Moonkin Raiding - Page 70

Lemzix · 04/26/09, 9:27 AM

Originally Posted by polarbearoughey

Faerie Fire(talented) -> Moonfire(glyphed) -> Starfire(glyphed)(until eclipse(wrath) procs) -> Insect Swarm(glyphed) -> Wrath(until eclipse passes) -> Moonfire

I also like that rotation, ill have to try it out. However I think ideally you would want to throw up IS before proccing eclipse so you dont have to waste a gcd of the eclispe uptime.

Lemzix · 04/26/09, 10:37 AM

After doing some test dummy runs with both starfire and wrath based eclipse rotations, ive drawn the following conclusions:

starfire eclipse rotation yields a pretty consistent 4.1 - 4.3k dps output, since when eclipse is up most of the starfires are critting for 9.7 - 12.5k (with both my illustration and broodmother stacks at full of course) and the damage variation in this rotation stems from how long it takes to proc eclipse.

wrath eclipse rotation has the potential to do moderately higher dps if you crit a lot when spamming it during eclipse, and do less dps if you dont. for me, 3.8k dps on the lower end of things, around 4.7k with lots of crits, so anywhere in between is predicted.

i think im going to stick to my starfire eclipse rotation. i really enjoy it, except when i get an unlucky string of wrath non-crits or crits that dont proc

. also, starfire eclipse + heroism or to a lesser extent power infusion = awesome.

Ebonwood · 04/26/09, 8:49 PM

So far, what I've gathered, is that the IS> MF > SF > Wrath (under eclipse) rotation is better, and that the t8 4pc is good, because there is no internal cooldown? I got that from someone with 4 pieces. I dunno though, if I'm wrong, correct me.

But anyway, if this is true, theeeen... Forget about haste, and stack crit. If you're decked out in crit gear, then the 5% bonus of t7 becomes... stupid. And with imp FF, and MF, Sf should crit enough to keep eclipse up a bunch. And under that, with enough crit, you should be throwing out tons of wrath crits. That's gotta be the best way to go about it. Haste pretty much seems dangerous now, because even if you're going .01 seconds under the GCD, your wasting points in haste that COULD be crit.

Just my idea. If anyone has numbers otherwise, please, show me. Because I don't want to be stacking the wrong points.
But I'm getting the feeling that Blizz wants Critkins back.

Horao · 04/27/09, 7:03 AM

So I've been trying to put together my 5 pieces that I would use for their tier slot pieces. I figure 2 T8 and 3 off set, so here's what I put together:

(Background - I'm pushing for a wrath cycle, high crit, lower haste, and keep enough hit on these 5 to keep it off my belt/boots/wrist)

Base set:
Head - [Collar of the Wyrmhunter]
Shoulders - [Conqueror's Nightsong Mantle]
Chest - [Raiments of the Corrupted]
Gloves - [Handwraps of the Vigilant]
Legs - [Conqueror's Nightsong Trousers]

Comparison to current T7.5 here

Gains:
38 Int
73 Sta
19 Spi
34 Crit
109 Hit
75 Spell Power

Loss:
620 Armor
48 Haste

It seemed like the Shoulders and Legs were the best two of the T8 pieces. Legs are clearly better than the other t8 because of the lack of spirit. If you didn't need hit, you could swap the Rainments of the Corrupted for the Umbral brute

So then I put together a set that used all 3 ilevel 239 pieces here, which pushes the 2 T8 pieces being used to the less desireable ones:
Helm - [Conqueror's Nightsong Cover]
Shoulders - [Shoulderpads of Dormant Energies]
Chest - [Conqueror's Nightsong Vestments]
Gloves - [Handwraps of the Vigilant]
Legs - [Leggings of the Enslaved Idol]

Over the base set above seen here, gains:
40 Int
10 Sta
126 Spi
285 Armor
7 Haste
9 Spell Power
3 Sockets

Loss:
61 Crit
56 Hit

So there's a significant spirit & moderate int gain, but a big loss to crit and hit. The 3 sockets gain could make up some of that, but overall I think this doesn't gain more because it replaces the zero spirit chest and helm with Nightsong for a big loss.

