more magic testing

**vivbiibasjiu** · 2015-06-05 00:10

Well that seems rather unbalanced. Wonder if SE is just lazy and puts their accuracy at 99999 since they figure no one will bother trying to evade it anyway

**Saevel** · 2015-06-05 04:57

Originally Posted by vivbiibasjiu

Well that seems rather unbalanced. Wonder if SE is just lazy and puts their accuracy at 99999 since they figure no one will bother trying to evade it anyway

It's been like that for years. Super NM's tend to have maximum accuracy regardless of what you do to them, it's SE's way of forcing you to deal with their NMs unbalanced moves. Of course I don't think they have realized yet what happens when you stack Vex and Attunement. Once they figure it out they will created NMs what ignore that and we'll have to figure out another way to deal with those same overpowered moves.

**Damane** · 2015-06-06 09:58

Originally Posted by Saevel

It's been like that for years. Super NM's tend to have maximum accuracy regardless of what you do to them, it's SE's way of forcing you to deal with their NMs unbalanced moves. Of course I don't think they have realized yet what happens when you stack Vex and Attunement. Once they figure it out they will created NMs what ignore that and we'll have to figure out another way to deal with those same overpowered moves.

this is not true because otherwise foil wouldnt be usefull at all (and gets overwritten by evadown moves). and foil lets you evade alot of stuff from Nm physical moves.

**Protey** · 2015-06-15 21:38

been getting weird results with carnage elegy vs magic/gear haste.

with carnage elegy only and no haste gear/spell, reraise came out with the expected 90 seconds.
with carnage elegy and haste spell (150/1024) was getting 81 seconds recast (expected).
with carnage elegy and 15% haste gear (152/1024 to be exact) was getting 88 seconds recast (expected recast to be closer to 81 seconds).

The 2/1024 difference between gear and magic haste should not make for 7 seconds recast difference (since gear haste is more, it would give a better result anyway).

It was like gear haste had a diminished effect when applied against elegy. For instance, even with 12% in gear haste I was still getting 90 seconds recast.

**Saevel** · 2015-06-16 05:13

Originally Posted by Damane

this is not true because otherwise foil wouldnt be usefull at all (and gets overwritten by evadown moves). and foil lets you evade alot of stuff from Nm physical moves.

Foil is in a separate category and has absolutely zero to do with physical evasion. It seems to be a flat percentage chance that the enemies special move, whether it's magic, physical or breath, will just flat out miss. Evasion on the other hand is subject to the same linear acc vs evade mechanic, it even has a level correction component in older area's. Mega boss's tend to have deliberately inflated acc values and has been that way since 75. Some older tanking strats involved using to do Ninja / Thief + double mambo + blind, before the utsusemi nerf and subsequent creation of ridiculously accurate targets.

**Byrthnoth** · 2015-06-16 05:59

Wrong, Foil is just evasion. Go check BG wiki's reference and keep reading for about a page.

**Damane** · 2015-06-16 14:42

Originally Posted by Saevel

Foil is in a separate category and has absolutely zero to do with physical evasion. It seems to be a flat percentage chance that the enemies special move, whether it's magic, physical or breath, will just flat out miss. Evasion on the other hand is subject to the same linear acc vs evade mechanic, it even has a level correction component in older area's. Mega boss's tend to have deliberately inflated acc values and has been that way since 75. Some older tanking strats involved using to do Ninja / Thief + double mambo + blind, before the utsusemi nerf and subsequent creation of ridiculously accurate targets.

this is wrong info from your side, foil is evasion. it gets overwritten by evasion down moves and it will have no effect when you have a -evasion down debuff on, similar to how haste has no effect when cast on someone that was hit by slowga

**Protey** · 2015-06-16 18:28

The testing just went SUPERNOVA!

They said that exact value Slow/2 testing would drive a man insane...but I did it I tell you! Sure, several kittens and other lower lifeforms were obliterated from existence just by observing the sheer awesomeness that was this testing.... but it was worth it!

Enfeebling Skill did nothing, it was all about dMND. The "Slow Amount" is the number over 1024 (i.e.: 75 dMND for Slow gives 300/1024).

Slow1:

Spoiler: show

From this data we see that the Slow Amount changes 2 every dMND above 0, then Slow Amount changes 1 every dMND below 0. Formula:

Slow = (2)(dMND) +150 for dMND > 0; Slow = dMND + 150 for dMND < 0; Slow Amount caps at 75 and 300.

Slow2 - 1 Merit:

Spoiler: show

From this data we see that the Slow Amount changes 1 to 2 every dMND. Formula:

Slow = Floor (116/75)(dMND) +244; Slow Amount caps at 128 and 360.

