Erekul + Shrink Ray Question

[Darit]

Given a lvl4 Parasitic Infusion (-60% Weapon Damage) and a Shrink Ray (-40% weapon damage), what would the end result be of a roughly 200 damage enemy autoattacker?

 

And if I used this tactic and a barbed plating, would it be what he *normally* would've done damage wise or what it would be after?

 

Trying to decide if Erekul+Shrink Ray >=< Erekul + Barbed Plating

 

On a side note, I miss the old Stabilizer 0.2 hp/mp thing lol

[NoWaterJustIce]

i think someone said that right now erekul and shrink ray are stacking linearly right now so it would make their weapon dmg 0 :/

[Mus]

It's supposed to stack with diminishing returns. So it should reduce the target's weapon damage to 24% of its original value. 200 damage would become 48.

 

Barbed plating returns damage before mitigation. But your target will only be attacking you with 24% of his weapon damage, so barbed plating won't be very effective. I'm pretty sure that's correct.

[sTeveN]

i've done a post before on this topic, if i recall correctly, shrink ray and erekul's parasite infusion don't stack with diminishing returns. so this combo can nerf nova's late game dmg to 0.5, which i believe is the minimum cap of weapon dmg.

[WhaleTits]

correct

[ginosaji]

Broken

[FruitNinja]

It's supposed to stack with diminishing returns. So it should reduce the target's weapon damage to 24% of its original value. 200 damage would become 48.

 

Barbed plating returns damage before mitigation. But your target will only be attacking you with 24% of his weapon damage, so barbed plating won't be very effective. I'm pretty sure that's correct.

 

That's not diminishing returns.

 

With just erekul W or shrink ray, it takes twice as many hits to kill you.

With both, it takes 4 times as many hits to kill you.

 

If this is the case, then it's not diminishing returns, you get a larger benefit from stacking weapon damage reduction.

 

 

However I think it works linearly but I'm not sure. It seems simple enough to test. That would be even worse than multiplicative stacking. Instead of going from 2 to 4 it goes from 2 to infinity.

[Rem]

its alrdy been tested works exactly like steven said

[Mus]

Nope it is diminishing returns, since the effect of the two combined is less than the effect of their direct sum.

 

I am aware that Shrink Ray and Erekul W are stacking linearly at the moment, but they are meant to be stacking with diminishing returns.

[FruitNinja]

Nope it is diminishing returns, since the effect of the two combined is less than the effect of their direct sum.

 

Says who?

 

The effect of each reduction effect is 50% of the current value regardless of how much you already have. It's what you're supposed to pay for.

 

Saying that's diminishing returns is silly. Diminishing returns implies that you shouldn't stack something, but like armor stacking, getting more only makes the targeted damage type exponentially less effective.

Do the math. If armor gave 50% reduction per point, and our test hero has 1000 hp, then we'd have:

0 Armor: 1000 physical damage needed to kill.

1 Armor: 2000 physical damage needed to kill.

2 Armor: 4000 physical damage needed to kill.

3 Armor: 8000 physical damage needed to kill.

4 Armor: 16000 physical damage needed to kill.

5 Armor: 32000 physical damage needed to kill. (Equivalent to 228 armor in AoS :D)

 

and so on. Damage reduction is the same. With 50% damage reduction, you need to do twice the work you had to do previously to kill the target.

 

Diminishing returns would be something like each point of STR giving less HP depending on how much STR you have. Multiplicative stacking is not diminishing returns, it is multiplicative stacking.

 

Linear stacking of percentage reductions gives INCREASING returns. It is not the baseline.

 

EDIT: This is implying that the damage reduction stacking is multiplicative. It is not, and I think this should be fixed. However, semantics are important :P

[Phyresis]

Time to troll a pub, seeing as everyone still plays agi only in pub.

[Mus]

Assuming these things stack as they are supposed to:

 

Applying Shrink Ray reduces weapon damage by 40%. Applying W reduces weapon damage by 60%.

 

Applying both reduces weapon damage by 76%, not by 100%. This is less than what you would get if they stacked linearly.

 

It works the same as armour (diminishing returns). You reduce the target's weapon damage by X% of it's current value, not it's initial value. That is precisely what diminishing returns means.

[FruitNinja]

Assuming these things stack as they are supposed to:

 

Applying Shrink Ray reduces weapon damage by 40%. Applying W reduces weapon damage by 60%.

 

Applying both reduces weapon damage by 76%, not by 100%. This is less than what you would get if they stacked linearly.