This is all without gems, so with gems (using 2 blue in each set to activate meta) base gains 107 SP / 37 crit, 2nd set gains 153 SP / 53 crit, and the 2nd set also gains 4 SP from int and 18 SP from spirit, totaling 175 SP / 53 crit for gems + gains from spirit. So overall a 68 SP and 18 crit gain, turning the real difference into:

Gains:
7 Haste
77 SP

Loss:
43 Crit
56 Hit

So the spellpower gain certainly outweights the crit loss, so it would really come down to hit. If your good on it the second set would be better. Even though the 2nd set (ilevel 239) is forced to use the nightsong helm which is terrible compared to wyrmhunter, the extra sockets on the legs / shoulders seem to make up for it.

dukes · 04/27/09, 7:45 AM

Originally Posted by Horao

Chest - [Aesir's Edge]

Shoulders - [Dragonslayer's Brace]

You need to check your item links. Also I'd recommend checking mmo-champion as there's a decent list of 239 loot there, including leather spell power chest/shoulders/gloves which might let you change which tier pieces you use for your 239 comparison.

Horao · 04/27/09, 8:11 AM

Originally Posted by dukes

You need to check your item links. Also I'd recommend checking mmo-champion as there's a decent list of 239 loot there, including leather spell power chest/shoulders/gloves which might let you change which tier pieces you use for your 239 comparison.

Fixed (stupid 10 key VS phone keypad number inversion error), thanks for the heads up.
I did scan through some of the new items on MMO champion briefly earlier, I'll definitely go back for a better look.

Edit - I looked again at MMO champion hard mode stuff and it was all stuff I had previously considered. The Frozen Glade gloves are not as good as the cloth, the chest has 2 extra sockets over the umbral brute chest but lacks the extra crit and a little bit of haste, and the shoulders were already listed, however that was one of the items I had originally put in the wrong item number when you viewed the list. The Vestments of the Blind Denizen would be much better if they dropped that 78 spirit and added some hit or crit.

Druidbomb · 04/27/09, 9:00 AM

I really hope some mod cleans the group of balance threads up.

By making one post with each rotation clearly laid out for newer druids and the spec/gear stat weights that favors that rotation. And maybe show some WWS' that accompany that spec/gear stat weights.

So say someone that thinks the starfire eclipse rotation is the best and has some wws' to go with it, and says hey we should stack haste then shows the stat weights so people can plug it in loot rank and start planning their BiS (best in slot) gear. (just an example)

This should cut all the different opinions down quite a bit and clean up all the threads. (always good to have people question the cookie cutter specs/rotations but we need to clean this up a bit)

Right now I'm seeing a trend of "you should be using a starfire eclipse rotation, or a wraith eclipse rotation" with a mix of dot uptime/priority without any actual numbers or wws' to back it up.

Lets see some real numbers and results rawr/calculators are great in a perfect world but lets be honest eclipse is so random it's hard to live in said world.

dukes · 04/27/09, 10:00 AM

Originally Posted by Druidbomb

Lets see some real numbers and results rawr/calculators are great in a perfect world but lets be honest eclipse is so random it's hard to live in said world.

The whole point of calculators / simulators is to model a best case situation for things in order to try and get the most effect out of your gear, spec and rotation. I don't quite understand how you can say that "eclipse is so random" and then say that rawr and calculators aren't worth using. Eclipse is random so real testing will result in a bias one way or another, and real testing is impacted by too many outside factors in order to isolate one individual aspect enough to say what is better dps.

