Enmity Testing (Part VIII)

May 28, 2008 07:22

Overview

In this part, I'll be going over basically everything involving "Cures" or the "-- C --" in the Enmity Table.  This part has been a long time coming and results turned out much more complex than I originally had anticipated.  Here's an overview of everything to be covered...

-  Does Curing targets for over 0 HP generate both CE and VE?

-  What variables are involved in the "cure formula"?  In particular...
            -  Caster and Target Levels
            -  Target Max HP
            -  Target Current HP
            -  Caster and Target Jobs

-  Generation of a formula to account for all possible variables.

Before beginning, I would like to specifically thank a couple of people.  This part of the testing proved to be much more difficult than the other parts.  I want to thank Shukudai and of course Ashira for helping me do the actual testing.  In addition, I'd like to thank the math and computer wizards on BG that helped to number crunch and find patterns in the raw data - particularly Aurik and Suiram.

Previous:  Enmity Testing (Part VII)

Next:  Enmity Testing (Part IX)

Enmity Table

__________________________________________________________________________________________________

__________________________________________________________________________________________________

Test 22A

Does curing a target for over 0 HP generate more than 1 CE?

H0:  Curing a target can generate more than 1 CE.
      H1:  Curing a target cannot generate more than 1 CE.

Testing Setup:

- 2 level 75 members of the same job and HP with access to Cure and Stoneskin (used Taru BLM/RDM x2)
- 1 player of any level to pull the mob - access to Silence.
- Use a "level 0" monster - we chose Bumblebee in West Sarutabaruta
- Access to Poison Potion (-60 HP)
- Make sure the player using Poison has no form of Regen (natural or gear)

- Player A uses a Poison Potion to lower -60 HP.
- Excess player pulls with Silence.
- Both 75 characters use Dispel once to balance enmity.
- Player A cures himself for 60 HP using Cure II, III, or IV.

- Check to see if more than 1 CE is generated by balancing hate on Player B.
- You can find the exact amount of CE generated in similar fashion to how the Enmity Table values were found.

H0:  More than 1 CE was generated from the 60 HP Cure (43 CE).
      H1:  Only 1 CE was generated from the 60 HP Cure.

This was a pretty straight-forward and obvious test.  I think the result is pretty obvious but I still put it in here for completion's sake.  The actual amount of CE generated from this 60 HP cure was 43 CE.  The actual number generated really is not of importance at this point though, since other possible variables have not been controlled.

Conclusions

Curing for more than 0 HP can generate more than 1 CE.

________________________________________________________________________________

Test 22B

Does curing a target for over 0 HP generate VE?

H0:  Curing a target can generate VE.
      H1:  Curing a target cannot generate VE.

Testing Setup:

- 2 level 75 members of the same job and HP with access to Cure and Stoneskin (used Taru BLM/RDM x2)
- 1 player of any level to pull the mob - access to Silence.
- Use a "level 0" monster - we chose Bumblebee in West Sarutabaruta
- Access to Poison Potion (-60 HP)
- Make sure the player using Poison has no form of Regen (natural or gear)

- Player A uses a Poison Potion to lower -60 HP.
- Excess player pulls with Silence.
- Both players use Dispel once to balance enmity.
- Player B cures himself for 0 HP 44 times (+44 CE).
- Player A cures himself for 60 HP (+43 CE from test 22A) using Cure II, III, or IV.

- If the mob turns to Player A then returns to Player B, VE was generated.
- If the mob did not turn, then no VE (or no noticable VE was generated).

H0:  Mob turns to Player A upon cure for 60 HP then returns to player B (about 4 seconds).
      H1:  Mob did not turn to Player A at all.

This is another pretty obvious result - again, I put this here for completion's sake.  This shows that VE was generated from curing the target for 60 HP - about 4 seconds for ~240 VE.  This value is only a very rough estimate and honestly does not matter at this point.  The test was only used to show VE was generated.

Conclusions

Curing for more than 0 HP can generate VE.

__________________________________________________________________________________________________

__________________________________________________________________________________________________

Test 23A

Does the % difference of current to max HP of the target affect the amount of CE generated by curing?

H0:  The % difference of the current to max HP of  target does not affect the CE gained by curing.
      H1:  The % difference of the current to max HP of  target does affect the CE gained by curing.

