This site makes extensive use of JavaScript.
Please enable JavaScript in your browser.
Live
PTR
10.2.7
PTR
10.2.6
Beta
Moonkin in 3.2
Post Reply
Return to board index
Post by
murakaz
The main reason for taking the talent is the haste, which (in terms of dps value) is actually worth more than spell power until you hit the soft cap on Wrath.
Since when? Did you mean worth more than crit? Nothing is worth more than Spell Power at any gear level.
Ever since 3.2, really. It's not really significant enough of a difference to split hairs over though, and CF allows you to worry less about it and itemize for it less.
http://elitistjerks.com/f73/t59457-moonkin_beginner_guide_updated_3_2_a/
Stat priority runs in the following order:
Hit rating (to cap) > Haste rating (to soft cap) > Spellpower > Crit rating > Haste rating > Intellect > Spirit.
Post by
curlymon
The main reason for taking the talent is the haste, which (in terms of dps value) is actually worth more than spell power until you hit the soft cap on Wrath.
Since when? Did you mean worth more than crit? Nothing is worth more than Spell Power at any gear level.
Ever since 3.2, really. It's not really significant enough of a difference to split hairs over though, and CF allows you to worry less about it and itemize for it less.
http://elitistjerks.com/f73/t59457-moonkin_beginner_guide_updated_3_2_a/
Stat priority runs in the following order:
Hit rating (to cap) > Haste rating (to soft cap) > Spellpower > Crit rating > Haste rating > Intellect > Spirit.
That's an entertaining thing to see since all the theorycrafting tools they use to get those numbers says SP is still king at all levels of gearing...
Scale Factors:
Druid_T7_58_00_13_W Int=0.54 Spi=0.45 SP=1.38 Hit=2.63 Crit=0.99 Haste=1.31 Lag=0.00
Scale Factors:
Druid_T8_58_00_13 Int=0.64 Spi=0.53 SP=1.58 Hit=3.31 Crit=1.21 Haste=1.31 Lag=0.00
It's close but not more powerful then SP. Both were done at 300 haste and leaving normal numbers as the default in the profiles.
Post by
Kaitain
Interesting. 300 haste is incredibly easy to reach though so most raiding moonkins will still spend almost all of their efforts chasing more spell power. Also strange that the beginners guide puts crit above haste (after soft caps) while the sim tool (presumably set to be at those caps) puts haste first.
Post by
curlymon
Interesting. 300 haste is incredibly easy to reach though so most raiding moonkins will still spend almost all of their efforts chasing more spell power. Also strange that the beginners guide puts crit above haste (after soft caps) while the sim tool (presumably set to be at those caps) puts haste first.
Rawr places Crit higher for haste after cap as well.
I use 300 haste for to ensure that haste cap was not met. Hitting the hit cap and exceeding it are fairly easy at all levels of gearing :P
Post by
QuestionC
Interesting. 300 haste is incredibly easy to reach though so most raiding moonkins will still spend almost all of their efforts chasing more spell power. Also strange that the beginners guide puts crit above haste (after soft caps) while the sim tool (presumably set to be at those caps) puts haste first.
This is a problem with the sim tool. I love Simulationcraft, but its scale factors are meaningless when any kind of cap is involved.
Post by
curlymon
Interesting. 300 haste is incredibly easy to reach though so most raiding moonkins will still spend almost all of their efforts chasing more spell power. Also strange that the beginners guide puts crit above haste (after soft caps) while the sim tool (presumably set to be at those caps) puts haste first.
This is a problem with the sim tool. I love Simulationcraft, but its scale factors are meaningless when any kind of cap is involved.
Um no... It's a very very accurate theory crafting tool and it
does
take into account stat caps... If it did not do this it would not be used at all... I'm not sure how you got this belief..
The numbers shown are under cap the haste cap. Once over the cap things swing far more in favor of Crit, as it should be.
Post by
QuestionC
Um no... It's a very very accurate theory crafting tool and it
does
take into account stat caps... If it did not do this it would not be used at all... I'm not sure how you got this belief..
The numbers shown are under cap the haste cap. Once over the cap things swing far more in favor of Crit, as it should be.
I get this belief from knowing how the program works, and am a little offended at the assumption that I'm clueless because you disagree with me.
