MoP BiS Arms Warrior

[Piglet]

to answer the question whether 6 armour matter to someone who already has 2650 armour, you need to calculate the probability of a situation where a player would not have died in a BG if he had this extra 6 armour. And then, what is the mathematical expectation of the number of battlegrounds that you have to play in order to encounter such a situation? Millions, I would guess?

 

[Goesid]

Piglet ur not bis w/o them.
It doesntmatter, true but since since the other pair is better plain and simple.

 

[Piglet]

do you understand, that 0.9999(9) is equal to 1? Lets start from there.

 

[Franciska]

Foreman's reduces physical damage by 0,021% or 1 physical damage over 1000 damage taken.

 

[Piglet]

Foreman's reduces physical damage by 0,021% or 1 physical damage over 1000 damage taken.

assuming the 0.0021% figure is correct, 1000damage*0.021% = 1000*0.00021damage = 0.21 damage


Thus, depending on how rounding works in WoW (and whether rounding errors are accumulated as opposed to discarded), it may never actually reduce damage.

You would need to crit him 0.50/0.21*1000 = 2381 physical damage (after resil, before armour) in order for the extra armour to reduce the damage by 0.5 which will be rounded up to 1. Need to check the max crits thread - what is the current physical crit record?

 

[Goesid]

How are you not getting this!?
YES IT DOESNT MATTER BUT FOREMANS ARE STILL BETTER EVEN IF IT WAS 1 ARMOR THEY WOULD STILL BE BETTER!

 

[Piglet]

Goesid, you don't understand what numbers are.

I am repeating this again for you:
0.9999999(9) and 1 are the same number! Exactly the same. You can not say that one of this numbers is greater than the other.

Another example: (1 mod 3) and (4 mod 3) are the same number. If two numbers behave exactly the same (when adding them, multiplying them, doing all other defined operations with them), then they are the same number.

Here, if two pieces of armor result in EXACTLY THE SAME damage reduction all the time. Then you can say with the absolute mathematical precision, that their armour values are the same.

Assuming the calculations above are correct, the 2 pieces of armour in question provide the same damage reduction up to 2381 damage. And as it is not possible, as far as I know, to crit more than 2381 (after resil), then the 2 pieces of armour in question ALWAYS provide the same damage reduction

 

[Sly]

There is a way to write 0.9999 that shows that it is a continuous 9 (I think it's a line over the second 9). The need for that shorthand shows that 0.99999999 is in fact NOT equal to 1. If it were equal to 1, it would be 1, not 0.999999. To be BEST in slot, you must first be best. An item with the same stats but more armor is better than an item with the same stats but less armor. Just like how serpent gloves are better than gold flecked, and the difference is only 2 armor. Might not really matter in the long run, but you can't say gold flecked are better than serpent's when clearly they aren't. Nor are they equal.

 

[Piglet]

0.999(9) notation means that 9 is cyclical
0.1212121212... is written down as 0.(12) or 0.1212(12)


In mathematical Number Theory, if two - seemingly different - numbers behave exactly the same, then they are the same number. If two pieces of armor give EXACTLY THE SAME damage reduction, then they have the same armour. So, mathematically, you can say that 2650 armour = 2656 armour. This is a strict equality, in a strict mathematical sense.

 

[Goesid]

edit: obsolete

 

[Kincaide]

assuming the 0.0021% figure is correct, 1000damage*0.021% = 1000*0.00021damage = 0.21 damage


Thus, depending on how rounding works in WoW (and whether rounding errors are accumulated as opposed to discarded), it may never actually reduce damage.

You would need to crit him 0.50/0.21*1000 = 2381 physical damage (after resil, before armour) in order for the extra armour to reduce the damage by 0.5 which will be rounded up to 1. Need to check the max crits thread - what is the current physical crit record?

Nice work... this reminded me of a quote that some of you may not be familiar with:

"But... this one goes to 11."

 

[Piglet]

We don't need no education

 

[Piglet]

Goesid, you should post about the topic, not about who is smarter.
You didn't understand any mathematical arguments I made, did you? Did you even try to?

