Level Correlation and it's affect on the Fairy and You

[Ghishlain]

Hello,

Preface: I am looking for theorycrafters to look at my methodology and data points and see if there is anything amiss with it. You can find my data on this Google Spreadsheet and my methodology further in this too massive post.

TLDR: Fairy growth is intended to get weaker as the Scholar levels, it's just most noticeable at this time due to the low stat growth from 50 to 60.

So, as of late on these forums, there's been some discussion about the power level of the Fairy as the Scholar gets higher and higher in level. A few posts on these forums and a few on Reddit have notably mentioned that our Fairy actually gets weaker as the Scholar's level increases.

Some have speculated this is based on the ilvl scaling being different due to the higher ilvl ramp up. Others have speculated that this is a bug that needs to be addressed by the S-E dev team. A few have mentioned this is an intended "nerf" to Scholar's overall power to try to bring them more in line with the other healers.

My thoughts after some testing? This is actually intended by the S-E development team. It's just more noticeable at this juncture due to two things:

1) Upon level up from 50+, the amount of stat points you gain is actually quite low compared to equal jumps in the pre-50 stat.

At level 44, a naked SCH with a level 1 book has the following stats:
8 MD
213 MND
166 DET

At level 50, a naked SCH with a level 1 book has the following stats:
8 MD
261 MND (+48 from level 44)
202 DET (+36 from level 44)

At level 56, a naked SCH with a level 1 book has the following stats:
8 MD
275 MND (+14 from level 50)
212 DET (+10 from level 50)

As Fairy stats are directly affected by your stats, that means their stats are also getting an equivalent amount of low boost.

2) For an i130 SCH, there's no real need to change gear until approximately level 55-56 with i136 gear in the Aery.

This mean gear bonuses have basically stayed static until this point, and it's when Scholar's first began to notice their Fairy doing notably less Embrace values than at 50.

So, how did I come upon my thoughts? You can find my data on this Google Spreadsheet and my methodology below.

Test #1: Determine how the Fairy "grew" as the Scholar leveled

My initial tests began with trying to determine how the Scholar Fairy grew as the Scholar grew. In order to maintain a controlled sample that wasn't influenced by gear, I basically took a level 1 book (8 MD only) and unequipped everything but my RSE (no stats across the board). I then began to use "minimum ilvl sync" in various dungeons to see what the Fairy's baseline stats would be. I also compared it to my own Physick numbers to see how it was affected.

Dungeons included the following:
Level 15 - Satasha
Level 24 - Thousand Maws
Level 35 - The Sunken Templ of Qarn
Level 44 - Dzemael Darkhold
Level 50 - Wanderer's Palace (I made sure to resummon the fairy after changing my equipment back to ilvl 0 as I had to meet the ilvl minimum to enter)
Level 55 - At the time I began contemplating how to test this, I was a level 55 SCH
Level 56 - My current SCH level

It should be noted at this time that with these data points, the following trends occurred:

Pre-50 - Both my Physicks and Embraces grew slowly over the course of these tests as my ilvl increased. Expected considering one expects to be stronger as they level higher.

Post-50 - My Physicks grew in strength but my Fairy was beginning to have notable dips in her Embrace power. In fact, she even got weaker as I got higher. But the fairy getting weaker was an anticipated result, given what was known at the time.

Overall, this trend went as expected given what was known.

Test #2: Level Correlation

It was clear that level was having a strange effect on the Fairy. Originally I had attributed this to the lower than normal MND and Determination gain from leveling in post-50. This lead me to believe there was some sort of level correlation and this thought was also suggested by Dervy in a train of thought on Reddit.

You can already sort of feel this is happening because there was no way in heck Fairy healing should get WEAKER when you level. Which lead me to believe that somehow her embrace was affected by the following formula:

( Healing Potency of Embrace ) / ( Level Correlation Factor )

It's a broad formula, but it does illustrate the point sufficiently enough as at this juncture, we don't know the details or coefficients that are affecting her.

Without being able to jump into level 50+ dungeons in an undersize and minimum ilvl setting, a friend whom I was bouncing ideas off of suggested we compare things with his level 51 WHM and 50 SCH versus my 60 WHM and 56 SCH.

Originally we tested on our WHMs, with his level results at level 51 versus mine at level 50 and 60. He also gave me some data points for his non-synced gear as well, though at this juncture it's just fluff until I can figure out if there's something to compare to.

As expected, the Cure numbers did get higher as the levels went up. Okay, no problem.

Then we struck upon what happens if we made out MND, MD, and DET completely the same. In theory, if level correlation did not exist, two WHMs of exactly equal MND, MD, and DET should have very similar Curing power.

So, we test with my level 60 WHM versus his level 51 WHM. He got equipment and melded materia until we had exactly the same stat lines.

Our results:
32 MD (we both had Walnut canes)
301 MND
222 DET

50 WHM Cure II average was between 636 to 637
60 WHM Cure II average was between 664 to 666

With that in mind, we both also did a similar test between his level 50 SCH and my level 56 SCH, once again with him getting gear to make up the difference in stats due to level different.

