EPT[Patch 3.10a: Worlds Balance] General Discussion - Page 71
| Forum Index > LoL General |
Carnivorous Sheep
Baa?21242 Posts
| ||
|
wei2coolman
United States60033 Posts
| ||
TheYango
United States47024 Posts
On August 27 2013 11:58 Zdrastochye wrote: If it gave 8% more AS, and 8 more damage on the passive for the same price, I could see it being an easy 3rd item on Kog vs double AP comps. It DOESN'T have those stats however, so I recommend it on NOBODY. Well the other problem is that the "rounded offensive stats at a competitive price" single item niche is already filled by both BotRK and Triforce now. There's not really room for Wit's unless one or both items get the nerfbat. | ||
|
Lord Tolkien
United States12083 Posts
On August 27 2013 11:52 obesechicken13 wrote: Brutalizer actually gives you more damage than other AD items. The buildup isn't that bad. It's still the late game I'd be worried about. Brutalizer falls off fast and many people don't turn it into anything and just sell it :/ Brutalizer scales with flatpen runes and %pen. The closer you get your target's armor to 0, the more effective and potent it becomes. BrutCleaver on Corki isn't that bad (it's actually pretty good), given his E procs it and already has a large innate (flat) shred. On other ADCs, only Ezreal's a major candidate for a Brutalizer, if only as a substitute to SotEL for full CDR. Maybe MF if she's just going all out on her ulti and getting a cleaver for the team. Lucian...seems an underwhelming candidate for it. Cleaver may work on him given how relatively fast he can stack it for his team, but it would probably gimp his damage. Definitely not wits, though. But then again, I don't really understand Lucian too well. Also, I take full GP10 seals on my Janna/Nami. Wouldn't recommend them though; flat armor too good/versatile on everyone. On August 27 2013 12:14 TheYango wrote: Well the other problem is that the "rounded offensive stats at a competitive price" single item niche is already filled by both BotRK and Triforce now. There's not really room for Wit's unless one or both items get the nerfbat. Wits is already a niche item to begin with, as a split offensive/defensive item. No one saw it with much frequency at all after the nerf it got with the Hecarim Patch (30MR to 20MR, 2000 to 2150 gold), and it's effectiveness on many top laners (who looked for AS and needed MR) got gutted. With the removal of Malady, Nashor's definitely got the better end of the deal, as the MR shred+gain takes place far too slowly for it to have much effect. And again, only a few champs who need AS benefit directly from the MR shred (Shy, P.Udyr, Shen, ???), and there are better MR options which don't need to stack up. The other major "AP autoattackers" like Kayle much, much, much prefer Nashors, and getting a Wits on top of that is rather unattractive at best. | ||
|
Aukai
United States1183 Posts
On August 27 2013 11:54 onlywonderboy wrote: This isn't even that bad, at least MoonBear has a detailed argument instead of him just calling PX shit. Honestly this has been pretty civil so far. I just feel like I'm watching MB argue with hodor or something, I would have called him shit by now. I only get upset that some new person may read this garbage and think its actually viable. | ||
|
TheLink
Australia2725 Posts
The closer you get your target's armor to 0, the more effective and potent it becomes. Is this true? I believed this for the longest time but then there was that Reddit post arguing that every point of resistance is as valuable as every other due to effective hp. Didn't understand it at the time and still don't now but he seemed to make some good points. The diminishing returns thing (and vise versa) is apparently an illusion. | ||
|
RagequitBM
Canada2270 Posts
| ||
|
Dusty
United States3359 Posts
I came into this thread expecting an actual discussion and now I have the taste of vomit in my mouth. Wit's end Lucian? armor/lvl seals in a 2v2 lane? anyways, I figured out how to make Lucian an actual champion; copy the mashme/saiph build, which is pretty much BT -> Brut -> Zephyr -> Cleaver. so strong | ||
|
ETisME
12285 Posts
his mobility and "poking" is quite powerful, making him not a bad solo champion as well. comparing him to the more traditional adc like vayne would be a bit unfair | ||
|
phyvo
United States5635 Posts
