WAR Dex and Parry testing!

[Ninjiitstu]

So after a great deal of debate on what DEX does, and the limited testing results we have from sources, I decided to spend today gathering some data for the discussion.

I ran tests with 5 different gear sets, registering 5000 hits per gear set. ( I actually did more, as it appears unless you are directly facing the mob, you lose a significant amount of parry, just fyi )

For simplicity, I used level 37 Wolfs near camp dragonhead. I could AFK on them with no issues, and they did not have any dot effects that would have skewed the numbers. Since Parry is a simplistic check, the value of parry % should not be a factor with mob level. I did have to do the math manually as the game uses a 1 hit skill check.

Test 1:

189 Dex
440 Parry

Registered 24.88% Parry rate.

Test 2:

216 Dex
456 Parry

Registered 23.19% Parry rate.

Test 3:

216 Dex
486 Parry

Registered 25.08% Parry rate.

Test 4:

246 Dex
486 Parry

Registered 28.63% Parry rate.

Test 5:

266 Dex
456 Parry

Registered 27.91% Parry rate.

Results Interpretation:

Contrary to the previous results given on the Methodology, it appears that the walls for DEX are much higher then the +-15 recorded previously. There is little to no difference between the first three tests, even with a 27 dex and 46 parry increase from test 1 to 3.

However, once hitting 246 DEX, using test 3's gear setup and +30 dex from attributes, we got a 3.56% increase in parry rate. I would interpret this as a +-1% for variable sake, so a marked increase of around 2.5% would be fairly safe.

Somewhere between 216 and 246 is a DEX wall that increases parry fairly significantly. I would posit that it is closer to 246 then 216, as even going up to 266 did not allow me to hit the next wall.

You can reach this cap by wearing 5 Gryphonskin STR/DEX accessories, and possibly reach the next wall with a few HQ Darksteel, most notably the pants allowing for 30~ dex to be melded onto them.

Someone else will have to do the math to see if they are worth it in EHP, but at this point I am now skeptical that Gryphonskin is BiS.

[Coramac]

Can you post the raw numbers? What do you mean you had to manually calculate?

Evasion has no effect on EHP.

[Ninjiitstu]

Can you post the raw numbers? What do you mean you had to manually calculate?

Evasion has no effect on EHP.

Those are the raw results. Hits / misses effect your parry rate. You cannot parry a miss. The APP takes the total attacks and does not take into account misses. So for example, if I got hit 5000 times, and parried 1254 of them, its 25.08% parry.

Evasion has a rather large effect on EHP, though I wasn't testing that and there is no real way to test it currently.

[Sephirah]

Evasion has a rather large effect on EHP, though I wasn't testing that and there is no real way to test it currently.

Technically the best mitigation is not getting hit in the first place.

(I'm agreeing with you btw, don't think I'm telling you you shouldn't actually be getting hit, lol, seeing "Dodge" pop up on mountain busters is a lot of fun)

[Coramac]

Those are the raw results. Hits / misses effect your parry rate. You cannot parry a miss. The APP takes the total attacks and does not take into account misses. So for example, if I got hit 5000 times, and parried 1254 of them, its 25.08% parry.

No, there's scant evidence of a multi-roll system.

Evasion has a rather large effect on EHP, though I wasn't testing that and there is no real way to test it currently.

Nope, it affects efficiency not EHP.

[Ninjiitstu]

No, there's scant evidence of a multi-roll system.

Go let a level 20 mob hit you for 30 minutes. If you think your parry rate is actually 6%, like the app shows, then there is no convincing you anyway. :-P

Nope, it affects efficiency not EHP.

That is not correct. Over the course of many encounters your static X% to dodge will DIRECTLY impact your EHP by the same %. It is the same thing as parry or block reduction, only more effective. But why are we even arguing this? I wasn't testing it to begin with.

Finally: It appears tomahawk is effected by BOTH STR and DEX. Getting some pretty sexy numbers from a skill that is otherwise destined to pull mobs. Not really relevant for PvE as you need to maim anyway for DPS, but might be a factor for PvP.

[RapBreon]

Finally: It appears tomahawk is effected by BOTH STR and DEX. Getting some pretty sexy numbers from a skill that is otherwise destined to pull mobs. Not really relevant for PvE as you need to maim anyway for DPS, but might be a factor for PvP.

As in; DEX also increases its damage on account of being ranged? Any ballpark numbers in comparison to say Heavy Swing? Because this is very interesting.

[Coramac]

How did you treat crits? The app doesn't count them in normal hits. I do no know whether it is calculated before or after parry. I have some data suggestive of before but nothing definitive.

There is some very odd things going on with these numbers assuming that, like block / parry mitigation, FFXIV only uses two decimals for rates. #5 really just doesn't fit. With the prior mentioned assumption, it looks like you cross three dex thresholds which I don't think should be possible given #2 results (not impossible, but unlikely). A possible consideration may be that Dex and Parry rating are not treated separately for the purpose of determining the rate. Although, that makes #2 hard to reconcile.

[Kaalan]

Just have to compare 2 with 5 and 3 with 4. 1 has no point and you gotta look at DEX with equal parry or else it doesn't mean anything.
But yeah, seems obvious that there's a threshold around 240. Thx for those numbers, will help with the gear (and this time it clearly puts DPS rings ahead of tank one)

[Coramac]

Just have to compare 2 with 5 and 3 with 4. 1 has no point and you gotta look at DEX with equal parry or else it doesn't mean anything.
But yeah, seems obvious that there's a threshold around 240. Thx for those numbers, will help with the gear (and this time it clearly puts DPS rings ahead of tank one)

Test #1 is critically important.

I agree with Ninjii, if anything, this makes the DPS rings worse rather than better.