Celtic Heroes

Papi's Guide to Metagaming
One of the things we were in encouraged to do in beta was to create guides to help the greater community so here is mine. Don't want to waste all that work.

Before we go any farther let’s take a moment to figure out just what in the world am I talking about. First of all this guide will not give you any secrets about the game itself. Well I do work one example but its not much of a spoiler. But it will help you understand and participate in the metagame.

• My definition: Using resources outside the game to assist and give you an advantage to help you play the game.
  The resources I use: Stopwatch, Microsoft Excel, algebra

Here is a simple example.

• What does more DPS, an ancient greatsword or an ancient great axe?
  ancient greatsword speed 4167 damage 82
  ancient great axe speed 4903 damage 99
  
  The game would have you fight say a Connach shieldmaster using each and then look to see how much health you have when the fight is over. The one that lets you have more health when it is all over must do more dps.
  The metagame solution is to understand that the speed is really in milliseconds so breaking out the good old calculator we find the greatsword does 82 damage in 4167 milliseconds or 19.67 damage per second while the great axe does 20.19 DPS.
  But there is a problem, when you killed the shieldmaster the sword did more dps. How can that be? Well you realize your sword ability is higher than your axe ability so naturally you are better with swords and thus the greatsword will have better DPS until you train up your axe ability. But how much does it effect it? Now we have an experiment to do. Determine how much ability changes your damage……..

The game provides the information, you take that information outside the game and study it. Take it apart. See how it ticks inside. And then use that information to build a more efficient character.

So how do you start metagaming. Well first you have to start with a question. Some of the typical questions we have to discover are:

• - Should I put more points into STR or DEX?
  - Should I buy a Golden Warmagic Helm (60k, +20 Elemental damage) or a Golden Quiver of Bounty (60k, +12 Physical damage)
  - Do I put more points into Giant Swing or Pummel?

Lets think about the first one. STR vs DEX. What do I need to know to answer this question?

• How does STR contribute to how many points of Damage I can do?
  How does my listed damage compare to what I am doing to the mob?
  How does DEX affect my % chance to hit?

Each of these questions is a different experiment. Let’s take the easy one.

How does my listed damage compare to what I am doing to the mob? The natural reaction would be to take your weapon and go hit a mob a bunch of times and write down every damage number that comes up. But before you do that, what do you know and what do you not know about that scenario?

• Well we know our own listed damage but we don’t know the mobs armor.
  We also know that damage changes if the mob is higher level than you.

It is very difficult to try to figure all of these relationships out at once. You need to isolate the different pieces to study them separately. So what you want is to have a mob your same level that you know their armor value. The answer is PVP. Go find a friend the same level as you and go set up your experiment. While you are at it grab a bunch of different weapons to use to see if that makes a difference. And have a number of different armor combinations too. So you go and whack away at your friend, and he you. You will find that the problem has layers
Before you even start make a few guesses at what you expect to see:

• - A range of numbers between some maximum number and some minimum number.
  - The max number should be my max damage.
  - But if he has armor on then I won’t ever do my max but something less.
  - The min and overall average will also be less with more armor.

That should tell you that in order to make your job easier you need to start with no armor at all and then add armor pieces and see how that changes things.
So you do your experiment with no armor, there is a max number and a min number and you can calculate the average which just so happens to be half way between the max and min, but other than that the numbers are all over the place in between with no rhyme or reason. No bell curve or anything just a bunch of random numbers in a range. And that is your key. The computer is using a random number generator to pick numbers between the max and the min. It picks about the same number above the average as it does below which is why the average is right in the middle. With no armor on you will find the max is very close if not your listed damage and the min is half of that number. You may also notice that every once in a awhile you will get a number slightly greater than your max. (I still don’t know why that happens.)
So the first part of your experiment shows:

• - My listed damage is the max I will do if the mob has no armor
  - The min is half my max
  - On average I will do 75% of my max (or half way in between the two)

By doing this portion of the experiment you can get an idea of just how many hits you need to record to get an accurate measure of the max, min and average hit. Maybe 100 is more than you need. Perhaps not.

