[Frost]Akraen's Comprehensive Guide to 5.4 Frost(Advanced)V.2

[Nitwit]

If you have 35% haste, and add 30% haste, you do not get 1.35*1.3 = 1.755.

Yes, you very well do.



Top is without bloodlust, and bottom is with bloodlust. 1.2606 * 1.3 = 1.63878

[pyrostorm9001]

Just a heads up. If you have based your spreadsheet on Lhiveras TheoryCraft-o-Matic Lite, there was a bug with FoF scaling with mastery that I reported to Lhivera a few days back.
It has now been fixed and unless im mistaken your spreadsheets are showing this same bug.

[Akraen]

You went from 26.06% to 63.87% which is 37.81% gain, or 7.81% above the flat percent added by the buff.

When I just hit my time warp at 13022 haste, I went from 39.78% to 81.7%, or 11.93% above the flat 30% added by the buff.

So 30% haste gives me 41.93% added haste. It gives you 37.81% gain. Time Warp therefore benefits me more than it benefits you.

- - - Updated - - -

Just a heads up. If you have based your spreadsheet on Lhiveras TheoryCraft-o-Matic Lite, there was a bug with FoF scaling with mastery that I reported to Lhivera a few days back.
It has now been fixed and unless im mistaken your spreadsheets are showing this same bug.

I verified the 2, 3, and 10 target numbers today. The image I uploaded to photobucket is older, though. I'll get that updated.

[Nitwit]

You went from 26.06% to 63.87% which is 37.81% gain, or 7.81% above the flat percent added by the buff.

When I just hit my time warp at 13022 haste, I went from 39.78% to 81.7%, or 11.93% above the flat 30% added by the buff.

So 30% haste gives me 41.93% added haste. It gives you 37.81% gain. Time Warp therefore benefits me more than it benefits you.

No, it doesn't. That's the whole point. It gives both of us 30% more casts (1.3X). It is, in fact, calculated in the way that muphrid and I are describing, and not by doing whatever the hell you seem to think it is.

If you have 35% haste, and add 30% haste, you do not get 1.35*1.3 = 1.755.

^ Correct calculation

You get (1.35*1.3) + (.3*.35), .105+1.755=1.86 or 86% haste

^ Incorrect calculation

[Akraen]

I understand fully how you're looking at this. But you're not getting it.

Do you cast Nether Tempest ticks? Do you cast the haste that is pushed to your WE? Do you cast the snapshot haste? Do you cast the mirror image snapshot?

No, you don't.

You're describing the math to calculate # of casts gained during a haste window. For that statistic, you are correct. When you are analyzing how haste impacts DPS and DPET, this is not applicable because we are looking at 1 cast, not a series of casts.

When you develop simcraft, this matters because you're looking at a series. This is a point independent of a set of data.

You're borderline trolling at this point.

[Nitwit]

Do you cast Nether Tempest ticks?

Replace the word "cast" with "tick," if it makes you feel better.

(arguments over semantics are not impressive and do not convince anyone to change their stance)

Do you cast the haste that is pushed to your WE?

No, *you* don't, but your water elemental *does* cast those spells.

Do you cast the snapshot haste?

What is this even supposed to mean? Refer back to replacing "cast" with "tick."

Do you cast the mirror image snapshot?

Again, the mirror images "do" cast the spells. I'd also be very surprised if they snapshotted stats, since (if they did), they'd be the only guardian/pet left in the game that still behaves in that manner.

You're describing the math to calculate # of casts gained during a haste window. For that statistic, you are correct. When you are analyzing how haste impacts DPS and DPET, this is not applicable because we are looking at 1 cast, not a series of casts.

When you develop simcraft, this matters because you're looking at a series. This is a point independent of a set of data.

The haste calculation is the haste calculation. When you focus on a single spell, it doesn't morph into something else. Using the numbers from the example earlier, if you have 35% haste baseline and cast time warp, the tick interval on your DoTs is (base tick interval)/(1.35*1.3), not (base tick interval)/[(1.35*1.3)+(.35*.3)]. That extra (.35*.3) term that you invented is not used in anything that I'm aware of other than your own calculations.

You're borderline trolling at this point.

I am not trolling. What's happening here is that several of us are trying to tell you that haste calculations don't work in the way that you think they do, and rather than accepting that feedback, you're lashing out with hostility. That is not the proper way to respond to criticism.

[Muphrid]

You went from 26.06% to 63.87% which is 37.81% gain, or 7.81% above the flat percent added by the buff.

When I just hit my time warp at 13022 haste, I went from 39.78% to 81.7%, or 11.93% above the flat 30% added by the buff.

So 30% haste gives me 41.93% added haste. It gives you 37.81% gain. Time Warp therefore benefits me more than it benefits you.

But these numbers are not consistent with the model you yourself just put forth. 1.2606*1.3 is exactly 1.6387. If we use your model below,

If you have 35% haste, and add 30% haste, you do not get 1.35*1.3 = 1.755. You get (1.35*1.3) + (.3*.35), .105+1.755=1.86 or 86% haste

we would get (1.2606*1.3)+(.3*.2606) = 1.6387 + .0781. You're just double-counting. Or, you're confusing yourself because for some reason you think the 30% from meta/hero should be adding and you're looking at that excess 7.81% and saying its "haste scaling itself". But it's not.


