1. Hello!

    First of all, welcome to MapleLegends! You are currently viewing the forums as a guest, so you can only view the first post of every topic. We highly recommend registering so you can be part of our community.

    By registering to our forums you can introduce yourself and make your first friends, talk in the shoutbox, contribute, and much more!

    This process only takes a few minutes and you can always decide to lurk even after!

    - MapleLegends Administration-
  2. Experiencing disconnecting after inserting your login info? Make sure you are on the latest MapleLegends version. The current latest version is found by clicking here.
    Dismiss Notice

Numerical analysis of HT dpm between hero and paladin

Discussion in 'General Discussion' started by wanpi12456, Jul 21, 2021.

  1. -ovv
    Online

    -ovv Horntail

    2,280
    904
    500
    Feb 23, 2020
    Male
    12:14 PM
    -ovv
    Beginner
    200
    Honor
    The difference is that you're giving these various situations all equal weighting when there are external factors you can utilize to influence the cancel rotations and timings. We've tested this in several runs and it has worked for us fairly well.
     
  2. OP
    OP
    wanpi12456
    Offline

    wanpi12456 Timer

    105
    13
    120
    May 19, 2019
    Male
    Taiwan
    3:14 AM
    PrEP
    Paladin
    191
    Beaters
    Yes I mean, you can decide the time point of 2nd head casting cancel.
    But the data from my post shows that time point do 0 influence to cancel situation.

    Another thing is, 2nd head has independent chance for 50%mc or 50%wc.
    This is the thing that you cannot control.
    As a result, whether you control time point or not, it goes to same results in average.
     
  3. -ovv
    Online

    -ovv Horntail

    2,280
    904
    500
    Feb 23, 2020
    Male
    12:14 PM
    -ovv
    Beginner
    200
    Honor
    Assuming that groups will eventually figure out how to control cancels into favorable conditions, you'd have to basically compare the two favorable conditions for either group:
    • Favorable condition for cleave - both heads are cancelled in sync, so you have ~1 minute of free fire damage on both heads until both are cancelled, to which you either have to swap sides to do ~10 seconds of damage on a single target head before swapping back.
    • Favorable condition for paladins - heads on one side alternate between cancels, meaning as soon as one head starts wc, the other head finishes wc. This allows you to optimize by dealing damage on heads without much wasted time crossing body or hitting wing.
    The thing is, favorable conditions for cleave is generally an unfavorable condition for single-target attackers, which means unless a majority of your party composition consists of cleave, the rest of your group will suffer due to having unfavorable conditions for single-target.

    I might be biased because I run with mostly single-target attackers, and we bring a hero once a week. In the runs with the hero, we still optimize our cancel rotations to favor single-target, so all I see the hero doing is running around to avoid cancels while our paladin and other single-target attackers stay on one side.
     
    • Agree Agree x 1
  4. OP
    OP
    wanpi12456
    Offline

    wanpi12456 Timer

    105
    13
    120
    May 19, 2019
    Male
    Taiwan
    3:14 AM
    PrEP
    Paladin
    191
    Beaters
    Let me reply seperately to accurate the concept i'm saying.
    1. Everyone cannot control cancel into favorable conditions.
    2. Because 2nd head's cancel is random, wc or mc.

    1. Before we talk about Favorable condition for cleave and for range, we should look at cancel situation (aka 2heads' cancel mechanism).
    2. Favorable condition for cleave = double wc + double mc = 1/2
    3. Favorable condition for range = 1wc + 1mc = 1/2
    投影片8.PNG

    1. Whichever "time point" is choosen, the results should contain a combination of "lucky(50%) + unlucky(50%)"
    2. 2nd head's cancel cannot be controlled as it is random.
     
  5. -ovv
    Online

    -ovv Horntail

    2,280
    904
    500
    Feb 23, 2020
    Male
    12:14 PM
    -ovv
    Beginner
    200
    Honor
    I’ll just leave you to play test it yourself, but there is a fairly straightforward way to manipulate cancels staying synced or not.
     
    • Agree Agree x 1
  6. xiaoyaoz
    Offline

    xiaoyaoz Balance Team Staff Member Balance Team

    194
    129
    196
    Jul 17, 2020
    Male
    8:14 PM
    I think his assumption is about taking consideration of both favourable and unfavourable conditions in the overall picture (since you can't really control how the head cancels), which may not be the best representation of the real situation, but gives a fair comparison between the different scenarios?

