I've been working w/Efejel's uberkin spreadsheet for a few months now (btw, kudos and much thanks, Ef). A while back I noticed I was having an issue with my dps supposedly increasing when selecting crit over hit. For example, selecting Adornment of Stolen Souls over Brooch of Unquenchable Fury provided me with a dps increase, even though it decreased my hit by 15 points under the cap (I'm sitting at 151). With another boomkin's help we figured it was because I had the Draenei racial & shaman totems selected. Once this was remedied, whenever I selected any gem/weapon/armor in the spreadsheet that decreased my hit, it displayed a corresponding dps decrease, as it should.
I am now experiencing the same problem again, without having the Draenei racial/shaman totems selected. No matter what I seem to change in the spreadsheet I get a dps increase by losing hit (i.e. putting myself below the hit cap) and gaining crit. I've tried it in the new version of the sheet and the previous version. I'm stumped. Any ideas, folks?
I think you'll find that it's not choosing crit over hit, it's choosing 23 crit and 2 spell damage over 15 hit. Hit isn't ALWAYS better than crit, especially as you approach or reach the cap as it becomes less useful, but point for point it's better. I find people tend to put too much weight on hit rating and lose sight of the fact that every stat has diminishing returns (except maybe spell damage not sure about that one) and that the lower the stat is the more you'll get for it per point. So your hit is high and your crit is probably low and that's why 2 spell damage and 23 crit wins out over 15 hit.
I was wondering after capping spell hit at 152 what should you focus on your gems to, spell damage, spell crit?
Good thing there's a handy dandy spreadsheet in the first post to help you answer the question. In general, Hit > Damage > Crit. Your gear will change that relationship.
I just wanted to show some hope to other moonkins that being a viable spec is possible.
Our WWS from our first night of illidan attempts: Wow Web Stats
p.s.I WTB an elemental shaman, our guild lacks shaman only 2 resto and 1enhancment as mains. I think i can take the warlock once i get a elemental shaman.
Looking into switching from resto to balance and was curious as to how good the deathsworm trinket is with 25% chance on starfire to proc 8 seconds of 150 + spell damage. Running a 2x mf 11x starfire rotation it would seem that theoretically this trinket would be up more than 50% of the time. Is it better than picking up a blessings deck? Other trinket is an icon until I can possibly get the za trinket.
Reading the last few pages makes me think that the suggested idea of making a new thread to deal with general moonkin theorycraft would be good. The original poster would ideally be able to keep it updated, so while I'll start to compile some information, I'm feeling like my interest in WoW is starting to wane, and it might be better to have someone more stable to do it.
General rule of thumb for gems:
Spell hit = 1.2 spell damage
Spell crit = 0.7 spell damage
Use whatever gem gives you the highest total value.
In the end, you still need to use the spreadsheet to determine what's best for you as the values above are only approximations.
---
The ashtongue trinket seems like it'd be pretty strong, if the trinket really has no internal cooldown per the last comment on wowhead. 8 seconds at a 25% rate can be approximated with 2 seconds each starfire, or 22 seconds total. Assuming 0 crits and no spell haste, your rotation would be 36 seconds. This is approximately a 61% uptime, or about a +91 damage effect. Of course a proc while the proc is already up does not gain the full benefit since it just resets the timer, but you are also very likely to get NG procs and have spell haste, so I think over all it'd be better than the blessings deck, which is basically a +80 damage trinket.
The order for the trinkets you listed is probably icon < blessings < ashtongue~=hex. I'm not sure how much the blessings deck runs on your server, but I'm not sure if it'd be worth the investment for the improvement over the icon if you're actively running ZA.
I think you'll find that it's not choosing crit over hit, it's choosing 23 crit and 2 spell damage over 15 hit. Hit isn't ALWAYS better than crit, especially as you approach or reach the cap as it becomes less useful, but point for point it's better. I find people tend to put too much weight on hit rating and lose sight of the fact that every stat has diminishing returns (except maybe spell damage not sure about that one) and that the lower the stat is the more you'll get for it per point. So your hit is high and your crit is probably low and that's why 2 spell damage and 23 crit wins out over 15 hit.
I used the neckpieces simply as an example. As far as I can see, when fighting a raid boss (level 73 elite), a drop in my spell hit by say 30 points, to 122 (which, IMO is pretty far away from the hit cap), should practically never result in a dps increase. There certainly are tradeoffs when choosing between any 2 stats, but it still doesn't explain the values the spreadsheet is displaying, and why it differs from what I have observed previously. I still suspect there is an error somewhere, and am hoping someone can point me in a specific direction.
