP) in 105/107 trials
- the 3rd hit only procced in 6/107 trials, and those 6 were all NNP overall for n5. If you assume even the minimum 36% proc chance for n5-3, the chance to get only 6 procs in 107 trials is in the region of 6x10-14
n3:
- every combination has been observed multiple times (NN, NP, PN, PP)
Spreadsheet: https://docs.google.com/spreadsheets/d/1_ZFiEDXpQFTtjHBgjyOXdG-PQW1j8zy8w7w3GK9iaIs/edit?usp=sharing
Videos: https://www.youtube.com/watch?v=UdntmwQl2rk
https://www.youtube.com/watch?v=62pGzyDR67U (might want to mute this one, forgot to strip the audio)
amapolamysteryWed, April 6, 2022 at 01:40 AM
are the videos meant to be private?
weirdfoolWed, April 6, 2022 at 06:41 AM
Oh no, let me change that now
weirdfoolWed, April 6, 2022 at 06:57 AM
Should be good now, although the second might run you an ad while it's in the process of muting
Finding
Hit registration occurs before evidence of it in the animation
Evidence
In the following video, I got hit out of Kazuha's 2-hit n3 before any visual indication that I'd hit the first attack (before the hitflash) - the hitflash occurs afterwards and there's an instance of damage
Video: https://youtu.be/wlm5yPNYQ1g
Significance
Despite not being specifically Echoes related, I do think the question of exactly when damage is registered is important for how Echoes works for characters with fast and/or multihit attacks.
Kazuha's n3 visually has only 10 frames between 3-1 and 3-2, and yet the two hits are independent of each other Echoes Proc-wise which runs contrary to how Echoes should work (within the 12 frame CD time). If 3-1 registers damage a few frames before the hitflash, this may fall in line with Echoes.
Hit registration occurs before evidence of it in the animation
Evidence
In the following video, I got hit out of Kazuha's 2-hit n3 before any visual indication that I'd hit the first attack (before the hitflash) - the hitflash occurs afterwards and there's an instance of damage
Video: https://youtu.be/wlm5yPNYQ1g
Significance
Despite not being specifically Echoes related, I do think the question of exactly when damage is registered is important for how Echoes works for characters with fast and/or multihit attacks.
Kazuha's n3 visually has only 10 frames between 3-1 and 3-2, and yet the two hits are independent of each other Echoes Proc-wise which runs contrary to how Echoes should work (within the 12 frame CD time). If 3-1 registers damage a few frames before the hitflash, this may fall in line with Echoes.
Finding
Ping affects how Echoes behaves
Evidence
Recorded video of the same set-up as my previous Kazuha test, but with latency artificially increased by 500ms. While I haven't transcribed the exact breakdown of procs into a spreadsheet, key things that are easy to see (in contrast to the normal latency testing) are as follows:
n1 and n2 have become linked - every pair of n1 and n2 in the 30+ strings captured have the same proc value (NN or PP)
n3 multihits are not linked
n5 multihits have new proc strings, i.e PPP, NPP, neither of which ever happened in the prior testing.
- It looks like 5-2 and 5-3 may now be linked instead of 5-1 and 5-2? In addition, the third hit proccing (xxP) seems more common? Further analysis needed for definitive statements on these two.
Video: https://youtu.be/h_R05z7UXqM
3
Ping affects how Echoes behaves
Evidence
Recorded video of the same set-up as my previous Kazuha test, but with latency artificially increased by 500ms. While I haven't transcribed the exact breakdown of procs into a spreadsheet, key things that are easy to see (in contrast to the normal latency testing) are as follows:
n1 and n2 have become linked - every pair of n1 and n2 in the 30+ strings captured have the same proc value (NN or PP)
n3 multihits are not linked
n5 multihits have new proc strings, i.e PPP, NPP, neither of which ever happened in the prior testing.
- It looks like 5-2 and 5-3 may now be linked instead of 5-1 and 5-2? In addition, the third hit proccing (xxP) seems more common? Further analysis needed for definitive statements on these two.
Video: https://youtu.be/h_R05z7UXqM
3๐1
The Golden ManWed, April 6, 2022 at 03:24 PM
Speculation/Hypothesis: The game uses an internal timer (the 0.05 second buff?) to treat AoE hits that happen within that timer as a single hit for the purposes of echoes. Linking is a result of the game treating successive hits as AoE hits (and thus as a single hit) for the purposes of Echoes because they happen within that internal timer. The internal timer is affected (extended, but also maybe cut short sometimes) by ping.