 

It works the same as armour (diminishing returns). You reduce the target's weapon damage by X% of it's current value, not it's initial value. That is precisely what diminishing returns means.

 

What definition are we using?

 

http://en.wikipedia....nishing_returns

http://tvtropes.org/...turnsForBalance

http://www.wowwiki.c...nishing_returns

[Mus]

I don't understand why you cannot see how this works. The definition you provided fits what I have said.

 

You reduce the target's weapon damage further and further, but you are removing less and less damage points each time.

[ginosaji]

Mus, I believe what's happening is that what you are saying is how it's supposed to work (diminishing returns), but that Erekul is bugged. There's a thread somewhere that came up recently about shrink ray stacking linearly, Shrink Ray and W gives 100% weapon reduction. I know that was a bug recently, don't know if it has been fixed or not yet.

[FruitNinja]

I don't understand why you cannot see how this works. The definition you provided fits what I have said.

 

You reduce the target's weapon damage further and further, but you are removing less and less damage points each time.

 

The problem with the way you're describing it is that 76% weapon damage reduction is over twice as good as 50% weapon damage reduction. It makes you take twice as many hits as the former. Each time you apply it, it gets better at doing what it's supposed to. That is the complete opposite of what diminishing returns in games is supposed to accomplish.

 

Armor is the same. Stacking armor makes you nigh-invulnerable to physical damage. This is why Justicar Tychus, and LZ are so ridiculously tanky--because their resistances stack with the usual tank builds they buy.

 

The numbers themselves don't matter at all. What you're describing can be reworded in a lot of ways to support my argument instead, as my previous example showed.

 

Mus, I believe what's happening is that what you are saying is how it's supposed to work (diminishing returns), but that Erekul is bugged. There's a thread somewhere that came up recently about shrink ray stacking linearly, Shrink Ray and W gives 100% weapon reduction. I know that was a bug recently, don't know if it has been fixed or not yet.

 

There is no confusion over how it actually works. I'm just arguing that the term "diminishing returns" doesn't apply and has never been applicable to AoS as the term implies that stacking these sorts of things results in wasted minerals.

[Mus]

If you're arguing that the term diminishing returns doesn't apply to aos then there isn't any point in me continuing to argue with you, I thought you were misunderstanding the way the effects stacked rather than the actual definition used.

[ginosaji]

The problem with the way you're describing it is that 76% weapon damage reduction is over twice as good as 50% weapon damage reduction. It makes you take twice as many hits as the former. Each time you apply it, it gets better at doing what it's supposed to. That is the complete opposite of what diminishing returns in games is supposed to accomplish.

 

Armor is the same. Stacking armor makes you nigh-invulnerable to physical damage. This is why Justicar Tychus, and LZ are so ridiculously tanky--because their resistances stack with the usual tank builds they buy.

 

The numbers themselves don't matter at all. What you're describing can be reworded in a lot of ways to support my argument instead, as my previous example showed.

 

 

 

There is no confusion over how it actually works. I'm just arguing that the term "diminishing returns" doesn't apply and has never been applicable to AoS as the term implies that stacking these sorts of things results in wasted minerals.

 

 

Ah, didn't read closely enough. Now that I have.... I believe your argument is one of semantics, or at least realativity. In absolute terms diminishing returns does exist in AOS, here's an example. 25% and 25% reduction on 500. (500)(.75)=375. (375)(.75)=281.25 The difference in the first reduction is 125, the second is 93.75. So in absolute terms, the overall damage reuction was less for the second application of 25% than the first. So, in fact the second application of 25% reduction has given you less damage reduction 125 the first application and 93.75 the second application. If you graphed an equation like this it would describe and exponential decay. Exponential decay and "diminishing returns" are synonimous terms, one from mathematics, the other from economics.

 

So, you do get less for you money. Example: 20% and 20% spell damage reductuion. Lets say a spell does 100 damage. The first application is the the 20% spell resist every hero gets. (100)(.80)=80. (80)(.80)=64. (64)(.80)=51.2. The reduction the first application was 20 damage and the reduction of the last of the two hypothetical item caused 20% reductions reduced damage by 12.8. For argument's sake say these items cost the same amount 2000 minerals. 2000/20=100 minerals/1 spell damage reduced. 2000/12.8=156.25minerals/ 1 spell damage reduced. Thats a huge increase in cost/benefit quotient, which is the definiton of "diminishing returns."