As far as I can see, when fighting a raid boss (level 73 elite), a drop in my spell hit by say 30 points, to 122 (which, IMO is pretty far away from the hit cap), should practically never result in a dps increase.
Why not? 30 hit is approximately 2.5% DPS, slightly more depending on how much hit you have. (In your example, it's about 2.525%, nothing extraordinary). Depending on how much spell damage you have, that's easily overcome. For example, in half-Kara gear, a [Runed Living Ruby] was worth more DPS for me than a [Veiled Noble Topaz], or a [Great Dawnstone], even though it was only 1 more damage than hit.
I used the neckpieces simply as an example. As far as I can see, when fighting a raid boss (level 73 elite), a drop in my spell hit by say 30 points, to 122 (which, IMO is pretty far away from the hit cap), should practically never result in a dps increase. There certainly are tradeoffs when choosing between any 2 stats, but it still doesn't explain the values the spreadsheet is displaying, and why it differs from what I have observed previously. I still suspect there is an error somewhere, and am hoping someone can point me in a specific direction.
There is a simple test. Don't change your gear at all, use the other section and go to the hit rating column, enter -1. if your dps doesn't go down your over the hit cap, (with all the buffs/auras you have selected.) If your dps does go down then your at or below the hit cap and it's just the other item has enough non-hit stats to make up for it.
This is a speadsheet, you can give it negative values and it will pretend you just added a gem that reduces your hit rating. The items don't have to be real.
Reading the last few pages makes me think that the suggested idea of making a new thread to deal with general moonkin theorycraft would be good.
The ashtongue trinket seems like it'd be pretty strong, if the trinket really has no internal cooldown per the last comment on wowhead. 8 seconds at a 25% rate can be approximated with 2 seconds each starfire, or 22 seconds total. Assuming 0 crits and no spell haste, your rotation would be 36 seconds. This is approximately a 61% uptime, or about a +91 damage effect. Of course a proc while the proc is already up does not gain the full benefit since it just resets the timer, but you are also very likely to get NG procs and have spell haste, so I think over all it'd be better than the blessings deck, which is basically a +80 damage trinket.
Agree on at least a thread that summarizes stuff. I think i said this 2-3 times already in this thread, but the ashtongue trinket just works differently. Your uptime might be correct, but its with the worst possible timing for Starfires. The SF it procs off receives no +damage of it and getting 3 SF in after it proced is not guaranteed. If you count 2.5 SF affected and overlapping procs its down to 40-50 depending on your haste and crit rates. This is for pure SF spam, it gets worse for GCDs used on IS or MF. It ends up being worse than blue trinkets.
Maybe Sunwell haste gear buffs this trinket slightly, but 2.4 also brings [Timbal's Focusing Crystal] which is not hard to get and in any case way better than this.
What about when you are in excess of 200 haste, and you will always get 3 starfires buffed by the trinket using starfire spam? If your normal casts are 2.65 seconds (207 spell haste), assuming minimal lag you could achieve 150 +dmg on 3 starfires. Then you would just have to calculate uptime of trinket. If ~1/3 starfires crit (easily attainable raid buffed in t6), your overall average starfire cast time is 2.5 seconds. Now Im horrible with math, so if someone else could please calculate the uptime if you have a 25% chance to proc the buff for 8 seconds, and you're casting at an average of 2.5 seconds, the we could see if it is decent given the proper endgame gear.
I'm a big fan of the spreadsheet, but there is one thing that has been bothering me for a bit, and I can't figure it out. I got curious about how everything worked, so I took a peek in the Cast Cycles tab after unhiding it. Most of the calculations make sense, but I'm confused by the DPS calculations for the MF, SFx3, W and MF, SFx4, W cast cycles. The calculation for the first cycle is
and the code for the second is similar. I'm confused by the use of Insect Swarm average cast time instead of the Moonfire average cast time. After careful deliberation, my brain surrendered, and I'm guessing that it is just a typo.
This all started because I was toying around with custom cast cycles, and I noticed that entering a custom cast cycle that is identical to a prefab cast cycle almost never results in identical dps for the two cycles. Exceptions (that I noticed) are Wrath spam, Starfire spam, and IS, Wx8. I'm not sure if the first issue is causing this, or if it is just due to non-optimal spell arrangement during custom cast cycles. If it is worth fixing (and I'm not entirely sure that it is), good luck. Thanks again for all of your work on this sheet, and for clearing up my confusion.
Reading the last few pages makes me think that the suggested idea of making a new thread to deal with general moonkin theorycraft would be good. The original poster would ideally be able to keep it updated, so while I'll start to compile some information, I'm feeling like my interest in WoW is starting to wane, and it might be better to have someone more stable to do it.
