Feelings on Dex?! (Gryphonskin accessories)

[Faeron]

With respect to bonus points, there's no way you can catch up -- you should put those points into STR or VIT. Secondary stats are not nearly as valuable as primary stats -- if you choose between +5 STR and +10 DTR, you'd be a damn fool to take the latter. If 1 DEX is worth 1 Parry, then you're still effectively comparing secondary to primary stats because you don't care about the primary portion. However, Gryphonskin or Rose Gold will both have the same VIT, either by melds or natively.

But with Gryphonskin you pick up a ton of str and dex, while maintaining VIT and only losing a few parry.

[Dhex]

If you've got the time and money to dump into Gryphonskin melding VIT/DET/Crit/Parry then go ahead - but doing it specifically for the DEX is stupid. The gain of having all these stats would be hardly worth the time, money, or effort. You can dance around this as an idea but what's the point? Are you trying to min/max - or is it you want a set you can use across all your DoW?

[Gamemako]

But with Gryphonskin you pick up a ton of str and dex, while maintaining VIT and only losing a few parry.

You would meld STR onto Rose Gold rings otherwise. It's really just the difference of DEX&SSPD on Gryphonskin versus any one other stat of your choosing on Rose Gold.

If you've got the time and money to dump into Gryphonskin melding VIT/DET/Crit/Parry then go ahead - but doing it specifically for the DEX is stupid.

So you would just forgo +Parry because "it's stupid"? If Dexterity affects parry rates, then there's really no difference. Both ilvl70 rings cost the same number of philo items.

[Dhex]

-

I ask because spending all this money on a set that is marginally better after a significant amount of melding is going to leave him dissatisfied - the gains aren't as huge as you think.

Just swapping your 30 point allotment around can show you you'd need more than just 30~ stats to make a change (regarding DEX/Parry/DET) - for example every 10.5 (alternating tiers basically 10/11) Skill Speed is .01 off your GCD for example. Thal's balls that's 100 SS for .1 ~ all MNK should stack that SS for posterity!

The same applies for DEX 29/30 DEX ~ 1/2% or +.01 to that there Parry Rating. If it's worth doing it to you - go ahead. Most people will look at that and scoff - as I do - because it's stupid to sink so much effort into something that will not only cost you time/effort/money/patience but it will likely be rendered obsolete when new content gear and new craft gear get's added a few months down the road.

It may cost the same Philo - but it costs gil or time to get Materia. It takes having a craft or a friend with a craft and the catalysts - and for what? Marginal increases - a more even distribution of stats across a wider range of stats. Minimal or unnoticeable gains.

I'm not going to spend the time to build up a set like this; but kudos to the OP or yourself if you do. (:

[AppleGrocer]

Honestly, I question the value of parry as a whole.

The only information I can find regarding parry rates is:
http://valk.dancing-mad.com/?page_id=179/#Parry

This is old stuff, and might not be up to snuff, in fact I kind of know it isn't. I need to test the numbers on low levels of parry (shown in the graph), but there's no way I'm getting a static .076% chance to parry per point.

I have ~532 parry on my War iirc, and my parry rate sits somewhere around 20% from my own little testing runs. So my best guess, pending verification of the low levels of parry numbers, is that there's either a huge drop off from the benefits gained per point of parry after ~300 points, or its worth almost nothing.

The scaling of Dex in relation to parry can only be worse...

[Ladon]

I wish people would understand that damage avoidance is pointless for a tank unless it is 100%.

If tank 1 has 7000 HP and a 30% avoidance rate, they are still by far the better tank than a tank with a 60% avoidance rate and 5000 HP. You simply don't tank stuff on probability.

Damage avoidance is a bonus but should never be considered over more HP.

[themitey1]

Here you go:

http://forum.square-enix.com/ffxiv/t...zation-for-WAR.

For your reading pleasure, plus additional links to his resources.

[CianaIezuborn]

In Valk's old testing STR gains to parry/block reduction jumps in tiers, and each tier is 1% reduction. Lets assume that holds true for DEXs effect on parry/block chance also. Let's then assume that going from the Allagan ring to a crafted ring would net you a jump in BOTH STR and DEX tier. Furthermore, lets assume that your block/parry rate and reduction are all currently at 25%. I'm making this as best case scenario as I possibly can, and these parry/block numbers are likely high. I have not done any testing on these rates, I'm just doing some theory crunching.

Note that I am leaving out the mitigation(def/shield oath) portion as this will multiply both numbers by the same amount and I'm trying to pull the percentage difference which will not effect the calculations. This is to slightly simplify the numbers.