**caveat** dMND of -74 will give 129 Slow Amount instead of the actual 130 and dMND of 31 will give 291 Slow Amount instead of the actual 292. No matter how much I tried manipulating the equation, I could not get it to fit better.

Slow2 - 2 Merits:

Spoiler: show

From this data we see that the Slow Amount changes 1 to 2 every dMND just like 1 Merit, and has just been shifted by 10/1024. Formula:

Slow = Floor (116/75)(dMND) +254; Slow Amount caps at 138 and 370.

**caveat** dMND of -74 will give 139 Slow Amount instead of the actual 140 and dMND of 31 will give 301 Slow Amount instead of the actual 302. No matter how much I tried manipulating the equation, I could not get it to fit better.

Slow2 - 3 Merits:

Spoiler: show

From this data we see that the Slow Amount changes 1 to 2 every dMND just like 2 Merits, and has just been shifted by 10/1024. Formula:

Slow = Floor (116/75)(dMND) +264; Slow Amount caps at 148 and 380.

**caveat** dMND of -74 will give 149 Slow Amount instead of the actual 150 and dMND of 31 will give 311 Slow Amount instead of the actual 312. No matter how much I tried manipulating the equation, I could not get it to fit better.

Slow2 - 4 Merits:

Spoiler: show

From this data we see that the Slow Amount changes 1 to 2 every dMND just like 3 Merits, and has just been shifted by 10/1024. Formula:

Slow = Floor (116/75)(dMND) +274; Slow Amount caps at 158 and 390.

**caveat** dMND of -74 will give 159 Slow Amount instead of the actual 160 and dMND of 31 will give 321 Slow Amount instead of the actual 322. No matter how much I tried manipulating the equation, I could not get it to fit better.

Slow2 - 5 Merits:

Spoiler: show

From this data we see that the Slow Amount changes 1 to 2 every dMND just like 4 Merits, and has just been shifted by 10/1024. Formula:

Slow = Floor (116/75)(dMND) +284; Slow Amount caps at 168 and 400.

**caveat** dMND of -74 will give 169 Slow Amount instead of the actual 170 and dMND of 31 will give 331 Slow Amount instead of the actual 332. No matter how much I tried manipulating the equation, I could not get it to fit better.

Conclusion: Slow 2 goes up 10/1024 (~1%) between each merit and is about 6-10% (depending on # of merits) stronger than Slow 1.

And in case anyone was wondering: capped dMND, empy body and hands, saboteur Slow1 gave 725 Slow Amount. Same setup with 1 merit in Slow2 gave 869 Slow Amount. Same setup with 5 merits in Slow2 gave 909 Slow Amount.

Also, I have heard people say before that there is a 100% slow cap, this is not true. 1 Merit in Slow2, empy body and hands, saboteur + Carnage Elegy had Reraise with 140 seconds recast.

Well then, at ~3 minutes per data point I am kind of worn out and so will take a break from any lengthy testing for awhile.

**Saevel** · 2015-06-17 05:41

Originally Posted by Byrthnoth

Wrong, Foil is just evasion. Go check BG wiki's reference and keep reading for about a page.

I did, there wasn't much due to limitations of Martel's evasion gear, and it was quickly swept away to another topic. It couldn't even conclude that it was actually just +physical evasion and I could of swore that it worked on magic based TP moves (resist states).

it gets overwritten by evasion down moves and it will have no effect when you have a -evasion down debuff on

This means absolutely squat. Phalanx is static -damage, Barrier Tusk is -15% DT II. Both take up the Phalanx slot yet reduce damage in different ways. Heck we could make an argument in regards to Flurry vs Slow seeing as ranged attacks aren't effected by slow to begin with. Dia and Bio share a similiar relationship, they do different things yet one can block another. This is just how SE coded certain status effect's to take up different slot IDs.

Anyhow I'm open to being wrong if there is sufficient information available or a solid rational argument, I'm human like the rest of you. And my original statement was that many mega boss's, have ridiculously high accuracy. So high that attempting to combine buffs and debuffs to lower them is pretty futile. SE's develops seem to of overlooked doing that to their magic accuracy and thus we now abuse Vex + Attunement.

**Fwahm** · 2015-06-17 06:09

Barrier Tusk isn't DTII, as it stacks multiplicatively with DT instead of additively.

