Guild Wars 2

Check “Conclusion” below to skip straight to the goods

Hello world!

So, I have a new idea on how Magic Find % works that has yielded much more accurate and predictable results than other theories flying around. You’ve all probably heard the discussions flying around Lion’s Arch and guild chats about how Magic Find % works. People say “it stacks” like little blocks on top of each other with the player needing to do nothing else other than measure the height of the pile when they are done with that stacking. I’m not sure about you, but this stacking idea doesn’t work for me. Before I address this though, I will talk about another commonly miscalculated statistic that may help to understand where I will come from later on:

Critical DMG %

Okay Berserkers, it’s your time to shine. You have all noticed the little number in the Hero’s screen underneath critical chance… that arbitrary little critical damage number that has been slowly rising throughout your progression to level 80. The game doesn’t do much to explain this number, so you may finally have your nice exotic set for endgame mayhem and interpret that “99” or whatevever it is as 99 more damage. Some may say it’s 99% more damage than before. I say it’s neither, and here’s why.

Being that this is a forum, I will skip the long table of base damage and critical damage hits I compiled back home. Rather, I’ll tell you that I did averaged about 400 damage per regular hit against medium armoured enemies and 335 per hit against heavily armoured enemies in the level 80 Orr area. My critical hits for these same enemies were, respectively, averages of 1050 and 840 (rounded because I’m not actually home at the moment while writing this).

So let’s do some quick calculations:

We assume that (Critical Damage)=(Base Damage)*(Crit Multiplier)… This is not true, but I will offer my explanation towards the bottom of this post. So with the 99% stacking theory of 99 critical dmg being 99% more damage than usual, we get

For heavy armoured enemy: Critical Damage=335*1.99=667
For lighter armoured enemy: Critical Damage=400*1.99=798

You may notice that these are considerably less than the critical damage values I mentioned above. So what’s the actual (Crit Multiplier)?

Medium Armour (Crit Multiplier) = 1050/400 = 2.63 (Approx)
Heavy Armour (Crit Multiplier) = 840/335 = 2.51 (Approx)

So this lead me to believe that it’s a logarithmic process of calculating the (Crit Multiplier) instead of a stacking one. Here’s the three ideas I came up with:

Idea 1: The bonus s multiply in increments as shown on armour
Idea 2: The bonus %s multiply in increments as shown on armour AND in traits
Idea 3: The bonus %s mutiply in 1 increments regardless of trait and armour increments
Idea 4: The critical multiplier is based on a continuous logarithmic progression

To test these ideas, I will list where my crit dmg % increments come from: +8% runes (3% and 5%), +18% from jewel upgrades (3% * 6), +5% from amulet, +1% from backpack, +6% from earrings (3% * 2), +6% from rings (3% * 2), +2% (boots), +2% (gloves), +2% (helm), +2% (shoulders), +3% (legs), +5% (chest), +30% (traits, 1% each), the rest from precision. This gives us:
31*1%, 4*2%, 12*3%, 2*5%, 14% from precision bonuses

And now to test these multiple ideas:

Idea 1 (Crit Multiplier): (1.01)(1.02)^2(1.03)^12(1.05)^2(1.14)(1.30)=2.45
Idea 2 (Crit Multiplier): (1.01)^31(1.02)^2(1.03)^12(1.05)^2(1.14)=2.54
Idea 3 (Crit Multiplier): (1.01)^99=2.68

You’ll notice that we’re getting closer and closer to the tested (Crit Multiplier) values. Notice that the multiplier is slightly less for heavy armoured enemies and more for medium armoured enemies. Since I never got a chance to test on light armour enemies, this leads me to believe that it’s even higher for those (higher than 2.68). Here’s where idea 4 comes in.

Idea 4 (Crit Multiplier):
P=Ae^(rt); where P=crit damage, A=base damage, r is the exponential constant, and t=crit dmg %

To be honest, I cannot make calculations for this idea yet because I don’t have trials at different crit dmg %s. I may or may not get back to this later. However, this is the one that I believe to be most correct and consistant with the fact that higher enemy toughness will impact your Crit Multiplier.

I’m hungry – should I get pizza or tacos for lunch today?

Amos.8760:

I’m hungry – should I get pizza or tacos for lunch today?

Tacos. Always.

I was thinking that too. Sorry to derail, but I was trying to read all of your post on a tacoless stomach. Wasn’t working. I’m going to go eat, regroup, gather my thoughts, and then come at this again later. Cause I really do want to learn more about this stuff.

