ARR Attack Magic Damage Test (beta3) Int, dtr & acc

[Xatsh]

Did test with determination as well this last weekend.

Not large enough to be exact figures, but in the test with drastic changes in determination we expected to see some change... it was literally an average dmg change of 0.14% per nuke average from +0 determination to +36 which is easily in the margin of error. (a few points of dmg more or less)

Used gear without int and to test DET increases to gage if there was a dmg increase or not. (did a sample of 30 nukes per each setup... yea not large enough yet but just started towards the end of last beta)

+0 +5 +10 +15 +22 +36

For each of these we shown almost no correlation to dmg output from thm using only blizzard 3 on level 31 mobs. With int, we saw drastic increase per added stat per test.

So I am not sure if determination effects dmg in another way but for now it seems determination has no impact on dmg output for mage nukes. Was going to test and see if it effects DoTs or maby some other factor in the dmg output. But int is vastly superior to det based on initial testing.

This will be important to know when materia comes into play with the low stat caps for materia gear. Most mage gear can only hold 1 int materia. So need to see if mp, spell speed , det, or Crit will be better.

[Laevetine]

I think that is the kicker is we can not stack enough Determination to get outside a +-5% margin.

So Its looking like its not doing much.

From melding a few +0s on over the weekend I am not getting my hope up for materia.

[Crevox]

Did you forget that cleric stance also increases "potency of attack magic spells by 10%"?

[Deli]

Did you forget that cleric stance also increases "potency of attack magic spells by 10%"?

who? me?
every single stone II was boosted by 10% ~= no boost.

unless the 10% was added only in special part of the formula, so it will be, for example,
damage formula for stone II = weapon dps * {0.014 [(idonno) log(int) + a]} * (1+10%) * success % * (how many cookies you eat) + {0.007 [log(dtr) +b]}


/muttering on, minds off topic...
btw, it's not completely guess tho...
I mean there are strong scientific reports, from K country, that shows eat cookies helps your dps in game...

ok that part is fake (or is it...?), but I don't think it's a linear formula. They want you to see the lv up benefit in the beginning of the game, so a shifted log plot is pretty good I think.


Come to think about it, dodge that cracks in that 2 seconds is way more important than spent 14 days to get some super shinny gears that add 5 more int...

Also...ifrit has strong att, garuda need lots acc, so titan... ya need more cookies...
/muttering off

[Deli]

after ~ 8 hours tests, updating the results...

[Hvinire]

This is some thorough research! We asked the dev. team if they could give us a comment on the use of the determination stat in relation to what you were researching.

They remarked that at high levels, you should definitely be able to feel the difference when you enhance DTR through gear.
At low levels though, you're right that there isn't much of an effect when increasing DTR.

[Nix]

This is some thorough research! We asked the dev. team if they could give us a comment on the use of the determination stat in relation to what you were researching.

They remarked that at high levels, you should definitely be able to feel the difference when you enhance DTR through gear.
At low levels though, you're right that there isn't much of an effect when increasing DTR.

Ahhh Cactoooos, you're awesome! <3

[HiirNoivl]

Oh Wow!

Hvinire just put down some seriously useful information!!! :O

Let's go to the leader board....

*ding*

The audience rates this dev post as "F A B U L O U S"

[Molly_Millions]

Good stuff! It confirms what's in my guide

http://www.abaddon.us/phpBB3/page.php?p=stat-guide

[Hulan]

Good stuff! It confirms what's in my guide

http://www.abaddon.us/phpBB3/page.php?p=stat-guide

Hey Molly, in that guide, you suggest Dex as a tanking stat for the damage mitigation. I did a quick and dirty Dex and Parry test this last weekend (that, I admit, is hardly a rigorous examination of the stats, but my time and resources were limited, I plan on doing a much more in-depth study in Phase 4). My preliminary results show extremely disappointing returns on Dex and Parry on block rate, with a nice bump in Parry rate from the Parry stat only.

Originally I posted the results hyperbolously claiming "Dex is worthless", and I regret that claim now, as it clearly has worth, but until I have time to test further, I found Dex to be a marginal boost to damage mitigation at best (at least at entry levels of Dex). I'll repost my findings here, though when compared to Deli's I'm afraid the professionalism is somewhat lacking. When I do a more in depth test with more data points, I'll calculate a line of best fit and R² values.