**Byrthnoth** · 2015-06-17 06:20

http://www.bluegartr.com/threads/115...=1#post5665305
75% hit rate with just Foil. (-20~25% land rate or +40 evasion over the floor)

http://www.bluegartr.com/threads/115...=1#post5665776
+44 evasion and Swordplay (+3 to +60 Evasion) plus Foil (-20~25% land rate or +40 evasion over the floor) moved him to approximately floored ability hit rate (~20% * (1-0.34 parrying rate)).

If Foil was a separate proc:
We know that he was getting 104 evasion for the last 90 seconds and less than that for the first 30 seconds. If Foil is a separate proc, we don't know what the monster's accuracy is. Worst case is that it's perfectly capped, in which case +104 evasion would bring the hit rate down to 43%. 0.43*0.75*0.66 = 21% land rate after a separate 25% Foil proc and Parrying. 13.6%+/-1.7% observed, so we can exclude that possibility.

If Foil was purely evasion:
It should have taken at least 100 additional evasion to reach the confidence interval (.15/.66=.23; 200*(.75-.23) = 104). So he would have been been capped 3/4 of the time. It's still not perfect, but it's definitely closer.

If anyone was going to repeat it they probably wouldn't use Swordplay or Parrying and I can't rule out Foil being an evasion multiplier (which would be most consistent with the results), but I think it's obvious that it isn't a static separate chance to evade.

**Damane** · 2015-06-17 08:09

@saevel i dont get your argument. foil is strait evasion up for physicsl moves. it will not allow to evade pure magical moves. you can go test that on AA mithra, you will evade most Physical WS with foil but you will allways take dmg from cloudsplittet even if its 0. Foil does nothing for pure magical moves (hybrid is something different) you will never evade but you can resist them which Run can do alot due to gear/traits/runes

**dasva** · 2015-06-17 10:10

Originally Posted by Fwahm

Barrier Tusk isn't DTII, as it stacks multiplicatively with DT instead of additively.

Yeah it's just a completely separate term from DT most likely some form of sdt

**Saevel** · 2015-06-18 04:48

Originally Posted by Byrthnoth

http://www.bluegartr.com/threads/115...=1#post5665305
75% hit rate with just Foil. (-20~25% land rate or +40 evasion over the floor)

http://www.bluegartr.com/threads/115...=1#post5665776
+44 evasion and Swordplay (+3 to +60 Evasion) plus Foil (-20~25% land rate or +40 evasion over the floor) moved him to approximately floored ability hit rate (~20% * (1-0.34 parrying rate)).

If Foil was a separate proc:
We know that he was getting 104 evasion for the last 90 seconds and less than that for the first 30 seconds. If Foil is a separate proc, we don't know what the monster's accuracy is. Worst case is that it's perfectly capped, in which case +104 evasion would bring the hit rate down to 43%. 0.43*0.75*0.66 = 21% land rate after a separate 25% Foil proc and Parrying. 13.6%+/-1.7% observed, so we can exclude that possibility.

If Foil was purely evasion:
It should have taken at least 100 additional evasion to reach the confidence interval (.15/.66=.23; 200*(.75-.23) = 104). So he would have been been capped 3/4 of the time. It's still not perfect, but it's definitely closer.

If anyone was going to repeat it they probably wouldn't use Swordplay or Parrying and I can't rule out Foil being an evasion multiplier (which would be most consistent with the results), but I think it's obvious that it isn't a static separate chance to evade.

I can get with it being a multiplier or somehow factored differently then straight acc vs evade. It really doesn't seem to be a static evasion boost because it seems to allow dodging of TP moves on things where no amount of -accuracy or +evasion help. It's also possible that it ignores or reduces the accuracy bonus some TP moves get.

you will evade most Physical WS with foil but you will allways take dmg from cloudsplittet even if its 0

You most certainly do "evade" magic moves, that's exactly how resistance rates work. Monsters magic accuracy is compared to your magic evasion the same way physical acc is compared to physical defense. A dice roll is made (RNG go), and if it fails (a miss). If there are multiple stages then it gets multiple rolls with each being lower damage, lower duration or some other reduced effect until it's final check is made and there results being a critical failure for damage or you just resisting the effect for status enfeebles. Player base magic evasion is a hidden C rank skill for every job, then you add on +magic evasion gear, after that you add on any +resist for that specific element for the final number that's checked against it's accuracy.

So when I say raise's evasion against magical TP moves, I don't mean producing a message of "miss" but rather some increase to magic evasion or flat ignore rate. I say this because with Foil up I've resisted / taken noticeably less damage from magic based special moves not aligned with the runes I've used. It made me scratch my head but whatever and continued playing.