I'm not sure how to test that precisely, though.
steno119Wed, April 6, 2022 at 03:49 PM
that's very likely true, you can see the same wording in other mechanics that do similar things
Cer0Thu, April 7, 2022 at 08:00 AM
https://youtu.be/hD2mKnfMsHA not so sure if you would consider it as a bug but, normal attacks first and second hit may or may not trigger the set at the same time. If they can trigger set independently, that should mean they count as seperate hits. If they count as seperate hits how come I can't proc in 4 NAs (8 hits total, should be 7*20%+36% way over guaranteed). Video starts with a double non-crit procs, next 4 NA's are non-proc crits, then a double proc followed by back to back one out of two hits proc. as a footnote: for this video 1589 is a non-crit proc, 1618 is a crit non-proc, 3865 is crit with proc.
ning echoes bug
GPU: NVIDIA GeForce RTX 2070 Super
CPU: Intel(R) Core(TM) i7-10700K CPU @ 3.80GHz
Memory: 32 GB RAM (31.86 GB RAM usable)
Current resolution: 1920 x 1080, 240Hz
Operating system:
CPU: Intel(R) Core(TM) i7-10700K CPU @ 3.80GHz
Memory: 32 GB RAM (31.86 GB RAM usable)
Current resolution: 1920 x 1080, 240Hz
Operating system:
borderlinefThu, April 7, 2022 at 11:39 PM
My understanding is that when normal attacks hit <0.05 seconds of each other, they will share the instance of echoes, meaning they will both proc or both not proc. When hits are 0.05<x<0.2 seconds apart, they will not share a proc and the second hit will never proc or increase the proc rate. Since the first 4 na were not simultaneous, they each only increased rate once. The 5th hit was simultaneous, so they shared the proc.
pray_primeThu, April 7, 2022 at 11:52 PM
wouldn't be surprised if that internal timer is affected by hitlag too, could check against a bow user
โฎโท๏ฝก ๐ป๐๐๐๐ธ๐ฝ ๐ป๐๐พ๐ ๏ฝกโทโฎFri, April 8, 2022 at 08:44 AM
Finding:
Additional data about Zhongli's spear kick, performed at 140-150ms ping
Evidence:
160 full combos (plus some partial ones because I can't click properly) against ruin guard. First three hits of n5 were linked, except for once instance where only the first two were linked.
Spreadsheet (includes links to videos): https://docs.google.com/spreadsheets/d/1sPWn7gCcHv3s30gEitw1WG7AWsuGx-nXe5PNdBSaFo0/edit?usp=sharing
Additional data about Zhongli's spear kick, performed at 140-150ms ping
Evidence:
160 full combos (plus some partial ones because I can't click properly) against ruin guard. First three hits of n5 were linked, except for once instance where only the first two were linked.
Spreadsheet (includes links to videos): https://docs.google.com/spreadsheets/d/1sPWn7gCcHv3s30gEitw1WG7AWsuGx-nXe5PNdBSaFo0/edit?usp=sharing
๐1
The Golden ManFri, April 8, 2022 at 06:33 PM
Since at 50-60 ms only Zhongli's first two spearkicks are linked, at 140-150 three are linked, and at 200+ all four are linked, it's probably safe to conclude that higher ping = more linking with Zhongli. The exact breakpoints would require more testing, though.2
2
xreejanSat, April 9, 2022 at 07:53 PM
after looking at the recording just realised that my ping fluctuated a lot3
Finding:
Additional data about Zhongli's with the Echoes Set, performed at 165-170+ ping (fluctuated a lot more)
Evidence:
35 Full N5D combos vs a Ruin Guard
First 3 hits of the spear kick were mostly linked
Spreadsheet:
https://docs.google.com/spreadsheets/d/1iWXdiUqkWMOlinDUm5yRGrhJA02i-cv1/edit?usp=sharing&ouid=105974977777829459167&rtpof=true&sd=true
Video:
https://www.youtube.com/watch?v=6l2tIOwWBHQ
Additional data about Zhongli's with the Echoes Set, performed at 165-170+ ping (fluctuated a lot more)
Evidence:
35 Full N5D combos vs a Ruin Guard
First 3 hits of the spear kick were mostly linked
Spreadsheet:
https://docs.google.com/spreadsheets/d/1iWXdiUqkWMOlinDUm5yRGrhJA02i-cv1/edit?usp=sharing&ouid=105974977777829459167&rtpof=true&sd=true
Video:
https://www.youtube.com/watch?v=6l2tIOwWBHQ
after looking at the recording just realised that my ping fluctuated a lot3โฟ.๏ฝก.:* ๐๐ถ๐๐พ | ๐ฆ๐๐ฑ๐ธ๐ธ๐ธ๐ธ๐ป๐ป๐ป *.:๏ฝก.โฟSun, April 10, 2022 at 04:18 AM
My contribution to Echoes
Disclaimer: Because I was dumb and didn't notice, my N3 non proc crit and proc non crit is the same value. Logically, due to my 5% crit rate it should count as a proc but I've chosen to disregard all N3 data instead, just in case.