If you remember doing science tests at school, you may remember listing variables and constants. In order to make any test actually worth doing, you need to be able to control variables so that they can be constant in any one test. When you consider a DPS test on a target dummy you have variables out of your control which may vary from zone to zone or day to day, and variables which vary from person to person. Server lag, computer lag, crit strings/crit rate, eclipse proc variability (using wrath to proc eclipse), frequency of button presses, reaction time. All of these are variables which would require a huge amount of testing in order to reduce the background effect enough to see small differences in DPS between the tested factor (rotation, gear, etc), which is why we use calculators and simulators. If you then want WWS's of data from boss fights you're introducing a whole new set of variables such as buffs gained, percentage of the fight spent moving, time in combat, percentage of the fight bloodlust is active, etc etc. It doesn't make sense in the current environment to back up arguments based upon in-game testing. The best approximation we can get is to test the simulators/calculators against extensive tests from in game to make sure they are simulating/calculating things correctly, and then using those simulators/calculators to provide DPS estimates for constant situations, allowing the change of a piece of gear or switching which DoT effect is cast first in a rotation and getting a meaningful answer.

With Rawr and Wrathcalcs available for modelling, there isn't much point in putting up individual posts per rotation/spec/whatever because every case is different. It's nice to have guidelines about what could be used, but it's easier for everyone if you treat yourself as an individual case.

Druidbomb · 04/27/09, 11:18 AM

Originally Posted by dukes

The whole point of calculators / simulators is to model a best case situation for things in order to try and get the most effect out of your gear, spec and rotation. I don't quite understand how you can say that "eclipse is so random" and then say that rawr and calculators aren't worth using. Eclipse is random so real testing will result in a bias one way or another, and real testing is impacted by too many outside factors in order to isolate one individual aspect enough to say what is better dps.

If you remember doing science tests at school, you may remember listing variables and constants. In order to make any test actually worth doing, you need to be able to control variables so that they can be constant in any one test. When you consider a DPS test on a target dummy you have variables out of your control which may vary from zone to zone or day to day, and variables which vary from person to person. Server lag, computer lag, crit strings/crit rate, eclipse proc variability (using wrath to proc eclipse), frequency of button presses, reaction time. All of these are variables which would require a huge amount of testing in order to reduce the background effect enough to see small differences in DPS between the tested factor (rotation, gear, etc), which is why we use calculators and simulators. If you then want WWS's of data from boss fights you're introducing a whole new set of variables such as buffs gained, percentage of the fight spent moving, time in combat, percentage of the fight bloodlust is active, etc etc. It doesn't make sense in the current environment to back up arguments based upon in-game testing. The best approximation we can get is to test the simulators/calculators against extensive tests from in game to make sure they are simulating/calculating things correctly, and then using those simulators/calculators to provide DPS estimates for constant situations, allowing the change of a piece of gear or switching which DoT effect is cast first in a rotation and getting a meaningful answer.

With Rawr and Wrathcalcs available for modelling, there isn't much point in putting up individual posts per rotation/spec/whatever because every case is different. It's nice to have guidelines about what could be used, but it's easier for everyone if you treat yourself as an individual case.

You completely modeled your post to form an attack making the assumption that I believe rawr or wrathcalcs is a bad tool which isn't the case, or my personal opinion.

I only asked for a clearly defined post on where balance druids should be heading in regards to stats/rotation now with 3.1 hitting.

And how is eclipse random? Well first off its RNG crit chance, ontop of a RNG proc chance. If that isn't random, then I must be a little out of touch on what RNG means. Also this is off topic, my original post was to ask for ONE solid clear post like all the other class spec's have had in the past.

