[dev1027] [dev1032] Status Allocation System + Damage Calculation Broken/Unbalanced

[Deus]

I didn't get any response in my last post so I am putting up the dev tags and writing out my conclusions clearly on this forum. Skip to #9 if you're impatient.

The following 3 posts offer empirical data supporting all the statements I am making in this post:

Test Central: Log Parse Results = MAX Attribute+Gear Effects
http://ffxiv.zam.com/forum.html?foru...7176564&page=1

Log Parse Results! Effect of (INT PTY RANK ELE)on THM Spells
http://ffxiv.zam.com/forum.html?foru...07363078848581

Log Parse Results! Effect of (INT PTY RANK) on Phantom Dart
http://ffxiv.zam.com/forum.html?foru...32715260171115
------------------------------------

1) Empirical Test Conclusion: Increasing Piety has ABSOLUTELY NO EFFECT OR CORRELATION TO INCREASING ACCURACY OR LOWERING RESIST RATES for Phantom Dart, Damnation, Scourge/II, or Shadowsear.

In none of my tests did Piety have any effect whatsoever, even with dramatic changes from 40 to 100 PTY. PTY is an utterly useless stat to invest in.

2) Empirical Test Conclusion: Increasing INT values does steadily increase Phantom Dart average damage. However it may not affect Weapon Skill (Damnation) or Nuke (Scourge/Scourge II) damage. Either the cap was met at a lower value in my test, or INT does not affect Weapon Skill or Nuke damage. Phantom Dart does not yet appear to cap at all.

Increasing INT with Phantom Dart increased damage steadily 0.25% (R43) to 0.5% (R48) PER 1 INT. See test for more details.

3) Empirical Test Conclusion: With Phantom Dart and Scourge, average non-critical and critical damage increases in direct proportion when the change in (PlayerRank/MobRank) increases. Further, there seems to be a point where the rank difference dramatically changes damage. This was observed on the Rank 59 mob.

All tests showed consistently that the R59 mob defensive abilities increased 20% from 58 to 59 while from 55 to 58 each step is only 3%. This proves the calculation system was unbalanced due to player and mob rank. This may be because the R59 mobs' innate stats were raised inconsistently.

4) Empirical Test Conclusion: Per #3 and #2 above, the effect on damage output of increasing your Rank by 1 is as influential as increasing your INT by as much as 16 in this test. Compound this with the players ability to load the INT stat and you have a very unbalanced system (see 5).

5) System Conclusion: Per #3 above, the status point allocation system forces the need to make player rank far too influential in damage calculation, just to mitigate the unbalance caused by loading up stats. The player's ability to so drastically change status parameters makes the effect of changing parameters nearly useless compared to changing rank. Being able to so widely set the range of INT is just plain sloppy and useless.

6) System Conclusion: Per #5 above, the status point allocation system makes any gear status bonus utterly useless, as the range you can increase status points makes a typical gear's +1 to +7 bonus pale in comparison. Per above, a +7 INT bonus inconsistently increases damage, and in the best case the increase is not even half the effect of +1 player rank. It was also untested if gear status points even contributed at all. It's true that increasing VIT and MND with gear bonus has no effect on HP or MP.

7) Appeal: Please produce a combat system where increase of player stats will directly and consistently impact damage output, and lower misses and resist rates.

8) Appeal: Consider smooth curve functions when players and mobs change their stats as rank changes. Dramatic inconsistent differences with just a small variable change will only confuse players. As evidenced, the rank 59 to 59 mob results in a 20% damage decrease compared to the typical 3% change as rank changes by 1.

9) Final Appeal: Implement the best solution, eliminate player and mob rank values from the calculations completely, and eliminate the status point allocation system completely. Rely solely on the primary stats (STR INT DEX PTY MND VIT) that are increased and FIXED accordingly depending on rank and class. This will enhance class (and mob) uniqueness.

This will dramatically increase the effect of raising stats with gear. Raising your STR by +10 via gear bonus should result in a damage output as if they had ranked up 2 levels. This will exponentially increase the value of elite end-game items.

It will create a strong baseline for balancing the calculation system by directly modifying ONLY the primary stats on players or mobs. The current system is too complicated as was admitted by Naoki Yoshida in a recent fansite interview. It worked just fine in FFXI. If it wasn't broken, don't fix it.

Players. if you agree with this assessment, please "like" the post and/or post here to ensure the dev team understands the importance of the issue. If you disagree, please note the # above and explain why.

[REDace0]

I really hope the system we get in the Big Battle Patch follows your suggestions or otherwise addresses your valid concerns. The devs have stated that the current formulas are too complex to balance with a practical amount of work, so the patch should bring very different and simpler formulas for damage. It sounds like relations between stats, regardless of rank, would make the most sense, even if I no longer get XP. Oh wait...

[Belial]

Liked, appreciate the time and effort you've put in, this gives some interesting insight and backs up my own trial and error testing. Fully agree, although with these issues in development, I imagine they have already created their own answer to the problem.

[Deus]

Tweaked my points a bit more. I hope they actually have created their own answer to the problem, and will share it with us.

[Jinko]

Totally agree with this post.

Ideally stats need to be given to the class as a fixed incremental per level, balancing this kind of system would be much easier not only for healing/damage output but for balancing monsters and NM's.

Removal of Physical levels also as they would no longer be relevant, this would at least go in-line with the 2nd poll results.

[Rydin]

Didn't the Delta rank ratio play a large part in FFXI's damage formula also?

[Zephy]

Ah, this is fantastic work. Thanks a lot. One thing is that I've seen with my own informal tests that CON spells do indeed scale with INT and with Elemental Allotment. I brought up the disparity in my own thread:

http://forum.square-enix.com/ffxiv/t...nd-As-Intended

But it's absolutely great that you've provided some solid data backing up my claims with respect to THM nukes. Your figures on PIE are pretty damning, I'm wondering it extends even into Bio/Dia/Poison, as SE really pushed the influence of PIE on DoT spells.

[Rydin]

Ah, this is fantastic work. Thanks a lot. One thing is that I've seen with my own informal tests that CON spells do indeed scale with INT and with Elemental Allotment. I brought up the disparity in my own thread:

http://forum.square-enix.com/ffxiv/t...nd-As-Intended

But it's absolutely great that you've provided some solid data backing up my claims with respect to THM nukes. Your figures on PIE are pretty damning, I'm wondering it extends even into Bio/Dia/Poison, as SE really pushed the influence of PIE on DoT spells.

No empirical data to prove it, but I've felt like DoTs definitely benefit from higher PIE
Also... now, most experiments are purely for entertainment purposes, since the formula's are changing.
For example.... Before the November patch VIT had a much bigger impact on damage taken....
And logging in after the patch, we all saw a serious decrease in our HP
So with each patch, their is a chance any previous experiments will be made irrelevant...

@OP I do commend your efforts and your grasp of the scientific method, especially in regards to having as few variables as possible....
Very nice work....

[Amineri]

I can supply some additional data for you (PM me if you want to see the data gathering spreadsheet).

For Conjurer Cures, each 2 points of Mind adds 1 point to cure amount, and each 3 points of Magical Potency adds 1 point to cure amount. No cap that I could discover (and if there is a cap, it is above the MP pool based on Mind).

Just these two value can completely predict cure amounts, and I have been unable to find any other factor.

[Deus]

Awesome I'd love to see your numbers. I can integrate it into my thread and give you credit for it. Really appreciate anyone who can contribute test data.