The next step is to look at the numbers for putting on armor.
Lets say you did 100 data points for each armor set and you come up with a table listing Armor vs max hit. How will you know what is the max hit? You use the results of your last experiment:

• Max hit = biggest number = 2 x smallest number = 4/3 x average number

If the numbers do not agree then go get more data
You should end up with a table that looks like this:
Armor Max
0 126
35 112
74 100
100 92
184 76
244 67
334 57
414 51
So now what do we do? Well we guess. What type of formula would behave this way? You make a few graphs of known formulas and see if any one looks promising.
You want increasing armor to reduce damage but notice the damage never goes to 0 no matter how much armor we have. Well the y=100/x looks promising but it drops off too fast and if x is 0 it explodes. The y=x^1/2 has the right shape but is upside down. And isn’t quite flat enough.

What about percents? Maybe the armor relates to a percent of damage? What does the data look like if we view it as a percent of max damage?

Armor Damage
0 100%
35 89%
74 79%
100 73%
184 60%
244 53%
334 45%
414 40%

If you used a few weapons you will see that the percentages are the same regardless of which weapon is used. So now we know we are on the right track.
The chart doesn’t look any different but notice the 50% mark is close to 244 armor. That’s almost twice our max damage. That may be a hint. So lets make another guess.

We want the formula to start at 100% when armor is 0 and reduce to 50% when armor is twice the max damage. Lets try 1-armor/4/max.
Well that gets us our 0 point and our 50% point but the shape is wrong and will go negative with enough armor. Well lets try see if we can get the shape of y=100/x.
How can we put this in the denominator and get the same points of 0 armor = 100% and armor = 2*max being 50%. We need the denominator to get bigger so the percent can get smaller so lets try.
1/(1+armor/max/2)
Hey that looks nice. So lets multiply that by out max and compare it to our data.
x.......y=1/(1+armor/max/2)....Data
0................126 ................126
35................111 ................112
74................ 97 ................100
100................90 ................92
184................73 ................76
244................64 ................67
334................54 ................57
414................48 ................51

That’s not bad very close and the right shape. We know our data can be off by a few points because we didn’t take a lot of data so it looks like we have it. Is it the exact formula? Maybe, maybe not, but is it close enough to use in making decisions? I think so. The other benefit is that the formula is not overly complicated. If you play with it a little you see that 2.2 actually makes it a much better fit than 2.0 but then you have to ask the question Why would OTM make it 2.2 instead of 2? The difference is negligible. It is a better chance that our data is a little off than OTM chose an obscure number just for the heck of it. And even if we are off its not by much.

So now what? Well we go find a few mobs our level and try to use this formula to calculate the armor on the mob. We do this enough time to decide if PVP and PVE use the same formula. We know there is another term in the formula that reduces your damage if the mobs level is higher but we cant figure that out until we know the basic formula it correct.

Well that is the basic process. Any question you have can be reduced to an experiment. Try to limit the number of factors that might effect your outcome and concentrate on one.

If you decide to look at how you calculate max auto damage you will realize there are a few pieces that have to be isolated in order to make it easier to figure out. There is a STR piece, an ability piece and then the physical damage is thrown in as well as the elemental damage.

So you try to set ability to zero and change your str with a rebirth book. Use weapons with just physical, just elemental (offhand while punching) and both but keep the data separate and look how the different weapons change the results.
So how do you keep your ability zero and have enough levels to play with STR? Well with a warrior maybe you have never used blunt and it is zero. Or spear. Lots of rangers never use spear.

Once you figure out how it works without ability then you can use a few characters and see how the different ability score changes your damage when you use the same weapons.

Divide and concur.

Well thats my introduction to the wonderful world of metagaming. Good luck and pm me if you have questions.

FYI if you think this is really cool, consider a science major.

MASTER PIECE!!! GOOD WORK PAPI!!!

Nice formulas they represent the situation

You dont get out much do you?

You dont get out much do you?

One of the things we were in encouraged to do in beta was to create guides to help the greater community so here is mine. Don't want to waste all that work.

He is following beta protocol issued by the Admin and apparently he is good at Algebra so he's decided to give a lesson on metagaming.

I have been longing for another thread like this. And it's looong

Nice work!

LOL!! I dont understand anything.. Im so stupid

What is the point of this posting? To show how smart u are? Nobody understand or care to read. Failed.

What is the point of this posting? To show how smart u are? Nobody understand or care to read. Failed.

This is very interesting stuff , it sucks you can't see that he put work into this as well as it's possible it does help someone , someone smarter than you obviously and smarter than me as well .

Bigots are the ones who knock stuff they don't understand .