Thread: Find Weakness

  1. #1

    Find Weakness

    I've been trying to find out the % damage increase FW provides, but testing it on a dummy is a pain, to say the least. Best i could come up with is ~23% on top of every special/auto-attack, while FW is up, but the low no. of iterations means there is a rather substantial error margin.

    Apparently there is no up to date portal/guide that explains armor mitigation mechanic, or if it is, I haven't found it. I presume that the physical damage mitigation doesn't scale exactly linear with armor rating, there has to be some form of DR tied to it. Perhaps a better question is does FW reduce the direct physical mitigation from armour, or just the armor rating?

    Is there anyone with the theorycrafting skills to lend a hand with this? I suspect Simcraft would provide the answer rather quickly, but I lack the know-how to set up a simulation, and doing dummy trials is an option I would rather avoid

  2. #2
    I wouldn't have thought it would vary too much. get a few specials in FW, few w/o FW, pop prep rinse repeat. etc.

    Should give you a good idea if you check through the combat log.

  3. #3
    Quote Originally Posted by Impa View Post
    Apparently there is no up to date portal/guide that explains armor mitigation mechanic, or if it is, I haven't found it. I presume that the physical damage mitigation doesn't scale exactly linear with armor rating, there has to be some form of DR tied to it. Perhaps a better question is does FW reduce the direct physical mitigation from armour, or just the armor rating?
    If physical damage mitigation with armor were linear, effective health would be a vertical asymptote (meaning much steeper than exponential). In order to keep effective health linear with armor, armor has diminishing returns in terms of mitigation. Armor penetration removes armor from the top, making it the reverse of armor mitigation, so it has increasing returns.

    70% reduced armor does not translate to 70% increased damage. In fact, raid bosses at level 90 have approximately 30% damage reduction from armor so 100% armor mitigation (ie, shadow blades) is about a 42% increase in damage. The last 30% from FW to shadow blades is probably about as effective as (if not moreso than) the 58% additional armor ignore from sunder armor to FW.
    Last edited by shadowboy; 2013-01-16 at 05:43 AM.

  4. #4
    I am Murloc! Ryme's Avatar
    Join Date
    Jul 2009
    Location
    In a field, somewhere
    Posts
    5,919
    Running sims, I'm getting an increase of about 38% to backstab and 67% increase to eviscerate. Sample size is poor though since you have to pull some work arounds in simc to "simulate an environment where FW is always up. I'll try and improve the model.
    I am the lucid dream
    Uulwi ifis halahs gag erh'ongg w'ssh


  5. #5
    Quote Originally Posted by Ryme View Post
    Running sims, I'm getting an increase of about 38% to backstab and 67% increase to eviscerate. Sample size is poor though since you have to pull some work arounds in simc to "simulate an environment where FW is always up. I'll try and improve the model.
    The damage increase to backstab and eviscerate should be the same, so something is wrong with your methodology. The armor provides a relative damage change so 38% to BS should yield 38% to eviscerate, too.

    In my quick simcraft test it shows a ~35% increase to both backstab and eviscerate damage. Actual result was ~32% for BS and 38% for eviscerate, but the standard error of the mean is pretty high due to small sample sizes. For example, in my FW test backstab was only used 31 times per iteration compared to the 172 times it's cast during the simulation with no find weakness uptime at all.

    The long and short of it is this: The damage contribution that find weakness provides is not easily quantifiable. It all depends on the current armor of your target, so sunder armor reduces the damage contribution of FW. Also FW will do less on targets with less armor.
    Last edited by shadowboy; 2013-01-16 at 01:12 PM.

  6. #6
    I am Murloc! Ryme's Avatar
    Join Date
    Jul 2009
    Location
    In a field, somewhere
    Posts
    5,919
    Heh, after searching through code, asking my personal god, pleading with my personal demons I noticed I read my output wrong
    I am the lucid dream
    Uulwi ifis halahs gag erh'ongg w'ssh


  7. #7
    Well, there's always that

  8. #8
    cheers for the help guys.

    i was under the impression that raid bosses have a standard amount of armor rating, so the question would probably boil down to:

    1. dps increase without sunder

    2. dps increase with sunder

    ..with both having a precise value.

    That being said, I'm happy with at least having a better ball-park figure, probabil somewhere around 35%, without sunder?


    One last question - although I wouldn't mind carrying the discussion further and refining the results - do you happen to know a good source/portal that explains mitigarion vs. armor stacking for MoP?

  9. #9
    The 35% is with sunder. The best way to test is to repeat the above experiment with and without the sunder debuff.

    If you are capable of running simcraft, just basically do this:

    Test backstab / eviscerate damage without find weakness by commenting out every line with ambush, then noting the damage-per-execute of backstab. You could get away with a 10,000 iteration simulation to get good values for this.

    To test backstab / eviscerate damage during find weakness, find every backstab/evis line in the priority list and add &target.debuff.find_weakness.up to the end of it. Uncomment the ambush lines so you actually get the FW debuff. You will want to use the 50,000 iteration option for this since the sim will only backstab/evis during FW so the counts will be very low.

    Do both of those steps with and without sunder debuff to find out the effect sunder has on debuff.

    And raid bosses generally have a standard armor level, but adds may or may not and plus adds are generally not level 93 and the physical mitigation by armor DOES depend on the relative level between attacker and attackee (meaning 20,000 armor on a level 93 target is not the same mitigation as 20,000 armor on a level 92 target).
    Last edited by shadowboy; 2013-01-16 at 07:22 PM.

  10. #10
    Why is this being done in such a convoluted way?

    As 90, attacking a 93: 46257.5 / (24835. + 46257.5) is base armor.
    Multiply the "24835." by 0.3 for Find Weakness.
    Multiply by 0.88 for Sunders.
    Or multiply by 0.264 for both Sunder and FW.

    Divide the resulting values by each other to get the relative changes (20% increase, 28%, whatever).

    https://github.com/dazer/ShadowCraft-Engine/wiki/Armor

  11. #11
    Quote Originally Posted by Pathal View Post
    Why is this being done in such a convoluted way?

    As 90, attacking a 93: 46257.5 / (24835. + 46257.5) is base armor.
    Multiply the "24835." by 0.3 for Find Weakness.
    Multiply by 0.88 for Sunders.
    Or multiply by 0.264 for both Sunder and FW.

    Divide the resulting values by each other to get the relative changes (20% increase, 28%, whatever).

    https://github.com/dazer/ShadowCraft-Engine/wiki/Armor
    I was doing it in a convoluted way because I don't remember the armor constants :P

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •