Thank you Beregren, but as far as the data you have presented is concerned, the numbers are only showing caps for HP and MP. We do not yet have conclusive evidence to say that all stats cap at the same numbers at specific ranks, hence the statement:
It's 82 for every class for every stat at Rank 22 across the board.
is only a hypothetical assumption.
Your reply however did provide much food-for-thought, and I am thankful for your observation that spell modifiers are not equal to stat caps. Before we can determine modifiers, it is then only rightful if we can know our actual stat cap figures.
How can we then go about doing that? A parser may make things convenient, but without one, manual recording of numbers can also do the trick, the only downside being that it is extremely tedious. Here are some ideas on how we can test each stat individually: STR
- Increases damage dealt by physical attacks
- Increases damage prevented when blocking attacks with a shield
To test STR, what we can do is to kill a certain number and same type of mobs, record all the damage per hit we dealt, then identify the minimum and maximum damage dealt from the figures. We then add 1 point into STR, do the same thing over again, until we get the minimum and maximum damage we are doing per increment in STR.
If we have a shield, we would also take note of the reduced damage numbers after successive blocks (I would think that is how shields work and how the information would be shown on screen, but I have yet to use a shield) and see if the numbers become lower as STR increases.
Should we include critical hits? I think these numbers are important and should be recorded as well, but we ought not to take them into account for the above experiment, because critical hits are primarily affected by DEX. What we may want to know is whether STR does have any effect on critical damage dealt though, and to see if the numbers will increase with STR. VIT
- Reduces damage taken from physical attacks
- Increases maximum HP
Since we already have the caps for HP, what's left to test will be the damage we take. Here, I think we ought to collect the data without the use of a shield, such that it does not take into account any additional defense modifiers. Similar to STR then, we kill a certain number and same type of mobs, record the amount of damage we received from mobs, then determine the lowest and highest damage received. We then add 1 point into VIT, and rinse and repeat the process. DEX
- Increases accuracy of physical attacks
- Increases chances of evading physical attacks
- Increases critical hit rate of physical attacks and resulting damage
- Increases parry rate for certain weapons
DEX proves to be the more tedious stat to work on. Firstly, we would want to take down the number of times we managed to hit a mob, the number of times we evaded, the number of times we get critical hits and the corresponding damage, the number of times we parried, and most crucial are the total number of hits and misses we damaged the mob, and the total number of times the mob tried to hit us. From here, we can then work out percentages of the various areas that are affected by DEX, then repeat the entire process. Moreover, the numbers would only be more accurate the more mobs we kill, so it's a whole lot of work. A parser would do wonders here, otherwise we would have no choice but to go the hard way. INT
- Increases damage dealt by magic attacks
In my opinion, INT is not broken at all, but perhaps the increase in damage is so slight that one cannot really feel much of a difference. Then again, with no solid numbers and purely by intuition, it is no wonder we tend to come to such a conclusion.
What we can do first of all is to set all our elemental attributes at 20, because at this point we haven't any clue how they affect spell damage, so it is safest to keep them at the same number. Next, we can cast one spell repeatedly on a certain type of mob over a certain number of kills, and much like STR, we take down the minimum and maximum damage dealt to them, not taking into account partial resists over here. Then we repeat the process with new values of INT. MND
- Reduces damage taken from magic attacks
- Increases maximum MP
MP aside, the only way that we can test MND and magic damage reduction would be to fight mobs that do magic damage, of which elementals would be the easiest. It would be important to fight the same type of elemental such that we only get attacked with one element type, increasing the accuracy of the test. Similar to VIT, we only need to record the minimum and maximum magic damage dealt to us by the elementals, yet besides that we may also wish to separate the data into the different magic attacks they do on us, assuming if they do that. Rinse and repeat with increasing values of MND. PIE
- Reduces chances of your magic attacks being resisted
- Increases your chance of resisting magic attacks of others
To experiment on PIE, the process is similar to DEX except it may be simpler. We only need to work out the percentages of: the number of times our spells are resisted to the total number of times we casted; the number of times we resisted enemy's magical attacks to the total number of times enemy used magic attacks on us.
I understand that spells also have chances of critically hitting mobs, yet there isn't any indication by SE where this would fall under. I would assume PIE in this case, so it may also be good to test if PIE actually does affect spell crits.
The above are just suggestions of how one can go about doing tests on the individual stats. So far, at least to me, it seems to be workable on paper. However, we ought also to remember other factors that may affect the figures, such as gear durability, and other attributes like defense, evasion, resilience, magic potency, etc. which we currently do not know how they function in relation to our primary stats.