I'm well aware that assessing the impact of haste on casting rotations is not this simple. But as far as I could tell, you based your "haste scales itself" on this very multiplicative principle. I know quite well that increasing casting speed gives more time to cast filler and lessens the proportion of execution time needed to put up DoTs. I've argued against that free scaling repeatedly. But taking haste from procs and haste from gear and multiplying them and saying, "see? you gain more haste when you have more haste, therefore you're getting extra scaling" is misleading. Don't talk about how much "more haste" there is in that scenario, because all we care about really is the bottom line. Look at the % DPS increase.

I freely admit that a 30% speed proc is probably not exactly 30% DPS. I'm pretty sure it's more than that, and and NT/LB benefit more than Frost Bomb does (particularly with snapshot).


At any rate, I'm going to get away from arguing over semantics of how to present this effect. Let me ask directly about the numbers themselves.

You posted that the haste build has, with meta or hero, 1.4736 casting speed multiplier. The mastery build has 1.4101 multiplier, for a relative multiplier of 1.4736/1.4101 = 1.0450.

You also post that with unglyphed IV those numbers would be 1.3157 and 1.2734 for a relative multiplier of 1.0332. Okay. This must be what you're talking about. The relative gap between the two builds increases (from 1.0332 to 1.0450) under the effect of increasingly strong non-haste-rating speed buffs. Are these numbers using the numbers in the formula I quoted above? Where did this formula come from? It's not a simple multiplicative formula, and it can't be gleaned from the character sheet. What is the theoretical or experimental basis for it?

[Tomey]

I'm sorry Muphrid, but you're incorrect. That is not how it works.

If you were correct, it means Lhivera is wrong, AMR is wrong, I am wrong, and the Blizzard character sheet is wrong.

If you have 35% haste, and add 30% haste, you do not get 1.35*1.3 = 1.755. You get (1.35*1.3) + (.3*.35), .105+1.755=1.86 or 86% haste
If you have 45% haste from gear, and add 30% haste, you do not get 1.45*.3 = 88.5% haste, you actually are at 102 haste. Which means you effectively gained 102-86 = 16% more haste by having 10% more haste before the buff happened.

DPS does not scale consistently or linearly with haste. You cannot say by adding x% haste you gain y% DPS. Some of our spells do not have a DPS increase as you gain haste. Some spells do, but at a decreasing rate as they approach their GCD (which is the waste Pete was referring to).

Conversely, some spells do not scale with mastery, fewer-- actually.

Based upon this haste, I have a glyph suggestion that may solve the GCD issue when Meta Gem or Time Warp is up.

If we take the 18960 haste rating, which is reachable and a break point for LB and NT, the current problem is that during Meta gem proc or Time Warp, we go under the GCD fairly significantly for a dps loss with our frostbolts. One of the other problems with this break point is that because of snapshotting dots and the high haste values, Brain Freeze procs occur at a rate that we cannot keep up with unless we cancel frostbolts, which is a dps loss. At the 18960 break point we have raid buffed 62.47% haste which translates with meta gem proc up to around 130% haste. At this level, would it not be prudent to run the FFB glyph, which I believe goes under the GCD at 125% haste? This is still under the GCD, but at a miniscule level in comparison to Frost Bolt, meaning less damage loss. Also, furthering the less damage loss, because we are constantly casting during these high haste uptimes FFB, we are utilizing more FFB procs from LB but specifically NT. care to comment on this suggestion?

[Akraen]

Believe whatever you want. Fine. You win. It's your forum.

I'm actually done with you people now.

PS: You're wrong.

[Deleted]

It's amazing how many people do not understand how haste works in WoW and can't even comprehend how they are missing it (not Akraen, others in this thread).

[mrgreenthump]

From the look of things.. this thread is starting to boil over. Please do cool it down a bit or we are forced to close it.

[Tomey]

From the look of things.. this thread is starting to boil over. Please do cool it down a bit or we are forced to close it.

^This

It would be nice if people stopped fighting and complaining and pointing fingers and actually provided CONSTRUCTIVE CRITICISM. Akraen is actually right about the haste rating and how it applies to the 2 people that want to keep arguing with him over haste rating, you are wrong. You can google these formulas, and you will find that he literally is copy pasting these and using them correctly.

[Muphrid]

I'm a bit mystified by the notion that the thread was "boiling over". No one has been uncivil with Akraen in my opinion. I respect the work he has put in to try to demonstrate the pros and cons of the various approaches to gearing. That having been said, the math underlying his conclusions must necessarily be sound for those conclusions to be considered valid.

This is the first I have heard of any non-multiplicative formula for haste rating and haste proc interactions. I consulted wowpedia's page on casting speed and found this:

Spell haste can be obtained from a number of sources. Some haste comes from Haste rating, which increases all types of haste, while other sources specifically benefit spell haste without increasing other forms.

