Hybrid View

Previous Post Previous Post   Next Post Next Post
  1. Hit Chance Data Collection  - 11-23-2016 #1

    Offline
    Iron Fist Your Bum
    Administrator
    864
    318
    Sep 2016
    #
    10 Year Veteran
    Wondering if anyone can get this for me.

    Accounts needed:
    The first account needs nearly the same attack, ranged, and magic level.
    The second account needs as high a defence level as possible.

    Attacker gear:
    For melee, just an abyssal whip (82 accuracy).
    For ranged, just a karils crossbow (84 accuracy).
    For magic, 80-85 accuracy (such as ahrims, sotd, god cape, and barrows gloves), spell doesn't (shouldn't?) matter.

    Defender gear:
    Defence and magic level need to be close (magic level factors into defence)
    Basically need gear for each style that matches eachother in terms of defence bonus.
    Given magic defence is a bit harder to come by, you're likely looking at karils to defend against magic, and adamant to defend against melee and ranged. This will be at the very low 100s for defence against each combat style.
    If you can get your magic defence higher then do so but match your melee and ranged defence as closely as possible.

    Make sure to use the accurate attack style. Take down as many successful hits as possible. The more hits, the more accurate it'll be. Whatever you do, make sure you do the same number for each combat style and to include to total hits you did for each style. Also include both players levels!
    For example, 50 out of 100 range hits were successful is the type of data I need from this.

    What is this for? This will for starters give me some direct data I can use to compare to my hit formula and adjust to match the results. Secondly, it'll tell me if any combat style has hidden boosts applied.



    TLDR
    What's important when testing on OSRS:
    • Include both players combat skill levels, and accuracy and defence bonuses used
    • Attacker needs similar attack/ranged/magic levels
    • Attack accuracy needs to be close for all three combat styles
    • Defender needs similiar defence and magic levels
    • Defender defence bonus needs to be close for all three combat styles
    • Same number of hits tested for each style (50, 100, whatever you want to do)
    • Include number of hits successful over total number of hits tested for each style


  2. Offline
    Donator
    228
    250
    Oct 2016
    #
    With Even Attack/Defence Roles

    Melee (82 Slash Attack VS 82 Slash Defence W/Equal Attack/Defence Levels)

    Test 1: 43/100 hits
    Test 2: 48/100 hits
    Test 3: 52/100 hits
    Test 4: 49/100 hits
    Test 5: 48/100 hits

    So 240/500

    Range (84 Range Attack VS 84 Range Defence W/Equal Attack/Defence Levels)

    Test 1: 50/100
    Test 2: 47/100
    Test 3: 48/100
    Test 4: 47/100
    Test 5: 53/100

    So 245/500

    Magic (90 Magic Attack VS 90 Magic & 90 Defence, +64 attack bonus vs +64 defence bonus)

    Test 1: 52/100
    Test 2: 47/100
    Test 3: 49/100

    So 148/300


    1 Role 50% bigger/smaller than the other


    Melee/Range (Equal Attack VS Equal Defence, but 84 Attack/Range Vs 42 Defence )

    Test 1: 65/100
    Test 2: 68/100

    Melee/Range (Equal Attack VS Equal Defence, but 21 Attack/Range Vs 42 Defence )

    Test 1: 17/100
    Test 2: 20/100

    Magic (90 Magic Attack VS 90 Magic & 90 Defence, +64 attack bonus vs +32 defence bonus)

    Test 1: 70/100


    Conclusions:

    • Range/Melee/Magic all use the same hitChance formula since results seem to be pretty much identical.

    • When attack/defence roles are equal you're looking at a 50% hitchance (Which would kind of make sense anyway).


    hitChance Formula

    Assuming you are still using THIS FORMULA, this would achieve what the results seem to be showing:

    If (accuracy > defence) {
    hitChance = 1.0 - (defence * 0.5) / accuracy;
    } else {
    hitChance = (accuracy * 0.5) / defence
    }
    I still don't understand what "return (hitChance *32.0)" is.

    Extra:

    Not saying these currently don't work correctly, this is just to check over and confirm they definitely are.


    • Make sure you're calculating magic defence role correctly

      A = ((Defence Level*Prayer)+Stance) * 0.3 + (Magic Level * 0.7)
      B = Equipment Bonus

      Magic Defence Roll = A * (B+64)


    Formulas I used for calculating attack/defence roles found Here
    Last edited by Iloi5ia89; 11-24-2016 at 02:31 PM.


  3. Offline
    Iron Fist Your Bum
    Administrator
    864
    318
    Sep 2016
    #
    10 Year Veteran
    Seems our PvP combat isn't as inaccurate as you think. Maybe everyone is just too used to normally PKing on OSRS where you're not usually fighting someone with 99 defence?

    Equal bonuses
    249/500
    242/500
    241/500
    238/500
    245/500

    64 vs 32
    78/100
    76/100
    72/100
    75/100
    74/100


    This is nearly our current combat formula



  4. Offline
    Donator
    228
    250
    Oct 2016
    #
    Definitely not, stuff on here is way to tanky compared to osrs. Noteiceably more otherwise so many people wouldn't mention it >.> I just don't understand what's wrong since your formulas look fine.

    You shouldn't be calculating normal defence and magic defence the same for sure, magic defence should be:

    (Defence level × Prayer × Other + 8 + Attack Style Bonus)*0.3 + (Magic level * 0.7)
    Last edited by Iloi5ia89; 11-24-2016 at 05:33 PM.


  5. Offline
    Iron Fist Your Bum
    Administrator
    864
    318
    Sep 2016
    #
    10 Year Veteran
    Quote Originally Posted by Jessie J View Post
    Definitely not, stuff on here is way to tanky compared to osrs. Noteiceably more otherwise so many people wouldn't mention it >.> I just don't understand what's wrong since your formulas look fine.

    You shouldn't be calculating normal defence and magic defence the same for sure, magic defence should be:

    (Defence level × Prayer × Other + 8 + Attack Style Bonus)*0.3 + (Magic level * 0.7)
    It is, but when calculating for both at level 99, there's no need to do that so I didn't include it in my test run code. I'm saying out of 500 hits, my code averages 250 successful hits using 84 accuracy and 84 defence. If yu think defence is too tanky, try ~80 accuracy (for a whip) against atleast full rune if not higher, and I'll compare results.


  6. Offline
    Iron Fist Your Bum
    Administrator
    864
    318
    Sep 2016
    #
    10 Year Veteran
    Images of it all




Closed Thread
Results 1 to 6 of 6