Some quick dirty numbers:
Sunseeker full i90 paladin with Allagan ring

HP: 7005
Parry: 25%/25%
Parry Reduction: 6.25%
Block: 25%/25%
Block Reduction: 6.25%
Total Reduction: 12.11%

EHP: 7005
AHP: 7970.13


VS

Sunseeker full i90 paladin with Crafted Ring

HP: 6930
Parry: 26%/26%
Parry Reduction: 6.76%
Block: 26%/26%
Block Reduction: 6.76%
Total Reduction: 13.06%

EHP: 6930
AHP: 7971

EHP difference: Allagan +1.08%
AHP difference: Crafted +0.015%

I may have overestimated the Allagan HP by a slight bit due to not being completely sure how the +3% group bonus works, however the fact of the matter is this:

EHP is more important than AHP. For any EHP increase to be sacrificed for AHP increase, the AHP increase must be vastly superior. On the scale of 5 to 10 times as much to even be entertained, and then only when EHP is deemed reasonably sufficient to the highest possible burst encounter. Given these numbers, that I believe are inflated towards Avoidance, I don't see there being any reasonable use for STR/DEX rings in the near future as an upgrade for tanking.

[Gamemako]

I ask because spending all this money on a set that is marginally better after a significant amount of melding is going to leave him dissatisfied - the gains aren't as huge as you think.

I never said they were large. I really don't know what you're on about here.

Just swapping your 30 point allotment around can show you you'd need more than just 30~ stats to make a change (regarding DEX/Parry/DET)... The same applies for DEX 29/30 DEX ~ 1/2% or +.01 to that there Parry Rating.

You clearly have not tested this at all. If you look at what meager data exists, Dexterity would be worth more than Parry. We do not have anything resembling a working formula at this time, but it's clear that "+0.01 to that there Parry Rating" is off by at least two orders of magnitude. Ninjiitsu's testing found an increase of 3.5% parry chance (relative 14.1% increase in parrying rate) by adding 30 points into dexterity.

None of the data is all that good, but for what's there, DEX adds a non-trivial portion to parry chance. We just don't really know how it works.

I wish people would understand that damage avoidance is pointless for a tank unless it is 100%.

That is incorrect. A tank who can withstand all hits in the game is ineffective if it cannot reduce healing load -- healers just go OOM and your party dies. eHP is not the end-all-be-all, it's just one facet of tanking. You must have aggro generation, burst eHP, and mitigation all if you are to be an effective tank.

In Valk's old testing STR gains to parry/block reduction jumps in tiers, and each tier is 1% reduction. Lets assume that holds true for DEXs effect on parry/block chance also.

That is not a valid assumption. It would be best to consider a possible build and see if reaching a parry strength tier is feasible, then consider a possible parry rate increase as a result. The real problem here is figuring out how much VIT is really needed, and however much you can avoid using can go into other stats. If you need all the VIT to withstand a hit, then it's a bit of a meaningless question.

EHP is more important than AHP.

AHP is not a valid metric. It would be something like Schrödinger's cat -- somewhere, that tank is both dead and alive.

[CianaIezuborn]

That is not a valid assumption. It would be best to consider a possible build and see if reaching a parry strength tier is feasible, then consider a possible parry rate increase as a result. The real problem here is figuring out how much VIT is really needed, and however much you can avoid using can go into other stats. If you need all the VIT to withstand a hit, then it's a bit of a meaningless question.

For my purposes, this is more than a valid assumption. Unless someone proves that avoidance effects from str/dex are MUCH higher than the values I assumed, then I'm not interested in str/dex stacking. Once ilvl100 gear is released, it'll need to be examined again, however until then unless someones measurements cause me to make orders of magnitude changes to my assumptions, I'm good.

This is equivalent to building a bullet proof vest out of paper. If I can see that one sheet of paper only absorbs a 1/100,000 of a bullets impact, then I can assume a 100 page book isn't going to stop a bullet. Until someone proves that I'm off by a factor of 1000 I'm not going to try and stop a bullet with a notebook.

AHP is not a valid metric. It would be something like Schrödinger's cat -- somewhere, that tank is both dead and alive.

I agree, it's not a valid metric when compared with EHp. That's why it needs to be ORDERS of magnitude greater than EHp (which is 100% valid and 100% reliable) for it to ever be considered over EHp. I'm not sure what we're arguing here. I said A > B, then you said B is a half dead cat?

Who has the DeathStar >> Sith lord>> Storm trooper picture thingy?