incorrect automaton ws damage

ISP:Comcast
Type of Internet Connection: Cable
Internet Connection Speed: 100mbps
Date & Time: Ever since we were given automaton ws stats
Frequency: Always
Character Name: Urmom
Race: Taru
World: Shiva
Job: Pup/whm
Area and Coordinates: All over the place
Party or Solo: Party or solo
NPC Name: None
Monster Name: Various monsters from high level content to level 0 starter mobs. But for the test level 0 rarabs outside of windhurst
In-Game Time: all the time

So for awhile now the damage the automatons do seemed pretty off compared to the stats given to us by the devs so decided to go test it on some low level mobs to ensure my att and str are capped on it.

Before I get into the steps of the tests and the results since the players probably have different terms and ways of understanding how damage is calculated I thought I'd give a brief overview of the math behind it so we are on the same page. Bear with me this will get a tad long

For ranged damage when at capped attack damage is the base damage * pdif. Pdif being the multiplier used that compares attack to target defense and in this case caps at a steady 3.0 (this is important because melee it has some amount of random variance). For a ranged ws it will be similar with base damage *pdif * ftp. Ftp being what we call the damage multiplier for ws which the devs have given to us at +x00%.

For autos regular ranged attacks base dmg is the sum of automaton ranged skill, agi, dex and fstr. With the skill, dex, and agi having multipliers while fstr being the amount added based on the difference between attackers str and targets vit. While for ws it's all of those and what players call wsc which is the stat modifiers for the ws which is how much of that stat you have multiplied by factor which was given to us as well.

Or in more formula like range dmg = pdif * (skill mod + agi mod+ dex mod+ fstr)

While the ws dmg will be range wsdmg = ftp*pdif * (skill mod + agi mod+ dex mod+ fstr+wsc)

So as you can see the only difference between the 2 is ftp and wsc. Given this I decided to raise my automatons agi in order to increase it's base dmg. And as per the formula above you can see that since most the variables are shared the increase in ranged ws dmg should be equal to the increase in the regular ranged dmg *ftp.

Or formula wise Δranged ws dmg = Δranged dmg*ftp

So by comparing changes in ranged and ranged ws dmg by adding agi you can determine ftp.

Similarly we can check the wsc by raising that stat however a little more math would be involved. Given the above formulas the simplified version would

Δranged ws dmg = ftp* (Δranged dmg +wsc*pdif)

As a remind the stats given to us for ranged ws and for fun flameholder are as follows

Flameholder
No Fire Maneuvers: Damage +25%
1 Fire Maneuver: Damage +100%
2 Fire Maneuvers: Damage +175%
3 Fire Maneuvers: Damage +250%

Arcuballista
Base: Damage +600%, Accuracy +100
Attribute Mod: DEX 50%
TP 1000: Damage +0%
TP 2000: Damage +300%
TP 3000: Damage +600%

Armor Piercer
Base: Damage +300%, Accuracy +100, Target defense -50%
Attribute Mod: DEX 60%
TP 1000: Damage +0%
TP 2000: Damage +150%
TP 3000: Damage +300%

Armor Shatterer
Base: Damage +500%, Accuracy +50, Additional Effect: Defense -15% (90 seconds), Attack +125%
Attribute Mod: DEX 50%
TP 1000: Additional Effect Duration +0%
TP 2000: Additional Effect Duration +33%
TP 3000: Additional Effect Duration +66%

Daze
Base: Damage +500%, Accuracy +150, Additional Effect: Stun
Attribute Mod: DEX 100%
TP 1000: Damage +0%
TP 2000: Damage +250%
TP 3000: Damage +500%

Given the difficulty in controlling the automaton I've decided to only do 3000 tp so that would put the ftps I plan on use at 13, 7, 6, and 11 for arcuballista, Armor Piercer, Armor Shatterer and Daze respectively

Step 1: Get 3k tp on ranged auto letting it shoot level 0 rarabs.

