Evasion---Hard Counters with nothing to recounter

better you dont discuss with that guy, he seems to want to be invicible with an evasion character who runs around with 2k Life lol. So forget about discussing anything with that person. Damn Trolls.

What needs to be done for eva: change the (eva/4)^0.8 factor to (eva/4) and remove eva's ability to check "hit" twice against crit attacks.

"

tmaciak a écrit :

Very good theorycrafting Autocthon, congratulation.

But small question. Did you have possibility to check those theories on content >70 lvl? Because my practice shows, that even on low level maps, my eva characters are wrecked comparing to my ar characters.

At the moment I'm leveling 3rd test eva character, this time I will be able to get Acrobatics (previously I rolled EB/CI characters so obviously I didn't want to reduce my ES) but I have little to no hope that it will work for me. That's why I will be passing only 3 point away from IR :)

Because of the nature of EV and monster damage there's a heavy gearblock going into higher content.

BAsically if you want to do the same content you have to spend double on gear, and it's gear that doesn't exist so you'll be spending 3-4 times as much time looking for upgrades. The upside is that maxed enfeeble+Blind is near immunity to damage as long as you get above 54% listed evasion, so it can let you skimp on actual EV rating in favor of raw HP and resists.

"

mazul a écrit :

What needs to be done for eva: change the (eva/4)^0.8 factor to (eva/4) and remove eva's ability to check "hit" twice against crit attacks.

WTF makes you think that's a good idea? EV doesn't need better scaling per rating point. EV characters can already pretty simply hit the EV cap, the issue is not how often they can evade but how well they take a hit they don't evade. Which is currently "not as good as ES, and not as good as AR without spending 5x the effort finding gear"

"

Autocthon a écrit :

"

tmaciak a écrit :

Very good theorycrafting Autocthon, congratulation.

But small question. Did you have possibility to check those theories on content >70 lvl? Because my practice shows, that even on low level maps, my eva characters are wrecked comparing to my ar characters.

At the moment I'm leveling 3rd test eva character, this time I will be able to get Acrobatics (previously I rolled EB/CI characters so obviously I didn't want to reduce my ES) but I have little to no hope that it will work for me. That's why I will be passing only 3 point away from IR :)

Because of the nature of EV and monster damage there's a heavy gearblock going into higher content.

BAsically if you want to do the same content you have to spend double on gear, and it's gear that doesn't exist so you'll be spending 3-4 times as much time looking for upgrades. The upside is that maxed enfeeble+Blind is near immunity to damage as long as you get above 54% listed evasion, so it can let you skimp on actual EV rating in favor of raw HP and resists.

"

mazul a écrit :

What needs to be done for eva: change the (eva/4)^0.8 factor to (eva/4) and remove eva's ability to check "hit" twice against crit attacks.

WTF makes you think that's a good idea? EV doesn't need better scaling per rating point. EV characters can already pretty simply hit the EV cap, the issue is not how often they can evade but how well they take a hit they don't evade. Which is currently "not as good as ES, and not as good as AR without spending 5x the effort finding gear"

To make Evade Hybrids far more viable. The scaling of evade is lackluster currently. You can hit the EV cap only against projectile attacks and only if you go to that part of the tree.

ES & Evade hybrids would certainly benefit from a such change and simultaneously solving the issue you seem to have with evade characters not surviving big hits.

By removing eva's ability to check "hit" twice against crit attacks, you help out the eva crit builds since that way they aren't as badly affected by lack of accuracy as they are now. Less need of accuracy means you can spend more points towards survivabiliy.

(Of course the eva values of monsters would have to be adjusted as well:P)

Edit: Also to hit the 50% chance to avoid attack from evasion rating alone vs lv77 enemies (i.e. vs roughly 530 accuracy), you need currently about 10.2k evade rating. Which is quite a lot if you want to get ES too on your gear:P.

Edit2: With the suggested change, you would of course need to readjust monster accuracy.

"

By removing eva's ability to check "hit" twice against crit attacks, you help out the eva crit builds since that way they aren't as badly affected by lack of accuracy as they are now. Less need of accuracy means you can spend more points towards survivabiliy.

Can't really say that I agree.
Those evasion builds usually have a high dexterity. A high dexterity will give you a high chance to hit. I have no pure +accuracy node, I got Finesse from Ranger and two 6% DMG/8% ACC nodes. That's it, and I have 86% chance to hit. Not much points invested here, I'd say.
Scaling the monster evasion up again would throw player's accuracy a few patches back, lowering it would make achieving 90%+ too easy.

"

mazul a écrit :

To make Evade Hybrids far more viable. The scaling of evade is lackluster currently. You can hit the EV cap only against projectile attacks and only if you go to that part of the tree.

What you mean to say is "I don't want to use Blind or Enfeeble"

Hitting Evade caps against all attacks is only a matter of actually using the secondary multipliers provided to Evasion by skills. The multipliers analogous to those used by Armour characters without a second thought.

"

ES & Evade hybrids would certainly benefit from a such change and simultaneously solving the issue you seem to have with evade characters not surviving big hits.

By removing eva's ability to check "hit" twice against crit attacks, you help out the eva crit builds since that way they aren't as badly affected by lack of accuracy as they are now. Less need of accuracy means you can spend more points towards survivabiliy.

EV/Crit builds don't hurt for accuracy.

