nakedduck

Fat fingered it. =X

It's just interesting to see. There's a lot of people that believe that the raw stats on sunwell gear would surpass the t5 bonus, but it's very interesting to see that ef's spreadsheet says otherwise. The additional 3 secs on the 2t6 bonus should help as well keeping the debuff up.

edit:

Another very interesting thing, it looks like 4t5 + 2 T6 > 6t6

 

Morpheen

Has anyone done much work with certain haste thresholds? I know that with natures grace, it can be difficult to predict casting rotations, but Ive been trying to find specific Haste levels that will allow that extra starfire to benefit from a proc. Currently Im attempting to see if ~400 haste will allow me to reliably get 6 starfires in my rotation, and sneak 4 starfires in a idol of unseen moon proc (crits willing). Not to mention over 393 Haste allows post crit starfires to cast in under 2 seconds =)

 

Destined

I was wondering what people thought of Ember Skyfire Diamond(14 spell dmg, 2% int). Is it possibly better than Chaotic Skyfire diamond(12 crit rating, 3% increased dmg on crits)? I'm thinking no, but I'm not 100% sure. Sorry if it's been mentioned before, I've read through most of the post but haven't seen anything come up about metas.

 

HawkenCL

Hey, thanks for the spreadsheet! So far its been very usefull. Im not sure how to add an item that isn't already listed though.

Im using Hammer of Righteous Might which is a good starter boomkin weapon, and pretty rare, but its not listed.

Any way to add it manually? Im new to spreadsheets. Thanks!!

ETA: nvm I found how to add the stats in. Thanks again.

 

Destined

A few notes about the spreadsheet:

It doesn't include Grand Magister's Staff of Torrents in it.
[Grand Magister's Staff of Torrents]

 

Ten

I did another pass at optimizing gear with the latest version and topped out dps at almost 2600 for a 6 minute fight (ie Brutallus). My unbuffed stats there are 107 hit, 20.02% crit, 491 haste, 1472 arc dmg (no, it is not hit capped, but 15.89% was the closest I could get). Chaotic Skyfire > Ember Skyfire at all levels of gear that I tried. My optimal rotation at this level was MFx1/SFx8 and the average starfire cast time was 1.93s (including 50ms latency), well under the 2.0s threshold for optimizing the class trinket.

It's worth mentioning that at this level of haste, mana consumption was surprisingly a non-issue. A shadow priest is worth ~25% of his dps in mp5, so a good shadow priest on Brutallus who does ~1400-1500 dps will be worth 350-375 mp5. Based on this, the level of gear you reach with these figures is capable (according to the sheet) of going for 10.5 minutes using a starfire only rotation without running out of mana, although this does drop your dps down to ~2525. This is without using mana potions, innervate, or brez.

What's interesting about high numbers like this is that for some reason, the spreadsheet values 10 haste gems over any other possible gems for certain spots, and it values varieties of pyrestones for other slots even when the colors are the same. I should probably note that I'm used to playing at ~50ms latency (.05 on the sheet) which is a HUGE factor in having such high dps numbers on the sheet.

A couple changes I made to the spreadsheet for approximating dps values were:
Set [The Skull of Gul'dan] to have 20/120*175=29.1666666 haste rating since I'd have it up every time I could by means of a macro on my Starfire key.
Did the same for the +dmg on [Hex Shrunken Head] (88.17 total) and the darkmoon crusade deck (80).
Put in a preliminary value for [Ashtongue Talisman of Equilibrium] that calculates based on average starfire cast time (>=4 casts in 8 seconds gives full value, less gives a % based on number of casts you can do in 8 seconds). Obviously this isn't valid for rotations that use wrath.
Changed spell haste rating to 15.7=1% since this is what wowwiki says.
Changed Moonfire cast time to =1.5/(1+GS_Haste/Sp_Haste)+Latency to account for the 2.4 gcd/haste change (I think it gets reduced just like 1.5s casts, but I could be wrong. If so, it should be 1+0.5/(stuff)+latency instead). I have no idea how you coded the other spells so I didn't tinker with them.