Haste stacks in a multiplicative manner. Haste rating is considered a single source, and all haste rating is combined additively with itself before being multiplied with other sources. For example:

[Borrowed Time] = 14% spell haste = 1.14
[Power Infusion] = 20% spell haste = 1.2
7% spell haste from haste rating = 1.07

1.07 * 1.14 * 1.2 = 1.46 = 46% spell haste

That's a multiplicative formula.

I went to wowhead's page on Sinister Primal Diamond and found this:

"My base haste as shadow is 25.29%... add in lust and the proc I'll be sitting around 80%"

"@falcon2099: Nops! You will have: (1.2529*1.3*1.3)-1)=1.117, and this is 111,7% Haste."

That, combined with the testing done already in this very thread, leads me to conclude that the multiplicative formula is correct.

So what is the basis of this "easily googleable" other formula that is being passed around?

[Tickspoon]

Akraen is actually right about the haste rating and how it applies to the 2 people that want to keep arguing with him over haste rating, you are wrong. You can google these formulas, and you will find that he literally is copy pasting these and using them correctly.

This claim by Akraen:

If you have 35% haste, and add 30% haste, you do not get 1.35*1.3 = 1.755. You get (1.35*1.3) + (.3*.35), .105+1.755=1.86 or 86% haste

is nonsense.

All secondary stats stack multiplicatively with each other. Haste is unique among secondary stats in stacking multiplicatively with itself. But that just means Bloodlust is multiplicative with Crit, Mastery AND Haste, hence it benefits them all equally.

[Deleted]

A very useful and informative guide Akraen. So thank you for that.

What would be useful to me (and possibly others) would be to add to it is some stat weights for the 3 builds that we can plug into AMR. I appreciate this is spoon feeding me a little bit, but it is a guide after all and it is the easiest way for me to try out the different builds,

I have a full haste build at the moment, but the biggest advantage for me of the mastery build would be to have a viable arcane offspec to try out as well.

Thanks again for the guide, and the immense amount of work in collating the figures.

[Akraen]

Not uncivil, just ignorant.



this is accurate. I fully yield to the possibility that I might communicate some math terminology incorrectly. I'm more of a student of English and business than I am of math.

However, what is incredibly frustrating about this argument with you people attempting to invalidate my thread is that you can go in game, take some gear off, hit heroism, note your haste, put some gear on, hit it again, and see exactly what I'm talking about.

During raid last night our moonkin validated what I said by removing moonkin aura and putting it back on. An entire raid of 25 people went "ohhhh that's how that works" because they didn't just get 5% tacked on.

I'm happy to deliver such a revelation to you all, but I am indeed correct and what's in that table is correct.

Thanks to all the PMs and whispers in-game I got from people. I'm glad there's a quiet majority lurking these forums that appreciates the work that goes into stuff like this.

[Salystra]

@Akraen - Thanks for the insight to the differering ways of working the frost builds, at the moment I'm going for the pure haste as although hectic I am loving the fast paced way it plays. I'm having issues working through the charts so if you could just give me a quick headsup it would be appreciated over what my haste cap should sit at and whether FFB glyphed would be a better option to ensure that BF procs always get consumed.


Although noticed you have now stripped your haste in favour of "pure int" gems?

As for the comment about "moonkin aura" its HELL in this build we had our SP healing last night for a few attempts and the amount of haste I lost was unreal lol

Thank you

[Deleted]

Hey Akraen!

Thank you for a very usefull guide, will come in handy now that I finally got back to playing frost. One question though; you say we continue to stack haste in the haste bild, though visiting your armory you have been gemming int. Is there a certain haste points you have used and then home for int or are you just trying out to spec to make out some sort of "best play"?

- - - Updated - - -

*point *gone *build *the

Harder than I thought to do this on the phone

[Salystra]

Hey Akraen!

Thank you for a very usefull guide, will come in handy now that I finally got back to playing frost. One question though; you say we continue to stack haste in the haste bild, though visiting your armory you have been gemming int. Is there a certain haste points you have used and then home for int or are you just trying out to spec to make out some sort of "best play"?

- - - Updated - - -

*point *gone *build *the

Harder than I thought to do this on the phone

He used to be pure haste gemmed I also asked the same question as you however.

[Akraen]

Hey there Fiscality, don't always go based off my armory. Sometimes I switch things around because I'm testing a theory or have a specific role in a fight.

Last night I wanted a good demo of a pure multi target set up with the haste build strategy but using intellect gems. The result was pretty incredible:
http://worldoflogs.com/reports/rt-ia...&e=4017#Akraen

I don't have any good analysis with math to describe intellect on frost bomb yet. Unlike what some people think I only post what I am 100% positive about.

It is a cool way to drop some haste for those who are uncomfortable though and remember that intellect scales with everything.

As for stat weights, no can do. AMR is updated to show haste correctly now I'm told. I'll add more to the guide as I can.

- - - Updated - - -

More incredible yet was how well the build also did single target. I think there is a big unspoken advantage to having an anchor rather than multidot, maybe I can ask Nathyiel to help explain that:
http://worldoflogs.com/reports/rt-iagzru3kq3jnx2v0/analyze/dd/source/?s=3386&e=4017&target=147