Tuesday, August 24, 2010

The True Sources of "Fun"

Ok, I've delayed this post long enough. Two weeks ago I took a look at synergy as a concept, and tried to show why synergy is not a good thing to ask for in a rotation. This isn't to say synergy is bad, but it's not really good either. If you look closely you will realize that synergy is neutral, because synergy is a tool that can be used to achieve other goals.

Some, like Relevart, have said they enjoy synergy by itself even if it doesn't materially impact the mechanics of their rotation, but I don't think that is the case for most of the champions of synergy. I think that most of the Champions of synergy are confusing the tool with the goal, and what they are really looking for is some combination of Complexity and Unpredictability. In this post I would like to take a look at those concepts and give my impression on where moonkin are in live and in beta.


In my experience, the complexity in WoW falls into one of two categories: pattern complexity or management complexity:

Pattern Complexity: If you're DPSing a single target without having to move, then your key presses will be determined by DPET and will fall into a pattern. Eventually everything will line up and the pattern will start over. When I talk about pattern complexity I am really talking about the length of that pattern.

Pattern complexity is achieved my having multiple buttons to press with varying durations and cooldowns. The best example I know for Pattern Complexity is the Shadow Priest rotation from TBC. In that rotation the player had two DoTs with different durations, two nukes with different cooldowns, and a channelled spell for filler. There were no hard and fast rules like "you should always cast Shadow Word: Death after Mind Blast." You just had to know what spells were available and which of them had the highest DPET.

Management Complexity: Where Pattern Complexity is about pushing the best button for that particular moment, Management Complexity is all about managing an outside resource, like mana, to maximize DPS. Management Complexity is more about planning then it is about reaction.

A very good example of Management Complexity is the Arcane Mage rotation. They only have three buttons to push regularly, but how they push those buttons determines how much DPS they do and how mana efficient that DPS is. If they use their highest DPS rotation they are likely to blow through their mana very quickly and lower their DPS over the entire fight. If they play it too safe then they could end the fight with mana that could have been used to increase their DPS.

The complexity of the Arcane Mage rotation doesn't come from having a lot of buttons to push and picking the one with the best DPET, but it comes from the choices they have to make. When is the best time to have a high DPS rotation? When is the best time to save some mana? How can I end the fight with little or no mana left?

Moonkin and Complexity:

On live, Moonkin have very little Pattern Complexity. We cast only four spells regularly, but we really only cycle through 2 of them at any given time due to how Eclipse works. The DoTs have different durations, but they are not so different that they are hard to keep up if you want to. Moonkin do have a little Management Complexity in that they need to think about when is the best time to cast our DoTs relative to Eclipse, but I don't think this is hugely complex by any means.

On Beta, Moonkin still don't have a lot of Pattern Complexity currently. We do have an additional spell to manage in Starsurge, but our DoT durations match up exact with out the glyphs. If the current beta when live we would still be cycling through just 3 or 4 spells.

However, what beta moonkin lack in Pattern Complexity they make up for in Management complexity. There are already a bunch of theories out there about how to maximize Moonkin DPS in Cataclysm. Is it best to cast Starsurge on cooldown to maximize up time or do you want to only cast it in Eclipse to maximize Eclipse for the nukes? Is it a good idea to extend Eclipse as long as possible so that the DoTs get the maximum benefit? So far these choices don't seem to have a huge impact on potential DPS, but they are choices we need to make to squeeze out as much DPS as we can, and that is a lot more choice then we have currently on live.

Final Thoughts on Complexity:

How much complexity is enough and how much is to much? That is really a matter of opinion depending on the individual giving the answer. However, I do think it is a little premature for some of the Beta players to say that the Moonkin rotation is boring or not complex enough.

Erdluf made a great comment on the Moonkin Repository that I agree with 100%. He said, "I'd like to see more of the PvE complexity come from the encounter, rather than from the rotation."

It's important to remember that our rotation doesn't operate in a vacuum. Most of the comments saying that the Moonkin rotation is boring are coming from people who have tested it against a target dummy. The moonkin rotation my very well be boring in that situation, but any experienced raider will tell you that DPSing a target dummy is nothing like DPSing an actual boss. In my opinion, I think we should wait until we see how it plays in an actual fight before we make any definite conclusions about the quality of the moonkin rotation.

Unpredictability and Randomness:

I know that unpredictability and randomness are bad words to some of you, but in my opinion they are really not all that bad in and of themselves. A little bit of randomness can break up the monotony of doing the same thing over and over again, and I think the current moonkin rotation would be quite boring with out the randomness of Eclipse. However, that doesn't mean that I think all randomness is good.

In my opinion good randomness break ups the pattern of your rotation but doesn't overly dominate you total DPS. I think an example of good randomness is the Fire Mage talent Hot Streak. It breaks up the pattern of casting Living Bomb and Fireball, but it doesn't dominate their DPS. It's value isn't dramatically reduced by poor reaction times or delays forced by the fight mechanics. A good example of bad randomness is what we see in the live moonkin build.