Amos.8760:

I was thinking that too. Sorry to derail, but I was trying to read all of your post on a tacoless stomach. Wasn’t working. I’m going to go eat, regroup, gather my thoughts, and then come at this again later. Cause I really do want to learn more about this stuff.

Feel free to skip it all and just look at the conclusion. That’s the only part that really matters if you want something you can actually relate with. The rest is just support to show where the equations actually came from.

Ran out and got some tacos. Mexican place just down the street from my office.

Now – I have been assuming magic find was something along these lines:

If I had an item or rune granting +3% magic find, I assumed I was getting this:
(Base Magic Find Chance)1.03 = Actual Magic Find Chance

Do you know what the base magic find chance is?

I would say that that’s close enough to the actual chance since you are low on MF overall. However, if you started trying to calculate it like that with much higher MF ability and many more stacks, then the difference between having something like 100 magic find and 200 magic find wouldn’t be 100% the original magic find but rather closer to 460%.

The point I’m trying to make is that though the game doesn’t explicitly state this anywhere, the drop rates seem to suggest otherwise.

Seems legit. I’ve not played a game with magic find in it before, so this is new to me. At first I assumed the base chance was low, and that I would only be adding small percentages to it. However, I’m only in my 40’s right now, and I’m getting buffs with higher percentage than I expected. I also thought this only applied to gear drops, but are you saying these also apply to blue crafting mats?

I’ve read kinda fast, but tried to grasp what your idea was.

I’m not testing right now, and I will when I can, but I think your theory is flawed.
You’re assuming the Critical Damage is ONLY what shows in your character sheet.

Most games in the past used the Critical Damage to always 2x the damage. Then, in MMOs, they decided it was too strong and a trend started with 1.5 the damage.

So a 1000 hit would crit 1500.
Assuming the base crit damage to be 1.5, you’ll have:

Heavy Armor 335 × 1.5 = 502

If you add your Critical Damage stats from gear, since they’re multiplicative on the damage, I’ll add the values (or you’d get some really insane numbers)

Heavy Armor: 335 x (1.5 + 0.99) = 837

Closer to what you showed at first.

Bah, math. I don’t give it much thought, I just hit stuff till it dies, and when a magic item drops I say “oooooh, shiny!” Then I sell it on the TP because I keep getting things that I can’t use.

For the Magic Find one, I’ve seen a couple threads and posts about it, and some testing data posted on reddit.

From what I could see, MF doesn’t increase the actual drops you have (if it does, it’s minimal), but instead, it increases the rarity AMOUNT of the items you get.

For example, if you were to play for 1 hour and get 100 items with 0% MF, you’d get the same 100 items with 100% MF in the same hour.
The difference is the quality of these items.
Instead of having a 1% chance to get a Rare Drop, you’d have 2% (with 100% MF) to have it. And so on down the rarity table, so you’d have less junk and more useful items.

That’s my theory though, from what I’ve experienced (playing with MF sets and not) and from the testing data I’ve seen.

You are absolutely right and duly noted.

However, it still only applies to heavy armour whereas medium and light will still have higher than calculated modifiers. So yeah, I can already trash that equation as being the right one (it’s still a good approximation though).

Still it leaves the question as to why lighter armoury are still taking higher than calculated damage.

50 pieces of data is nowhere near enough to realistically establish a trend.

Tiny data sets simply don’t work. If I tossed a coin 3 times and it came up heads each time, should I conclude that there’s a 100% chance of it coming up heads or should I do many more coin tosses?

If you want to do more accurate calculations, go to the pvp mist area and get some STEADY weapons from the second tab (i think) on the weapon vendor…. They offer no damage range and hit for low amounts…. this will give you some more accurate numbers to deal with instead of dealing with damage range on other weapons….

As far as I know… there is no difference between light/med/heavy armor in dmg calculation other than the simple fact that lighter armor has less armor on it… they don’t follow any special rules…

The only set with 50 samples was the one on the Reddit post. I collected hundreds. Thank you for the reminding me about the basics of sampling though

Oxnan.2459:

As far as I know… there is no difference between light/med/heavy armor in dmg calculation other than the simple fact that lighter armor has less armor on it… they don’t follow any special rules…

Agreed:

(P + M * 35) * WS * SC / (T + D) = Total Damage

P = Power, M = Might Stacks, WS = Weapon Strength, SC = Skill Coefficient, T = Toughness, D = defense

Source: http://www.guildwars2guru.com/topic/60838-math-damage-reduction-toughness-and-vitality/

Oxnan.2459:

If you want to do more accurate calculations, go to the pvp mist area and get some STEADY weapons from the second tab (i think) on the weapon vendor…. They offer no damage range and hit for low amounts…. this will give you some more accurate numbers to deal with instead of dealing with damage range on other weapons….