On August 27 2013 12:52 TheLink wrote: Is this true? I believed this for the longest time but then there was that Reddit post arguing that every point of resistance is as valuable as every other due to effective hp. Didn't understand it at the time and still don't now but he seemed to make some good points. The diminishing returns thing (and vise versa) is apparently an illusion. It's diminishing returns in the sense that if you have lots of armor it's more cost effective to build HP than more armor. The reddit argument is equivalent to saying that whether the target has 190 or 10 armor you're reducing 10 armor so therefore it's the same in-game value, but that's just incorrect. When targets have low armor flat pen will increase your damage by a higher %. Building AP/AD increases your damage the same % regardless of the target's resistances and building % pen is more gold effective against high resistances. It's as simple as gold efficiency and the damage returns (that is, the in-game value) you get out of it. | ||
|
Scip
Czech Republic11293 Posts
| ||
|
Lord Tolkien
United States12083 Posts
On August 27 2013 12:52 TheLink wrote: Is this true? I believed this for the longest time but then there was that Reddit post arguing that every point of resistance is as valuable as every other due to effective hp. Didn't understand it at the time and still don't now but he seemed to make some good points. The diminishing returns thing (and vise versa) is apparently an illusion. It is entirely true. Every point of armor/MR increases your EHP by 1% of your maximum HP, so with 50armor/mr, your EHP is 150% of your max HP, at 150armor/mr, 250% max HP, etc. It's fairly straightforward. Given the multiplicative nature of HP*Resistances, however, Resistances become less cost-effective versus health the more you stack (and vice-versa). Stacking 300armor/MR with only 2k HP is less effective than 200armor/MR with 3k HP, for instance (8k EHP vs 9k EHP). Etc etc. Resists also increase your effective (flat) health regen. Similarly, flat pen's effective damage increase is more valuable at lower resistance levels than at higher ones. Let's say I have 100 flat penetration (for ease of calculation). If the target has 100 armor/MR, I'm doing true damage to them, and the 100 flat pen is effectively doubling my damage, by reducing their EHP from 200% to 100%. If they have 200armor/MR, the effect is only 300% to 200%, or a 50% increase in my damage versus that target. If they have 300 armor/MR, it goes from 400% to 300%, or a 33.3-% increase to my damage. You're getting the same reduction in EHP from max HP from every point of armor penetration: however, if their EHP is buffered by high resists, the penetration is less effective. Previously in S1 & 2, given the way penetration was calculated (Shred, then Flat, then %Pen) and general cost effectiveness of resistances vs health, people stacked resistances over health, due to the weak nature of flat pen and the way it falls off (given it is then worth less when also acquiring %pen). With S3 and the shifting of the calculation of %penetration before flat penetration, %pen amplifies the effectiveness of flat penetration. So having, % penetration increases the effectiveness of your flat pen instead of reducing its value. For instance, if I had 50% pen and 50 flat pen, previously against a 100armor/MR target, the flat pen would reduce their armor/MR to 50, then %pen would reduce it to 25armor/MR, whereas now %pen would reduce their armor/MR to 50 then flat would reduce it to 0. EDIT: Diminishing returns is not an illusion given opportunity cost. If we have 2kHP and 200 Armor/MR, and we can spend gold to either buy 100armor/MR or 1k HP, which would be the better choice? If we had 3kHP and 100armor/MR? Buying Resistances increases the effective value of HP, making HP more cost effective. Vice versa is also true. This doesn't factor in penetration or %hp damage (the former favors HP as a defensive stat, the latter favors resistances). But that is the gist of it, and pretty much all you need to know about pen/resistances/ehp. | ||
|
PrinceXizor
United States17713 Posts
| ||
|
obesechicken13
United States10467 Posts
