User talk:Suiram

From FFXI Wiki



General

Tests

Multiple-Target Damage Reduction

In this test, I set out to confirm the damage reduction from area of effect Elemental Magic spells hitting multiple targets, as well as this term's place in the overall magic damage calculation. The test subjects were Huge Hornets in North Gustaberg. These are known to possess an INT value of 5, although I have personally verified this as well.

Initial Results

In the first phase of this test, I cast Thundaga 3 eleven times, on increasing numbers of total targets per cast. As a Hume BLM/RDM with 5 merits in INT and 5 merits in Lighting Magic Potency, I equipped myself with INT+28, MAB+8 with no staff, to bring my total INT to 105 and MAB to 1.5. This kept the dINT value at 100 throughout this phase of the test. Testing was done on a neutral day of the week, and no weather was present. For reference, the applicable subformula of the magic damage formula (using "[ ]" as shorthand for the floor function throughout) is:

[[(dINT * M) + V] * MAB] = [[(100 * 1.5) + 697] * 1.5] = [847 * 1.5] = [1270.5] = 1270

Targets Damage Possible Multiplier Possible V
1 1270 1 697
2 1015 .8 527
3 952 .75 485
4 888 .7 442
5 825 .65 400
6 762 .6 358
7 697 .55 315
8 634 .5 273
9 571 .45 231
10 507 .4 188
11 507 .4 188


This suggests, as I have indicated in the third and fourth column of the table, that for two or more targets there is either a fractional decrease in total damage from an additional multiplier, or that there is a preprogrammed alternate V value for multiple targets. It also suggests that for 10 or more targets the decreased damage per target remains the same. Furthermore, if it is in fact an additional multiplier, it is calculated prior to MAB, since:

[[847 * 1.5] * .8] = [1270 * .8] = 1016

while

[[847 * .8] * 1.5] = [677 * 1.5] = 1015

the value achieved in the test.

Determining the Type of Reduction

In this phase of the test, I equipped myself with a total of 104 INT (dINT = 99) and 1.5 MAB, and cast Thundaga 3 on two hornets. If the reduced damage results from an additional multiplier, then the resulting value should be:

[[[(99 * 1.5) + 697] * .8] * 1.5] = [[845 * .8] * 1.5] = [676 * 1.5] = 1014

Whereas, if the reduced value results from a different V value for multiple targets, then by the above result, the calculation for the damage equation gives:

[[(99 * 1.5) + 527] * 1.5] = [675 * 1.5] = 1012

The result was 1014, confirming that the decrease results from an additional multiplier in the damage calculation.

Determining the Placement of the Term with Respect to the Overall Formula

The parts of the of damage formula ordering which have so far been confirmed are that the calculation of D precedes the resist term, which precedes the staff bonus term, which precedes the Day/Weather term, which precedes the MAB term, and that the Magic Burst term also precedes the MAB term. I next checked the placement of the AoE reduction term with respect to the staff bonus term, however as you will see that its relationship to the staff bonus term is implied by its relationship to the resist term, I will omit the results of my staff bonus tests in the interests of brevity. For this next test, I forced my INT to 109 (dINT = 104) and MAB again to 1.5. I then continued to gather pairs of bees and cast Thundaga 3 on them until one resisted the spell. For a 1/2 resist, the outcome if the resist term was calculated prior to the AoE reduction term would be:

[[[[(104 * 1.5) + 697] * .5] * .8] * 1.5] = [[[853 * .5] * .8] * 1.5] = [[426 * .8] * 1.5] = [340 * 1.5] = 510

while the outcome were the AoE term calculated first would be:

[[[[(104 * 1.5) + 697] * .8] * .5] * 1.5] = [[[853 * .8] * .5] * 1.5] = [[682 * .5] * 1.5] = [341 * 1.5] = 511

The result was 511, confirming that the multiple-target damage reduction is calculated prior to the resist term.

Comments

Please feel free to ask any questions or make any comments regarding this test in this subsection.


M and V data

The goal for this test was to confirm the values for M and V for all of the Black Magic damage spells. Over three separate tests I confirmed the values for every spell in this family except for the Ancient Magic series II spells. While I used the data to determine the values independently from previous figures, I will present it here simply in a collection of tables, and anyone concerned with the accuracy of the results can verify the values in the table with the formula and the screenshots of the damage in the chat log. The MAB I used was 1.5 throughout, as that would ensure that the damage would increase by at least one point per change in dINT after flooring. The relevant portion of the magic damage formula for the entire series of tests is [[dINT*M + V] * 1.5], where "[]" indicates flooring.


