Atma Testing

[Mojo]

Zagex & I tested some of the elemental attack atma as the values listed on wiki, Studio Gobli, and JP wiki were not consistent. We found the following values.

Atma Name				Element	Potency
Atma of the Cosmos		Dark	40
Atma of the Banisher		Dark	10
Atma of the Claw		Earth	30
Atma of the Smoldering Sky	Fire	30
Atma of the Lone Wolf		Fire	30
Atma of the Beyond		Light	30
Atma of the Beyond		Ice	30
Atma of the Undying		Ice	Unknown
Atma of the Sellsword		Light	10
Atma of the Lion		Thunder	30
Atma of the Blinding Horn	Thunder	30
Atma of the Noxious Fang	Water	40
Atma of the Tusk Terror		Water	20
Atma of Gales			Wind	30

Neither of us had Atma of the Undying, so if someone can test that it would be nice. Atma of the Holy Mountain does not seem to add any light potency. Furthermore, it's not listed under the potency atma on Studio Gobli and the description on JP wiki says it adds light elemental accuracy. I believe that the potency is a typo and that it's actually accuracy (or it might just be broken.)

Also, Zagex also tested Atma of the Gryphon's Claw.

Info on Griffon Claw with ranged WS
3 tests showed you need to equip a 2h weapon in order for GC to work on ranged WS. Ranged weapons themselves do not count as 2h.
No main, No GC - 1280
No main, GC - 1280
Staff main, GC - 1536

So it does seem that it will work for ranged weaponskills such as Wildfire as long as you have a 2 handed weapon equipped.

[Kirschy]

I can get some testing on Undying later tonight. I think it might be worth having a thread or at least a post just for atma effect and links to the testing. A lot of stuff is contradicting info is thrown around, and there's a lot of test/info scattered all over the place. (ie; VV's DA+ rate)

[Kirschy]

Tested Undying. Unfortunately it was during Firesday so I had to do enough sample to get /w and w/o day proc:

Base:
608 (Normal)
547 (Day Proc)

Undying:
730 (Normal)
657 (Day Proc)

Seems like 20% to me~

[Yugl]

Atma of Griffon's Claw do work on ranged ws if you are main handing 2 handed weapon. Tested using wildfire on shadow lizard. All tests are done on the same exact mob on windsday (neutral day) without any damage enhancement.

Test 1, Gun only:
823

Test 2, Gun only, atma of griffon's claw:
823

Test 3, Gun + Staff, atma of griffon's claw:
987
987

Test 4, Gun + Staff:
823


823 * 1.2 = 987.6. Seems pretty definite. This makes smoldering/lone wolf/griffon claw the best atma for wildfire while brewing.

Taken from RQ Thread.

Edit: Also, any comments on how to organize this thread (If at all)? I imagine it will serve more utility if done in an indexed fashion.

[Byrthnoth]

Atma of the Future Fabulous is likely 50 MDB, if people weren't sure before. Less than perfect test on my end, but it came out pretty close...

Popped T1 Goblin in Grauberg, Firaja for 2000.
Popped him again with Atma on, Firaja for 1337.

So it's probably 50 MDB

[Mojo]

I did some testing on the Fast Cast atma today using Migawari. The first column is just the single atma, the second column is with the addition of a Loquacious Earring.

Atma			Recast1	Recast2
Merciless Matriarch	114	112
Rescuer			117	116
Revelations		111	110
Tusked Terror		117	116

The values for Minor, Major, and Superior are probably 2%, 5%, and 7% (on recast, double for casting time.) I was expecting to maybe see something like 2.5%, 5%, and 7.5% but the data clearly shows that the 2.5% and 7.5% don't fall in line with what was observed. There also seem to be several typos on the atma.

Atma			Fast Cast	Expected	Observed
Lightning Beast		-		Minor		-
Tusk Terror		2%		Minor		Minor
Merciless Matriarch	5%		Superior	Major
Future Fabulous		-		Minor		-
Ambition		-		Minor		-
Rescuer			2%		Major		Minor
Revelations		7%		Superior	Superior

For anybody who might be wondering, Revelations is MP +10% and +10 MP/tick Refresh. The Refresh only activates when your HP (before cruor buffs) is below 50%. Dunno if this was tested already.

