Tuesday, August 3, 2010

Why Genesis Still Sucks.

In my last post I said, "The only way to make Genesis significant would be to increase the buff significantly or increase our base DoT damage significantly." Some readers seem to have taken issue with at statement and posted comments in to support of Genesis. They implied that since DoTs are now impacted by Haste and Crit that DoT damage would significantly increase and thus Genesis would be come valuable.

I can't say whether or not Genesis will become a part of most Moonkin builds in Cataclysm. There are still to many moving parts to say that definitively, but I can say one thing. If the current version of Genesis is a part of most Moonkin builds in Cataclysm, it is not because it is a strong talent. Here's why.

DPS vs DPET:

I want to clarify what these two acronyms mean. I've seen a little confusion on the forums and such and want to make sure we are talking about the same stuff.

DPS - Damage per Second: DPS is a summary measurement. It tells you how well you did over a period of time. It's good for evaluating changes in rotation and spec, but it tells you very little about how to play your class without making those experimental changes.

DPET - Damage per Execution Time: DPET is purely a decision making tool. It tells you what spell to cast next. It's not as good at measuring overall damage because it is focused on a single spell. It can also cause you to overestimate the overall impact of a spell if you don't' understand the spells DPS value.

In this post I am going to be using DPS because I want to look at the overall impact of the Cataclysm changes. Yes, DoTs have very high DPETs. That is a good thing because we wouldn't cast them otherwise, but DPET doesn't tell us what they contribute to our overall DPS.

The Goal of the Test:

As I said above, we are still in a relatively early stage of development. It is impossible to say what will go live in a few months. However I want to gauge the impact of the Haste and Crit changes we've seen in the beta. To make things a little simpler, I am going to apply the changes to a level 80 moonkin. I will take a look at the current DPS of my DoTs. Then look to see how that will change with the Haste and Crit scaling of Cataclysm. I am not going to use all of the changes we've seen in Cataclysm because I think it would just muddy the waters. In the end I hope to show how haste and crit will in prove out DoTs. This in turn should allow us to estimate how significant DoTs will be for the Moonkin Rotation and finally show what impact Genesis will have.

Assumptions:
  1. For these calculations I am going to base the stats off of a fully raid buffed Graylo. I will use 4650 spell power, 56.88% crit chance, and 36.40% haste without NG.
  2. I am also going to exclude Glyph of Insect Swarm, Glyph of Moonfire and Glyph of Starfire from the calculation. A big glyph revamp is coming and who knows if these glyphs will exist. Also, GoIS and GoMF would only lower the value of Genesis since it is an additive bonus. GoSF is excluded because it would have no impact on the calculation.
  3. I am going to apply NG at 15% rather then 20% since it has been specifically changed by Blizzard. I am also going to assume a 100% uptime. It's unrealistic, but it's easier and favors Genesis.
  4. I am including the damage bonus of Imp Moonfire but not the crit bonus. I do this to keep the damage consistent between WotLK and Cata, but not to over value how Moonfire will scale with Crit when the crit bonus is not available in Cata.
The Current DPS of DoTs:
IS per tick DPS = ((215+(4650*0.2))*1.2831)/2 = 734.57 DPS
MF per tick DPS = ((200+(4650*0.1302))*1.2831*1.2)/3 = 413.38 DPS
These values probably look low to you. That is because they are. First they exclude the glyphs, but it is also good to remember that this is a DPS look. Thinks that extend a DoTs duration don't increase the DoTs DPS even though they do increase the DoTs value. I also want to define the 1.2831 number I used. That is a combination of Earth and Moon (raid), Earth and Moon (self), Arcane Empowerment, and Master Shapeshifter.

DoT DPS with Haste and Crit:

IS per tick Damage = ((215+(4650*0.2))+((215+(4650*0.2))*(0.545*0.5688)))*1.2831 = 1925
IS tick Duration = 2/(1.3640*1.15) = 1.2750 seconds
IS per tick DPS = 1925 / 1.2750 = 1509.80 DPS

MF per tick Damage = ((200+(4650*0.1302))+((200+(4650*0.1302))*(1.09*0.5688)))*1.2831*1.2= 2009
MF tick Duration = 3/(1.3640*1.15) = 1.9125 seconds
MF per tick DPS = 2009 / 1.9125 = 1050.46 DPS

As you can see having our DoTs scale with Haste and Crit has a huge impact. It would cause the DPS of Insect Swarm to increase by 205%. The DPS growth of Moonfire is even bigger at 254%. Now lets add Genesis to the Equation.