Finding:
More data on Zhongli's string for Echoes. (80 ish ms ping) I'm getting a consistent amount of 3 links in the spear kick, so @213133712072638465 you can prolly add in an extra threshold.
Bonus: I had a point in my testing where I disconnected entirely. It resulted in a 4 linked spear kick + N6 proc. Since it only happened once, I can't say it as hard evidence but it's a good chance it can contribute to the timer/ping theory. Included in the sheet at the bottom with vid, the disconnect happened around 00:10 (as you can tell by the ping meter and the sudden appearance of numbers as I reconnect). For as far as I have, this is my only occurrence of a 4 linked spear kick and technically 5 procs in a row.
Evidence:
37 full NA strings + 1 vid where I disconnected out of pure luck.
Spreadsheet:
https://docs.google.com/spreadsheets/d/1_HVfqu5x1tf5xifNFtqyxM5acPC2e4h0hF5ZObkj0Ug/edit?usp=sharing
Disclaimer: Because I was dumb and didn't notice, my N3 non proc crit and proc non crit is the same value. Logically, due to my 5% crit rate it should count as a proc but I've chosen to disregard all N3 data instead, just in case.
Finding:
More data on Zhongli's string for Echoes. (80 ish ms ping) I'm getting a consistent amount of 3 links in the spear kick, so @213133712072638465 you can prolly add in an extra threshold
.Bonus: I had a point in my testing where I disconnected entirely. It resulted in a 4 linked spear kick + N6 proc. Since it only happened once, I can't say it as hard evidence but it's a good chance it can contribute to the timer/ping theory. Included in the sheet at the bottom with vid, the disconnect happened around 00:10 (as you can tell by the ping meter and the sudden appearance of numbers as I reconnect). For as far as I have, this is my only occurrence of a 4 linked spear kick and technically 5 procs in a row.
Evidence:
37 full NA strings + 1 vid where I disconnected out of pure luck.
Spreadsheet:
https://docs.google.com/spreadsheets/d/1_HVfqu5x1tf5xifNFtqyxM5acPC2e4h0hF5ZObkj0Ug/edit?usp=sharing
kirkie.Sun, April 10, 2022 at 08:42 AM
has anyone tested if echoes is affected by hitlag yet?
7
7
Memento MoriMon, April 18, 2022 at 01:53 AM
Hello from AyatoMains! We on the TC team there have compiled data with a lot of help from the community on EoO average proc rates at varying ping levels with Ayato (and are still adding more as people submit their counts). You can find the individual sheets linked here, as well (in case you want to play around with the data yourself). (If you're willing to contribute, that would be very much appreciated. Feel free to DM me for info)
You're welcome to share this, so long as you credit us of course.
Finding:
EoO proc rate appears to fall off a little bit after 100-110 ping.
Evidence:
https://docs.google.com/spreadsheets/d/1OJDpktn0sk9K3Bw7o13alZRqan5AZUkmrXWQ3RdPuhs/edit?pli=1#gid=0
Notes:
Please do not use the individual proc rates at any specific ping level as an indicator for the general proc rate. i.e. there's one entry of a proc rate of 47% at very low ping, due to personal variability, this only indicates that the avg proc rate is within error range of the expected avg of 50.2%.
5
1
You're welcome to share this, so long as you credit us of course.
Finding:
EoO proc rate appears to fall off a little bit after 100-110 ping.
Evidence:
https://docs.google.com/spreadsheets/d/1OJDpktn0sk9K3Bw7o13alZRqan5AZUkmrXWQ3RdPuhs/edit?pli=1#gid=0
Notes:
Please do not use the individual proc rates at any specific ping level as an indicator for the general proc rate. i.e. there's one entry of a proc rate of 47% at very low ping, due to personal variability, this only indicates that the avg proc rate is within error range of the expected avg of 50.2%.