Testing Setup:

- 2 level 75 members of the same job and HP with access to Cure and Stoneskin (used Taru BLM/RDM x2)
- 1 player of any level to pull the mob - access to Silence.
- Use a "level 0" monster - we chose Bumblebee in West Sarutabaruta
- Access to Poison Potion x1 (-60 HP) and Venom Potion x1 (-300 HP)
- Make sure the player using Poison has no form of Regen (natural or gear)

- Player A uses a Poison Potion to lower -60 HP.
- Excess player pulls with Silence.
- Both players use Dispel once to balance enmity.
- Player A cures himself for 60 HP using Cure II, III, or IV.

- Using Player B to re-balance hate and determine the amount of CE generated from the 60 HP cure.
- Repeat the process only using a Venom Potion (-300 HP) on the player to get a different % HP.
- Check to see if the amount of CE generated varies.
- This can be verified at varying % HPs upon curing.

H0:  Amount of CE generated by the 60 HP cure did not vary at varying % HP.
      H1:  Amount of CE generated by the 60 HP cure did vary at varying % HP.

This test was placed to eliminate the possibility that the current HP is a factor in how much CE is generated.  A common belief by some people was that if your Cure cures a higher percentage of the target's current HP, more hate is generated - this is in fact false.  If you cure someone on the verge of dying for 60 HP, or the same person near max HP for 60 HP, you will get the same CE result.

Conclusions

The current HP % of the target of the cure does not affect the amount of CE gained.

________________________________________________________________________________

Test 23B

Does the maximum HP of the target affect the amount of CE gained by curing?

H0:  The maximum HP of the target does not affect the CE gained by curing.
      H1:  The maximum HP of the target does affect the CE gained by curing.

Testing Setup:

- 2 level 75 members of same jobs but different maximum HP, both with access to Cure and Stoneskin
- I specifically chose a Taru BLM/RDM and a Mithra BLM/RDM for HP difference of a good 200+.
- 1 player of any level to pull the mob - access to Silence.
- Use a "level 0" monster - we chose Bumblebee in West Sarutabaruta
- Access to Poison Potion x2 (-60 HP)
- Make sure the player using Poison has no form of Regen (natural or gear)

- Player A uses a Poison Potion to lower -60 HP.
- Excess player pulls with Silence.
- Both players use Dispel once to balance enmity.
- Player A cures himself for 60 HP using Cure II, III, or IV.

- Using Player B to re-balance hate and determine the amount of CE generated from the 60 HP cure.
- Repeat the process only using the Poison Potion on the other player at varying max HP.
- Check to see if the amount of CE generated varies.
- This can be verified at any max HP variation.

H0:  Amount of CE generated by the 60 HP cure did not vary at varying max HP targets.
      H1:  Amount of CE generated by the 60 HP cure did vary at varying max HP targets.

This is a test to show a slight variation to the previous test.  In the previous test, it was shown that the current HP of the target (how close one is to dying) does not affect the enmity generation of cure.  This test shows that the maximum HP of the target does not affect enmity generation on cure either.  This is actually a very good thing, since HP is really hard to control.  In every test after this, the maximum and current HP of everyone is not controlled.

Conclusions

The maximum HP of the target of the cure does not affect the amount of CE gained.




________________________________________________________________________________

Test 23C

Does the job of the caster or target affect the amount of CE generated by curing?

H0:  The job of the caster or target does not affect the amount of CE generated by curing.
      H1:  The job of the caster or target does affect the amount of CE generated by curing.

Testing Setup:

- 2 level 75 members of different jobs, both with access to Cure IV and Stoneskin (used RDM and PLD/RDM)
- 1 player of any level to pull the mob - access to Silence.
- Use a "level 0" monster - we chose Bumblebee in West Sarutabaruta
- Access to Venom Potion x2 (-300 HP)
- Make sure the player using Poison has no form of Regen (natural or gear)

- Player A uses a Venom Potion to lower -300 HP.
- Excess player pulls with Silence.
- Both players use Dispel once to balance enmity.
- Player A cures himself for 300 HP using Cure IV.

- Using Player B to re-balance hate and determine the amount of CE generated from the 300 HP cure.
- Repeat the process only using the Venom potion on the other player of different job.
- Check to see if the amount of CE generated varies.
- This can be verified using any job set, although you may be limited to Cure III in some cases.