Conquistador · 04/27/09, 11:19 AM

I think the real comparison comes down to 3 possible sets (with regards to tier slots)

4pc T7.5 + Leggings of the Enslaved Idol (this will be our base set for comparisons, as most people have all but the legs at this point)
2pc T8.5 (Legs+Shoulders) + 3 well itemized ulduar items
2pc T8.5 (Legs+Gloves) + 3 well itemized ulduar items

Beginning Assumptions:

Metagem is Chaotic Skyflare Diamond
All Red sockets and sockets which ignore socket bonuses will be filled with Runed Scarlet Rubies
Yellow sockets which we are matching will be filled with Veiled Monarch Topazes
Blue Sockets which we are matching will be filled with Purified Twilight Opals
I will not consider any profession perks like JC
2pcT8 is about equivalent to 4pcT7 in terms of overall DPS increase

T7.5 (Base Set)-
Head: [Valorous Dreamwalker Cover]
Shoulders: [Valorous Dreamwalker Mantle]
Chest: [Valorous Dreamwalker Vestments]
Hands: [Valorous Dreamwalker Gloves]
Legs: [Leggings of the Enslaved Idol]

Important Stats:
Int - 354
Spirit - 35
Crit - 252
Haste - 116
Hit - 147
SP - 476
Sockets - 5r/1y/2b
Socket Bonuses (assuming only needing 2 blues for meta): +12 int, +6 Haste, +9 SP

T8.5 (Legs/Shoulders) -
Head: [Collar of the Wyrmhunter]
Shoulders: [Conqueror's Nightsong Mantle] (ignore socket bonus)
Chest: [Robes of the Umbral Brute] (ignore socket bonus)
Hands: [Handwraps of the Vigilant]
Legs: [Conqueror's Nightsong Trousers]

Important Stats:
Int - 395
Spirit - 113
Crit - 264
Haste - 150
Hit - 110
SP - 534
Sockets - 0r/3y/4b
Socket Bonuses (assuming only needing 2 blues for meta): +23 SP

Gemming Summary (not factoring in professions):
We can assume that 2 blues from each set will be Purified Twilight Opals so that will be a wash. The additional 2 blues on this new set will be ignoring socketing bonuses so they'll be filled with Runed gems just as 2 of the red sockets with our base set. Each set will have one Veiled, so those are a wash. That leaves us with 3 red gems with the base set, and 2 yellow gems with our comparison set. Assuming that the red sockets get runed, and the yellow sockets get Veiled, we lose an additional 39 SP and gain 16 haste. (A second option would be using 2 Runed gems in the gloves instead of two Veiled and ignore the socket bonus, which would adjust our total gains in the following manner: +13 SP, -16 Haste)

Losses (including gems and socket bonuses):
Hit - 37
10% crit to our two main nukes from 4pc T7

Gains (including gems and socket bonuses):
Int - 29
Spi - 78
Crit - 12
Haste - 44
SP - 33
15% bonus to eclipse

T8.5 (Legs/Shoulders) Summary:
We gain a fairly decent amount of stats assuming we can afford to lose the hit. The hit should be easily recoup-able through the other pieces. It should be noted that the loss of the 4pc T7 will decrease eclipse procs by a fairly decent amount which will in turn lower the value of the bonus of the 2pc T8. This set should be at worse equivalent and at best a small upgrade.

T8.5 (Legs/Gloves) -
Head: [Collar of the Wyrmhunter]
Shoulders: [Shoulderpads of Dormant Energies]
Chest: [Robes of the Umbral Brute] (ignore socket bonus)
Hands: [Conqueror's Nightsong Gloves] (ignore socket bonus)
Legs: [Conqueror's Nightsong Trousers] (ignore socket bonus)

Important Stats:
Int - 392
Spirit - 104
Crit - 189
Haste - 150
Hit - 167
SP - 535
Sockets - 1r/1y/5b
Socket Bonuses (assuming only needing 2 blues for meta): +16 SP

Gemming Summary (not factoring in professions):
We can assume that 2 blues from each set will be Purified Twilight Opals so that will be a wash. The additional 3 blues on this new set will be ignoring socketing bonuses so they'll be filled with Runed gems just as 3 of the red sockets with our base set. Each set has a red socket which will be filled with a Runed. The yellow socket in the new set is paired with a blue socket so we wont be filling it with a Veiled and will instead be filling it with a Runed which we can count against one of the red sockets in the base set. The only socket left is one additional Yellow socket in the original set which means this comparison set loses an additional 9 SP and 8 Haste.