[Kirschy]

There's strong evidence that certain pieces of fast cast won't show fractional % on recast. ie; 3% fastcast shows 1% on recast instead of 1.5%, or 5% fastcast shows 2% on recast instead of 2.5%. But when combined you get 8% fast cast, and 4% on recast.

I'll try to find some time this weekend to write up info/examples.

[Mojo]

I've been messing around with some Fast Cast equipment. To begin with, the base recast of a spell is multiplied by 2.0 if you're weakened. It stacks with the 1.5 Hasso/Seigen multiplier, thus Hasso or Seigen + Weakness = base recast x 3.0. Fast Cast/Haste can still be applied to recast even with the presence of those modifiers. I tested this with various n.25, n.5, and n.75 base recast time spells with Hasso/weakness and such and everything was peachy. Recast is also capped at 255 seconds (heh.)

Either way, I know most (if not all) of this was already evidenced else where, I just thought it would be a good start to verify it. I assumed that Fast Cast is measured in n/1024, like most other stuff in the game. I've been dieing a lot on my brother's WHM, the results are below.

Spell		Status		Gear								Recast Time	Minimum	Maximum	Range	Minimum	Maximum	Spread
Reraise		Weakened	Incantor Stone							118		9	17	9	0.833	1.667	0.833
Reraise		Weakened	Augur's Gaiters							118		9	17	9	0.833	1.667	0.833
Reraise		Weakened	Orisons Locket							117		18	25	8	1.667	2.500	0.833
Reraise		Weakened	Incantor Stone + Orison Locket					116		26	34	9	2.500	3.333	0.833
Reraise		Weakened	Incantor Stone + Augur's Gaiters				117		18	25	8	1.667	2.500	0.833
Reraise		Weakened	Augur's Gaiters + Orison Locket					115		35	42	8	3.333	4.167	0.833
Reraise		Weakened	Incantor Stone + Augur's Gaiters + Orison Locket		114		43	51	9	4.167	5.000	0.833
Reraise		Weakened	Fast Cast I							114		43	51	9	4.167	5.000	0.833
Reraise		Weakened	Fast Cast I + Incantor Stone					112		60	68	9	5.833	6.667	0.833
Reraise		Weakened	Fast Cast I + Augur's Gaiters					112		60	68	9	5.833	6.667	0.833
Reraise		Weakened	Fast Cast I + Orison Locket					111		69	76	8	6.667	7.500	0.833
Reraise		Weakened	Fast Cast I + Incantor Stone + Orison Locket			110		77	85	9	7.500	8.333	0.833
Reraise		Weakened	Fast Cast I + Icantor Stone + Augur's Gaiters			111		69	76	8	6.667	7.500	0.833
Reraise		Weakened	Fast Cast I + Augur's Gaiters + Orison Locket			109		86	93	8	8.333	9.167	0.833
Reraise		Weakened	Fast Cast I + Incantor Stone + Augur's Gaiters + Orison Locket	108		94	102	9	9.167	10.000	0.833
Cura II		Weakened	Incantor Stone							79		1	12	12	-	1.250	1.250
Cura II		Weakened	Augur's Gaiters							79		1	12	12	-	1.250	1.250
Cura II		Weakened	Orisons Locket							78		13	25	13	1.250	2.500	1.250
Cura II		Weakened	Incantor Stone + Orison Locket					77		26	38	13	2.500	3.750	1.250
Cura II		Weakened	Incantor Stone + Augur's Gaiters				78		13	25	13	1.250	2.500	1.250
Cura II		Weakened	Augur's Gaiters + Orison Locket					76		39	51	13	3.750	5.000	1.250
Cura II		Weakened	Incantor Stone + Augur's Gaiters + Orison Locket		76		39	51	13	3.750	5.000	1.250
Cura II		Weakened	Fast Cast I							76		39	51	13	3.750	5.000	1.250
Cura II		Weakened	Fast Cast I + Incantor Stone					75		52	64	13	5.000	6.250	1.250
Cura II		Weakened	Fast Cast I + Augur's Gaiters					75		52	64	13	5.000	6.250	1.250
Cura II		Weakened	Fast Cast I + Orison Locket					74		65	76	12	6.250	7.500	1.250
Cura II		Weakened	Fast Cast I + Incantor Stone + Orison Locket			73		77	89	13	7.500	8.750	1.250
Cura II		Weakened	Fast Cast I + Icantor Stone + Augur's Gaiters			74		65	76	12	6.250	7.500	1.250
Cura II		Weakened	Fast Cast I + Augur's Gaiters + Orison Locket			72		90	102	13	8.750	10.000	1.250
Cura II		Weakened	Fast Cast I + Incantor Stone + Augur's Gaiters + Orison Locket	72		90	102	13	8.750	10.000	1.250
Flash		Weakened	Incantor Stone							89		1	11	11	-	1.111	1.111
Flash		Weakened	Augur's Gaiters							89		1	11	11	-	1.111	1.111
Flash		Weakened	Orisons Locket							88		12	22	11	1.111	2.222	1.111
Dread Spikes	Normal		Loquacious Earring						178		6	11	6	0.556	1.111	0.556