As far as I know… there is no difference between light/med/heavy armor in dmg calculation other than the simple fact that lighter armor has less armor on it… they don’t follow any special rules…

I phrased it wrong. It’s not the armour itself that affects damage received but rather the toughness rating. Toughness does not remove a set amount of damage per hit but rather a percentage, so yes it does follow special rules that will lessen the effectiveness of a critical.

Did you find anything else on critical ccalculations as well Jestunhi?

check this out i went and purchased 6 loto tickets no winner on all 6
that means there is a 100% chance to lose at loto
hmm should of thought of that before i spent my money

For Faux Sake.8457:

Did you find anything else on critical ccalculations as well Jestunhi?

Nope, sorry. I happened across that link when looking into how to balance toughness vs vitality, but I don’t see anything on crit damage on that forum.

Well the thing you should be looking at is ttk vs mf

If you are sporting 200 mf how much slower are you killing mobs then in your full reg armor and just using food
I think you’d find that lets say in cursed shore time 100 mobs with and w/o magic find

In essence what the theory is having killed more mobs also increases the opportunity to get rarer drops

Not just time to kill, but damage too.

As the quality of loot is effected by how much of the damage you did to a mob (or so I read, should be simple enough to test by equipping a lvl 1 weapon and comparing with a level-appropriate weapon – but for much more than 50 kills!)

Dang. Id accept the crit=base(1.5+crit dmg %) if it was actually like that consistently.

For Faux Sake.8457:

Dang. Id accept the crit=base(1.5+crit dmg %) if it was actually like that consistently.

At a chance, based on critical chance (player statistic) and other modifiers (like the Fury boon), a normal hit may become a critical hit, which results in a damage multiplier of minimum ×1.5 (plus Prowess). It is currently unknown if there is a cap to this damage multiplier.

http://wiki.guildwars2.com/wiki/Damage#...from_critical_hits

So:
Damage * (1.5 + Prowess) = Crit Damage

:edit:

Not sure exactly how prowess is stored, whether it’s a decimal, a percentage, or a larger figure which needs to be divided first, etc.

Jestunhi.7429:

For Faux Sake.8457:

Dang. Id accept the crit=base(1.5+crit dmg %) if it was actually like that consistently.

At a chance, based on critical chance (player statistic) and other modifiers (like the Fury boon), a normal hit may become a critical hit, which results in a damage multiplier of minimum ×1.5 (plus Prowess). It is currently unknown if there is a cap to this damage multiplier.

http://wiki.guildwars2.com/wiki/Damage#...from_critical_hits

So:
Damage * (1.5 + Prowess) = Crit Damage

:edit:

Not sure exactly how prowess is stored, whether it’s a decimal, a percentage, or a larger figure which needs to be divided first, etc.

I realize this is what it says on the wiki, and it holds perfectly true for enemies with higher toughness and such. All you need to do is attack a dummy or something with lower armour and you will see immediately that the damage does not exactly fit that equation (if you are assuming crit dmg % to be prowess). Just not understanding where the extra damage is coming from is all.

And to everyone else, I’ll go ahead and edit this thread now that I’m on a computer. Jestunhi and others are right that critical damage is not exponential.

I still do believe magic finding to be exponential though. There have been so many trials with various magic finds that I’ve looked at or done myself that it seems next to impossible to be coincidence after all this time.

This thread is useful and relevant to my interests. Keep up the good work!

Also, I just read your post mangarrage…

I don’t really have the willpower to test something like that, but it’s definitely a cool thing to look into. I know I’d be interested in what that equilibrium between power and magic would be.

Also, for your MF testing, not sure if you know this, but there is anti-farming code in the game that I don’t think anyone has details on really. I have been affected by this before and it is definitly there. Was farming a bugged event that waves of enemies kept coming non-stop for an hour+ and after a while the mobs stop dropping loot completly locking you out until you stop farming completely for a period of time (30min I think I waited before getting loot again)

This is for GW1 and not 2… but GW2 probably has something very similar
http://wiki.guildwars.com/wiki/Anti-farm_code

I have done some tests with crit damage and steady weapons in the mists.
I’m using the elementalist staff, so the bouncing attack hits all 3 golems at once and I can parse the combat log easier.
These are the results.
0 critdamage:
http://i.imgur.com/MztGX.jpg
30 critdamage:
http://i.imgur.com/go29x.jpg
50 critdamage:
http://i.imgur.com/tmxm5.jpg

You can see for yourself by doing the math, but the critical multipliers are:
1.5 for 0 critdamage, 1.8 for 30 critdamage and 2 for 50 critdamage.
This would suggest that critdamage indeed just gets added to your base 1.5 multiplier.