I think this is a great idea. I've had questions before that I've wanted to research, only to stare at the daunting 36 pages on this thread, and try to find the answers elsewhere. :P
Sounds great. It'd probably be a good idea to link to this thread as well as the one Efejel started on the wow forums since that actually covers quite a lot of info too. WoW Forums -> Balance One-Stop Shop
And as a friendly reminder, please read this post, as the values I speak of in that post are in the current 2.3 mana regen model, and you're missing out on some free mp5: http://elitistjerks.com/f31/t11582-d...34/#post644254.
and the code for the second is similar. I'm confused by the use of Insect Swarm average cast time instead of the Moonfire average cast time. After careful deliberation, my brain surrendered, and I'm guessing that it is just a typo.
Indeed, this is most likely a typo that has survived since the inception of the sheet. Much <3 for taking the time to dig into the mechanics.
As far as custom vs. default values varying: DoT vs. cycle duration is the main cause. For default cycles, all ticks are assumed to tick out. For custom cycles, it checks to make sure you've padded in enough average casting time of other stuff and penalizes you if you haven't.
Oh, and pre-emptive response: the average cast time of MF & IS vary b/c I INCREASE the average cast time of IS based on SPELL MISS, the assumption being it's inexpensive enough and useful enough to recast when resisted whereas with Moonfire, spell miss is translated into damage reduction--if you're resisted, you'll try again later.
"Electronic communities build nothing. You wind up with nothing. We are dancing animals. How beautiful it is to get up and go out and do something." - Kurt Vonnegut
Oh, and pre-emptive response: the average cast time of MF & IS vary b/c I INCREASE the average cast time of IS based on SPELL MISS, the assumption being it's inexpensive enough and useful enough to recast when resisted whereas with Moonfire, spell miss is translated into damage reduction--if you're resisted, you'll try again later.
Is this generally the ideal way to treat the two dots? I tend to run through the entire cycle misses or no, simply because I'm generally already casting the next spell before my UI updates the boss debuff list for me. I may end up switching to this method if I can figure out how to get BigWigs to add a bar for each of my dots (to help with clipping), since I tend to cast IS before MF, so restarting the cycle is easier on IS.
Is this generally the ideal way to treat the two dots?
For the most part, yes. Based on years of experience rather than any sort of equation. Simply put, there will be interrupts and movement in at least 75% of boss fights, and so if you're using DoTs you want your most efficient one up as often as possible.
For pure min-max I try to keep to two diff rotations: IS-MF->Starfire spam, so that during the MF GCD I can determine if my IS "stuck" or not, or: MF-IS->Wrath spam so I can try and sneak in a Starfire if I notice during the IS GCD that NG proc'ed on MF. This also plays pretty well with where these rotations are used; the first is more boss friendly with the IS lacking initial damage giving the tank more time to build threat, and and the second for trash where on all but taunt-immune mobs that won't one-shot me in the case of a resist--assuming I'm with my tanks who know me well--a pre-emptive Starfire joins that MF giving the tank a nice burst of threat when he taunts.
"Electronic communities build nothing. You wind up with nothing. We are dancing animals. How beautiful it is to get up and go out and do something." - Kurt Vonnegut
This is a very basic question and im looking for a basic answer. Im fully aware it may be a bit more complicated with different varying factors but try and base your response on the data given. Ive been discussing the age old question of wrath vs starfire with some fellow moonkin raiders on my server and im getting mixed answers. So heres the scenario:
Givens:
1500 spell dmg
Infinite mana ie unlimited
Stationary for the duration of the fight( one target one fight nuke as fast as you can)
Same amount of crit and hit
You only cast wrath or starfire (no IS of MF)
Which does more dps? Starfire or Wrath?
Very simple question just trying determine which spell does the best dps in a simplistic setting
Very simple question just trying determine which spell does the best dps in a simplistic setting
1) Is this question aimed at 2.4 or live?
2) Is Curse of Shadow on the target, and if so is it 10% or 13% dmg?
Live + no CoS = depends on your crit rate, latency, and skill with timing casts through said latency
Live + 10% = depends on your crit rate, but starfire is probably better (the incidental crit in gear at that +dmg level should easily be high enough)
Live + 13% = starfire is better
2.4 + 10 or 13% = wrath should be better
"Electronic communities build nothing. You wind up with nothing. We are dancing animals. How beautiful it is to get up and go out and do something." - Kurt Vonnegut
I'd appreciate it if IFF could be added as a rotation option. I know I can just reduce the dps by 3.5%, or 5%, or whatever, but it seems like it would be an easy thing to add, and would make the spreadsheet all the more self-contained.
03/02/08, 6:01 AM
Similar Threads