In the remote off chance that you don't know how it works, here's a summary.
1. Add 125 Haste to the character. Check the DPS.
2. Subtract 125 Haste from the character. Check the DPS.
3. Take the difference, divide it by 250.
There are some other subtleties to it (special handling for Hit and Expertise, various options), but that's the gist of it. It's just a simple formula with none of the insane math that would be involved in detecting soft caps.
So why the confusion? Are you looking at an older version perhaps?
Post by
curlymon
Um no... It's a very very accurate theory crafting tool and it
does
take into account stat caps... If it did not do this it would not be used at all... I'm not sure how you got this belief..
The numbers shown are under cap the haste cap. Once over the cap things swing far more in favor of Crit, as it should be.
I get this belief from knowing how the program works, and am a little offended at the assumption that I'm clueless because you disagree with me.
In the remote off chance that you don't know how it works, here's a summary.
1. Add 125 Haste to the character. Check the DPS.
2. Subtract 125 Haste from the character. Check the DPS.
3. Take the difference, divide it by 250.
There are some other subtleties to it (special handling for Hit and Expertise, various options), but that's the gist of it. It's just a simple formula with none of the insane math that would be involved in detecting soft caps.
So why the confusion? Are you looking at an older version perhaps?
It is in essence a simulation why would a creator leave an inaccuracy in their simulation? They would not release a broken simulation and just leave it that way. Try more then one version yourself. The simulation has the mechanics hard coded into it. I does not have special functions to make sure a soft cap would be enforced... The math inherent in the calculations would cover that on it's own...
As for the subtleties of math... finding a soft cap does not even remotely need complicated math. Predicting one mathematically can be fairly complicated but anyone who understands the math already can easily determine that one would be present and even a close set range of numbers that you would find it within...
The scale factors are only off if you do large value changes to a stat. Specifically one that would have you going into or out of "soft capped" as a state.
I did not imply you knew nothing about the program or nothing in general. I'm only saying that SimulationCraft would not be so widely used if it was not accurate.
Post by
QuestionC
It is in essence a simulation why would a creator leave an inaccuracy in their simulation? They would not release a broken simulation and just leave it that way. Try more then one version yourself. The simulation has the mechanics hard coded into it. I does not have special functions to make sure a soft cap would be enforced... The math inherent in the calculations would cover that on it's own...
....
...
Look man, I'm honestly a computer programmer who got his info from reading the source for Simulationcraft (sc_scaling.cpp). Beyond that, I got my major by writing a thesis on the density of primes and... yea, whatever. I'm friggin' qualified to comment on whether or not that math required to determine a soft cap through simulation would be 'insane math'.
I am not bull$%^&ting you when I say that the scaling factors calculation in simulationcraft has some serious intractable problems.
Trust me. Please trust me when I say that all Simulationcraft does is take a fairly large spread and takes the average over it. It has no concept of a soft cap.
Post by
curlymon
Trust me. Please trust me when I say that all Simulationcraft does is take a fairly large spread and takes the average over it. It has no concept of a soft cap.
The scale factors are only off if you do large value changes to a stat. Specifically one that would have you going into or out of "soft capped" as a state.
Sounds like we are saying the same thing... degree or not. Following the math to find a soft cap is easy. Finding the exact spot it would occur at is complicated. Recognizing that one would be in effect is not.
As its stands now... ~400 haste is the soft cap. If the scaling calc moves to either side of this then yes indeed it would be skewed. BUT we know this is present and that is why I took a value for haste that was 100 point lower then the cap itself. I don't want this to be an argument, So let me reiterate:
I did not imply you knew nothing about the program or nothing in general. I'm only saying that SimulationCraft would not be so widely used if it was not accurate.
Recognizing a flaw is a good thing but that does not invalidate a program.
Post by
116387
This post was from a user who has deleted their account.
Post by
curlymon
Easy way to think about it when looking at gearing.
Haste should be taken over crit until the softcap of ~400. After which crit should be taken over haste. Both have value and should not be avoided by any means either before or after the softcap. Spellpower is always worth more point for point then either of these two.
Post Reply
You are not logged in. Please
log in
to post a reply or
register
if you don't already have an account.