Eva E/Ring

the ocelot trousers back it up and my damage output lets me hold hate. In abyssea i use razed ruin + gnarled horn + alpha omega for DD, if im tanking tanking i replace it with mounted champion. I hold hate just fine.

Hate caps, doesn't take long to cap, enmity-7 won't really change that. Once capped it's not going to affect you at all.

i hate to be "that guy" but you parsed this on pld.... pld's evade is absolute trash regardless...so doesn't that kind of kill the experiment? plds are definitely not known for evade tanking and i'd think adding some skill+ to their already abysmal skill may not even help that job

Isn't that the point, "that guy"? Take something that has no native evasion traits to test evasion on. If you take a job that is already evasive and add more evasion to it the results will be less obvious.

The point was to compare the difference between skill and straight stat, only.

Pretty sure I saw a test a long time ago that said the same thing, 1 evasion -> 1 skill.
Dont remember how much less it was (not much), but yeah, evasion earring = trash.

Look at page 1 for my test. I agree, evasion earring is garbage.

Yeah, that showed the same thing as your test, sorry :P
I dont remember where I saw that other test, but it was years ago >_<

Maybe it's just ignorance, but I think I'd use Patrician's Cuffs over Slither Gloves (+6 eva skill vs. +9 evasion).

I blink tank on DNC, not sure if there are any high +evasion boots; but I use Emissary Boots (+5 eva skill/parry skill).

Is there a rule of thumb on when to stop stacking evasion skill and go for +evasion?


This is what I use currently on DNC.
Kila is a wip on Added.: Acc Down
And I'm working on finding a new earring to nix the Suppa.

No, 1 evasion -> 1 evasion skill.

Hmmmm, OK.

I do need to work on the set; but it's been decent to me so far. Those Slither Gloves do look nice. As well as that +12eva Beret (idr name).

Any recommendations on feet over Emissary?

I have dnc as well and before my slither i had patricians, i sold my patrician cuffs and dont regret it one bit. the added subtle blow is a bonus.

for feet emissary still about the best you'll get for dnc

I prefer ballerines.

Forgive me if I misunderstand your conclusion, but it was evasion>skill?

Assuming that's what you implied, it looks like an incorrect conclusion as your two samples fall within Margin of Error(3%~) due to the low sample size.

I went ahead and replicated your test for my own purposes.

In my sample, the two stats more or less came out as the same increase overtime. Over the parsing period my evason fluctuated from between 52.70~ and 52.90~ on both sets. It's pretty much safe to say they're equal if added solely, not sure how they interact if both are added in the equation.

If you care about my methods it's under the spoiler:

I like the added parry bonus from emissary~ but i do have ballarines for TP

If I might ask, what mobs were you letting hit you at Lv.90? If you are evading at 50% on BLM, something is iffy...

Same mobs as you, and I was on RDM not BLM.

Edit: My evasion is capped @280 before gear.

Right, I was thinking RDM but my fingers typed BLM.

Well the difference between Evasion and Evasion Skill might not be linear. You're base evasion skill will be higher with the level increase. Also, there is a significant difference between Lv.75 and Lv.90 vs the same mobs.

Maybe I'll head back there once my PLD hits Lv.90 and compare my own results using the same test. If they show inconclusive results like yours I'll have to find some high EM low DC mobs to run the original test on again. Because honestly, who cares about the difference in evade rate vs low EP or Too Weak mobs? In a perfect setup and perfect test, I'd be able to perform this +10 Evasion and +10 Evasion Skill against every level of mob and every mob family. That goes way beyond the scope of this test.

As far as I'm aware there's no level correction on evasion.

The /check rating shouldn't matter the least as my evasion is neither at the evasion floor (20%) or the ceiling (95%). Your test may have had evasion floor as it falls within the Margin of error, however I doubt it did. :P

Disregarding that, the only thing that matters should be the fact they both raised evasion rate, and by the same amount. Factors such as how much it went up, base evasion, on what mob, what color the moon was shouldn't matter as the test was designed to see the percent increase of the two stats increasing independently of each other.

However I'm amused at my tests being "inconclusive" and you feel the need to rerun the original test despite mine having 10 times the sample size and larger variable increase. The evasion base change is likely nothing more than the 5~ evasion you gain on each level.

It wasn't meant to be condescending or disrespectful. I simply want to recreate the same setup using the same gear at the new level to verify my own results seeing your new results come to light. Nothing more.

Fair enough. :)

I always thought that Evasion VS Evasion Skill acted in a way similar to Attack VS STR (just as an exmple), where the 1st woud reach a Cap sooner while the latter allowed you to push to a higher tier.

Naive, I know ._.'

Btw Hagino, those are some nice gear builds you have there, congratulations :D

Also, just to toss it out there. According to Wiki there is a Level Correction penalty imposed. Not saying to treat it as gospel, but just something to think about.

http://wiki.ffxiclopedia.org/wiki/Hit_Rate#Level_Correction

Pld's evade rate being low means nothing he tested it in a controlled enviorment where only 1 variable changed, the boxers and the o hat. Thats howa test works.

Now if he did a test on easy mobs like this with thief good chance hed be capped on both setups main reason a job like pld is easier to test things.

This is essentially the same test I did but I did way more than 1000 hits a test. I also want to point out the margin of error created by random chance. Both setups (10 eva and 10 eva skill) in my test showed approximately a 5% boost in evade rate. Which is essentially what your test show also.

Think of eva skill more like accuracy and combat skill gear. In the case of these we all know accuracy is slightly more potent than combat skill(for acc%) at high levels. This however is much more easy to test.

Honestly my test and these other guys tests show that eva basically = eva skill. However is is also possible at high levels 1 eva skill = .9 evasion.

Good to know, I thought there was but couldn't find anything to confirm it so I thought I was mistaken.

I'm not a statistics major, obviously. What would be a good number of attack rounds to shoot for when I do this test again to compare results? I thought 1000 attack rounds would be enough, but I was called out now for having a low sample size.

Thank you, I do believe that its non linear and that after a certain point straight evasion > skill

in my evasion set i currently sit at 386 skill

Its not that ur sample is low, its a good test but a 3% difference can easily just be a margian of error, there are a couple things you can do to make the test a little more stable/ conclusive.

Try 20 eva gear vs 20 eva skill gear, the difference in results would be more pronounced if straight eva was in fact better.

Or you can simply let the test run longer and watch the parse waiting for your evade rates to stop fluctuating as much.