H0:  Amount of CE generated by the 300 HP cure did not vary between job caster / target
      H1:  Amount of CE generated by the 300 HP cure did vary between job caster / target

This test shows that it does not matter what job you are curing with or what job the target you are curing is when it comes to the enmity generated by cures.  Practically, this means that you don't get less CE per the same HP cure just because you are a say WHM; likewise, you don't get more CE per the same HP cure just because you are a PLD.  You can verify this yourself with various job sets in this test if you are not convinced.

Conclusions

The target's job does not affect the amount of CE generated by curing.

________________________________________________________________________________

Test 23D

Does the job level of the caster affect the amount of CE generated by curing?

H0:  The job level of the caster or target does not affect the amount of CE generated by curing.
      H1:  The job level of the caster or target does affect the amount of CE generated by curing.

Testing Setup:

- 1 character level 75, another not 75, both with access to Cure and Stoneskin (used 70 SMN and 75 RDM)
- 1 player of any level to pull the mob - access to Silence.
- Use a "level 0" monster - we chose Bumblebee in West Sarutabaruta
- Access to Poison Potion (-60 HP)
- Make sure the player using Poison has no form of Regen (natural or gear)

- Player A (level 75) uses a Poison Potion to lower -60 HP.
- Excess player pulls with Silence.
- Both players use Dispel once to balance enmity.
- Player B (level 70) cures player A for 60 HP using Cure II, III, or IV.

- Using Player A to re-balance hate and determine the amount of CE generated from the 60 HP cure.
- From test 22A, the CE generated when using 2 level 75 characters was 43 CE.
- Check to see if the amount of CE generated matches 43 CE.

H0:  Amount of CE generated by the 60 HP cure was 43 CE.
      H1:  Amount of CE generated by the 60 HP cure was not 43 CE.

This test shows that the level of the caster does not affect the amount of CE generated from the cure.  It turns out to be 43 CE, the same as the first test using 2 level 75 characters.  The next test is almost the exact same except it will test the level of the target, not the caster.

Conclusions

The caster's job level does not affect the amount of CE generated by curing.

________________________________________________________________________________

Test 23E

Does the job level of the target affect the amount of CE generated by curing?

H0:  The job level of the target or target does not affect the amount of CE generated by curing.
      H1:  The job level of the target or target does affect the amount of CE generated by curing.

Testing Setup:

- 1 character level 75, another not 75, both with access to Cure and Stoneskin (used 70 SMN and 75 RDM)
- 1 player of any level to pull the mob - access to Silence.
- Use a "level 0" monster - we chose Bumblebee in West Sarutabaruta
- Access to Poison Potion (-60 HP)
- Make sure the player using Poison has no form of Regen (natural or gear)

- Player A (level 70) uses a Poison Potion to lower -60 HP.
- Excess player pulls with Silence.
- Both players use Dispel once to balance enmity.
- Player B (level 75) cures player A for 60 HP using Cure II, III, or IV.

- Using Player A to re-balance hate and determine the amount of CE generated from the 60 HP cure.
- From test 22A, the CE generated when using 2 level 75 characters was 43 CE.
- Check to see if the amount of CE generated matches 43 CE.

H0:  Amount of CE generated by the 60 HP cure was 43 CE.
      H1:  Amount of CE generated by the 60 HP cure was not 43 CE (46 CE found).

This test shows that the level of the target does indeed play a role in the amount of CE generated by the cure.  It seems that the amount of CE for the same HP cure goes up as the target's level decreases.  This is the final test I am going to show regarding possible variables to the amount of CE generated.  I'm sure there are others but I did not feel there was a big reason to exhaust every possible variable when many don't make too much sense.  Feel free to test yourself though.

Conclusions

The target's job level does affect the amount of CE generated by curing.

__________________________________________________________________________________________________

__________________________________________________________________________________________________

Test 24A

Determine how target level affects the CE generated by curing.

Testing Setup:

- (Player A) 1 level 75 RDM/BRD - access to Dispel, BRD songs, Cure, and Phalanx II
- (Player B) 1 level 75 SCH/WHM - access to Rapture, Cure IV, and Divine Seal
- (Player C) 1 player of a certain level to be determined - subjob /WHM for access to cure
- Use a "level 0" monster - we chose Bumblebee in West Sarutabaruta
- Access to Poison Potions (-60 HP) or Venom Potions (-300 HP)
- Make sure the player using Poison has no form of Regen (natural or gear)
- Make sure enmity enhancement is set to 0 for both Player A and Player B

- Player C uses combinations of Poison or Venom Potions to lower HP - more the better
- SCH/WHM can use Poisona to cut poison off short, leaving Player C with very little HP.
- Stoneskin on both Player A and Player B