Step 2: Use maneuvers as needed to force the various ws and activate flameholder. Note the only maneuver that effected dmg was thunder to force daze as dark doesn't effect dmg at all and strength from fire was already capped.

Step 3: perform steps 1 and 2 with an additional 50 automaton agi.

Step 4: record damages and analyze results to figure out ftps

Step 5: Perform steps 1 and 2 with an additional 60 dex.

Step 6: record damages and using the ftp from step 4 determine actual ftp of daze

Results:

Ranged damage without thunder maneuver
Range damage: 1551
Agi +50 ranged dmg : 1665


Ranged damage with thunder maneuver
Range damage: 1560
Agi +50 ranged dmg : 1674
Dex +60 ranged dmg: 1641

As you can see +50 agi increased dmg by 114 and +60 dex increased dmg by 81

Arcuballista:
without flameholder

ws dmg: 15912
+50 agi dmg: 16887
Damage increase: 975
Expected increase: 1482 =114*13

with flameholder
1 Maneuver
ws dmg: 17136
+50 agi dmg: 18186
Damage increase: 1050
Expected increase: 1596 = 114* (13+1)

2 Maneuver
ws dmg: 18054
+50 agi dmg: 19158
Damage increase: 1104
Expected increase: 1681.5 = 114 * (13+1.75)

3 Maneuver
ws dmg: 18972
+50 agi dmg: 20133
Damage increase: 1161
Expected increase: 1767 = 114*(13+2.5)

Armor Shatterer:
without flameholder
ws dmg: 6966
+50 agi dmg: 7416
Damage increase: 450
Expected increase: 684 = 114*6

with flameholder
ws dmg: 7254
+50 agi dmg: 7725
Damage increase: 471
Expected increase: 712 = 114*(6+ .25)

Armor peircer:
ws dmg: 8568
+50 agi dmg: 9093
Damage increase: 525
Expected increase: 798 = 114*7

Daze:
ws dmg: 13596
+50 agi dmg: 14421
Damage increase: 825
Expected increase: 1254 = 114*11


As you can see the expected dmg for all of those is a just under 50% higher than the actual. This means either the ftp values given us including the ones for flameholder on at least ranged ws are all 50% higher than the actual value, there is some weird extra modifier for ranged ws that isn't a part of ranged attacks that lowers it's value, or some parts of the ranged dmg calculation are being ignored or used at a lower value when calculating ranged ws dmg

Now to check a stat mod using the lower ftp values determined above
Daze:
ws dmg: 13596
+60 dex dmg: 14784
Damage increase: 1188
Expected increase: 1914 =11 *2/3* (81+(3*60*100%)).

Note: 2/3s is the modification in dmg determined above 81 was the increase in ranged damage from dex and 100% is the perfect of dex counted for wsc and 3 is pdif which needs to be multipled by wsc before added to the damage increase from ranged attack.

So decently lower rearranging it we can try to calculate what percent is being used to be:
Damage increase = ftp*2/3*(range dmg increase+(pdif*dex added*x%))
(Damage increase/ftp*3/2 - ranged dmg increase)/(pdif*dex added)= x%
(1188/11*3/2 - 81)/(3*60)= 45%

So a little less than half the stated mod. It's close enough that probably is just 50% with some rounding issues or possibly some level correction that probably still exist on pets that doesn't on players anymore. Should note while I don't have exact numbers very basic testing of Arcuballista's stat mod (stated to actually be 50%) puts it at the same. This explains why Arcuballista always does decently more damage than Daze despite the fact that the slightly lower ftp on daze should be made up with the much higher stat mod.

I haven't gotten around to testing melee ws yet since they are more difficult but I'd imagine they need looking into. As is there is a straight 33% reduction on the damage of all ranged ws and Daze on top of that has a lower than expected stat mod that together make it's actual damage with a that dex build to be about half of what it should be