The actual value must be greater than the minimum decimal listed but can be equal to the maximum. Thus if a minimum decimal value could be evenly converted to n/1024, then it could not have that value as an n minimum (this occurred twice.)

Either way, there is probably a lot that could be gathered from this data alone, however I found it much less confusing to just take some simple things away and do more tests.

Combining test data yields the following.

Gear			Minimum	Maximum	Range	Minimum	Maximum	Spread
Fast Cast I		43	51	9	4.167	5	0.833
Incantor Stone		9	11	3	0.833	1.111	0.278
Augur's Gaiters		9	11	3	0.833	1.111	0.278
Orison Locket		18	22	5	1.667	2.222	0.555
Loquacious Earring	6	11	6	0.556	1.111	0.555

I then did a few tests with Aero III.

Aero III	Normal	Augur's Gaiters		24	11	50	39	0.99009901	4.95049505	3.96039604
Aero III	Normal	Loquacious Earring	24	11	50	39	0.99009901	4.95049505	3.96039604

This seems to indicate with almost complete certainty that those two pieces of gear are either 11/1024 or 0.01 Fast Cast.

Bludgeon	Weakened	Incantor Stone + Loquacious Earring	23	12	21	9	-	2.127659574	2.127659574
Waterga		Weakened	Incantor Stone + Loquacious Earring	12	12	20	8	-	2.040816327	2.040816327

This data is what kind of weirds me out. If it is in n/1024, then Incantor Stone is only 9/1024 (a first.) This may be the case, but it just seems unlikely I guess that it's not 10/1024 or 11/1024. Everything works out fine if you use 0.01 instead of decimals, so maybe it's an n/100?

Either way, I need to wait until my brother gets to sea before I can test Loquacious Earring at the same time as Augur's Gaiters. Maybe somebody else can do a bit of testing also.

[Yugl]

Hey. Kirschy's post here http://www.bluegartr.com/threads/103...t=#post4531216

Dunno if that links right since I use like 50 posts per page or something but it's supposed to be this post

There's strong evidence that certain pieces of fast cast won't show fractional % on recast. ie; 3% fastcast shows 1% on recast instead of 1.5%, or 5% fastcast shows 2% on recast instead of 2.5%. But when combined you get 8% fast cast, and 4% on recast.

I'll try to find some time this weekend to write up info/examples."

Regarding this, I did a lot of testing on cure cast -% back in 2008, or maybe it was 2009 don't remember. But either way I discovered that that a small increase in cure cast-%, while sometimes it would not cause a change in the recast or the animation of the spell, it would still cause the spell to land earlier serverside. I had to test this by dualboxing two characters, binding a macro on my g15 to both instances, adjusting the cure cast -% on each of them so the recast was the same and the animation was the same, however the one with slightly higher cure cast-% gear landed the cure first 90% of the time. I attribute the 10% to random internet, server lag. So the recast does not appear to be perfectly tied to the increase in casting speed. Oddly, a change in the recast time always coincided with a change in the animation speed.

Thought that might be useful/relevant to Kirschy.

I seem to remember checking the animation and recast on stuff like curaga IV with different amounts of cure cast -% And I think could alter the point where increasing cure cast -% took effect on the animation/recast by using dark arts on whm/sch. However, it could be that my findings were opposite of that and I could not make the point where increasing cure cast -% changed the recast/animation. I really can't remember which way it was, so this isn't terribly helpful. But it is still an idea that might warrant some testing if FC works the same as cure cast -.