DoT DPS with Haste, Crit, and Genesis:

IS per tick Damage = ((215+(4650*0.2))+((215+(4650*0.2))*(0.545*0.5688)))*1.2831*1.06 = 2040
IS tick Duration = 2/(1.3640*1.15) = 1.2750 seconds
IS per tick DPS = 2040 / 1.2750 = 1600.00 DPS

MF per tick Damage = ((200+(4650*0.1302))+((200+(4650*0.1302))*(1.09*0.5688)))*1.2831*1.26= 2109
MF tick Duration = 3/(1.3640*1.15) = 1.9125 seconds
MF per tick DPS = 2109 / 1.9125 = 1102.75 DPS

So, Genesis increases the DPS of IS by 6.00% as expected in this model. Moonfire is increased by only 5.00% when you factor in the additive bonuses of Moonfury and Imp MF.

Why Genesis Still Sucks:

So far so good right? Our DoTs have improved by more then 200%. Genesis now improves them by 6% instead of 5%. Genesis has got to be good now right?

Wrong.

DoTs are only a portion of our DPS. One of the key reasons Genesis sucks is that our DoTs make up a relatively small portion of our DPS. In ICC WotLK, DoTs are about 13% of a moonkins DPS. If the improvements go exactly as I've outlined above DoTs would increase to about 29% of our DPS.

So, the 6% from Genesis would only impact 29% of our rotation. Not only that, Genesis has a bigger impact on IS then it does on MF. In the end, a 6% increase to our DoTs ends up being about a 1.64% increase to our over all DPS. Given that this is an idealized situation that is a pretty small increase. It is a less then 1% per talent point and is even worse considering that the number of Talent points have been cut in half.

Adding glyphs would limit the significance of Genesis. Remember that glyphs and Genesis are additive bonuses. This means that if the Glyph of Insect Swarm becomes favored again, the 6% increase from Genesis gets eclipsed by the 30% increase from the glyph. For Moonfire, I already showed above how just Imp Moonfire and Moonfury reduces the buff from 6% to 5%. Now think about how another 75% from Glyph of Moonfire would limit the value of Genesis.

The Glyph of Insect Swarm is not typically used current, but adding the Glyph of bonuses reduces the overall DPS increase of Genesis from 1.64% to 1.40%.

If the traditional glyphs are excluded then the DPS of our DoTs is greatly reduced. Excluding the Glyph of Moonfire would result in a 16% decrease in the overall DPS from Moonkin DoTs and thus reducing the impact of Genesis. If you exclude the impact of Glyphs the over all impact DPS impact of Genesis shrinks from 1.64% to 1.40% as well.

So, glyphs or no glyphs Genesis is looking more like a 1.4% DPS increase instead of a 1.64% increase.

TL-DR:

There is no doubt that having our DoTs scale with Haste and Crit will dramatically change how we value those spells. However, that does not mean every talent associated with a DoT is a good talent.

Some people are speculating that the increased value of our DoTs will breath new life into the much ignored talent Genesis. Unfortunately, the Math does not back that speculation up. In a very optimistic scenario I estimated the overall DPS bonus to be 1.64%. That is just 0.55% per talent point which is very low when compared to most of our other talents.

The most telling figure is that even if DoTs made up 100% of our DPS and Genesis would still only represent a 6% increase to our DPS or 2% per talent point. The average value of talent points is about 1% currently. Since Cata has half the number of talent points the average in Cata is probably 2%. Therefore, in the most unrealistic of circumstances the best that Genesis can be come is average.

It is very possible that the current version of Genesis could become a standard part of most raiding moonkin builds. If it is, its because we lack any better options, not because it is a strong talent.

7 comments:

Erdluf said...

When they announced they'd be reducing the talent count, I assumed they'd be making them stronger, point-for-point. That may have been a bad assumption.

Furor is stronger (because Int is stronger). Genesis is stronger, but still terrible. BoP is stronger, in a weird sort of way. Solar Beam, Lunar Shower and Fungal Growth are very strong utilities, but are pretty weak "turret PvE" abilities.

There isn't any other Balance talent that feels stronger, point-for-point than its LK version (or something it replaced).

I don't mean to whine. I really hope to have some fun with our new "utility" and other changes.

I guess I'm saying that even though a min/max PvE build may include Genesis, I really hope and expect to spend my points elsewhere (sometimes it is good to be a casual player).

Relevart said...

Both of our main nukes have lost significant strength and when Eclipse is implemented, it will come at a lower value than before. So using todays percentages to talk about the damage distribution is wrong. I'm carrying a static spellpower close to what you have raid buffed over on the Beta server and I haven't seen a Starfire crit over 16k. Wrath is a horrid spell that has a greater MPCT and lower DPCT than Starfire, making it an even poorer choice. The DPS of both Starfire and Wrath is incredibly poor over on Beta right now and the dev silence in response to our feedback is a cause for concern.

With the introduction of Starsurge and a scaling Eclipse model that can be boosted through mastery, I am not expecting our baseline spells to become much more powerful. This is to separate Balance druids from other druids casting offensive spells.

