I'm not going to point fingers; I like arguing math with math.
What is this "something" and what is this "other external influence"?
I'm not aware of any MATK bonus that is reliant on some sort of external influence while being achievable with more MATK. It doesn't even make sense.
It would basically require a card that said "if you have more than 1100 MATK, or have external influence X, +100 MATK."
It is this kind of sketchiness which causes a lot of confusion in people, leading to my earlier post. 1100 (stat, I presume) MATK is irrelevant to whether or not %MATK or +MATK is beneficial, but it depends on the actual bonuses in question. I can imagine that for your build and your stats 1100 MATK is your magic breakpoint at which this happens, but no math or proofs have been posted as to why this is so.
I politely request the "mere maths" you speak of.
And that is why people are getting confused. Even if that 1100 MATK barrier turned out to be universal, your point is moot without proof as to why it exists.
That is an entirely different concept. DEX and LUK are bound variables in a very well-defined system. It is mathematically simple to determine an optimal amount of matk from any given stat points, distributed between dex and luk. And because I like proofs:Benefit = floor(DEX/5 + LUK/3) StatCost(DEX) + StatCost(LUK) <= N For Stat < 100: StatCost(Stat) = ∑(Stat,k=1) [Floor((k-1) / 10)] + 2 For Stat >= 100 StatCost(Stat) = (∑(99,k=1) [Floor((k-1) / 10)] + 2) + (∑(Stat-99,k=1) [Floor((k-100) / 5)]*4 + 16)
More or less that (it's not exact, and I probably missed a few points here and there since I didn't test it, but that's the gist of it). From there it becomes a two-variable system. Or you can do it iteratively.
What you're describing as a "barrier" is as much a barrier as a car has a "60MPH barrier" because it's the point where it stops going faster when you push the pedal. There may very well be a reason for it, and there very well may be a solution, but - without proofs - it's relative to that car and shouldn't be pushed around to other people without justification.
I assume this is your justification for your 1100 MATK number. Please, do post your graphs, I'd love to see them. I'm not sure I understand this paragraph at all, because I'm not sure how statistics factor into this. You're a scientist, please back up your claims. I'd simply like proof as to why you call this 1100 MATK point special and, from the way you say it, universal.
Firstly... I have yet to see any calculation or whatsoever using any data to remotely back up what you are saying. I did post the calculations I was based off at the beginning of the thread.
Hmm, where do I start? Oh yah, if you happen to read my quote, you can tell that the same way, you can save points by balancing DEX/LUK since it will give you the same amount of MATK by using less stat points (tried with INT, but INT has its unique way in which every 2 points you will see an increase of 2 pts instead of 1 thanks on how decimal points are not taking into consideration)...
I did post the math behind the MATK, the reason why I mentioned that after 1100matk you have the need to increase +MATK even further is quite simple actually, a hint, try using %matk to calculate the matk output and then try replacing the %matk with +matk and see how much increase you get from each of them... then compare and get your own conclusions.
At 1100matk, you will have that each %matk point, it will value 11matk towards the end damage... now if you happen to replace the it for +matk instead, you will need at least a card/accessories to grant you +11matk or more for it to be replaced by %matk. Rata card is the latest one that will give you +10matk towards equipment but once you pass that mark, +matk will became pointless because the matk output will be less than the expected even using the same slot.
Therefore, once you reach 1100matk, the required +matk to replace %matk will be +11 in which, there are few items to grand such thing while there are better options to give you %matk instead... that plus adding the new gear coming from kRO, increasing +matk will be futile and obsolete.
Lets see... from this:
Benefit = floor(DEX/5 + LUK/3) StatCost(DEX) + StatCost(LUK) <= N For Stat < 100: StatCost(Stat) = ∑(Stat,k=1) [Floor((k-1) / 10)] + 2 For Stat >= 100 StatCost(Stat) = (∑(99,k=1) [Floor((k-1) / 10)] + 2) + (∑(Stat-99,k=1) [Floor((k-100) / 5)]*4 + 16)
I believe you can tell what would be the "breaking point" or "balance point" in where the amount of points used for either DEX or LUK became relevant... if not, there's a hint, at certain point, there's a margin of 3 stat points that will make one better than the other for the same amount of MATK. I think you have the grasp of it, I do hope so.
The "proof" that you request of was quite there from the very beginning... hence why I did say that there was no need to mentioned anymore (because it was quite obvious).
Again, if you didn't understand what I said...
1100*0.01 = 11 right? so that means, after 1100matk, you will need to add >+11matk to actually get something better than +1%matk.
There are 2 ways to do so, remember that equipment MATK follows its own unique formula and by that, it needs another modifier depending on the weapon level you are holding (that not including misc matk of course).
Most buffs will add ONLY Misc MATK and as such, it won't be modified by Weapon Lv., the only way for +MATK to actually beat %MATK after the 1100 mark (in which, each %matk will be +11 or higher), is for it to be classified as Equipment MATK and as I mentioned before, there are just few ones that will accomplish this, sadly, iRO just have just couple of them atm.
Of course, remember that %MATK will multiply by themselves instead of adding, and that's when my second "argument" kicks in...
The math and graphs for this is quite simple really... in fact, its simple distribution by having a media of 50% in which, the closest to it, the better output will you get.
Hows that you ask? simple!
1.5*1.5 = 2.25.
Now, what will happen if the %MATK is far from the media (50%):
1.3*1.7 = 2.21
As you can see, you will have less and less matk every time you get farther than the media..
How do you know if that applies to RO or not? Simple... trial/error is always the best way to try new things... by exchanging gears, replacing the %MATK you will get the same experience (always use a buffer of 10 tries so you can get better results).
And educated guess that you can tell by just looking at the numbers.
Any question, suggestion or something else, feel free to post it here =)
PS: I'm not posting the entire data because it won't be fun, but with the background info, I'm pretty sure most of you guys can follow from it.
PS 2: Most MATK modifiers out there will only affect Misc MATK, hence it has a random probability for it to reach max output... the probability for it to reach its max output is quite simple actually... P(A) = N1(favor)/N2(total)
∂e−∂xdx = ∂ 1∂ e−∂x ∗0 = 1
f(x) = ∂e−∂x
Please, kindly, define what you mean with "min barrier", "break point", and "external factor." And "fiction" - does that mean "not real" or "not precise" or "not defined" or what?
Repeating what you said without any clarification is terrible form in any sort of discussion.
I politely requested your math which you refer to in determining the 'barrier' you defined about at 1100 MATK.
read above... sorry, forgot to add it to my last post xD
Edited by Wizard, 19 March 2013 - 07:39 PM.