Parry...

[Casper]

If people don't want to take my results into account, I'll just do the most obvious test, and show point per point what my Rage of Halone combo can do exact number wise, with and without the Crit/Determination gear. I'll get all the statistically high and low numbers in , without any buffs at all.

Alright. I did my own testing. I went in no stance, no buff, on dummy 50, and i used fast blade (no halone combo or anything, just fast blade) repeatedly. I did this until i had what seemed to be max and min, after about 50 use, so while i could miss one point at the beginning or the end of the range, i think i got it all overall. I discounted crit and SS for now, just watching the effect of det and str:

Test 1: Str 364, Det 211: range 153 - 169
Test 2: Str 377, Det 211: range 157 - 173
Test 3: Str 372, Det 266: range 161 - 177

The progress from 1 to 2 and 2 to 3 is about 4/160 = 2.5% increase overall, and 5% from 1 to 3.

A 13 str increase corresponded to a 2.5% increase, which means 5.2 str would improve my damage by 1% overall, or that each point of str is a bit less that 0.2% increase in dps.

You can also see that 13 str increase my dps by 4, and 8 str and 55 det increased it by exactly 8, which means that 26 str is roughly the same than 8 str and 55 det, or that 18 str is roughly the same as 55 det. So 3.05 det is roughly equivalent to 1 str. It would thus take 15.86 det to increase damage by 1% according to this test.

Now regarding skill speed, 10 to 11 SS reduce your GCD by 0.01, which is an increase of 0.4% in your dps. So you need about 26 SS to increase your dps by 1%.

Regarding crit, you need about 14.5 crit rate to increase your crit % by 1%, which increase your dps by a bit less than 0.5%. So about 30 crit rate are necessary to improve your dps by 1%.


The difference of gear you proposed is trading:
115 parry (115 / 13 = 8.84 parry % chance, assuming you are a pld and block 21% of the time with weathered noct shield, and parry for 24%, comes down to a 1.68% damage reduction)
and 1 str (0.18% damage)

to win

24 SS (0.92% damage increase)
46 Det (2.9% damage increase)
45 crit (1.5% damage increase)

The four damage chance compound for a total change of 1.015 * 1.029 * 1.0092 * 0.9982 = 1.052, so a total of 5.2% increase in damage, which should come to about 6 dps increase if you are hovering around 120 to 130 dps.


Now you have hard numbers. While some of these numbers are not 100% accurate, they are close enough. Could there be a 20% difference from what i said ? Sure. Here you go. There might be a 7 dps increase from your 130, i'll even give you 8. Could there be a 400% difference to go from 5% increase to 25% or something. Nope. Nope. Nope. Sorry, just no way.

I urge you to do the test yourself and post your results back in there. Use your 2 sets (you obviously have more leeway on dps gear, i don't have that many options), go on the lvl 50 dummy west of coerthas, get the whole range for unbuffed unstanced fast blade with both sets, and provide your numbers. We should be able to tune a bit more finely the equations i used. In any case, i am quite doubtful you will come to a different conclusion than me. I don't know how you obtained the result you posted, but there seems to have been a massive bias somewhere.

Now the question of wether 1.68 % DR is worth 7 dps is up to you. For me, it is, all day long, but it is only my personnal opinion here.

[bokchoykn]

Now regarding skill speed, 10 to 11 SS reduce your GCD by 0.01, which is an increase of 0.4% in your dps.

Skill speed only increases your damage output for Weapon Skills, not auto attacks. It would actually be less than 0.4%. It also consumes your TP at a faster rate, so in a fight where you can run out of TP, its value is somewhat diminished.

115 parry (115 / 13 = 8.84 parry % chance, assuming you are a pld and block 21% of the time with weathered noct shield, and parry for 24%, comes down to a 1.68% damage reduction)

1.68% damage reduction against attacks that can be parried. Not all attacks can be parried. FYI, 25% of the damage our Paladin took in Turn 8 this week came from unparryable sources, and these are the sources that are most likely to kill him.

Also, because of the unpredictable nature of parries, the DR is actually nearly useless in terms of increasing your Effective HP, since this extra 1.68% may or may not occur when you're about to get one-rounded by a sudden burst from the enemy. You can't rely on Parry to help you here. You can only rely on MOAR VIT. For that reason, 1.68% DR from Parry is much less valuable than a constant 1.68% DR. Because Parry DR works this way, its only benefit is saving MP for your healers in the long run. Not for actually increasing your effective HP. This is very, very important to realize about Parry and block.

Anyway, whether I would rather take a 5.2% increase to damage output over a 1.68% damage reduction from Parry depends on the fight and what role I'm playing, but for most content in the game, I would honestly probably take the damage.

Anyway, this test is a very extreme case where you're trading 113 points of Parry for the equivalent in offensive stats and not simply minor gear decisions here. The differences either way are minimal and there are merits to both. So why are people so obsessed with Parry-based BiS lists? I've been wondering this since 2.0.

Casper has concluded that he likes the all-Parry build. ZDamned concluded that he likes the all-Damage build. My conclusion is WHO CARES?! If the opposite ends of the f---ing spectrum equate to 1-2% DR vs 6-7 DPS, we are splitting hairs here. Strength and Vitality are what actually matter. Have enough Accuracy but not too much or else you're wasting stats. The rest is so inconsequential.

Thanks for doing the testing, Casper. This is great info. I'll try something similar on my Warrior.