- Player B  pull with Silence (B: 201 CE)
- Player A uses Silence three times (A: 2 CE)
- Player A uses BRD song on both Player A and B, but not hitting Player C 5 times (A: 203 CE, B: 201 CE)
- Player B uses Divine Seal then Rapture (A: 203 CE, B: 203 CE)
- Player B uses Cure IV on Player C for X amount of HP cured (up to 1292 HP with Apollo Staff)

- Player A uses various hate tools to try to rebalance CE.
- Best tools to use are Dispel (320 CE), BRD Song (20 CE per target hit), Cure for 0 HP (1 CE)
- Note the 1) level of player C, 2) the amount of HP cured by Player B on Player C, 3) CE generated

________________________________________________________________________________

Test 24B

Determine how target level affects the VE generated by curing.

Testing Setup:

- This test requires that you know the amount of CE generated by a cure for X HP at some target level (Test 24A)

- (Player A) 1 level 75 RDM/BRD - access to Dispel, BRD songs, Cure, and Phalanx II
- (Player B) 1 level 75 SCH/WHM - access to Rapture, Cure IV, and Divine Seal
- (Player C) 1 player of a certain level to be determined - subjob /WHM for access to cure
- Use a "level 0" monster - we chose Bumblebee in West Sarutabaruta
- Access to Poison Potions (-60 HP) or Venom Potions (-300 HP)
- Make sure the player using Poison has no form of Regen (natural or gear)
- Make sure enmity enhancement is set to 0 for both Player A and Player B

- Player C uses combinations of Poison or Venom Potions to lower HP - more the better
- SCH/WHM can use Poisona to cut poison off short, leaving Player C with very little HP.
- Stoneskin on both Player A and Player B

- Player B  pull with Silence (B: 201 CE)
- Player A uses Silence three times (A: 2 CE)
- Player A uses BRD song on both Player A and B, but not hitting Player C 5 times (A: 203 CE, B: 201 CE)
- Player A uses various hate tools to gain the amount of CE predicted +1 (A: 204 + X CE, B: 201 CE)
- Player B uses Divine Seal then Rapture (A: 204 + X CE, B: 203 CE)
- Player B uses Cure IV on Player C for X amount of HP cured (A: 204 + X CE, B: 203 + X CE)

- The mob will turn to Player B for some period of time, then revert back to Player A - note this time in seconds
- The amount of VE can be estimated by the number of seconds to revert multiplied by 60 VE
- Note the 1) level of player C, 2) the amount of HP cured by Player B on Player C, 3) VE generated




________________________________________________________________________________

Data Collection and Analysis

I am going to now present a table which will show the data actually collected by these tests.  Please note that I did not go around collecting this data for all 75 levels.  I only did enough to get a pretty good estimation of what was going on.  Below is a general table which includes only raw data of all levels collected.  There will be other tables which include some analysis, but this one includes purely raw data.


Level  
          HP Cured          CE                  VE    190327~ 19202100333--470200--8130288--10160320~ 192011175350--12195371~ 222014240436--15265481--25300444--27320457--37282341--40340388~ 240052380370--53380370--56628584--70418321--71585450--751029748~ 4530

I only did a couple of VE value tests since they are a bit annoying to pull off.  Anyways, this is just the raw data set.  Everything after this will use this data in some form or fashion.

________________________________________________________________________________

Relationship Between CE and VE Generated

This one was pretty easy.  All I did in the table below is cut out the values without any VE data in it.  The result is a pretty small data set of 5 levels - 1, 10, 12, 40, and 75.  Even though the data set was small, it did cover a large range of levels, including the high and lows.  Included is also the ratio of VE to CE (simply divided one by the other).


   Level

HP Cured

CE

VE

VE / CE
     190327~ 19205.871610160320~ 19206.000012195371~ 22205.983840340388~ 24006.1856751029748~45306.0561

What we see here is that the amount of VE gained relative to the amount of CE gained is 6 to 1.  This ratio seems pretty consistent throughout all level ranges in this small data set.  Based on this information, I assume from here on out that finding the amount of VE gained is as simple as finding the amount of CE gained and multiplying by 6.

Note that there is some margin of error.  Remember that all VE calculations have some error due to experimental testing limitations.  This is why all the values hover around 6 but do not all hit 6 on the dot.  I feel that the values are reasonably close enough to assume that they are indeed 6 times the CE value.  From here, I'm going to focus on CE.

