A DPS Simulator 2 - SimFF

[enil]

Here's a quick one off parse I did.
http://i.imgur.com/arNvXnP.png
And my modified version of the base script. (modified rotation and my current stats)
http://pastebin.com/QWv05mM6

One thing I notice directly comparing the numbers - what did you set leg sweep cd to? there's a significant difference between the parse and sim (for reference leg sweep is 20 second cd).

[pandabearcat]

Idk. I was probably not paying attention while writing the numbers down. All the more reason to dump all that data into a spell database file so people can change it when I make dumb mistakes.

I don't have the source in front of me so I cannot check.

Numbers seem off. Prob have to revisit formulas and/or check to make sure how stacking buffs works.

Relative damage contributions seem actually fairly okay, so rotation and/or general potency check is about right.

[Kenji1134]

I found the neural network toolkit in Matlab. Fed it my data, trained it, got very low errors... so now how do I get a meaningful equation of out it?
It seems like all the neural network does is train itself in order to process similar data. Is there any way to get the kind of equation we're looking for out of that?

[pandabearcat]

http://www.mathworks.com/matlabcentr...nction-for-gen

I didn't read this cause if I was doing it I would just feed it into the neural network every time =D but this is what you want probably.

[Kenji1134]

But that kind of defeats the purpose doesn't it?
We can already make models which match the inputs to under 0.5%, but they are only good for the input range, and maybe another 3-5% outside of the input data.
Granted since it is not overly difficult to feed new data into the model, we could continue to update the model to obtain stat weights.

But the real equation would be so nice to have. =)

[pandabearcat]

Not sure what the question is.

If you mean "can we somehow guarantee we will have the right formula, the one the devs use" the answer is no.

If you mean "can we get a formula that will fit the data perfectly" the answer is yes.

In reality, there is no difference between the two as long as we can test for the accuracy of inter/extrapolation.

Its sort of how you can get the taylor expansion of any function, there is a polynomial approximation with some error.

You want less error, you put in more terms (in our case, perhaps more hidden layers, and more data to train).

[Judge_Xero]

Thought this might be a good place to ask, but I'm curious what the relation is for DET.

Currently testing on my MCN it takes 74 Det to increase my damage by 1%. Will this always be a linear increase? i.e. 74 = 1% (despite DEX/WD Changes?)

I ask because I was looking at the relationship for all stats and both SPD and Crit have exponential gains, where as DET just starts out better.

Crit is just plain ridiculous, and I estimate by a 25% increase over the base, the effectiveness of each point will double.

SPD seems to be better than both Crit/DET out the gate, until you factor in not having an increase to AA damage, and the exponential gains are much lower than Crit.
-----------------------------------------------------------------------

My curiosity about DET is due to the way Bard now works. Since in it's DPS stance it essentially does not have AA, and has a natively high Crit chance through Straight Shot.

So if it scales well off both the exponential gaining stats then I could see it being quite the power house if you are able to stack them both high enough.
(Crit seems to plateau with DET @ about a 5% increase at which point it just increasingly gets better)

Another point I was considering is the straight shot buff, which is natively 10% X Base Multiplier of 1.45 = 4.5% damage increase.

But....say @ a 25% increase in Crit (+~1056 Crit which doesn't seem that far off) you would have 10% X 1.7% = 7% damage increase.

----------------------------

SPD example +25 = 0.01 2.49s VS 2.5s = 0.4% increase
SPD example +25 again = 2.48s VS 2.49 = 0.4016% increase

So on a class without AA, point for point SPD should start out higher than DET and further increase with each increase in reduction to the GCD.

[Dervy]

Yeah. 73 DET = 1%. Or, 1%/73 = 0.00013, which is my DET coefficient. It's the same as me doing 1/7290, which is what I actually use for my Damage formula. So you're dead on track. It's the reason why @Arisue_Neetsha uses my 0.000137 in his new updated JP formula. He ended up with the exact same conclusion as you have with his Bard.


I've got bard weights done already. Well, the only problem is sorting out potential GCD clipping with Bloodletter once we get really, really high GCD... So I cannot accurately calculate Skillspeed, but it's higher than a Monk and Dragoon.

Off the top of my head, they were:

WD: 11
DET: 0.14
CRT: 0.223
SS: 0.135

Anyway, I've sorted out my AA formula for Dragoon now, which does take into consideration Auto-Attack Delay! And surprisingly, it's more accurate than my WD Damage model for my Dragoon!

=(((WD/3*AA_DELAY)*0.029+1)*(STR*0.1484)*(DET*0.0001472+1)*BUFFS)-1.87

Or, if I convert it to whole numbers,
=(((WD/3*AA_DELAY)/34+1)*(STR/6.8)*(DET/6795+1)*BUFFS)-1

[Judge_Xero]

Yeah. 73 DET = 1%. Or, 1%/73 = 0.00013, which is my DET coefficient. It's the same as me doing 1/7290, which is what I actually use for my Damage formula. So you're dead on track. It's the reason why @Arisue_Neetsha uses my 0.000137 in his new updated JP formula. He ended up with the exact same conclusion as you have with his Bard.

Wow, thanks for your quick reply. I didn't even think to look at your blog for that sort of information and could not find the Bard Guide page anywhere. I should just set your link as a Favorite lol.

I starting looking into it because I was trying to rationalize SE's reason for not having the MCN Turret scale off SPD (+Damage or otherwise) and through that I came to the conclusion that through the psudo AA damage that the Turret does (MCN has AA outside of Gauss Barrel, but I think they wanted the Turret to be considered the AA damage for MCN while in Gauss i.e. Turret is ~20% MCN damage similar to other classes with a baseline AA % of damage)

All that just to push MCN in a different ideal gear stat priority than BRD, as that 20% difference off the top for SPD puts it so far below DET that it will never catch up.

[Dervy]

Don't think that stuff is ony blog anyway. Well, it is. But not explained. I'll be in putting a lot of stuff there this week. Bard, Monk, White Mage... Summoner maybe, and how AAs affect weights. The bard weights are with WM on btw, because who cares about it being any other way (lol!)


And yeah. It'll be weighted lower for a MCH due to your Turrets, but.. it won't be much. I mean a Bard without WM has an SS weighting of 0.103. If anything, SS will be around 0.09 for an MCH as you only have one DoT.

How does DET/WD/DEX scale with Turrets?