**Saevel** · 2015-06-18 04:50

Originally Posted by dasva

Yeah it's just a completely separate term from DT most likely some form of sdt

I just call it DTII because there hasn't been a term created for it yet, though seeing as the tier II terms are now standardized for above the cap additives it might be wise to create a standard term for it. It takes the place of phalanx yet instead of being static reduce it's a percentage one.

**Damane** · 2015-06-18 07:39

which magical moves in particular did you resist with only foil up? i havent noticed foil increase resistance much at all. runes combined with barspells and gear did the trick for me. cloudsplitter will never get fully resisted. i have never witnessed a full magical movebeing annihilated just by foil. i have seen tough wonky stuff on hybrid moves, but noone knows exactly which moves are hybridmoves or puremagical moves. i still believe and stand by the fact that foil only works for physical moves (to anextent maybe hybrid moves)unless proven otherwise by you.

i know how magic resistance works but foil does shit for that in terms of magic evasion agianst tp moves. there is just too many instances where vex or attunement do the trick and foil does shit. (example balladonna charm in delve 2)

**Saevel** · 2015-06-18 09:27

which magical moves in particular did you resist with only foil up?

cloudsplitter will never get fully resisted. i have never witnessed a full magical movebeing annihilated just by foil.

i know how magic resistance works but foil does shit for that in terms of magic evasion agianst tp moves.

The above quotes demonstrates you don't.

You still think "miss" = "fully resist" = "does nothing", which isn't how resistance works. It's multiple stages, not a 1 one or 0.

Magic damage demonstrates this the easiest.

Take a nuke that would have 1000 damage, here is the resist stages

Check 1
If "hit" then it deals 100% damage / effect, 1000 Damage
If "miss" then roll check again

Check 2
If "hit" then it deals 50% damage / effect, 500 Damage
If "miss" then roll check again

Check 3
If "hit" then it deals 25% damage / effect, 250 Damage
If "miss" then critical resist and deals 12.5% damage / resisted effect, 125 Damage

That is a four state spell, which is most damage spells. Enfeebles tend to have two or three states instead with the duration being the reduced value though the jury is out on paralyze.

Because of this, Foil will never have you "resist" something like cloud splitter, instead you'll just get more 50, 25 and 12.5% values then you'd have otherwise. Also because its the same Mag Acc vs Mag Evade in each step, once your go over / under 50% it has a dramatic result in either direction. Also enfeebles that are "additional effect", like virtually all of the TP moves in existence, do a separate magic acc check then the move itself.

So what I've noticed is that with foil up, while I do tend to take lower damage from moves not aligned with my Runes + Barspell + valliance, its' mostly the enfeebles that stand out. Your buffed against Thunder. Earth or Fire but the NM also has a stun / paralyze effect, which seems to last less time or just not take effect whenever Foil is up. I really noticed it fighting D and VD Titan, buffing against Earth does jack sh!t for Geocrush's stun additional effect, but with Foil up that effect seems to either not proc or only last a very short duration even when the move hits me. If I had let foil fall and he hit with my Geocrush, it was very noticeable as I stood around waiting for it to wear off. Anyhow that's just an observation, what's likely happening is Foil is doing something weird to the acc / magic acc of the TP moves instead of being a straight +Evasion / +Magic Evasion.

**dasva** · 2015-06-18 10:13

Originally Posted by Saevel

I just call it DTII because there hasn't been a term created for it yet, though seeing as the tier II terms are now standardized for above the cap additives it might be wise to create a standard term for it. It takes the place of phalanx yet instead of being static reduce it's a percentage one.

I just call it sdt since it is multiplicative with dts and this is the only form of multiplicative percentage dmg reduction we have

**Protey** · 2015-08-30 02:16

Small test of Temper II:

500 rounds of 500 enhancing magic gave 92 triple attacks or 18.4% Triple Attack rate.

500 rounds of 569 enhancing magic gave 129 triple attacks or 25.8% Triple Attack rate.

Assuming increase in potency is fairly linear, we have the approximate equation:

Triple Attack Rate = floor(Enhancing Skill - 328.43)(37/345)

or about 1% TA rate per 10 enhancing skill.

Which means a RDM can get up to about 48% triple attack rate with high enhancing skill + TA gear.

**Byrthnoth** · 2015-08-30 09:04

RDM Meree: Finally actually a thing?

Thread: more magic testing

Thread Tools

more magic testing

Similar Threads

"Magic Critical Hit Rate" Testing

Blue Magic and Sea Gorgets - Testing?

Formal Magic Accuracy Testing

Afflatus Solace and Magic Defense Bonus tests

Bah! Too Many Rangers. Need More Thieves!

Latest Threads

Up & Coming Threads

Hottest Threads