Shaking the habit

There have been numerous great guides and threads full of theorycrafting on the value of various mana regain talents we have, converting each talent point into a static MP/5 value so we can easily compare those and create the talent choices wisely. But this has been broken up somewhat by our new 4piece T10 bonus. With 0.5 seconds off the base channeling time of Mind Flay, we are casting many more Mind Flays than before.

In another angle, the amount of mana we gain from Spirit has been reduced substantially because all the good gear we want has no (or nearly no) Spirit on it. Intellect also has a much better regen value than Spirit in raids, since Replenishment, Shadowfiend, Dispersion and Innervate are all based on your maximum mana.

So, we learn to cope with less Spirit in our mana regain as well as the fact that the amount of DPS gained from it will not increase. In fact, we will likely see it removed in Cataclysm.

What does this imply?

This implies that very soon some of us may elect to get rid of Meditation, and instead pick up another talent to give similar (and improved) mana efficiency: Focused Mind.

  • Meditation (3/3): Allows 50% of your Spirit-based regen to continue while casting.
  • Focused Mind (3/3): Reduces the mana cost of Mind Control, Mind Sear, Mind Flay and Mind Blast by 15%

I have been breaking my head over this, and re-written this post multiple times. Math’s not my strongest as I have stated before, and this is going to be extremely geeky. Also, since Meditation has always been considered “mandatory”, people will advocate taking it along with Focused Mind – but the fact of the matter is that mana is not an issue for us.

The core here is whether at a certain level of gear, and a certain (low) level of Spirit, Meditation is finally matched or outdone by Focused Mind. Will those of us who have plenty mana with Meditation, now find their mana pool life expectancy increased even more by respeccing?

Some base information

The amount of MP5 gained from Spirit is not very hard to check – it’s listed in your characted sheet. You can add some to account for Prayer of Spirit, but don’t expect miracles. I personally am stuck between 75 and 85 MP/5 from Spirit while casting.

As our Haste improves, we cast more spells per time frame. Take for instance a spell that takes 2,5 seconds to cast. In 5 seconds (the time it takes to recover 1 “unit” of MP/5) we can cast 2 of these spells. Once we have 100% Haste (cutting the cast time in half), each of those spells will take 1,25 seconds to cast. In 5 seconds, we will now cast four of these spells. So we just spend double the mana, but out Spirit-based regen remains the same!

Haste going up not only consumes more mana by allowing you to cast more spells in the same time, it hits your Spirit-based regen hard because that mana regen is only time-based, it does not scale. Mana cost reductions on spells (like Focused Mind) then become more attractive. Since it applies to each spell you cast, it will actually give you more benefit the more spells you cast.

Effects of Focused Mind without Haste

Focused Mind 3/3 will give you a 15% reduction on Mind Flay, Mind Sear, Mind Blast and Mind Control. How much is that exactly? Note that these costs do not include the 6% cost reduction on all Shadow spells from Shadow Focus, which would skew the results slightly less towards saving mana with FM.

Cast time Mana cost Cost/5s Mana Cost: FM Cost/5s
Mind Blast 1.5 657 597 524 476
Mind Sear 5 1082 1082 864 864
Mind Flay 3 348 209 278 167

Mind Blast

The first thing we can look at is Mind Blast. This is the easiest to calculate, since its 5.5 seconds (talented) cooldown means that we can assume it is cast every cooldown, and then calculate how much MP/5 it is worth. With a cooldown of 5.5 seconds, will spend 597 mana per 5 seconds casting it. With 3/3 Focused Mind this becomes 476 mana per 5 seconds.

The difference between these two is 121, meaning you save 121 mana per 5 seconds with 3/3 Focused Mind and casting Mind Blast on cooldown.

Mind Sear

We can observe Mind Sear best in a specific scenario: mowing down trash. A good example is the ICC trash before Marrowgar or Deathwhisper, where there’s practically no time to drink as the raid rushes forward to get through. In this situation, without any haste, Mind Sear will take 5 seconds to cast and require 1017 mana. This also means 1082 mana spent per 5 seconds. Gosh, that’s a lot.

With 3/3 Focused Mind this becomes 864 mana spent per 5 seconds, saving 218 Mana per 5 seconds.

Mind Flay

This is actually the hardest to calculate, since you won’t cast Mind Flay “on cooldown” or “spam it”, rather you will weave it between Mind Blasts and keeping your DoTs up. Still, we have to divine some way of finding out how much we can save on it.