Test I

This was done in North Gustaberg against Huge Hornets. I varied my dINT to 100, 101, and 102 for each spell in this test. I actually tested the series II single-target spells here as well, to compare with the second test and ensure that I didn't hit the inflection point for adding INT, but they agree with the values in Test 2 below until Stone II, so I will omit them here.

TABLE I
Spell V M 100 101 102
Stone III 210 1.5 540 541 544
Water III 236 1.5 579 583 540
Aero III 265 1.5 622 624 627
Fire III 295 1.5 667 669 672
Blizzard III 320 1.5 705 706 709
Thunder III 345 1.5 742 744 747
Spell V M 100 101 102
Stone IV 381 2 871 874 877
Water IV 410 2 915 918 921
Aero IV 440 2 960 963 966
Fire IV 472 2 1008 1011 1014
Blizzard IV 506 2 1059 1062 1065
Thunder IV 541 2 1111 1114 1117


Test II

These tests were performed with Nivlakian's help against Aht Urhgan Attercops. He cast Blizzard IV or Thunder IV with fixed INT and MAB, which we used to calculate the individual spider's INT, then I adjusted my INT to achieve a dINT of 0 and 5 for each spell. Please note, the tests for single-target Aero I/II and Stone I/II were done on Windsday, so when referencing the screenshots use the lower Aero damage and the higher Stone damage.

TABLE II
Spell V M 0 5
Stone 10 1 15 22
Water 16 1 24 31
Aero 25 1 37 45
Fire 35 1 52 60
Blizzard 46 1 69 76
Thunder 60 1 90 97
Spell V M 0 5
Stone II 78 1 117 124
Water II 95 1 142 150
Aero II 113 1 169 177
Fire II 133 1 199 207
Blizzard II 155 1 232 240
Thunder II 178 1 267 274
Spell V M 0 5
Stonega 56 1 84 91
Waterga 74 1 111 118
Aeroga 93 1 139 147
Firaga 120 1 180 187
Blizzaga 145 1 217 225
Thundaga 172 1 258 265
Spell V M 0 5
Stonega II 201 1 301 309
Waterga II 232 1 348 355
Aeroga II 266 1 399 406
Firaga II 312 1 468 475
Blizzaga II 350 1 525 532
Thundaga II 392 1.5 588 598
Spell V M 0 5
Stonega III 434 1.5 651 661
Waterga III 480 1.5 720 730
Aeroga III 527 1.5 790 801
Firaga III 589 1.5 883 894
Blizzaga III 642 1.5 963 973
Thundaga III 697 1.5 1045 1056
Spell V M 0 5
Freeze 526 2 789 804
Tornado 552 2 828 843
Flood 630 2 945 960
Flare 657 2 985 1000


Test III

When analyzing the data, I noticed a problem with the recorded images for Quake and Burst, so I returned to North Gustaberg to retest those spells. I used dINT of 80 and 100 for this test.

TABLE III
Spell V M 80 100
Quake 577 2 1105 1165
Burst 603 2 1144 1204


Image Record

Aero4 Stone4: http://img234.imageshack.us/img234/8914/a4s4sa1.jpg

Aero1/2 Stone1/2 dINT0: http://img72.imageshack.us/img72/4813/a12s12dint0yp5.jpg

Aero1/2 Stone1/2 dINT5: http://img72.imageshack.us/img72/5862/a12s12dint5ps4.jpg

Aeroga 1/2/3 dINT0 dINT5: http://img408.imageshack.us/img408/8692/aga123dint0dint5vi1.jpg

Freeze dINT0 dINT5: http://img61.imageshack.us/img61/4497/bamdint0dint5xt7.jpg

Blizzaga 1/2/3 dINT0 dINT5: http://img148.imageshack.us/img148/2826/bga123dint0dint5nj1.jpg

Fire3 Aero3: http://img316.imageshack.us/img316/7208/f3a3dy9.jpg

Fire4 Water4: http://img316.imageshack.us/img316/8836/f4w4kl6.jpg

Fire1/2 Water1/2 dINT0: http://img148.imageshack.us/img148/8330/f12w12dint0sa2.jpg

Fire1/2 Water1/2 dINT5: http://img48.imageshack.us/img48/1888/f12w12dint5qp2.jpg

Firaga1/2/3 dINT0 dINT5: http://img120.imageshack.us/img120/8990/fga123dint0dint5hg6.jpg

Quake Burst dINT80 dINT100: http://img120.imageshack.us/img120/2333/samtamdint80dint100of2.jpg

Stonega 1/2/3 Waterga1/2/3 dINT0: http://img127.imageshack.us/img127/8796/sga123wga123dint0kl8.jpg

Stonega 1/2/3 Waterga1/2/3 dINT5: http://img116.imageshack.us/img116/3370/sga123wga123dint5gs8.jpg