Our results:
75 MD (we both were using Book of Diamond)
317 MND
212 DET

50 SCH Physick average was between 719 to 722
56 SCH Physick average was between 738 to 741

Okay, this works for different levels of SCH, good.

Then we compared Embraces:

50 SCH Embrace average was between 445 to 446
56 SCH Embrace average was between 379 to 380

Holy cow what a difference. So, this clearly proves that level is affecting the Fairy in some capacity where external stats would not.

Like before, we made sure our Fairy's were resummoned to ensure their stat line snap shot was equal to our stat line.

Test #3: Level Correlation pre-50

After a night's rest, I wanted to see if this negative level correlation trend actually existed before in a pre-level 50 state. To prove this, I did something similar to both test #1 and test #2

1) I used minimum ilvl sync to get into dungeons at a low level to get a baseline for a specific level
2) I adjusted my gear in a lower level dungeon until my MND and Determination matched the stats I had in the higher level dungeon.

Simple and straight forward. Here are the results:

Level 15-16 (Sastasha and Tam-Tara)
8 MD
61 MND
49 DET

Level 15 Physick had an average of 63.5
Level 16 Physick had an average of 60.5

Level 15 Embrace had an average of 38.5
Level 16 Embrace had an average of 35.5

Level 15-17 (Sastasha, Tam-Tara, and Copperbell)
8 MD
64 MND
52 DET

Level 15 Physick had an average of 67.5
Level 16 Physick had an average of 62.5
Level 17 Physick had an average of 59.5

Level 15 Embrace had an average of 40.5
Level 16 Embrace had an average of 36.5
Level 17 Embrace had an average of 38.5
*While the difference between 16 to 17 was an increase, level 17 is still weaker compared to 16. I theorize that the reason for this particular anomaly was because of stat point allocations into MND.

Level 24-28 (Toto-Rak and Haukke Manor)
8 MD
113 MND
89 DET

Level 24 Physick had an average of 98.5
Level 28 Physick had an average of 93.5

Level 24 Embrace had an average of 58.5
Level 28 Embrace had an average of 51.5

Level 32-35 (Brayflox and Qarn)
8 MD
150 MND
119 DET

Level 32 Physick had an average of 115.5
Level 35 Physick had an average of 107

Level 32 Embrace had an average of 65.5
Level 35 Embrace had an average of 61.5

Level 41-44 (Stone Vigil and Dzemael Darkhold)
8 MD
213 MND
166 DET

Level 41 Physick had an average of 171.5
Level 44 Physick had an average of 167.5

Level 41 Embrace had an average of 97.5
Level 44 Embrace had an average of 92.5

So, not only did Embrace's get negatively affected by the level difference, Physick's were also affected in the same way. The player aspect of this seems to be particular to pre-50 as it's clear a level 60 WHM has a higher Cure versus a level 51 WHM and a level 56 SCH has a higher Physick versus a level 50 SCH.

To me, this particular trend basically went unnoticed because it was always important to gear up more and more as you leveled. As long as a player remained at the appropriate gear level for their actual level, it was never clear that your powers actually got weaker as you leveled.

In conclusion - based on this research and testing, there has always been a negative trend to player power based on their level that has gone basically unnoticed until it came smashing into the spotlight with the "fairy changes". This was also emphasized by the fact that apparently player curve went positive in post-50 while fairy curve remained negative, thus making that gap even more notable.

Will S-E actually adjust the Fairy curve to suit? I guess that's only something they will decide. However, I do believe the curve was intended and the Fairy power just happens to be a side effect of this design.

If there are any thoughts or discrepancies that need to be asked, please ask. I believe the testing methodology is sound but want to be sure I get enough input from everyone to ensure I've covered all the bases.

Special thanks for Jeph Irving, my testing and theory craft buddy who helped me out with this and to n15n / Kyani Jawantal on Sargantanas for beginning the frame work on Reddit that sparked my interest. Also thanks for all the myriad of posters who brought this up in their random comments on here and Reddit that got the brains in my cog testing.

I hope you found this post enlightening and insightful.

[Bhuni]

This has been accepted as a bug and will hopefully be fixed soon.

http://forum.square-enix.com/ffxiv/threads/242504-Arcanist-Pets-Scaling-Issue

[Ghishlain]

This has been accepted as a bug and will hopefully be fixed soon.

http://forum.square-enix.com/ffxiv/t...-Scaling-Issue

Just a few things:

1) Do we happen to know which part of the bug report has been accepted? There are three individual reports there, after all.

2) To quote the rules of the bugs forum:

Bug reports moved to Accepted Bugs will be investigated and confirmed.
Depending on the issue, there may be cases where confirmation may take more time.

As a result of investigating and testing, reports determined to be bugs will be moved to the Confirmed/Planning to Address forum and those determined to be a normal feature of the game will be moved to the Working as Intended forum.