AYTM 4 Echoes Testing Compilation
Sheet1
AYTM 4 ECHOES OF AN OFFERING PROC RATE TESTING COMPILATION
It appears that the proc rate for 4 Echoes is dependent on the player's ping. To better explore this, we will be compiling all our testing here!
Hopefully through our collective efforts we'll be able to build a better understandin...
AYTM 4 ECHOES OF AN OFFERING PROC RATE TESTING COMPILATION
It appears that the proc rate for 4 Echoes is dependent on the player's ping. To better explore this, we will be compiling all our testing here!
Hopefully through our collective efforts we'll be able to build a better understandin...
51
valorha | Echoes PikeliTue, April 19, 2022 at 03:30 PM
Finding:
In continuation with this, lower framerate might have an effect on the behavior of the set on Zhongli, particularly the spearkick, possibly lowering the proc rate.
Evidence:
I recorded 2 encounters with the same enemy, with the same set pieces and weapon, but on 1st encounter I had 60 fps, and on second 30 fps, totaling 36 combos for the 1st encounter and 33 for the 2nd.
In 60 fps, 17/18 times, the 4 hits of the spearkick procced together.
However, in 30 fps, at most only 3 hits of the spearkick did so.
spreadsheet (2nd tab): https://docs.google.com/spreadsheets/d/10yMcbn1AkAHiyaPcwqddvX3Int-atCX5wJVtROjuvDs/edit?usp=sharing
video: https://youtu.be/jxNZBW_-pTc4
2
1
In continuation with this, lower framerate might have an effect on the behavior of the set on Zhongli, particularly the spearkick, possibly lowering the proc rate.
Evidence:
I recorded 2 encounters with the same enemy, with the same set pieces and weapon, but on 1st encounter I had 60 fps, and on second 30 fps, totaling 36 combos for the 1st encounter and 33 for the 2nd.
In 60 fps, 17/18 times, the 4 hits of the spearkick procced together.
However, in 30 fps, at most only 3 hits of the spearkick did so.
spreadsheet (2nd tab): https://docs.google.com/spreadsheets/d/10yMcbn1AkAHiyaPcwqddvX3Int-atCX5wJVtROjuvDs/edit?usp=sharing
video: https://youtu.be/jxNZBW_-pTc
Testing echoes on Zhongli (lag 200+ ms)
Echoes testing
valorha#6262,P = proc,video link,https://www.youtube.com/watch?v=dE-1SAl1E-g
211 - 215 ms,N = no proc,part 2,https://youtu.be/gkC8XjL0fmM
Expected damage for verifications,Avg,47...
valorha#6262,P = proc,video link,https://www.youtube.com/watch?v=dE-1SAl1E-g
211 - 215 ms,N = no proc,part 2,https://youtu.be/gkC8XjL0fmM
Expected damage for verifications,Avg,47...
421
pkach | @ for noille questionsTue, April 19, 2022 at 10:36 PM
Did you guys happen to find any data regarding linking? i.e. if the first hit procs Echoes then a second "linked" one always procs it as well, or if the first one doesn't proc a second "linked" one never will
I was skeptical at first but it's happening too often in too many samples... it's uncanny, doesn't seem to be that affected by ping either seeing as it happens across different ping ranges
I was skeptical at first but it's happening too often in too many samples... it's uncanny, doesn't seem to be that affected by ping either seeing as it happens across different ping ranges
Memento MoriTue, April 19, 2022 at 10:40 PM
Were you seeing this only for the very first hit, or throughout the sequence?
All the data is available in the spreadsheet, so if you want to check anyone's runs for that, it's easy to do.
All the data is available in the spreadsheet, so if you want to check anyone's runs for that, it's easy to do.
Reply to Memento Mori: Were you seeing this only for the very first hit, or throughout the sequence?
pkach | @ for noille questionsTue, April 19, 2022 at 10:43 PM Memento Mori: Were you seeing this only for the very first hit, or throughout the sequence? For the first hit of the sequence and for the spearkick hits
I'll take a look at the spreadsheet later when I'm on PC, just wanted to know off-hand if you guys noticed anything
Memento MoriTue, April 19, 2022 at 10:45 PM
Ah, we kinda disregarded the first 3 hits of every sequence in the proc chance calcs for Ayato due to his Namisen stacks and clone pop making numbers very inconsistent there, especially since we crowdsourced data.