Also, you need to be smart yourself in order to be able to tell, who is smarter ;-)

 

[Lil]

idc if it's not bis or not lol.

Lynx>Prisoner Boots

 

[CripzBlood]

Why not Stone Guard Greaves?

For me in arms or ret i always counted 1strength as 1pt and 1crit has 1pt too, and stamina as 0.5pt.

Foreman legs = 10,5 pts

Stone Guard Greaves = 11pts

Scales of Aku´mai = 12pts.

without counting with armor.

 

[Hashbrowns]

Goesid, you should post about the topic, not about who is smarter.
You didn't understand any mathematical arguments I made, did you? Did you even try to?

Just... no.

More armor is just that, more armor. It's pretty cut and dry.

Just because you can assess limits in math or physics doesn't mean it applies to this. Especially if Franciska post is correct, 0.021% is quite a lot compared to the 15 0's you were throwing behind your decimal place as an example.

I don't know where you are going with crit, that doesn't really matter. It’s Total HP/(100-(res%+ phy dam red%)). That’s your total effective health when fighting a melee class that’s behind you.

I'd agree we'd have to see where WoW cuts off for rounding for the formula above. But until then I think it's safe to assume what is shown on the tooltip as an armor increase will in fact reduce damage taken.

 

[Piglet]

Here, if two pieces of armor result in EXACTLY THE SAME damage reduction all the time. Then you can say with the absolute mathematical precision, that their armour values are the same.

Assuming the calculations above are correct, the 2 pieces of armour in question provide the same damage reduction up to 2381 damage. And as it is not possible, as far as I know, to crit more than 2381 (after resil), then the 2 pieces of armour in question ALWAYS provide the same damage reduction

What exactly is it, that you don't understand in the text above?

 

[Goesid]

Goesid, you should post about the topic, not about who is smarter.
You didn't understand any mathematical arguments I made, did you? Did you even try to?

Also, you need to be smart yourself in order to be able to tell, who is smarter ;-)

Okey let me lay it down to you on a 5th grade level so that you might understand.


You say it becomes better after 2381 damage?
You say there is no damage higher than that?

Yes...

Did you count 24s?
Did you count gemmed 24s?
Did you count FCing stacks?
Did you count 24 feral druids?
Did you count berserking?

No you did not because if you did you would understand that your pretty much hammering a nail into yourself with these incorrect posts...

 

[Nakedplumber]

BFD legs > Foremans > Poor people Foremans

 

[Vandiir]

You would need to crit him 0.50/0.21*1000 = 2381 physical damage (after resil, before armour) in order for the extra armour to reduce the damage by 0.5 which will be rounded up to 1. Need to check the max crits thread - what is the current physical crit record?

That assumes the WoW program rounds with respect to whether the decimal is greater or less than .5. It's also possible that the WoW client just takes the ceiling of that calculation and used it as it's approximation (in situations rounding to integers). For example, the ceiling of 200.01 is 201, and thus approximated 200.01 as 201, even though 200.01 is closer to 200. Unless one can confirm that it's rounded with respect to .5, or approximated through other means, one cannot be precisely sure.

----------------------------------------------------------------------------------
For those disagreeing or not understanding what he's trying to say, he's taking into account round-off error, or whether the difference is noticeable enough to not disappear in outputs due to round-off errors.

Here's a round-off error example, with a simple program [like WoW, or others] calculating and giving outputs rounded to integers, and [supposedly] rounded with respect to .5:

4000(.021%)=.84 is above .5 and below 1. So it's rounded up, and thus according to the program, 4000(.021%)=1.
However, 1000(.021%)=.21 is less than .5, so according to the program, 1000(.021%)=0.
So, according to the program or rounding scheme, 1000(.021%)=0 implies 1000(.021%)+1000(.021%)+1000(.021%)+1000(.021%)=0+0+0+0=0, even though the program would say 4000(.021%)=1.

So, according to the program, 4000=1000+1000+1000+1000, yet
4000(.021%)=1 ≠1000(.021%)+1000(.021%)+1000(.021%)+1000(.021%)=0.