I'm not sure what you're looking for out of Genesis, but there will always be a weakest DPS talent. Currently, I don't think that's Genesis. But even if it is, where else would you want to place the points? Frankly, the problem with DoTs isn't the damage they do, but its the lack of synergy within the rotation. DoTs will always seem to play a "small" role in the overall DPS due to the fact that to get the most out of them, we are required to let them run for a long time, effectively placing a cooldown on them.

If I were to focus my ire on a talent buffing DoTs, it would be either Lunar Shower or Blessing of the Grove, both of which are problematic in terms of actual benefit within the Moonkin rotation.

Graylo said...

@Erdluf

I made the same assumption as you, though I am not real worried about the current design of the talent tree. If this tree went live then I would have big concerns, but I think they are still playing with the numbers at this point.

@Relevart

The trigger for my post was that both you and Qieth commented that the value of our DoTs has increased significantly. Given the context think you are impling that Genesis is now a strong talent because our DoTs are significantly stronger.

In my opinion that is an incorrect conclusion because are a relatively small portion of our rotation.

I also disagree that using today's percentages is wrong. I know it isn't ideal, but it is a good indication of what Blizzards design goals have been historically. Of course, history isn't necessarily an indication of future design, but so far the only thing we've seen that indicates any kind of shift are a few early Beta builds which could easily be untuned and misleading.

At this point I think it is a little early to assume that Nukes are going to represent a smaller part of our rotation. They have only just started putting numbers to Eclipse. Mastery Rating was only added recently as well. Not to mention that you can't level to 85 yet and no testing has happened in actual raid situations as far as I'm aware. Using the current live percentages may not be ideal, but using the current beta numbers is probably equally misleading given the stage of development.

I'm not looking for anything out of Genesis actually. As long as our DPS is competetive I don't really care how we get there. My point is that Genesis should be evaluated honestly and realisticaly. The arguement that DoTs are now strong there for Genesis is now strong is completely incorrect. Stronger DoTs has improved Genesis, but it has been so weak that even doubling or tripling is value does little for the relative value of the talent at least in historical terms. I don't care if we pick it up or not. I will pick it up if it is the best available talent, that doesn't mean it isn't a weak talent.

All that said, it is quite possible that my expectations are out of wack. Blizzard could vary well be trying to lower the impact that talents have on player DPS. In such a situation then Genesis may seem more balanced. However, that will be hard to determine until we see a more polished design for Moonkin.

Gevlon said...

At first, I assume you have 56.88 Crit CHANCE and not Crit RATING.

Also, the value of DoTs depend on the number of targets. Due to their high DPET, they are preferred when several (but not many) targets are present, for example LK+valky, LK+ghost or Sindy+tomb.

We don't know how much Blizzard will rely on multi-target fights, therefore how much DoTs will be worthy.

Meiko said...

@Gevlon

Not necessarily. Right now they're only bigger DPET when they run the whole duration on the target. As in, theres not a single tick of the dot lost. I haven't actually found any adds in ICC(except Sindra tombs, but they die too fast anyway) that live longer than the dot duration is. So right now for moonkins the smartest thing is to keep spamming nukes.

paperclip said...

I think the assumption that the per talent point damage budget is going up may be wrong. The way I understood the move to a smaller talent tree was that half of the talent tree budget was moved into mastery (or specialization, or whatever they are calling it now). Thus each remaining talent still has the same budget of ~1%/point. The mantra of "more choice/more fun" further suggested that utility rather than simple dps might consume more of the budget.

That doesn't change your analysis of genesis of course.

Relevart said...

Ah, I am not implying that Genesis is a strong talent, by any means. I do think it has improved in worth and I also think that it's not a huge worry for Boomkins at the moment. What I mean by that is, in an ideal world I'd like to see more out of the talent. But as I mentioned, I think there are bigger black holes in our talent tree that need to be considered such as Euphorias interaction with Eclipse while the Eclipse buff is active, Lunar Shower/Blessing of the Grove being too weak, Starfall lacking oomph, Owlkin Frenzy still being shaped as a PvP talent (unless how it procs is significantly changed, Moonkin Form lacking any true interaction, and Fungal Growth having no forseeable PvE application (maybe for fights like Faction Champs...)

I gotta say that I think the design goal for DoT spells have changed. I think the fact that they allowed all DoTs to crit and be affected by haste indicates that they are anticipating a higher contribution of damage from these spells. But in reality, any spell that only buffs our DoTs will seem like a small 'per point' increase due to the nature of DoTs (the implied cooldown I spoke of earlier). The wording of the Eclipse tooltip at this moment implies that the respective DoT will benefit from the Eclipse bonus as well. This will help our DoTs stay competitive. With so much being pushed on DoTs (or so it seems), they could become too dangerous outside of PvE settings were Genesis to get buffed significantly.

On the whole, I'd much rather see Genesis carry over something like Improved Insect Swarm (although I think that should also be inaccessible to other specs, so it's a bad place for it). A flat percentage on DoT damage will likely always feel small and seem under-represented on a per point value.