Thunder3 Blizzard3: http://img239.imageshack.us/img239/7406/t3b3yc0.jpg

Thunder4 Blizzard4: http://img73.imageshack.us/img73/3755/t4b4to2.jpg

Thunder1/2 Blizzard1/2 dINT0: http://img73.imageshack.us/img73/4273/t12b12dint0ia4.jpg

Thunder1/2 Blizzard1/2 dINT5: http://img73.imageshack.us/img73/564/t12b12dint5wu3.jpg

Thundaga1/2/3 dINT0 dINT5: http://img360.imageshack.us/img360/5417/tga123dint0dint5lo0.jpg

Water3 Stone3: http://img46.imageshack.us/img46/696/w3s3gd8.jpg

Flood Tornado Flare dINT0 dINT5: http://img46.imageshack.us/img46/8493/wamaamfamdint0dint5hd7.jpg


Comments

Please feel free to ask any questions or make any comments regarding this test in this subsection.


Amir Set Bonus

Test I

This test was performed with Alexei, a 75PLD/37NIN in full Amir, and myself a 75BLM/37NIN. PLD's base INT was 54, and used a Snow Ring(+5) and Melon Pie+1 (+5) to raise his INT to 64. My base INT is 75 and I equipped Andvaranauts(-7) and an Ocean Sash (-4), lowering mine to match. The test was performed on Windsday, so there was no chance of day bonus interference in the recorded values. An Aw'aern's Thundaga III did 1106 damage to me, and 1019 to the PLD. This eliminates the likelihood of the bonus being an increase in Magic Defense Bonus, as MDB+9 should have resulted in 1014 or 1015 damage to the PLD, while MDB+8 would have been 1024 or 1025. However, the numbers would be consistent with a -8% (20/256) reduction in the Target Magic Damage Adjustment term, since floor(1106 * (236/256)) = 1019.

Test II

Proceeding on the indication in Test I that the set bonus is a 20/256 reduction in the Target Magic Damage Adjustment term, we set out to confirm this hypothesis by stacking the Amir set bonus with a known reduction in that term, Shell III (48/256). If the Amir bonus is a reduction in Target Magic Damage Adjustment, it should stack to make a 68/256 reduction, or 188/256 = 73.4% of the damage done to me. This test was performed on Iceday, and the spell used was Firaga III, so again there was no risk of day bonus altering the values. The damage dealt to me was 946, while the damage to the PLD was 694. Examining the possible equations with the known values,
floor(946 * (188/256)) = 694
floor(floor(946 * (236/256)) * (208/256)) = 708
floor(floor(946 * (208/256)) * (236/256)) = 708

Confirming our hypothesis.

Image Record

Test I: http://img187.imageshack.us/img187/7290/img20070125020623jm7.jpg

Test II: http://img187.imageshack.us/img187/2073/img20070125024928kf0.jpg

Comments

Please feel free to ask any questions or make any comments regarding this test in this subsection.


Wizard's Roll

Test

Tested Wizard's Roll bonus to MAB with BLM in party as 75BLM/37NIN with Chubbs, a 75COR/37THF. I had total of 1.5 MAB from Job Trait (32), Merits (10) and gear (Igqira Weskit 6, Yigit Turban 2). Damage from Blizzard 4 with no roll was 1056, making base damage for the spell 704. Relevant section of the magic damage formula is floor(704 * (1.5 + X)), where X is the MAB addition from Wizard's Roll expressed as a fraction of 100.

Roll Damage MAB
None 1056 1.50
1 1098 1.56
2 1105 1.57
3 1112 1.58
4 1112 1.58
5 1154 1.64
6 1119 1.59
7 1126 1.60
8 1133 1.61
9 1091 1.55
10 1133 1.61
11 1168 1.66


Image Record

Base, 1, 3, 9: http://img222.imageshack.us/img222/1067/wizard0139dm9.jpg

8: http://img409.imageshack.us/img409/3689/wizard8mx4.jpg

5, 6: http://img237.imageshack.us/img237/515/wizard56zw5.jpg

2, 7: http://img222.imageshack.us/img222/6391/wizard27zs8.jpg

11: http://img471.imageshack.us/img471/4972/wizard11us4.jpg

4, 10: http://img237.imageshack.us/img237/6453/wizard410vc8.jpg


Comments

Please feel free to ask any questions or make any comments regarding this test in this subsection.

When did you do the Wizard's Roll tests? Thanks.--Genosync 05:31, 12 October 2007 (CDT)

September 30, 2007, the same day I added the data here and on the Wizard's Roll page. I think the first image should have the /clock stamp. -Suiram 00:58, 13 October 2007 (CDT)