EDIT: Just because I thought it was funny, I tried "debuffing" the spreadsheet to see how much dps these stats would yield. For the same 6 minute fight, it comes close to 850 dps with IS,SFx4, running oom after ~3 minutes even while using potions and innervate. In a 1 minute fight where mana was not an issue, it's almost possible to get to 1750 dps.

 

• Adoriele

Something about this doesn't add up. It's quite literally impossible for [Quick Lionseye] to be worth more than [Great Lionseye], and yet you're not hit capped. I'm not sure I'd believe the answers quite yet.

 

Sorail

It is true if he factored an elemental shaman in his buffs as that sets him at 145 hit and a [Great Lionseye] puts him over the hitcap. Would be probably good to know with what kind of buffs he calculated.

 

Janne12

I got a few questions about moonkin dps and what gems I should value over others. Reason I do this is because I recently changed some of my 4 crit rating % 5 spell dmg gems, 5 spell dmg & 6stam gems to quick lionseyes (+10 spell haste rating) and did horrible dps in the raid zones / instances we did. around 1k raid dps using moonfire - starfirex4 rotatio.

How much spellhaste I should have? Or did I do horrible mistake when even bothering by equipping spell haste gems? What spell rotatio would be best? I assume it's moonfirex1 starfire x4. I don't usually have a manaproblem (as long as they give me shaman or shadowpriest, last raid it was elemental shaman).

I have unbuffed:
1270 spell damage
112 spell haste
23% crit (with moonkin aura, not counting talents)
hitcap

 

Aenerion

For some reason my comp. wont allow me to open the spreadsheet. Does anyone know why that is?, I'm trying to download it and it will not open

 

Silverclaw

I recently purchased Panzerkin.com and would like to help the community by compiling all known moonkin information and theory crafting to one place. This is a work in progress (just got the name a little bit ago and haven't had time to work on it yet) so nothing is there atm (though in the next few weeks I would like to change that).

I would like to reproduce some of this information there. As well as host some moonkin bloggers (talking about strats for raiding, pvp, and guides for helping the new moonkin out there.)

I bought the domain cause I was tired of looking around for information at 10 different sites. I can host files, movies, and pretty much anything on my site (I have 15,000 Gigabytes of monthly bandwidth with 1,500 Gigabytes of disk space available) I would like it to be a one stop shop for Balance Druids.

*** If you are interested in helping please contact me ***
Efenjel if you could let me know if I can host the spreadsheet as well as reproduce (edit and reformat) the information contained here)

My email is "silverclaw" at "panzerkin.com"

Thanks!

 

Audioslave

Ten, just wondering what gear set up you used to achieve those numbers.

Any chance of a link or a upload of your spreadsheet data?

 

Sorail

The problem may be in your spellrotation as you don't let the moonfire dot run out. With 100-ish haste your SF will be close to 2.8 sec without NG proc meaning you should use an MF, SFx6 rotation because of the 2p T6 bonus. Try that, you should get better results.

 

Luntishara

I may be misunderstanding, but shouldn't there be "tiers" of haste. What I mean is this (all just examples and math obviously not meant to be accurate just to make a point): if you have lets say 100haste and it adds 1more Starfire into a rotation that was MF,SFx4 into MF,SFx5; However if you have 150haste it does not squeeze an extra spell into the rotation at some point until you hit lets say 200. This is ofcourse if you are not changing it to MF, SFx5,Wrath which would then increase your dps with the extra haste. If this is true then and you are at a point inbetween those "tiers" of adding a Starfire should one not go with some more Crit since the little bit of extra haste is wasted on not squeezing in an extra spell into the rotation, or does one simple add Wrath to it, or even yet just stand waiting for the last tick of MF to tick off before restarting the rotation? I know the spreedsheet can automatically calculate the best rotation, but for gearing goals a standardized mark of levels reached, ie 151spell hit, 1507damage, etc... Hope I explain my thought process well enough.

 

Janne12

I have to actually check this out

Thanks for replies and help btw

 

Benita

Tiers of haste only apply to 4t5 where you cant let dots wear off before SF hits, for 4t6 it's better to just cast SF and after MF dropped off for 0-SFcasttime seconds to reapply it as the dps of both is quiet close. Standing around or using a wrath doesn't seem to improve my dps at least. With crit as a another random factor in rotations this could be the right approach for all but 4t5 and haste wouldn't scale in tiers except based on the simplification of spreadsheets.