So I haven't looked into anything about that.
pkach | @ for noille questionsTue, April 19, 2022 at 10:51 PM
Understandable, ty for sharing 
ren718.exeThu, April 21, 2022 at 04:28 AM
Finding:
Ayato's second slash and first C6 strike seem to be linked. If the second slash is buffed by echos, the first c6 strike is also buffed. In turn, if the second slash is not buffed, the first strike will not be buffed as well. The second strike seems to have its own proc and trigger towards echos.
Evidence:
https://docs.google.com/spreadsheets/d/1c1KO48Aj9ED1VeQqKDIbIB6uwOroQXQ65gTIoLxqFgU/edit?usp=drivesdk
https://youtube.com/playlist?list=PL48qQ9xM-_LM7JurvKvoQLgfKCw2Dz8ri
Ayato's second slash and first C6 strike seem to be linked. If the second slash is buffed by echos, the first c6 strike is also buffed. In turn, if the second slash is not buffed, the first strike will not be buffed as well. The second strike seems to have its own proc and trigger towards echos.
Evidence:
https://docs.google.com/spreadsheets/d/1c1KO48Aj9ED1VeQqKDIbIB6uwOroQXQ65gTIoLxqFgU/edit?usp=drivesdk
https://youtube.com/playlist?list=PL48qQ9xM-_LM7JurvKvoQLgfKCw2Dz8ri
Ayato Echos Tests
Sheet1
Results
Procs (no c6),748,Procs (c6),848
Slashes (no c6),1517,Slashes (c6),1717
Avg proc rate,49.31%,Avg proc rate,49.39%
Procs missed in successsion (C6) WIP
1,2,3,4,5,6,7,8,9
NOTES,(These are based on my observations; Do not take as fact)
Results
Procs (no c6),748,Procs (c6),848
Slashes (no c6),1517,Slashes (c6),1717
Avg proc rate,49.31%,Avg proc rate,49.39%
Procs missed in successsion (C6) WIP
1,2,3,4,5,6,7,8,9
NOTES,(These are based on my observations; Do not take as fact)
Proc rate for excluded c6 may not be acc...
Bobrokrot [RU] Genshin AcademySat, April 30, 2022 at 06:15 PM
Theory: 4 piece effect of an artifact set Echoes of an Offering triggers around 50,2% of the time in theory assuming the set works exactly as written.
Evidence: based on Markov chains theory, described in the pdf.
Significance: set doesn't actually work as written so it's useless
16
5
Evidence: based on Markov chains theory, described in the pdf.
Significance: set doesn't actually work as written so it's useless
Attachment: echoes_theory_EN.pdf
(166.5KB)
165Reposting it from the other ticket.
eriri๐ธSun, May 15, 2022 at 10:01 AM
Finding:
Turns out, Echoes does proc the buff off-field, but still does not gain stacks off-field.
Evidence:
Video 1 - Reference of Echoes proc/nonproc damage in the proceeding evidence.
Noncrit without Echoes Proc - 693
Noncrit with Echoes Proc - 1321
Crit without Echoes Proc - 1223
Crit with Echoes Proc - 2331
Video 2 - Echoes proccing the buff off-field on the first hit. (Proof that the buff procs off-field)
Video 3 - Echoes not proccing the buff off-field after five consecutive off-field hits. (Proof that the buff does not gain stacks off-field)
https://youtu.be/GrJWJz4EPOA
https://youtu.be/FngwQux_BkY
https://youtu.be/p7JdplFz6og10
3
1
Finding:
Turns out, Echoes does proc the buff off-field, but still does not gain stacks off-field.
Evidence:
Video 1 - Reference of Echoes proc/nonproc damage in the proceeding evidence.
Noncrit without Echoes Proc - 693
Noncrit with Echoes Proc - 1321
Crit without Echoes Proc - 1223
Crit with Echoes Proc - 2331
Video 2 - Echoes proccing the buff off-field on the first hit. (Proof that the buff procs off-field)
Video 3 - Echoes not proccing the buff off-field after five consecutive off-field hits. (Proof that the buff does not gain stacks off-field)
https://youtu.be/GrJWJz4EPOA
https://youtu.be/FngwQux_BkY
https://youtu.be/p7JdplFz6og
1031valorha | Echoes PikeliSun, August 14, 2022 at 05:33 PM
Sharing my latest findings and testings, sadly I couldn't complete all the tests I had intended due to lack of time and shifting of priorities. These are already a few months old, but if someone wants to pick things up or just for the sake of archiving these.