At this time it is being investigated. We don't know if this is a confirmed bug on S-E's end yet, just that they noted something is "amiss"

3) Do we know if the poster in question actually resummoned her fairy after changing gear? Fairy stats are a snap shot of stats based at the time of summoning. If you change gear, you'll need to re-summon your fairy to get the new stats.

4) There's a Reddit thread with information posting that weapon scaling does indeed change the stats of your Embrace.

5) I'm not talking about weapon level / MD scaling in my post, I'm talking about actual level correlation and how it has been affecting both the SCH and Fairy. I truly don't think it's an "ilvl scale issue" or gearing issue.

[Edit] Also realized I can't grammar *Glances at thread title* Oh well.

[Amaret]

2 Question:

1. Why do you compare a 300 potency skill (Embrace) with a 400 potency skill (Physic) instead with adlo? (300 potency)
2. Why aren't you using a comparison with a SCH pre Heavensward with a SCH now?

[Ghishlain]

2 Question:

1. Why do you compare a 300 potency skill (Embrace) with a 400 potency skill (Physic) instead with adlo? (300 potency)

When testing was first conducted by n15n / Kyani Jawantal on Sargantanas, they used Physick as their base line heal for SCH to compare to Embrace. I decided to just follow in that path initially, but it was beneficial in one aspect - it gave me a baseline heal to use for pre-30 dungeons as you can't use Adlo in those instances.

Also, as I continued to test more and more I began to realize I was testing less about the relationship between Physick to Embrace for the level. Instead, it became about the relationship between Physicks for different levels and Embraces for different levels as level correlation was showing its importance.With that in mind, I needed t osee how the spell values began to differ from level, thus spell type became less and less important.

If you want do want to compare Adlo and Embrace, you can pretty much just take my Physick results and reduce its value to 75% power to get the approximate power of Adlo and then recalculate the percentage.

2. Why aren't you using a comparison with a SCH pre Heavensward with a SCH now?

I'm not quite sure I follow what you're asking.

If by meaning "pre Heavensward SCH" you mean a SCH in pre-51 dungeons, I did do my testing in those dungeons with ilvl sync active.

Please elaborate.

[Amaret]

No i mean with SCH heal statistics before the changes hit the Server.

[Irving]

No i mean with SCH heal statistics before the changes hit the Server.

Most of the healing changes have been noticed post 50, as in 51+. Most people are testing now due to the healing drop off observed in Eos/Selene at 51+.
It is impossible to have done said healing statistics when we did not even know of this scaling problem pre3.0.

[ckc22]

IMO it's a silent nerf. They did it on purpose and were quiet about it.

My fairy used to embrace for = the heal component of adlo at pretty much all ilvls/gear.

My embrace now does about 1300 while my adlo does close to 2k heal.

Expect to get a "eff off scholars" message when they address this "accepted bug."

[Ghishlain]

I've added data points for a level 51, 52, and 57 SCH. I'll try to get the rest as I go. I also realize I should probably indicate the "+MND/DET" increase needed to match two levels so it's clear on the data sheet.

Most of the healing changes have been noticed post 50, as in 51+. Most people are testing now due to the healing drop off observed in Eos/Selene at 51+.
It is impossible to have done said healing statistics when we did not even know of this scaling problem pre3.0.

^ Pretty much this, to answer you second question, Amaret. Unless I can turn back time, I won't be able to do those tests.

IMO it's a silent nerf. They did it on purpose and were quiet about it.

My fairy used to embrace for = the heal component of adlo at pretty much all ilvls/gear.

My embrace now does about 1300 while my adlo does close to 2k heal.

Expect to get a "eff off scholars" message when they address this "accepted bug."

Funny enough, it's technically not a nerf. Looking at that data I compiled, the Fairy was always meant to grow "negatively" and the difference was to be made up in stats in the gear. If two Scholars have the same MND / Determination / MD but one is a lower level than the other, the lower level SCH will have more powerful spells and Embraces.

This seems to change when you hit post 50+ for the actual caster and is actually either staying even or getting stronger per level, but the Fairy themselves seem to stay on the same negative curve. This is where you're seeing the larger and larger divide as the two curves that were kinda in parallel pre-50 begin to part in opposing directions post-50. Admittedly, I didn't do the high level tests with a level 1 weapon, so the fact that the higher level SCH / WHM has a better cure may be because of the way the stat curve works for the weapons and will probably do some more testing some time this week, comparing my 60 WHM to a 50 WHM.

But as you say, this could just be S-E's way of "nerfing the Fairy" by not touching their growth but positively adjusting the growth of the SCH to put more emphasis on the caster versus the Fairy. Time will tell.

[ckc22]

But as you say, this could just be S-E's way of "nerfing the Fairy" by not touching their growth but positively adjusting the growth of the SCH to put more emphasis on the caster versus the Fairy. Time will tell.

Q: Does the fairy heal significantly less than it used to?
A: Yes

That's all I need to know. It's a nerf and I expect nothing from SE based on what they did for us in the expansion.