A Resistance Gear Test

Gear: Stormsoul Ring, Caster is RDM99 Capped Skill, Chatoyant staff, dINT50, testing Break.



Test 1, Stormsoul Ring x1:

11/1001
1.09% Resist Rate


Test 2, Stormsoul Ring x2:
45/1100
4.09% Resist Rate



Is there some kind of non additive effect going on? So assuming with both rings on it is a 2%~ rate, 4% is kind of strange.

Could it be that with both on you get two checks to resist, both with a 2% rate?

(I'm rather confused by this.)

So any input is appreciated.

http://www.measuringusability.com/wald.htm

Sample 1: 0.0059~0.0198
Sample 2: 0.0306~0.0544

We don't really know how resists work, but we do know that Resists are strange in Ballista. They might even be non-linear, who knows? To find out, you'd want to determine the real resist rate elsewhere and then see how Ballista affects it.

Well I have no base traits, only the rings, and the ring seems to have a base rate by itself of 1%.
When I put two in the effect seems to be amplified.

In Ballista resistances are "enhanced".
I think the effects are non-linear, or multiply.

From the results above what do you predict? That you get multiple checks per piece?



It isn't your fault, but I'm not quite sure what your post was getting at. The numbers.

Sample one can be anywhere between .59% and 1.98%, sample two can be anywhere between 3.06% and 5.44%.

Well I'm compiling data on every piece individually at 1000 casts, then I'm going to test a few variations to see if anything strange happens.


I'll probably need help interpreting the data I collect.

*Thank you for the website, it helps out quite a bit.

My first guess would be exponential gains, similar to haste.

(1/(1-.015) - 1)x100

(1/(1-.03) - 1)x100

Would be the closest I can think of and falls within margin of error. Most likely the .5 gets floored, too in the first sample.

I'd be very surprised if it was anything but additive under normal conditions. Ballista isn't normal conditions, though.

The above is what I have been theorizing for a long time. When I get more individual pieces I'll begin with various setups.

Don't exactly have enough information to rule it out completely, though, and it is SE. If he was willing to do it again with augmented black/dark earrings, assuming they're 1% per resist petrify, then we could at least see if stormsoul are more than 1%.

Well to be honest I believe gear operates the same.

The augments in PvP are, job traits are greatly enhanced, and you gain diminishing returns to 99% of enfeebling spells.




I'm 350 casts into unfettered ring, it seems to be 13~%, we'll see when I get to 1000. Then I'm testing a +3 earring, then two +3's.

Unfettered Ring:
902/1011
9.8% - I'd safely say 10.

have you put any effort into testing levels of wind resist+? id also be curious the check order of that, mnd/int difference and raw resist silence

I'm not confused it's seems to be 5% resist rate oer ring, but due to confident interval, that would be hard to tell exactly.

I'll suggest you to read my magic resist mage on the wiki. All spells get a base landing rate. The problem is that they get partial resists, and break is no different. If you assume that break as one half-duration resist state then

change of no resist : p
chance of 1/2 resist : p(1-p)
chance of full resist "resist message" : (1-p)^2

If you backsolve the equation you find
p=90% landing rate with one ring
p=80% landing rate with two rings
since the cap is 95% I think it's likely -5% per ring, as in:

one ring would give (1-0.90)^2= 1% full resist
2 ringd would give (160.85)^2=2.25% full resist
would need higher % to know the exact value but basically there is no problem with the effect being non linear as the expression of full resist rate is quadratic ( (1-p)^2).

If you wanted to observe a linear difference you should have registerd the number of 1/2 duration resists too.

It gives Resist Petrify job trait, which seems to bypass magic accuracy checks and just has a straight % chance to full resist. So it doesn't play in to the magic hit rate system at all.

And ...? Those job traits affect landing rate. All they do is change the upper cap value.

Like this:
* You cast Break on something with 10% Resist Petrify
* 10% of the time it "Resist!"s
* 90% of the time your MAcc is compared with its MEva to find the Magic Hit Rate, and then the resist state for the specific spell is chosen.

So it's affecting the resist distribution by adding a bunch of full resists. These "Resist!"s are nicely split out in the log, so it isn't even hard to measure.

A clue might be in the elemental resistance behaviors. For example, adding earth resist will increase your resistance against petrify, so it's not hard to imagine they function similarly as well.

With elemental resistances, nobody has really pinned it down before, but all testing has indicated that there are "tiers" of elemental resistance. For example, let's say you're facing a 1000 damage earth attack. Let's say that at 120 earth resist, you only take 600 damage instead. What folks realized is that you can go up to 140, or even 180 earth resist and still take maybe only 500 damage. However, if you can make it to the tier of around 200-220 earth resist, you'll suddenly start taking 200 damage instead.

So it's not completely tiered, but it's not linear or even a simple curve of increasing returns, as far as we could tell.

I wonder if resist ailment effects are the same, in that your second Stormsoul is putting you in the next tier of resist and that's why you're seeing a larger-than-expected jump?

Player break has no partial resist.

Resists are always 1/2, 1/4, 1/8, or 1/16 in terms of magic damage. It's not that a certain point gets you a huge jump, it's that as you add resistance you have a higher chance of 1/8 and 1/16 resists, which appear to be huge differences in damage. Pretty sure Pchan HAS pinned it down at the fire bomblet NM, or at least done some decent testing there.

I don't believe there's any testing supporting it being tiered, either.

Obviously it does, go learn to play FFXI.

1/16 resist doesn't exist.

Fire resist work like magic accuracy. +1 fire resit on you = -1 mac on target.

It really doesn't.


Byrth is right about the bypassing of the magical hit rate formula.


It seems that in order to get any benefit form elemental resist you'd need to stack a very large amount. It does not seem to be linear.


Neither of these do however.