 

Luntishara

What I mean by "tiers" is not Tier sets and really nothing to do with my question. My questions is that haste seems to act differently than hit and crit do on spell casting, particularly when dealing with a rotation.

As you increase your spell hit, it will steadily increase your dps until you hit the cap as more spells land. Crit will increase your dps as well as any spell that has crit ability is affected. Niether will effect the rotation ingeneral. Meaning no matter what your spell hit or spell hit is; if MF,SFx4 is your most effective rotation. When you add spell hit and crit (leaving spell damage out of the conversation for hopefully obvious reasons) will both just increase your dps and not change the rotation.

However, Haste at certain points, at least in my thought process, seems to be only at certain levels that it seems to maximizes its efficency when one can add a spell into the rotation. Any haste in excess if squeezing an extra spell in would be wasted and be better spent in going with more crit until it can be raised to a point to squeeze another spell into the rotation. That is why I am asking my question above and proposed the question I did. "Tiers" of haste may have been a poor choice of words, rather levels of haste or something better that I can not think of at this time.

Anyways, I am not at the point of many of the T6 moonkins here in this thread. However, my way of thinking makes just purely stacking haste (after reaching spell hit and spell damage goals) a waste of some points that once can put into crit. Rather should one just get to certain levels of the haste and then just add crit until one can reach another level of spell rotation.

I hope my attempts to explain my question is more clear. I apologize for any confusion.

 

Benita

I did understand you right, my point is the same. Unless you rely on timed procs like MF idol or SF BT trinket or you use 4t5, haste will always have a beneficial effect, even if its only 1 point. Perfect timing in rotations can't be achieved anyway due to the randomness of crits, lag and the human factor.
Spell rotations are a simplification model and therefor the "fitting in" issue is only showing on spreadsheets. E.g. 2xMF/13xSF is already trying to approximate the real behaviour better

 

Morpheen

Some of the issues of haste tiers is that depending on your lag, the actual haste required to hit such a threshold can be completely different than theory, and depending on your crits, natures grace is going to give you multiple tiers depending on how many times you crit during that rotation. However, Ive been working on this in hopes of finding thresholds where haste provides the greatest benefit (that extra starfire during a proc).

I unfortunately dont have the math backround, so I just plugged in haste numbers until I got the spells to cast in the allotted time. I looked at the BT trinket, idol of unseen moon (which I also used for Hyjal ring haste calcs), and casting rotations with and without IS. What I found was considering you are in t6, 105+ haste should be your first threshold. This allows 5 starfires after 1 moonfire with only 1 crit required per rotation, 3 starfires during the BT trinket uptime with 1 crit, and 3 starfires during an idol of unseen moon proc assuming no crits. So once you decide you are going to use haste, shoot for 105+ to get the max benefits.

Unfortunately the next tier where there were multiple benefits from an allotted amount of haste was at 314 haste (which is quite steep, so only endgame moonkin will be able to reliably achieve this). The nice thing about 314 haste is that with Drums of battle, you can hit 394, which is another tier (tho with less to offer). 314 haste allows for 5 starfires in a rotation with MF and IS without any crits, 4 starfires during a idol of unseen moon proc with 3 crits, and a mf1/sf6 rotation with 3 crits. Pop drums, and mf1/sf6 takes only 2 crits, and idol of unseen moon can have 4 starfires with only 2 crits. And if you are lucky enough to have every spell crit at 393 haste, you can get 4 starfires during a BT trinket proc.

As Benita mentioned, haste is always beneficial, and the randomness of lag, crits, and human error means that this is just a template, the actual haste required to achieve these goals ingame might be far higher. However, 105 and 314 seem the best benchmarks if you are considering stacking haste.