    I'm not sure what is this fairly straightforward way of manipulating cancels but from what I gather there are:
    1) having different starting point
    2) increases head available skills to cast (by focusing head until below certain threshold so that it has more summons skill or dispel skill)

    The first one basically is a roll of dice and hope that after u hit (for example) left head and wait 45 secs before you hit mid head that their cancels are different, but tbh, that's the same as hitting both head the same time and just hope they are different?

    The second one is kinda something to desync it, since if one head has more skills to cast, it may have a longer time not immediately casting cancels once it's off cd, which causes the cancels to be less consistent as before, but it's also something you can't really control and may end up favour/unfavouring you. If i'm not wrong, this is the way that people try to do when the first method turns out unfavourable to them while in the case the first method is favourable, people will spread the damage as much on the head to maintain same amount of skills to avoid desync? Not sure if that's considered controllable since it takes quite some time to desync when method 1 is unfavourable?

    Is there other ways to manipulate cancels that you mind sharing that is more controllable? :eek:
     
  7. OP
    OP
    wanpi12456
    Offline

    wanpi12456 Timer

    105
    13
    120
    May 19, 2019
    Male
    Taiwan
    3:14 AM
    PrEP
    Paladin
    191
    Beaters
    Most of the words above are correct.
    For futher explanation of "which may not be the best representation of the real situation".
    We should keep in mind that:
    "there is 50% chance to you to get the best condition, but there is also 50% chance to you to get the worst condition."
    (In other words, It is all about 2 heads' Cancel mechanism.)
    Another reason is, you cannot control 2nd head cancel.

    1) having different starting point
    If you are talking about different starting point, here is the pic.
    Please look at 2 yellow circles.
    it shows that whatever you choose the time point for 2nd head's cancel, they have same results.
    Keep in mind: "you have 1/2 chance to get better cancel condition for range, but you have also 1/2 chance to get worse cancel condition for range."
    投影片9.PNG

    2) increases head available skills to cast (by focusing head until below certain threshold so that it has more summons skill or dispel skill)
    This is totally discussing different directions.
    When you adjust the total time of wc+mc, it leads to the results in Part4.
    → As the total time of wc+mc increases (0%~100%), a single-target attacker get (0~50%) benefits.

    I will not further explan the following 2 concepts:
    (1) how does total time of wc+mc affect 2 heads' Cancel mechansim
    (2) how does the 3 heads' Cancel mechanism affect range and cleave
    Until people 100% understand what is the 2 head's Cancel mechanism.

    Edit:
    I mean : before we expand the assumption/results to 3 heads, we should know the 2 heads' Cancel mechanism first.
     
  8. -ovv
    Online

    -ovv Horntail

    2,280
    904
    500
    Feb 23, 2020
    Male
    12:14 PM
    -ovv
    Beginner
    200
    Honor
    I won't comment any further on this as I'd rather have the community discover this process on their own, and I like that different groups are workshopping different ideas. However I'll add that you're almost there with #2. Just to add - we've noticed our runs have gotten noticeably faster since employing this strategy, but we don't have enough run data to distinguish if it's just a series of luck or if our strategy is working as intended. Nonetheless, we've had consistent results thus far and will continue to run this way until results show otherwise.
     
  9. OP
    OP
    wanpi12456
    Offline

    wanpi12456 Timer

    105
    13
    120
    May 19, 2019
    Male
    Taiwan
    3:14 AM
    PrEP
    Paladin
    191
    Beaters
    Update:
    1. Delete original summary in Part4.
    2. Add a more appropriate summary in Part4.
    3. re-write (8-4).
     
  10. OP
    OP
    wanpi12456
    Offline

    wanpi12456 Timer

    105
    13
    120
    May 19, 2019
    Male
    Taiwan
    3:14 AM
    PrEP
    Paladin
    191
    Beaters
    More explanation about Best or Worst case for pure range HT squad.
    ---
    For range attacker, range favor the cancel situation with "1mc 1wc (rotate)",
    which look like this:
    投影片10.PNG
    ---
    But, because you cannot control 2nd head's cancel, it might has 50% chance fail to the double wc like this:
    投影片11.PNG
    ---
    When we are talking about expected value, we have to consider the chance for each case.
    So we have to multiply each case by 1/2 chance.
    It will look like this:
    投影片12.PNG
    When the total time of mc+wc = 100%, this result can be applied to other "different time point" cases.

    warning
    This is when we discuss in total time of wc+mc = 100%.
     

Share This Page