On August 27 2013 12:52 TheLink wrote: Is this true? I believed this for the longest time but then there was that Reddit post arguing that every point of resistance is as valuable as every other due to effective hp. Didn't understand it at the time and still don't now but he seemed to make some good points. The diminishing returns thing (and vise versa) is apparently an illusion. No one who's run the math is confused about it. Every point of resist is as good as the last. But let's say you have target a with 1000 health and 10 armor. Effective hp = 1000*(100+10)/100 = 1100 Your dps is 100, and then you deal 100*(100/(100+10))~91dps after armor Now if you have 5 armor pen, you get 100*(100/(100+10-5))~95dps If you have 10 armor pen, you get 100*(100/(100+10-10))~100dps Even though the difference between the 2 is always 5 armor pen, you get a slightly higher %dps from the 105 to 100 armor than from the 110 to 105 armor. That's how I always rationalized it. The flat effective health increase is the same for every point of resists you get. But the percents aren't the same. | ||
|
Slusher
United States19143 Posts
when you combine flat pen it becomes more of a post math figure, but after you calculate it all, getting that extra % means more and more dps as you approach 100%. I hope that makes sense. | ||
|
UnKooL
Korea (South)1667 Posts
| ||
|
sylverfyre
United States8298 Posts
On August 27 2013 12:52 TheLink wrote: Is this true? I believed this for the longest time but then there was that Reddit post arguing that every point of resistance is as valuable as every other due to effective hp. Didn't understand it at the time and still don't now but he seemed to make some good points. The diminishing returns thing (and vise versa) is apparently an illusion. It's true that 1 point of armor is +1% of maximum health as Effective HP. The thing is, that calculation doesn't work in the same way in reverse, so 1 point of flat ArPen is worth MORE when it reduces someone from 1 armor to 0 armor than it is when it reduces someone from 101 armor to 100. (This happens because the operation involves division, and thus is not communitave or associative) | ||
|
Amui
Canada10567 Posts
Less armor you have, the more each point of armor matters. The corollary to this is that the more armor you're shredding, the more each additional point of armor pen is worth. No way to get around this unless you change the mechanic which works the same across almost every game with ehp/arpen I've ever played. | ||
|
justiceknight
Singapore5741 Posts
On August 27 2013 13:21 UnKooL wrote: For anyone interested sAviOr is diamond 3 in kr server. he's streaming right now on afreeca, and someone else is restreaming it on daumpot. he mains ad and he blocks/ignores almost everyone in each game if they talk about him. his stats: http://op.gg/summoner/userName=ibiza beach stream link pls ^^ | ||
|
upperbound
United States2300 Posts
On August 27 2013 12:52 TheLink wrote: Is this true? I believed this for the longest time but then there was that Reddit post arguing that every point of resistance is as valuable as every other due to effective hp. Didn't understand it at the time and still don't now but he seemed to make some good points. The diminishing returns thing (and vise versa) is apparently an illusion. It's not an illusion. Each point of resistance increases your EHP by 1% of your maximum HP, not your maximum EHP. Still, this is definitely better proved semi-rigorously rather than conceptualized; Okay, so the formula for your damage multiplier against an enemy is 100/(100+armor-armorpen). Let's assign an arbitrary value x to armorpen. If armorpen gave truly even returns to scale, the ratio of adding 1 additional point to armorpen should be independent of this starting value, x, considering all values of x>=0. That doesn't happen, though. Proof: dmult(x)/dmult(x+1) = [100/100(100+armor-x)]/[100/100(100+armor-x-1)] = (99+armor-x)/(100+armor-x). This ratio clearly depends on the value of x; you can't manipulate this to get the x to cancel on top and bottom. If the damage multiplier were an exponential function instead, you might see even returns on armorpen. Additionally, it's easy to see that this ratio is less than 1, meaning that dmult(x+1)/dmult(x) > 1, meaning that each additional point of armor pen is more effective than the last. So yeah, if you look at the math, each point of armor pen is both (1) not equal in value; (2) always worth more than the last one for any non-negative value of armor pen. EDIT: Kennen'd by several | ||
| ||