 

spi

Hi. I just had a question about idol swapping. Many of you mention the idol switch macro. Specifically this one.
/cast moonfire
/equip Idol of Unseen Moon
I do get this to work even if I switch the idol AFTER moonfire is casted (it is reproduced consistantly). However if you pay attention closely, the macro triggers GCD twice. What I mean is that, once moonfire goes off, you see another GCD is being triggered afterwards losing about .1second or so. This macro has the benefit of having the last starfire before moonfire refresh be affected by the starfire idol. I was wondering if anyone was able to measure or calculate time lost using this specific macro and how adversely it'll affect the DPS compared to switching it during last starfire cast before refreshing the moonfire DoT? Would you recommend to sticking with idol switching right after the last starfire of the rotation?

 

• Adoriele

Having not grabbed the idol yet (Thinking about picking it up tonight, I have the badges, but that new badge dagger is just so sexy), this is mostly guesswork, but I'll bet the .1sec lag you're noticing is very heavily tied to your latency, i.e. the UI automatically starts the GCD when you hit Moonfire, then waits for a response back from the server to confirm (hence why, if you try casting Moonfire in the middle of a Starfire cast, you'll see a mini-GCD if you have a reliable way of seeing it, such as Quartz). For the Idol swap however, I'll bet it doesn't start the GCD until after it gets the response back, which would reset it and hence give you the minor discrepancy. So, it's not gonna be a hard-and-fast rule about how it's going to affect you, as it'll change depending on if you're in Australia or New York. It also won't always be consistent.

That said, it would be pretty simple (and accurate) to model it as an effective addition to the cast time of Moonfire, since really that's all the GCD entails, though it wouldn't be affected by Haste. In fact, I may have to add a new section in the article dealing with latency, though it'll be a long time before I'd be able to get to it. In all, it's probably a justifiable risk if you're in a situation where swapping is a good idea anyway. Moonfire is a pretty good spell to spend a GCD on, even with a large amount of Haste, and the minor addition is probably far outweighed both by the added chance to gain the Idol and NG procs, and the actual DPS that Moonfire provides.

 

spi

Thank you for the response. I always do use moonfire in the rotation, but assuming that i have less than 80ms ping, and that the delay of second GCD triggering is the same as my latency, i would guess using this macro may not be as beneficial as switching the idol out as last starfire in the rotation is being cast. I'd like to hear opinion if someone thinks otherwise.

 

• Adoriele

Depends. 55*1.32= 72.6 damage on one Starfire (not including crit) vs. adding your latency to Moonfire's cast time. Assuming .1s ping (which will only affect the MF GCD when you swap, since we're in the middle of a long boss fight, and you're really good at watching quartz and timing your cast chains), 1000 spell damage, 25% crit in-form, hit-capped, no haste, using a rotation which allows MF to drop during the last SF cast (i.e. not using 4T5), and no T6 bonuses, you're running MF, SFx4. You'll have an average of 973.6 DPS if you swap idols early, and an average of 973.22 DPS if you swap on the Moonfire cast, adding .1s to the GCD. I'd say that's just about as negligible as it gets, though it's going to be more of a problem when Haste enters the equation, as that .1s won't change, but the rest of your casting time will, so it'll have more of an effect.

 

spi

Wow, interesting. So it doesn't make a huge difference. I do currently have 201 haste and 4/5 T6 and around 1200dmg unbuffed. I was doing some tests on Dr. BOOM just now to practice idol switching. Using the macro to switch idol out is easier than trying to switch during the last starfire (i cast 5 or 6 starfires before refreshing mf i let moonfire fully tick off). I think I make less error using the macro compared to not using one and it looks like the delay is just about as much as the latency is (about 10th of quartz GCD bar gets recasted). From sounds of it i suppose even with haste and all it won't make a huge difference. When i get heroism+drums i just chain cast starfires anyway without casting moonfire. Thanks for doing the math! I think i'm gonna stick with the macro for now.

 

• Adoriele

No no. Haste will make a big difference. Without Haste (I'll use MF, SFx4 again because it's convenient, same stats as above), average cycle cast time is 12.9s instead of 12.8s when you use the macro, DPS loss of .78%. Adding in your 201 Haste, same cycle has an average cast time of 11.46s instead of 11.36s, DPS loss of .88%, or a 12% difference. Also, the additional 55 Spell Damage from IIotM will be less of a factor when you get more than 1000, which I'm guessing you have, so the benefit of having it equipped on the last SF will be less and less, especially as you add more SFs into the rotation.