The only way I can think of is to assume that we will fire “5 second bursts” of Mind Flay. In reality we will do either 2 Mind Flays consequtively, or maybe 3, but not many of us master the art of the MF clipping. When we consider one of such 5 second bursts, we are assuming a scenario that we put up our DoTs, cast Mind Blast, flay for 5 seconds, then Mind Blast again. This we repeat. Not perfect, but it’s a start.

What happens here is that in 5 seconds we burst through 167 mana instead of  209, but we spend one 5-second window without using Mind Flay at all. So we halve the benefit we get from FM on Mind Flay. Half of the 42 mana benefit is 21, so we have a 21 Mana per 5 seconds benefit in this scenario.

But wait! We now have 4pT10 – our Mind Flays become 2,5 seconds base, how convenient! In a 5-second burst we can now cast 2 Mind Flays, costing us 696 mana. With 3/3 Focused Mind we use 556 mana instead, for a 140 mana difference. Halving this gives us a 70 Mana per 5 seconds benefit.

See how the 4-piece bonus changes this all?

Just considering this (really bad) use of Mind Flay, we have 70 MP/5 equivalent on Mind Flay from using Focused Mind – and you can add the 121 MP/5 you effectively get from saving mana on Mind Blast as well.

Now I can see my math being way off – let’s face it that this is very basic napkin math simply comparing base casts and costs. But I know that since I have about 80 MP/5 raid-buffed now from Meditation, this would be already equal to the half the bonus I’d get from Focused Mind. In other words, it’s worth the trade even with 50% math error.

An example of such difficulties I could foresee is the connection between Improved Spirit Tap, getting a Spirit bonus which improves the value of Meditation. It also adds to the amount of mana regained in combat – this however is easy to ignore, because that happens regardless of having Meditation.

And with Mind Sear spam on trash, the results are even greater.

Conclusion

When your Spirit-based regen starts to drop below 100 MP/5 in combat raid-buffed, it becomes worth it to look into Focused Mind instead. This probably requires you to have 4piece T10, and will scale up with the amount of Haste you have.

In other words, the more Haste you have, the more it becomes a priority to save mana per cast, rather than get a steady flow of mana.

If you have better math skills than me (I know some of you do) feel free to run the math – this is hazardous ground for me theory-wise, and I know some people might hang me up a high tree for daring to drop Meditation. In my opinion, it might not be a bad idea to just respecc to 3/3 Focused Mind and see whether my time until OoM improves.

I guess that’s the only test whether I am on to something here.

About these ads

5 Responses to “Shaking the habit”

  1. Hmm this is an interesting theory and I can definitely see how all this extra Haste is burning through my mana pool quite quickly.

    Perhaps you should take a few actual logs and see just how many of each spell was cast and then calculate how much mana was spent (with and without focused mind).

    Just an idea.

    • That seems like a good thing to try – this post is more sort of the introduction to the whole idea. I will likely try it out first on target dummies, to see how long it takes to go OoM.

      After that, I’ll try and compare a few logs of Lich King tries. These are the only real mana-intensive fights for me, and because it’s often a whole night of tries, it’s reasonably accurate.

      Still, whether I do believe that personal experience will show whether it saves me mana or not, I would still need to try and find out at what level of haste (going up) and level of spirit (going down) one talent or the other becomes the kicker.

      • As an addition to the above, I tested my MP/5 from Spirit in a raid environment yesterday, and due to Prayer of Spirit and Blessing of Kings I had about 120 MP/5.

        So when I will do my tests, this is the number to beat for a talent shift to be worthwhile to me.

  2. Nata –

    Great post. I play Disc as my primary spec, but use a haste/regen/Gheal build, as opposed to the typical SP/Crit build. Your conclusion that “. . . the more Haste you have, the more it becomes a priority to save mana per cast, rather than get a steady flow of mana” is exactly what I have concluded with respect to stacking 900+ haste in Disc.

    I typically use the Spark of Hope (http://www.wowhead.com/?item=45703) on my progression fights, because the mana savings is better than the regen of much higher ilevel trinkets.

    Vess

  3. [...] Natarumah at Twisted Faith recently theorised that there might be a point at which a heavily hasted Shadow Priest would “gain” more mana during combat by talenting into Focused Mind instead of Meditation. He’s absolutely right to ask the question. [...]

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

Follow

Get every new post delivered to your Inbox.

Join 93 other followers

%d bloggers like this: