1. This site uses cookies. By continuing to use this site, you are agreeing to our use of cookies. Learn More.
  2. Discord Support team Facebook l2central.info

Experiment 182 Ultimate Servitor Share

Discussion in 'Archived experiments' started by Palkah, Dec 22, 2018.

  1. Palkah

    Palkah User

    Joined:
    25.08.15
    Messages:
    502
    Likes Received:
    295
    [​IMG]
    Goal: to understand how Ultimate Servitor Share transfers the P. Critical Damage stat to the Servitors, so that it's possible to say which SA and Augment (P. Atk or P. Critical Damage) is better for a Wynn Summoner in PvE.

    This experiment was inspired by a player request:
    During the experiment, a Wynn Arcana Lord 110 was using https://l2wiki.com/Summon_Commando_Cat_-_Wynn_Arcana_Lord to attack an Aeore Cardinal 110. The passive skill https://l2wiki.com/Superior_Weapon_Mastery_-_Wynn_Arcana_Lord has been learned.

    Here are the results:
    [​IMG]

    Let's draw some conclusions.

    Test 1: This is a point of reference. There are no buffs on Master or Servitor. "Total P. Atk" means the values displayed in character windows.

    Test 2: Here, the https://l2wiki.com/Ultimate_Servitor_Share_-_Wynn_Arcana_Lord Level 4 (max) is used. The buff says that it transfers 70% of Master's P. Atk to the Servitor. Let's check it:
    8698[servitor] + 0.7*4732[master] = 12010.4.
    The mistake of 1 P. Atk point is caused by the game: in reality, all of the character's statistics are Floats, they just get rounded for easier display. So far, so good.
    Conclusion: 70% of Master's Total P. Atk is added to the Servitor's P. Atk.

    Test 3: Servitor was buffed with https://l2wiki.com/Mass_Servitor_Might_-_Wynn_Arcana_Lord to see if buffs increase the Servitor's Total P. Atk, or the P. Atk before the transfer.
    Option 1: 12011*1.17 = 14053 -> makes no sense
    Option 2: 8698*1.17 + 0.7*4732 = 13489.06 -> confirmed.
    The transferred P. Atk is probably treated as Static P. Atk on the Servitor (we'll get to that later).
    Conclusion: any buffs increasing Servitor's P. Atk are applied first, and the transferred P. Atk is added later. This is true for all stats transferred by Ultimate Servitor Share.

    Test 4: Master was buffed with https://l2wiki.com/Horn_Melody_-_Iss_Hierophant Level 4 (28% P. Atk). Let's see how his P. Atk increased:
    4732*1.28 = 6057. Huh. The ingame result was 5613 O_O. This is because character's P. Atk is made of two parts:
    1. Static P. Atk - this is a sum of every item/passive that says it gives an exact value of P. Atk (+1000, +222 etc). The Wynn Summoner had 1584 Static P. Atk from Max Level of https://l2wiki.com/Superior_Weapon_Mastery_-_Wynn_Arcana_Lord
    2. Non-static P. Atk - this is base character P. Atk (depends on the weapon, level, and STR) increased by all the passives and percentage buffs. If a buff says it increases P. Atk by 20%, then it increases this part only.
    These two parts are added to make the Total P. Atk visible in character window.
    Quick calculation for Master:
    Non-static P. Atk = Total P. Atk - Static P. Atk = 4732-1584 = 3148.
    Applying the 28% buff:
    3148*1.28 = 4029.44
    Adding back the Static P. Atk to get Total P. Atk:
    4029.44 + 1584 = 5613.44 -> same as the measured result.
    Conclusion: any buff/item increasing P. Atk by a percentage increases only the Non-static portion of character's P. Atk.

    Test 5: Stats as in Test 2, but https://l2wiki.com/Baium's_Ring was added for 15% P. Critical Damage. The USS transfers 15% of character's P. Critical Damage. Logically, this would mean that if we add 15% to the Master, 0.15*15%=2.25% is transferred to the Servitor.
    Servitor's Critical Hit dealt 4151 damage in Test 2, and 4244 in Test 5. So, how much was the damage increased?
    (4244-4151)/4151 * 100% = 2.24%. The difference is again due to Floats rounding, otherwise it looks okay.
    Conclusion: none yet.

    Test 6: This time, two effects of 15% P. Critical Damage were added. If we use results from Test 5, we could assume that it will increase the damage by 2.25% twice, so 1.0225*1.0225=1.0455 -> 4.55%.
    Let's see:
    (4352-4151)/4151 * 100% = 4.84% This means that the damage has increased more than expected.
    [​IMG]
    Turns out, the Koreans from NCSoft used the easiest way for P. Critical Damage transfer. First, the P. Critical Damage bonus is calculated for the Master, and then the 15% of that total value is transferred to the servitor. This means that the more P. Critical Damage effects the Master has, the more his stats will increase with a new P. Critical Damage effect, causing bigger bonus transferred to the Servitor. Let's check if it works:
    Total P. Critical Damage on Master:
    1.15*1.15=1.3225 -> so the increase is 32.25%
    Transferred value:
    0.15*32.25% = 4.8375% -> same as the result from live test.
    Conclusion: P. Critical Damage from all of the Master's items/buffs is calculated on the Master, and then 15% of that is transferred to the Servitor via USS.

    The rest of the tests were done to confirm these conclusions, so let's skip analyzing them.

    Answering the question
    The question was: "Which SA and Augment (P. Atk or P. Critical Damage) gives more damage to the Servitor?"
    We've learned few things:
    1. The more P. Critical Damage on Master, the more P. Critical Damage is given by the next item, and bigger bonus is transferred to the Servitor.
    2. Buffs to increase Master's P. Atk only increase his Non-static P. Atk, so the more Static P. Atk we have, the less Total P. Atk a new buff gives us.


    The answer to the question is: it depends. Using the conclusions from this experiment, I have created a calculator which should tell you what is better for your setup. If you are too lazy even for that, you can use a general rule of thumb: if you're stacked, P. Critical Damage is probably much better than P. Atk.

    The calculator can be found here on the tab "Wynn SAs and Augments"


    This is what the calculator looks like:
    [​IMG]

    You can use it on any setup -> no buffs/full buffs/everything else. This is because the P. Atk buffs effectiveness is only dependent on ratios between Static P. Atk/Total P. Atk and Total P. Atk / Total Servitor P. Atk. I require information about your current Might SA and P. Atk Augment in older to "remove them" from your stats, so that the SA/Augment effectiveness can be always calculated for Wynn with no SA/Augment. Static P. Atk is something you have to calculate manually from your passives and items, there is a tool-tip with ideas where to look for them.
    The final section is "Critical Damage gear" which is required to calculate effectiveness of P. Critical Damage SAs and Augments. If I forgot any effect (except SA/Augment) available to Wynn Summoner (the Master) that increases P. Critical Damage, please let me know and I'll add it.

    After filling in the data, you will get multiple tables like these:
    [​IMG]

    The percentages in the tables mean exactly "how much bigger will my Servitor's damage be if I add this SA/Augment".
     
    Last edited: Dec 22, 2018
    SnoWyz, Oscar, 21Sav and 1 other person like this.
  2. Captive7

    Captive7 User

    Joined:
    26.07.14
    Messages:
    177
    Likes Received:
    37
    Keep going with this good work friend ;)
     
    SnoWyz and Palkah like this.
  3. giann0z

    giann0z User

    Joined:
    20.08.18
    Messages:
    6
    Likes Received:
    1
    That's art!
    Very nice work.....!
    Thanks a ton ;)!

    Concerning the interpretation of results based on the values, Patk still seems more robust, imho. Ppl have to take into account that a critical hit occurs at 50% probability (at best), so w/e bonuses u get from critical dmg they are applied only half of the time.
    So if u have to chose between the 2 patk seems better to my eyes and i think gives better dps output.
     
    Last edited: Jan 6, 2019
  4. Palkah

    Palkah User

    Joined:
    25.08.15
    Messages:
    502
    Likes Received:
    295
    Even if critical hit happens 50% of the time, the damage from that crit is multiple times bigger than a regular hit. On a Titan it's easy to achieve a situation where critical hit deals 40x more damage than a regular attack (and that's without all the items), so it should be achievable by the Servitor too, despite less amount of possible effects for Critical Damage.
     
  5. zauliuz

    zauliuz User

    Joined:
    26.03.14
    Messages:
    2,852
    Likes Received:
    442
    summoners will be way stronger in next update where they can use their dots in transform @Palkah maybe you can explain, or test or something, how works that sharing equipment. It's only for def stats for pets, or it also transfers something else, like pve damage from pve weapon, and other stuff.
     
  6. Palkah

    Palkah User

    Joined:
    25.08.15
    Messages:
    502
    Likes Received:
    295
    AFAIK https://l2wiki.com/Sharing_Equipment_-_Wynn_Spectral_Master transfers only the Dark/Bloody option of the weapon - so 15% PvE/PvP Damage. Last time I checked it didn't transfer 5% PvP bonus from SA, or any PvE/PvP bonus from a belt. I wanted to test it properly but everybody is on vacation xD
     
  7. zauliuz

    zauliuz User

    Joined:
    26.03.14
    Messages:
    2,852
    Likes Received:
    442
    and what about crit dmg rins? they also not work on pets? :(
     
  8. Palkah

    Palkah User

    Joined:
    25.08.15
    Messages:
    502
    Likes Received:
    295
    Nop, only the 15% from USS.
     
  9. giann0z

    giann0z User

    Joined:
    20.08.18
    Messages:
    6
    Likes Received:
    1
    I'm gonna say it once more...this is a fantastic test!
    I've been longing for something like that for a long time (I should have asked earlier :D)!

    I tried your calculator on my stats and i reached the following conclusions:
    1. Patk build is still better overall and imho should be prioritized in one way choice (like weapon augment)
    2. Critical dmg bonuses add a surprising amount of dmg boost (you always have to keep in mind that the boost is applied in 50-60% of the cases) and it surely adds significantly to servitors' total dps (i tend to believe the opposite bfr this test :D).
    3. Critical dmg scales very nice as @Palkah mentioned but in a normal situation where u don't have POM+SOS+GOG+CAT+WAR2, patk still has the lead. For extensive RB killing (where u can get all the aforementioned) critical dmg build seems to have the edge.

    [​IMG]
     
    Palkah likes this.
  10. iKamus

    iKamus User

    Joined:
    12.01.19
    Messages:
    2
    Likes Received:
    0
  11. Palkah

    Palkah User

    Joined:
    25.08.15
    Messages:
    502
    Likes Received:
    295
    If you want to increase something by 10%, you multiply that something by (1+percentage/100). Example:

    200 increased by 10% should be 220.

    200 * (1 + 10/100) = 200 * 1.1 = 220
     
  12. iKamus

    iKamus User

    Joined:
    12.01.19
    Messages:
    2
    Likes Received:
    0
    oh i see, and this work for all stats? like m.def, p.def , hp...
     
  13. Palkah

    Palkah User

    Joined:
    25.08.15
    Messages:
    502
    Likes Received:
    295
    yup, for the non-static parts at least
     
  14. giann0z

    giann0z User

    Joined:
    20.08.18
    Messages:
    6
    Likes Received:
    1
    @Palkah
    Link is dead. Is it possible we have it online again cause it's very usefull?
    Thanks in advance.

    BR,
    G.