Moonkin and Unpredictability:

On live, the moonkin rotation is very unpredictable, but not necessarily in a good way. I have gone over the problems of Eclipse before, but I think it has to main issues. First, Eclipse is way to significant to moonkin DPS. A little bit of bad RNG and our DPS goes from being competitive to being sub par. Second, it isn't resistant to interruption or poor reaction time at all. Since it is a timed buff rather then a charged buff, every second you are not using it is DPS lost. Combine this with the significance issue and you've got a recipe for how to incorporate randomness poorly.

On Beta, the moonkin rotation is almost completely void of randomness or unpredictability. The Euphoria talent does make the number of casts to proc and use Eclipse a little unpredictable but it will stall always fall into a two or the cast range.

A lot of the players on beta are upset about this, but I find it a little amusing. In the end it feels like Blizzard is damned if they do and damned if they don't. For months some players have said "eclipse is to random," and once they take the randomness out another group of players come out of the woodwork and say "eclipse is to predictable." For me personally I would like to see a little more randomness in the rotation, but I could be happy without it as long as the moonkin rotation is interesting enough with good fight mechanics and complexity.


In my opinion, synergy is unimportant as long as the moonkin playstyle is interesting given the complexity and unpredictability of the rotation in relation to the complexity of the fight. In WotLK, moonkin have had relatively little complexity, but high degree's of randomness. In the Beta, moonkin have done a complete one eighty, and have higher amounts of complexity but almost no unpredictability.

The fact that this is such a significant change, can be worrisome given the potential for error. However, I worry that some players are over reacting and making conclusions based upon limited information and trials. I would like to see how the moonkin rotation plays in an actual fight before making a conclusion on the quality of the moonkin playstyle.


Jen said...

I never really understood people who said moonkin rotation is boring... I'm by no means an expert (it's my rarely played offspec), but I found the mage rotation far, far easier. I was pretty crap at mage-stuff, I admit, but at least I could get the base rotation right, something I can't say about moonkin. I still don't know what the optimal time for refreshing DoTs is...

Sorry for not adding anything very useful, but I keep being intrigues by "moonkin = easy" opinions.

(Oh, and I like your TL;DR and don't think it's stupid.)

Blockfire said...

thanks for making a very concise and educated description of arcane mages not being two button ret paladins (two buttons instead of four).

Sean said...

The moonkin == easy crowd is pretty much right. However, every rotation/priority for a class/spec is easy on its own given time. Anyone who plays their class for more than a month or so can master it pretty easily.

It's exactly as you said - the complexity comes from the fights themselves, or adapting to a screwup within the party. Example: It's more impressive to see an elemental shaman switch gears and start healing in an emergency than it is to see that same player's mastery of his 3 button rotation. It's more impressive to see a feral druid pop out, brez a healer, innervate them, then go right back to their "complicated" cat dps rotation than it is to see a cat druid that refuses to leave form.

Anonymous said...

I have to agree with Sean. What I found really fun about being feral was tanking in Arc. Especially when the other tank picks up frosty, I pop out, brez and innervate a healer, then go back to bear and take frosty back. What is really fun though is, watching the other tank die so I'm the only tank, and popping my health, pop out, brez, innervate, then back into bear without dying and still down the boss. That is fun.

Anonymous said...

To further agree with Sean, I think the joy for me as a moonkin is to do things that other casters simply can't. I've played moonkin ever since the form was released (1.8) so I'm thankful for the evolution of the spec from spamming starfire to an actual rotation.

However, what makes me come back to the moonkin (I play many other classes) is that utility. Rooting/Cycloning mobs in place when things get out of hand; intelligent use of typhoon to interrupt casts and/or save a party or raid member; popping critical heals on the tank or a raid member (aside from the token tranquility drop) - this is what makes the spec rewarding for me and why I'll continue with it into cataclysm vs. my lock, mage, or hunter even though they may wind up being more competitive on dps.

The moonkin, when played to its full potential, is a very rewarding experience when you bring in the utility and that is something only an encounter can provide you the opportunity for; banging your head against a target dummy for hours is boring for any class IMO.

Randur said...
This comment has been removed by the author.
Randur said...

complex rotation of a moonkin

ps: too bad i can't attach this image to my comment

Pewpewkitty said...

I'm curious as to why people are suddenly so bored with all specs and rotations unless there are a million things going on. I remember back in BC people would play a mage or a lock, where they have 1-2 button rotations that are very simple, and nobody would complain. I have to agree with graylo here saying that the encounter should provide the difficulty and not the rotation. The rotation is just a simple way to sort out the people that lack multitasking abilities.

Ivan said...

Pewpewkitty: Imo, theres an easy explanation. Boss fights on TBC were actually boss fights.
Nowadays we have fights with mobs that have big health pools...