I currently can't retest this myself as I don't have access to my account til wednesday(hopefully). If you still need more info on it by the time i get my account back, I will be happy to do some testing.

PM from Shaddix.

[Mojo]

Cure spells have much smaller recast times, and given what we have observed about Fast Cast, it's no surprise to me that it may sometimes not make a difference on their recast. What's really starting to bake my noodle is this.

Flash		Weakened	Augur's Gaiters							89		1	11	11	-	1.111	1.111
Aero III	Normal		Augur's Gaiters							24		11	50	39	0.990	4.950	3.960
Flash		Weakened	Orisons Locket							88		12	22	11	1.111	2.222	1.111
Reraise		Weakened	Augur's Gaiters + Orison Locket					115		35	42	8	3.333	4.167	0.833

The first two tests seem to imply with certainty that there would be 11/1024 Fast Cast on Augur's Gaiters. The third test then gives a maximum of 22/1024 for Orison's Locket while the fourth test, if adjusted for Augur's Gaiters, gives a minimum of 24/1024.

D:

[Yugl]

I think Kirschy experienced this before where adding pieces of gear together would produce different levels of fast cast from wearing them individually. Although I have to admit, I'm not certain what the Min/Max/Range refer to. Difference in recast time?

[Mojo]

The minimum is the least amount of Fast Cast required on the piece (or combination of pieces) to produce the observed results (assuming that the way we model Fast Cast is correct.) Minimum decimal values are open boundary, meaning that the actual value cannot be that specific value but can be all values above it. I considered this when coming up with the binary minimum values (n/1024), so all binary minimums are closed (they can be equal to the listed value.) The maximum values for both decimal and binary are closed (equal to or less than.) The range listed for binaries is simply how many difference binary values could have produced the observed results and the spread for decimals is the amount of decimal space that could produce the observed results. Both of these are just indicators of precision.

[Byrthnoth]

I'm sure this was already known, but for some reason I'm under the impression that it isn't common knowledge.

HP Atma Buffs are calculated off your pre-Cruor Buffs HP, regardless of what order you get them in.
Base HP with no buffs: 1214
Base HP with Cruor Buffs: 2488 (+105% from Abyssite and zone bonus I assume?)
Ebon Hoof, Cruor Buffs: 2852
Sanguine Scythe, Cruor Buffs: 2731
Ebon Hoof, Sanguine Scythe, Cruor Buffs: 3095
Sanguine Scythe, no Cruor Buffs: 1456

Okay, so several things to notice-
1) Sanguine Scythe gives 242~243 HP (+20% base) no matter when I add it
2) Ebon Hoof gives 365 (+30% base) no matter when I add it
3) If the formula is this, they don't floor it until the end:
Atma'd/Cruor'd HP = HP + HP*{Cruor and Abyssite Buffs} + HP*{Atma %}
That's shown by the 2731, which should only be 2730 if they floored each step


It's the same with MP-
Cruor Buffs and zone bonus, +60 MP in gear - 93 MP = 60*155%
Cruor Buffs, Zone Bonus, Atma of the Banisher (tested earlier today to be 5% without Cruor Buffs) - 96 MP = 60*155% + 60*.05


Makes using HP and MP Atma seem kinda silly.

[Mojo]

I always cringe when I see people using HP atma when they don't even have all of the merit Abyssites. Should also be noted that the base HP used includes gear and food.

[Byrthnoth]

Yeah, otherwise my MP+60 example would have failed.

Do you happen to know if Ducal Guard activates on attacks that would push you below 50% HP? If so, it's a much better one-shot-safe than Ebon Hoof or Sanguine Scythe.

[Byrthnoth]

I hadn't seen the testing done, but I asked someone last night and they told me Savior's absorb rate was 10%. Being a contrary taru, I confirmed it:
48989 Hits taken, 5021 hits absorbed = 10.25% Absorb rate, assuming it procs absolutely last (and it probably does)

Those 50k hits (lets assume 40k were to my face) got me from 290 to 306 Parrying, mostly EP/DC mandies with the occasional EM. Recommendation: If you want good skillups, level your mandies up to IT first.