"

(Of course the eva values of monsters would have to be adjusted as well:P)

Edit: Also to hit the 50% chance to avoid attack from evasion rating alone vs lv77 enemies (i.e. vs roughly 530 accuracy), you need currently about 10.2k evade rating. Which is quite a lot if you want to get ES too on your gear:P.

Edit2: With the suggested change, you would of course need to readjust monster accuracy.

EV characters (including EV/ES) can realtively easily crest a +200% EV bonus.At this point Grace alone is worth 4k towards that EV and without any multipliers for EV on gear you're looking to another 2k+ from only base EV on the gear. That's 6k Evasion already.

"

Delirii a écrit :

"

By removing eva's ability to check "hit" twice against crit attacks, you help out the eva crit builds since that way they aren't as badly affected by lack of accuracy as they are now. Less need of accuracy means you can spend more points towards survivabiliy.

Can't really say that I agree.
Those evasion builds usually have a high dexterity. A high dexterity will give you a high chance to hit. I have no pure +accuracy node, I got Finesse from Ranger and two 6% DMG/8% ACC nodes. That's it, and I have 86% chance to hit. Not much points invested here, I'd say.
Scaling the monster evasion up again would throw player's accuracy a few patches back, lowering it would make achieving 90%+ too easy.

No high dextery alone won't give you a high chance to hit. You only get 2 accuracy per dex. You need accuracy from gear or nodes as a crit build.


Let's take a lv90 character as example:
You get (90 -1)*2 accuracy = 178 from leveling up

Assume that you got high dextery, i.e. 300 dextery. That means you get 300*2 = 600 accuracy from that.

=> A total of 778 accuracy.

vs lv77 mob: chance to hit = 778/(778 + 248) = 75.83%
(edited since the new factor is 248 and not 305.7)

Furthermore, instead of that 6% dmg 8% acc node, you can usually take a straight 12% dmg node; i.e. there is an opportunity cost involved in getting those 2 acc nodes.
-----------------

Evade builds hang around a place where it is rather easy to get crit nodes.

The factor you get from accuracy and crit is (all written in non-"%" form):
"Chance to hit"*{1 + "Chance to hit"*"weapon crit chance"*(1 + "Increased crit strike")*
(1.5*(1 + "Increased crit mult") - 1)}

By instead changing it to:
"Chance to hit"*{1 + "weapon crit chance"*(1 + "increased crit strike")*(1.5*(1 + "increased crit mult") -1)}

you would buff the damage of crit builds (indirectly evasion builds since most evasion builds would go crit due to those nodes being near).
---------------

So now what's wrong with how evade scales?

Well simply put, if you double your evade rating, your "increased effective hp vs evadable attacks" is only multiplied by factor 2^0.8 = 1.74

If you double armour on the other hand, your "increased effective hp vs armour-reduced attacks" is multiplied by factor 2.

Look what happens if we quadruple instead:
4^0.8 vs 4*1, i.e. 3.03 vs 4

And for that reason I want the 0.8 exponent removed.
-------------

"

mazul a écrit :

---------------

So now what's wrong with how evade scales?

Well simply put, if you double your evade rating, your "increased effective hp vs evadable attacks" is only multiplied by factor 2^0.8 = 1.74

If you double armour on the other hand, your "increased effective hp vs armour-reduced attacks" is multiplied by factor 2.

Look what happens if we quadruple instead:
4^0.8 vs 4*1, i.e. 3.03 vs 4

And for that reason I want the 0.8 exponent removed.
-------------

That's wrong for armour.

Damage Taken= (Damage*Armour)/(Damage*12+Armour)

If you have 50% "Damage Reduction"

50 = (100*1200)/(100*12+1200)

If you double your armour your new damage taken is (and your new DR)

(100*2400)/(1200+2400) = 33 Damage Taken
66% DR

Your eHP hasn't been doubled. It's been increased by 50%. In the case of evasion the exponent is to make Dex "better" with evasion than Str without breaking the curve. there is NOTHING wrong with evasion scaling.

"

Autocthon a écrit :

"

mazul a écrit :

---------------

So now what's wrong with how evade scales?

Well simply put, if you double your evade rating, your "increased effective hp vs evadable attacks" is only multiplied by factor 2^0.8 = 1.74

If you double armour on the other hand, your "increased effective hp vs armour-reduced attacks" is multiplied by factor 2.

Look what happens if we quadruple instead:
4^0.8 vs 4*1, i.e. 3.03 vs 4

And for that reason I want the 0.8 exponent removed.
-------------

That's wrong for armour.

Damage Taken= (Damage*Armour)/(Damage*12+Armour)

If you have 50% "Damage Reduction"

50 = (100*1200)/(100*12+1200)

If you double your armour your new damage taken is (and your new DR)

(100*2400)/(1200+2400) = 33 Damage Taken
66% DR

Your eHP hasn't been doubled. It's been increased by 50%. In the case of evasion the exponent is to make Dex "better" with evasion than Str without breaking the curve. there is NOTHING wrong with evasion scaling.

I didnt write eHP doubled, I said "Increased eHP" is doubled. Compare to going from 40% global crit chance to 80% global crit chance.

Futhermore, in almost all cases you are better off just having enough str, agi and int to equip your gear and gems. Any extra is usually a waste due to the opportunity cost.