Oh that's what you mean. You should really be very specific otherwise I have to make a lot of assumptions about what was only likely conveyed.

10/16/2012 03:28 PMPosted by

Jynus But don't forget mastery actually has increasing returns the more you get compared to other stats as it's a % of a % increase to throughput.

This is still incorrect. The premise is somewhat consistent, but the implication is not.

At 700 mastery gives the same hps increase as 2000crit when healing someone near death. And 2000 mastery gives far superior hps than 4000crit. Thats increasing returns mate.

Of course Mastery does massively more HPS per point at 0% HP, otherwise Mastery would be a horrible stat.

They are linear functions. Let's hypothesize a similar scenario. Let the function y(t)=2t represent the number of cookies I have eaten at any given time. Let the function z(t)=3t represent the number of cookies you have eaten at any given time.

So at t=0, we have both eaten 0 cookies. Similarly at 0 Mastery or 0 Crit, we both have no HPS gain.

At t=5, I've eaten 10 cookies and you've eaten 15 cookies. You've eaten 50% more cookies than I have.

At t=25, I've eaten 50 cookies, and you've eaten 75 cookies. You've eaten 50% more cookies than I have.

At t=9000, I've eaten 18000 cookies, and you've eaten 27000 cookies. You've eaten 50% more cookies than have.

We went from only having eaten 5 more cookies than me to having eaten 9000 more cookies than me! Yet you always eat a constant 50% more than me. The actual returns aren't increasing: for any t+1, you only eat 3 more cookies at every given t.

Now let's consider another example, using MoP values. Assuming no base stats:

We have 20000 Mastery and 0 Crit. At 0% HP, we do 100% more HPS relative to the base.

Add 5000 more Mastery: we now do 125% more HPS relative to the base.

Instead add 5000 Crit: we now do 126.7% more HPS relative to the base.

In this case, Crit succeeded Mastery.