Losses (including gems and socket bonuses):
Crit - 63
10% crit to our two main nukes from 4pc T7

Gains (including gems and socket bonuses):
Int - 26
Spi - 69
Haste - 20
Hit - 20
SP - 57
15% bonus to eclipse

T8.5 (Legs/Gloves) Summary:
This is probably worse than our other comparison set (with the shoulders). The additional hit probably isn't necessary and we lose a substantial amount of crit in the process. In addition, compared to our other comparison set we gain 23 SP (including SP from spirit) but lose 24 haste which is an insubstantial increase when considering the huge crit loss. It should be noted that the loss of the 4pc T7 will decrease eclipse procs by a fairly decent amount especially considering the loss of all that crit, which will in turn lower the value of the bonus of the 2pc T8. This set is probably a worse than our base set with T7 items.

Overall Conclusion:
In my opinion 2pcT8 (legs/shoulders) >= 2pcT7 > 2pcT8 (legs/gloves)
It shouldn't be this complicated to figure out upgrades.

• Adoriele · 04/27/09, 12:02 PM

Originally Posted by Conquistador

It shouldn't be this complicated to figure out upgrades.

That is, in fact, Blizzard's intent. They don't want a very rigid upgrade path, as they believe it makes the game less interesting. They like making players make decisions, even tough ones. They like loosely forcing players to play around with different item sets to find out what works best for them. Agree with that decision or not, it is how they itemize. It works out decently, too. If there was a very rigidly defined upgrade path, people wouldn't be as concerned about which Eclipse rotation was better, as the answer would always be the same. There would be no attempts at innovation.

Lord BEEF · 04/27/09, 12:04 PM

Have there been calculations done on the power of 4pc t8 now that we know there's no internal cooldown? I haven't had a chance to see if the tools model it yet

thedopefishlives · 04/27/09, 12:07 PM

Originally Posted by Lord BEEF

Have there been calculations done on the power of 4pc t8 now that we know there's no internal cooldown? I haven't had a chance to see if the tools model it yet

Rawr has, from the beginning, assumed a 5% proc rate and no internal cooldown. The valuation of the set bonus under these conditions is ~80 DPS in full BiS T8-25 gear, depending on the selected options. In particular, Rawr seems to be very optimistic on Solar Eclipse as gear level increases.

Conquistador · 04/27/09, 12:07 PM

Originally Posted by Adoriele

That is, in fact, Blizzard's intent. They don't want a very rigid upgrade path, as they believe it makes the game less interesting. They like making players make decisions, even tough ones. They like loosely forcing players to play around with different item sets to find out what works best for them.

Fair enough, but I'd rather the decision be "Which Ulduar items do I want to use" and not "Is the gear I've had for the past 3 months still better than the new (and theoretically superior) item choices I now have". If the only debate were between 2pcT8 (legs/shoulders) and 2pcT8 (legs/gloves) I would love that. Perhaps I phrased it wrong when I said it shouldn't be this complicated. I don't mind that its complicated per se, just that its complicated to ascertain if new, higher iLvl gear is better than old gear.

• Adoriele · 04/27/09, 12:13 PM

Originally Posted by Lord BEEF

Have there been calculations done on the power of 4pc t8 now that we know there's no internal cooldown? I haven't had a chance to see if the tools model it yet

Napkin, but assuming the 5% proc rate is correct you'd average one proc every 20 ticks, or 40s, assuming 100% uptime, which is unlikely. Call it 80% uptime, which is more likely, and you get 1 every 50s. It's worth half of a Starfire when it procs, and Starfire average including Crit rate is, say, 8.5k, 10k during Eclipse. That makes the bonus worth ~100 DPS if you manage to proc it during eclipse every time, 85 DPS on the low end. For comparison, 4T7 is worth ~150 DPS, 2T8 about 165.