________________________________________________________________________________

Relationship Between HP Cured and CE Generated

Based simply on the raw data, we can tell that there is a general trend that the lower the target level, the more CE generated for the same HP cure.  We can generalize the relationship by simply saying...

[ CE ]  =  [ CE Modifier ] x [ HP Cured ]

Where the "CE Modifier" changes depending on, and only on, the target of the Cure's level.  Our goal was to find how this "CE Modifier" changes and varies with the target's level using the data.  We went about this by taking this general equation and isolating the modifier in the equation...

[ CE Modifier ]  =  [ CE ]  /  [ HP Cured ]

The raw data provides both the CE value and HP Cured at various levels tested, so we can estimate the modifier by simply dividing one by the other.  We must remember, however, that the game must round off.  We made the assumption that the game does not round, but "cut off" the final value's decimal, making it a whole number.  With this assumption, we can provide an accurate RANGE for the CE modifier at various levels.  The actual modifier cannot be truly found, but with enough precision, a pattern can emerge.

[ CE Modifier (Minimum) ]  =  [ CE ]  /  [ HP Cured ]

[ CE Modifier (Maximum) ]  =  ( [ CE ] + 1 )  /  [ HP Cured ]

The minimum and maximum value for the CE Modifier for any given data set can be found using the equations above.  The maximum is found using the assumption that the final CE value was X.9 repeating.  Below is a table of the data along with CE modifier minimums and maximums.

LevelHP Cured
CECE Modifier
(Minimum)
CE Modifier
(Maximum)
CE Modifier (Guess)      1903273.63333.64441 / 0.25021003333.33003.34001 / 0.2754702002.85712.87141 / 0.35081302882.21542.22311 / 0.450101603202.00002.00631 / 0.500111753502.00002.00571 / 0.500121953711.90261.90771 / 0.525142404361.81671.82081 / 0.550152654811.81511.81891 / 0.550253004441.48001.48331 / 0.675273204571.43811.43131 / 0.700372823411.20921.21281 / 0.825403403881.14121.14411 / 0.875523803700.97370.97631 / 1.025533803700.97370.97631 / 1.025566285840.92990.93151 / 1.075704183210.76790.77031 / 1.300715854500.76920.77091 / 1.3007510297480.72690.72791 / 1.375




The last column (in purple bold) shows the "guess" value that was found.  I'm not going to go too deeply into how we came up with this type of pattern - if you are interested in this, check the Enmity Testing Thread on BG forums (again, thanks to Suiram and Aurik).  Basically there are a few interesting characteristics of the Mod Ranges that led to the "guess" values that we came up with...

-  The curve is NOT smooth.  Notice levels 10-11, 52-53, and 70-71.  They have same mod ranges.
          -  The curve looks roughly like a log(x) or an inverse function.
          -  All "guess" values, which follow "nice numbers" fit well into the provided CE mod ranges.

What we came up with is that you can get "nice" numbers if you use "divisors" instead of multipliers.  This sort of "fitting" is not error proof; however, I feel comfortable enough with the estimation to post it.

__________________________________________________________________________________________________

__________________________________________________________________________________________________
Conclusions

Here is the summary of conclusions reached in this post...

- Curing for more than 0 HP generates both Cumulative (CE) and Volatile (VE) Enmity.

- There are only 2 variables which affect the amount of CE and VE generated on a Cure.

- The amount of HP actually cured
                - The level of the target being cured

- The amount of CE and VE generated by a Cure of any HP at any target level can be found by...

[ CE ]  =  [ CE Modifier ] x [ HP Cured ]

[ VE ]  = 6 x  [ CE Modifier ] x [ HP Cured ]

- The [ CE Modifier ] varies with the level of the target being cured.
                          - The [ CE Modifier ] at level 75 is 0.7273.
                          - A list of [ CE Modifiers ] at all levels can be found here.

The list of "CE Modifiers" can be found at Enmity Table (Curing).  Practically, I think that this really only matters at level 75 in most cases.  As far as what this data means and how it can be used to better job play?  I'll leave that up to the reader beyond this note.

-  Cure IV for 390 HP generates 283 CE and ~1700 VE at base enmity enhancement (for you PLDs)

Also, sorry, there is no Pic of the Day or Drama Thread, this post is too long!

__________________________________________________________________________________________________

__________________________________________________________________________________________________

enmity testing!

Previous post Next post
Up