The numbers don’t quite match for the attacks on the heavy golem with 50 critdamage, but there’s a reason for that. I had to patch that picture together from different samples, because getting a bouncing hit where all 3 hits don’t crit is pretty hard with berserker’s equipment and 30 points in air. Then you look at one trait that Air has, Weak Spot, and you will see that there’s a chance to apply vulnerability on critical hits. That is what screwed up the values a bit.

As for why the actual damage values differ between my shots with 0 and 30 critdamage and the shots with 50 critdamage, I had to move spots because of thieves testing out their heartseeker deeps. Turns out the different golem spots actually have different armor values!

Oxnan.2459:

Also, for your MF testing, not sure if you know this, but there is anti-farming code in the game that I don’t think anyone has details on really. I have been affected by this before and it is definitly there. Was farming a bugged event that waves of enemies kept coming non-stop for an hour+ and after a while the mobs stop dropping loot completly locking you out until you stop farming completely for a period of time (30min I think I waited before getting loot again)

This is for GW1 and not 2… but GW2 probably has something very similar
http://wiki.guildwars.com/wiki/Anti-farm_code

Another thing to look into. But yeah, I don’t farm… anything I have for magic find is just a collection of strike marks saying that I passed through a particular area and kill x amount of those creatures. Record the number of drops and move on. So, my statistics aren’t from single encounters… Now that you put it that way, it may put my mob drops on the higher end of the spectrum. Even so, they are consistant with testimonials I’ve heard from other players. Really not sure what to make of these things now.

But hell, that’s the fun in trying to model stuff. You think you have a good answer, learn the parts that are wrong, and fix it for a better one later. Rinse and repeat.

spreadsheets.9805:

I have done some tests with crit damage and steady weapons in the mists.
I’m using the elementalist staff, so the bouncing attack hits all 3 golems at once and I can parse the combat log easier.
These are the results.
0 critdamage:
http://i.imgur.com/MztGX.jpg
30 critdamage:
http://i.imgur.com/go29x.jpg
50 critdamage:
http://i.imgur.com/tmxm5.jpg

You can see for yourself by doing the math, but the critical multipliers are:
1.5 for 0 critdamage, 1.8 for 30 critdamage and 2 for 50 critdamage.
This would suggest that critdamage indeed just gets added to your base 1.5 multiplier.

The numbers don’t quite match for the attacks on the heavy golem with 50 critdamage, but there’s a reason for that. I had to patch that picture together from different samples, because getting a bouncing hit where all 3 hits don’t crit is pretty hard with berserker’s equipment and 30 points in air. Then you look at one trait that Air has, Weak Spot, and you will see that there’s a chance to apply vulnerability on critical hits. That is what screwed up the values a bit.

As for why the actual damage values differ between my shots with 0 and 30 critdamage and the shots with 50 critdamage, I had to move spots because of thieves testing out their heartseeker deeps. Turns out the different golem spots actually have different armor values!

No kidding?! That’s great lol. Makes me wonder if actual players have different armour values depending on where you hit them. Headshots, anyone?

For Faux Sake.8457:

No kidding?! That’s great lol. Makes me wonder if actual players have different armour values depending on where you hit them. Headshots, anyone?

That’s not what he is saying, there is many test golems in the mist, and some heavy golems don’t have the same armor as heavy golems located somewhere else. He had to move to different areas since there was ppl in his old spots.

Something you may want to try, just an idea for test Crit Dmg is go and unlearn your skill at a trainer and Don’t respec, leave it empty. That Way you wont have to worrie about extra varriables. Hope this helps. I’m also woundering if Crit Dmg has a cap or not like the rest. Still testing it out but i can reach 113% Crit Dmg but not sure if its even counting the extra 13% or not.

so what % of a rare do i get running at 310% Magic Find? 10% or only 4%? Nice information by the way. Been farming for months just for fun. Lol.

Question, I currently have 143% MF just accounting for armor and trinkets. (the amulet is ascended with 20% MF)

Would it be worth the 18% loss in MF in order to switch over to level 78 exotic zerker armor? (I’m currently using rare explorer gear) Make that a 13% loss instead, since I can add exotic opal jewelry to my trinkets to get back 5%.