Finding:
Further data that could point to the set working as intended but be subject to ping/possibly low fps/character multihits quick enough to take advantage of the lag between server and client (like Zhongli, or indicates that his spearkick is simply working differently compared to other multihits).
Evidence:
-A typical N5D of Zhongli results in a higher proc rate than a full Ayato E sequence at 200+ ms, because of the spearkick all proccing together, while Ayato has no multihits (1500 hits Zhongli has 47% proc rate vs 1573 hits Ayato has ~34% proc rate)
-Spacing hits seems to result in higher proc rate (around the theoretical one): ~50% by doing N1J vs 47% for N5D with Zhongli (777 hits)
-Less data so just to observe a pattern: doing full combos with Ayato NAs results in a possibly higher proc rate than with E thanks to 2-hits on N4 --> need more data and to compare it with other characters to see if multihits with higher ping usually link and therefore results in a higher proc rate
-More data from different testers on Zhongli that support the theory that the lower the ping, the lower the proc rate and vice versa (more ping = more spearkick hits linking, though we haven't found the higher cap where it becomes less)
https://docs.google.com/spreadsheets/d/1k1rPyayo-JLpRw2j5EYrJw3vREI0jKPu2LU9ZSpIZ1Q/edit?usp=sharing
Finding:
Further data that could point to the set working as intended but be subject to ping/possibly low fps/character multihits quick enough to take advantage of the lag between server and client (like Zhongli, or indicates that his spearkick is simply working differently compared to other multihits).
Evidence:
-A typical N5D of Zhongli results in a higher proc rate than a full Ayato E sequence at 200+ ms, because of the spearkick all proccing together, while Ayato has no multihits (1500 hits Zhongli has 47% proc rate vs 1573 hits Ayato has ~34% proc rate)
-Spacing hits seems to result in higher proc rate (around the theoretical one): ~50% by doing N1J vs 47% for N5D with Zhongli (777 hits)
-Less data so just to observe a pattern: doing full combos with Ayato NAs results in a possibly higher proc rate than with E thanks to 2-hits on N4 --> need more data and to compare it with other characters to see if multihits with higher ping usually link and therefore results in a higher proc rate
-More data from different testers on Zhongli that support the theory that the lower the ping, the lower the proc rate and vice versa (more ping = more spearkick hits linking, though we haven't found the higher cap where it becomes less)
https://docs.google.com/spreadsheets/d/1k1rPyayo-JLpRw2j5EYrJw3vREI0jKPu2LU9ZSpIZ1Q/edit?usp=sharing
Valorha's echoes testing (200+ ms)
Compil findings per character
valorha#6262
ping range (ms),fps,Character,Test,Proc rate,Sample size (hit count),Observations,Link
~211 to ~215
~215 to ~218
~255 to ~259,60,Zhongli,Combo N5D,47,07%,1500,All spearkick hits link (proc/no proc together) 98% of the data
Many cases of more than 5 con...
valorha#6262
ping range (ms),fps,Character,Test,Proc rate,Sample size (hit count),Observations,Link
~211 to ~215
~215 to ~218
~255 to ~259,60,Zhongli,Combo N5D,47,07%,1500,All spearkick hits link (proc/no proc together) 98% of the data
Many cases of more than 5 con...
nzpiefaceTue, August 16, 2022 at 06:35 AM
do you guys want an explanation of echoes wrt ping but without evidence?
itzsomebody | chrisTue, August 16, 2022 at 06:36 AM
yeah go ahead. just mark it as missing evidence or whatever
nzpiefaceTue, August 16, 2022 at 06:38 AM
[I have evidence but I cbf finding it 
and the numbers weren't from me so just assume it's missing]
Echoes procs are either decided by the server, which is why ping affects it. The server does not know when the attacks happen on the client, so the timings for the stacks happen on the server, which is why stack gain/loss and the effect being active are all influenced by ping, as ping is the duration of time it requires for the client to send to server and server to client.
and the numbers weren't from me so just assume it's missing]Echoes procs are either decided by the server, which is why ping affects it. The server does not know when the attacks happen on the client, so the timings for the stacks happen on the server, which is why stack gain/loss and the effect being active are all influenced by ping, as ping is the duration of time it requires for the client to send to server and server to client.
basilissaofnutsWed, November 9, 2022 at 06:30 AM
wait what the fuck
this place was last active 2 months ago
TiBotWed, November 9, 2022 at 06:32 AM
Creating transcript...