I’ll never understand what you are saying. You seem to be saying that since a flat stat value can go on infinitely there is not a cap. The cap lies in the conversion to the % which cannot exceed 99.9% except CDR which is 550%. There is a limit you will never exceed a point at which your damage will not increase, that’s a cap.
I'm saying that flat stats aren't capped because they aren't capped. This is a true statement: no matter how high you stack a flat stat, adding one to that flat stat increases the percentage value that flat stat converts to. Because you can always increase the number, there's no cap.
Yes, there are percentage values that flat stats cannot reach. Those are asymptotic limits in the math. But the flat stats themselves are not capped - short of the limits the game engine allows numbers to even reach.
You say that flat stats cannot exceed 99.9%. That's false. At CR 100 a hypothetical flat stat higher than 1998000 would exceed 99.9%.
This is not just a semantic distinction, because it cuts to the heart of what the flat stats system even does. The point is the game implements a diminishing returns system in which no matter how high you stack up bonuses to a flat stat, you can always stack more. It is just that the benefit drops the higher you go. But the benefit is NEVER zero. It might become too low to be useful in the judgment of the player, but if the player wants to (and has the capability to) stack an attribute higher they can choose to do so, barring other limits outside the flat stat system. That's why the system was put in place in 12.0 - to moderate the effects of high stacking without actually implementing a hard cap on the values. No matter what players do, they cannot run into a cap in the flat stats system. They can always stack more value for more benefit.
They can not always milk more value, there is a cap, It was explained by Kabam themselves when flat values were introduced, when asked about RR reaching 100% crit, it was explained he would be able to reach 99.9% and no more. The formula you are using must be missing a value somewhere or something else about it must being misunderstood.
They can not always milk more value, there is a cap, It was explained by Kabam themselves when flat values were introduced, when asked about RR reaching 100% crit, it was explained he would be able to reach 99.9% and no more. The formula you are using must be missing a value somewhere or something else about it must being misunderstood.
Well, a couple things. First, I was the one that wrote the original DR and CR guides and spreadsheet calculators on the forums and the subreddit. The formulas and calculations came from discussions with, err, DuckSlug who is the author of the Auntm.ai site you referenced. Those formulas have been verified by a lot of other players, as well as accurately predict the champion attribute screens (which added percentage value calculations after the guides were written).
Okay then explain to me why your own spreadsheet and **** slugs calc both will not exceed 99.9% nor 550% no matter the value entered.
I can't explain that, because when I enter really big numbers into Auntm.ai's stat calc DR calculator I get numbers higher than 99.9%. 3,000,000, for example, generates 99.93%. That agrees with my spreadsheet, which it would have to as my formulas originally come from DS.
I'm honestly not sure what you're looking at, that generates the numbers you're citing. Perhaps you're using someone else's tool that has a clipping roundoff or something. But no spreadsheet I wrote did that, and neither of DS's calc tools does that as well.
I assumed 100% was rounding given Kabam themselves stated it cannot hit 100% or exceed 99.9% Change 99.9 to 100%, w/e the semantics employed are silly. And a fallacy fallacy does disprove the assertion of a cap.
Anyways this was about crit damage rating having a cap of 550% which the calculators mentioned do not exceed.
Can crit damage rating increase damage infinitely? No, just like other flat value stats they hit a point at which nothing can be done to improve the effect.
@DNA3000 I hope this doesn't seem rude, but how do you know all this? Are you a professional mathematician or something along those lines? Sorry if that's too personal of a question to ask, but I'm curious since your posts are always so precise.
@DNA3000 I hope this doesn't seem rude, but how do you know all this? Are you a professional mathematician or something along those lines? Sorry if that's too personal of a question to ask, but I'm curious since your posts are always so precise.
The mathematics involved in the diminishing returns system isn't complicated, at least not in absolute terms. But no, I'm not a professional mathematician. I'm an engineer by education, and an information systems architect by paycheck.
As to how I know how the system works, I had hunches, and I had them confirmed by a redditor that has a very consistent history of determining the mechanics of the game accurately. If you do this long enough and often enough, you can tell when someone is guessing and when someone knows, and in this case I know he knows.
There are actually a lot of ways to make a system like this, but only a few ways you typically see in games, because while game design uses a lot of math, it has few mathematicians. X / (X+K) is extremely common. Also, the simpler the math, the easier it is to implement in game engines and clients. You wouldn't want a system like 2 atan(stat)/π or something if it had to be calculated inside a critical combat loop.
Comments
Yes, there are percentage values that flat stats cannot reach. Those are asymptotic limits in the math. But the flat stats themselves are not capped - short of the limits the game engine allows numbers to even reach.
You say that flat stats cannot exceed 99.9%. That's false. At CR 100 a hypothetical flat stat higher than 1998000 would exceed 99.9%.
This is not just a semantic distinction, because it cuts to the heart of what the flat stats system even does. The point is the game implements a diminishing returns system in which no matter how high you stack up bonuses to a flat stat, you can always stack more. It is just that the benefit drops the higher you go. But the benefit is NEVER zero. It might become too low to be useful in the judgment of the player, but if the player wants to (and has the capability to) stack an attribute higher they can choose to do so, barring other limits outside the flat stat system. That's why the system was put in place in 12.0 - to moderate the effects of high stacking without actually implementing a hard cap on the values. No matter what players do, they cannot run into a cap in the flat stats system. They can always stack more value for more benefit.
Also, Kabam mentioned the formula when they discussed challenge rating: https://forums.playcontestofchampions.com/en/discussion/2317/challenger-rating-explained. Although that post focused on challenge rating, it noted the DR formula because that's the only way to understand the effects of challenge rating.
I'm honestly not sure what you're looking at, that generates the numbers you're citing. Perhaps you're using someone else's tool that has a clipping roundoff or something. But no spreadsheet I wrote did that, and neither of DS's calc tools does that as well.
Anyways this was about crit damage rating having a cap of 550% which the calculators mentioned do not exceed.
Can crit damage rating increase damage infinitely? No, just like other flat value stats they hit a point at which nothing can be done to improve the effect.
As to how I know how the system works, I had hunches, and I had them confirmed by a redditor that has a very consistent history of determining the mechanics of the game accurately. If you do this long enough and often enough, you can tell when someone is guessing and when someone knows, and in this case I know he knows.
There are actually a lot of ways to make a system like this, but only a few ways you typically see in games, because while game design uses a lot of math, it has few mathematicians. X / (X+K) is extremely common. Also, the simpler the math, the easier it is to implement in game engines and clients. You wouldn't want a system like 2 atan(stat)/π or something if it had to be calculated inside a critical combat loop.