Help Finish Finding Accuracy Values

**Blind_Archer** · 12-12-2011 02:29 AM

You're gonna have to use at least 3 cats for the accuracy to show up vs Archers I believe...
(1.4 x 2cats = 2.8kills...still not enough... 1.4 x 3cats = 4.2kills..)

**ShadowBunny** · 12-12-2011 02:51 AM

@Oldhag - Not theory, fact. :P Geoff said that in a post somewhere along the line.

@Blind - Read my post in between old's :P

**oldhag** · 12-12-2011 05:12 PM

Originally Posted by ShadowBunny

@Oldhag - Not theory, fact. :P Geoff said that in a post somewhere along the line.

@Blind - Read my post in between old's :P

ya well they also say that they are improving the game lol. so ya...

**Macluuu** · 12-26-2011 02:55 AM

pikes vs hc = 0.34 accuracy !

it means 1 pike kills 0.34 hc

cheers

**confud** · 12-26-2011 04:07 AM

Oh Dear!

Are people still trying to work this stuff out?

When will people learn that there is no one single fixed formulae for troop A VS troop B, and that it constantly varies on a sliding scale depending on the number and ratio of each troop type, along with other influencing factors? (except in the case of ranged troops)

If you attack 1 Pikeman with 1 cavalry, the cav kills the pike 1:1, if you send 1000 Cavs at 1000 pikes, the ratio is different, if you send 200K cavs at 200K pikes, the ratio is different again. Not only does it vary when the numbers are the same on both sides, but it also changes further still as the numbers on each side vary.

Someone just posted "pikes vs hc = 0.34 accuracy !" Really?, try sending 100K pikes at 1 million HC and see how that works out for ya!, it doesn't!

And in each case, the actual kill ratio's are affected by research, knights strengths, attack / defence boosts and guardians.

There is obviously a base formula that the battle system engine works from, based on a pre-determined amount of troops on each side.
The game mechanics then apply permutations to this base formula, based upon the actual numbers on both sides, and the other factors that affect kill ratios as mentioned above.

You'll also probably find that the base formula starts at a point where it is impossible to replicate in battle, i.e. all research levels being zero, which in the game would make it impossible to actually train most of the troops in the first place.

The only way you will ever get to a point where you can predict the outcome of a battle with 100% accuracy, is if you calculate the base formula first, and to do this you need to spend a serious amount of time replicating the same battles on a large scale, with a slight variation each time, and reverse engineering the results.

Apart from ranged only battles, there is no Troop A VS Troop B = Kill Ratio C.

In 99% of actual battles in the game, the No of A varies, the No of B varies, and the things that affect the efficiency of A and B varies too, so its obvious that C also varies too, and is not a fixed value.

Every "accuracy value" or "kill ratio" posted in this thread, could be proved inaccurate by varying the amounts of troops on each side.

My suggestion is get a rough idea of what kills what, and get stuck into fighting and playing the game instead of sitting up all night working on spreadsheets when you've started with flawed principalities of the battle system in the first place.

ShadowBunny:
I really don't mean to sound dismissive of what you are trying to achieve, I think you are a great moderator and contributor to this forum, but what people are trying to do here is in fact over-simplify, a battle system that is far more complex than it can appear on the surface sometimes.

1000 VS 1000, is not the same as 10,000 VS 10,000 and is certainly not the same as 199,999 VS 349,472 which is the reality of battles in the game, the troops vary, the circumstances vary, the results vary, and so does the ratio of troop A VS troop B.

**fdiniz** · 12-26-2011 05:12 AM

You have to understand when to use the accuracy values calculated here. It is a fixed value, but it can't be used alone to determine the outcome of a battle. And the 0.34 value for Pikes vs HC is wrong. The correct accuracy (already calculated here) is 1.2. It doesn't mean that 100 pikes will kill 120 HC. It means that the pikes attacks HC AS IF there are 20% more of them. Then you have to use attack, defense, health, range, guardian, knight, research, etc... to calculate the actual kill ratio.

**confud** · 12-26-2011 05:37 AM

Originally Posted by fdiniz

You have to understand when to use the accuracy values calculated here. It is a fixed value, but it can't be used alone to determine the outcome of a battle. And the 0.34 value for Pikes vs HC is wrong. The correct accuracy (already calculated here) is 1.2. It doesn't mean that 100 pikes will kill 120 HC. It means that the pikes attacks HC AS IF there are 20% more of them. Then you have to use attack, defense, health, range, guardian, knight, research, etc... to calculate the actual kill ratio.

Ok, i'm trying to understand what you are saying here:

If you attack 1000 HC with 1000 pikes, you will kill a small handful of HC, and the pikes will die.

If you attack 100,000 HC with 1000 pikes, not a single HC will die, but all the pikes will.

How can you say that the value is the same in both cases?

**fdiniz** · 12-26-2011 06:08 AM

Originally Posted by confud

Ok, i'm trying to understand what you are saying here:

If you attack 1000 HC with 1000 pikes, you will kill a small handful of HC, and the pikes will die.

If you attack 100,000 HC with 1000 pikes, not a single HC will die, but all the pikes will.

How can you say that the value is the same in both cases?

The kill ratio may not be the same (certainly won't, in this case), but the accuracy will. They are not the same thing. As I said before, you have to understand how accuracy works in the calculations.

**ShadowBunny** · 12-26-2011 08:03 AM

Confud - To find the results of a battle takes a rather simple formula.

# of attacking unit sent * (accuracy/# of attacking unit needed to kill defending unit) = number of defending unit killed ..... (within a round)
Then of course if the defending unit is within range of attacking back, you use the same formula with the ratios reversed.

The only part the research/knights/boosts/guardians come in to is finding the number of units needed to kill another unit.

If you want to doubt that battles can't be mathed at a constant rate assuming the accuracy values are found, start a new thread and post a hypothetical battle that you can set up and conduct yourself. I will tell you what the results will be within the confines of your variables, and so long as you follow each variable to the letter I can predict your battle with nearly 100% accuracy. The only time I would have even a slight issue is if there is an accuracy value that has not been found yet. Which is why we're looking for all of them.