 

Whether these diminishing returns are a "waste of minerals" or not is a different argument. One that I think is based situationally (opposing team comp, etc.) instead of absolutely. You could run an equation of each hero individual hero and item pairing to see where the efficient market position is. I'll leave that up to someone who know something about math though.

[FruitNinja]

So, you do get less for you money. Example: 20% and 20% spell damage reductuion. Lets say a spell does 100 damage. The first application is the the 20% spell resist every hero gets. (100)(.80)=80. (80)(.80)=64. (64)(.80)=51.2. The reduction the first application was 20 damage and the reduction of the last of the two hypothetical item caused 20% reductions reduced damage by 12.8. For argument's sake say these items cost the same amount 2000 minerals. 2000/20=100 minerals/1 spell damage reduced. 2000/12.8=156.25minerals/ 1 spell damage reduced. Thats a huge increase in cost/benefit quotient, which is the definiton of "diminishing returns."

 

You're forgetting that you can take more spells in the first place.

 

If you have 1000 hp then two 500 damage spells will kill you. Factor in 20% spell resist and you reduce 200 damage from the first two, plus 50 damage out of the one that kills you. You have 1250 effective hp, and block 250 spell damage.

 

Adding in another 20%, you take 320 per spell, resisting 180 each. It takes about 3 spells to kill you, so you block, in this case, 540 damage.

 

Out of that 540, more was due to the second spell resist item than the first.

 

Exponential decay and "diminishing returns" are synonimous terms, one from mathematics, the other from economics.

 

In completely different contexts.

 

In economics, you are trying to maximize profits. Your costs and revenues fluctuate with each other.

In AoS, you are trying to bring your opponents' health to 0.

[Rem]

Give it up fruit its diminishing returns no matter how you try and spin it. To show you an example on non diminishing returns,

 

1000 base hp 500 dmg spell 20% resist you resist 100 dmg

 

1000 base hp 500 dmg spell two 20% resists you resist 180 dmg

 

1000 base hp 500 dmg spell 40 % resist you resist 200 dmg

 

 

180 is clearly lower than 200 therefore 2 20% resist items are not equal to 1 40% therefore stacking lower resist items is of less value than one large resist item. Very definition of diminishing returns.

[FruitNinja]

Give it up fruit its diminishing returns no matter how you try and spin it. To show you an example on non diminishing returns,

 

1000 base hp 500 dmg spell 20% resist you resist 100 dmg

 

1000 base hp 500 dmg spell two 20% resists you resist 180 dmg

 

1000 base hp 500 dmg spell 40 % resist you resist 200 dmg

 

 

180 is clearly lower than 200 therefore 2 20% resist items are not equal to 1 40% therefore stacking lower resist items is of less value than one large resist item. Very definition of diminishing returns.

 

Give it up because your example is the exact thing I disproved? Should I "give up" because you've taken to restating things without additional reasoning?

 

You're ignoring that you resist more damage overall with stacked spell resist. You can't take it on a spell-by-spell basis, because your enemy needs more spell damage to kill you, resulting in more damage being blocked. Isn't that the purpose of spell resist in the first place?

[Rem]

Fruit you should give up because your talking out your ass and have no idea what everyone else is trying to drive into your skull despite being shown to be wrong by several people myself just being the last.

 

The example I gave you holds true no matter how many spells there are. If the sum of its parts is more than the whole, (in this case 20% resist + 20% reduces less dmg than 40% ) then it is diminishing returns.

 

So long as Two 20% is less than one 40% this is the only thing that matters.

[FruitNinja]

The example I gave you holds true no matter how many spells there are. If the sum of its parts is more than the whole, (in this case 20% resist + 20% reduces less dmg than 40% ) then it is diminishing returns.

 

20% spell resist gives you 25% more effective hp.

 

Two +25% effective hp items gives you 56.25% effective hp.

 

1/.64 (2x20% SR) = 1.5625

 

Do you see what I did here? The numerical values working together prove nothing. You can word it as spell resist but I can just as easily word it as something different and magically, it becomes the opposite of what you consider diminishing returns! If you use the previous criteria of using total damage blocked, then my argument is proven anyhow. You can't win here by rephrasing things to be more convenient for yourself.

If your hp lost remains constant, spell resist blocks more spell damage based on how much you stack. Considering that INT scaling is linear, there is no way on Earth you could call this diminishing returns, and I don't see any point in continuing this argument if you refuse to understand.

[Rem]

Yeah Ok I'm gunna follow everyone else and give you up as a hopeless imbecile....

[ginosaji]

Math is hard....

 

Mods please move this to www.reddit.com/r/logic

 

gg