2157 Hits taken, 111 Absorbs for Aquatic Ardor, so 5% Physical Absorb rate. Probably the same for magical.


Other Notes on Absorbing Damage:
* It does not appear to be affected by Physical Damage Taken gear.
- Test: Ducal Guard with all the mandies in Altepa. I canceled Phalanx and let them take me down to low HP, then put it back on. With 50% PDT below 50% HP, I should have noticed a dramatic decrease in the amount of absorbed HP if it counted. I didn't.
* You appear to absorb the damage you would have taken without PDT gear.
- Test: Same setup, canceled Phalanx as before without Terra's Staff on this time. The range of Absorbs (15 to 33) was almost the same as the range of damage (15 to 34)

So yeah, you just absorb the damage that you would have taken with no PDT gear and no Phalanx. I have not documented a crit hit, so I'm not sure if you can get crit-healed, but I'm using the 5% Atma and these mandies have a <1% crit rate on me.

One last random fact. It's possible to absorb counters.

[Byrthnoth]

I get that Atma of the Hero is approximately +15% Crit damage (44 Crit with +8% damage from traits, 50 crit with Atma and trait). Can anyone with a higher base damage Ranged weapon halp me out? My stupid Jinx Discus and I are tired and my resolution is terrible.

[Mojo]

There may be some correlation between evasion and the proc rate on Lethal Triclip/Ruinous Scythe cruor buff removal. I finished my Bukhis Wings a while ago, so I'm not sure if/when I'll be doing more, but I did notice that my buffs were almost never removed when evasion tanking (Yonin, hybrid evasion set, Kurayami and Cloak and Dagger, sometimes Mambos.) It could have easily just been luck but if someone plans on doing Bukhis anytime soon, it might be worthwhile to investigate.

[Byrthnoth]

I'm just going to drop this here. Someone else figure it out! Reraise on WAR/WHM. Anything over a minute is weakened. I threw capped gear Haste in on a whim to try and dissociate Future Fabulous' Haste and Fast Cast.

1:45 = Omni, 27% Haste gear
1:43 = Omni, future fab, 27% Haste in gear = 60*1.75*.99 = 1% Reduction from Fast Cast
1:58 = future fab, 0% Haste in gear = 60*1.99*.99 = 1% Haste, 1% Reduction from Fast Cast
1:57 = Ambition, 0% Haste in gear = 60*2*.98 = 2% Reduction from Fast Cast
1:42 (weakened) :44 (unweakened), Ambition, 25% Haste = 60*1.75*.98, 60*.75*.98 = 2% reduction from Fast Cast
1:39 = Merciless Matriarch, 27% Haste + Omni = 60*1.75*.95 = 5% reduction from Fast Cast
1:54 = Merciless Matriarch, 0% Haste = 60*2*.95 = 5% reduction from Fast Cast
1:52 = Merciless Matriarch, Future Fab, 0% Haste = 60*1.99*.94 = 1% Haste, 6% total reduction from Fast Cast
1:51 = Merciless Matriarch, Ambition, 0% Haste = 60*2*.93 = 8% total reduction from Fast Cast

Kirschy has, in the past, expressed a suspicion that Fast Cast is only applied to recast in whole % point increments, similar to how Killer effects affect damage. I failed to find two "odd" values above to prove it, but I can go back and test with another one some day. Point is, MM is 10% Fast Cast, Future Fab is 2% Fast Cast (or 2.5%, if that's a real thing), and Ambition is likely 5% Fast Cast.

[Byrthnoth]

Okay, so I've confirmed it, I think.

Lightning Beast = 1:57 Recast (-2%)
Ambition = 1:57 Recast (-2%)
Lightning Beast + Ambition = 1:54 Recast (-5%)
Lightning Beast + Tusked Terror = 1:54 Recast (-5%)
Any two + 27% Haste = 1:39 Recast (60*1.75*.95)
Merciless Matriarch = 1:54 Recast (-5%)
Merciless Matriarch + 27% Haste = 1:39 Recast (60*1.75*.95)

Any two 5% Atma (-2% recast alone) become the same as Merciless Matriarch (-5% recast alone) when combined. Looks like it's pretty much the same as Killer Effects.