Jump to content


Photo

Moonkin DPS Spreadsheet


  • This topic is locked This topic is locked
1925 replies to this topic

#1 Efejel

Efejel

    Piston Honda

  • Members
  • 246 posts

Posted 02 May 2007 - 12:39 AM

Current production release: Updated 2008.04.23
0.80c: Moonkin DPS - v0.80.xls - FileFront.com (notes)
Suggested version: Updated 2008.08.18
0.81alpha: Moonkin DPS - v0.81alpha.xls - FileFront.com

>>>FREE Spreadsheet Viewer (OpenOffice.org)<<<

Please also visit the "ThinkTank!" http://elitistjerks....-druid_balance/

FAQ
Q: My spreadsheet shows an estimated XXXX dps, but I can only pull off YYY dps despite my best efforts. What's going on?
A: DON'T PANIC. The spreadsheet tends to over-estimate. Make sure you've got the correct buffs selected, the latency is set correctly, and your fight duration is accurate. Now assume a 4-5% DPS loss if you're responsible for Faerie Fire. Each addt'l 10-15 seconds not casting (looking for the corpse of that damn rogue you have to B-rez, for instance) on a 6 minute fight can be another 5% DPS loss.

Q: Why do I sometimes see odd (and/or prime) spell rotations that don't factor out correctly, like Mfx2, Sfx11?
A: The spreadsheet is based on averages. A rotation like Mfx2, Sfx11 on the spreadsheet becomes Mfx1, Sfx5, Mfx1, Sfx6--depending on when you get the expected number of NG procs, etc. To best emulate these numbers in game, you simply want to keep your Moonfire on the mob without overwriting the last tick (sometimes called "clipping"), refreshing each time it's off without delaying your Starfire casts.

Q: Your spreadsheet shows I'm capable of X and my buddy's spreadsheet for his {mage, s-priest, ele/enh shammy, lock, hunter, rogue} suggests he should be doing less DPS, but he consistently beats me on the meters. Why?
A: Spreadsheets use different methodologies & make different assumptions, the numbers they provide are relative and aren't directly comparable. That being said, if you're getting spanked by your buddy in spite of posting WAY BETTER spreadsheet numbers, and out-gearing him in practice, try recording your fights (Fraps if possible, WWS if not) and look at how close you're coming to actual ideal consumable usage & cycle timing.

Q: How do you calculate for activated trinkets/potions and proc trinkets/set-bonuses/rings/idols?
A: I don't. Except for predictable mana restore effects, which can be easily averaged, the effects of proc-based (Mag's Eye, Quag's Eye, TLC, Sextant, etc.) and activated (Icon of the Silver Crescent, Scryer's Bloodgem, etc.) trinkets and the like are IGNORED. You'll pretty much be left to your own devices & the opinions of other experienced Druids to assign a numeric value to these effects, especially ones which are fight & spell-cycle dependent. Addt'l note: the "Moonfire" idol option is for the pre-BC "Idol of the Moon".

Q: The spreadsheet shows higher DPS with item X over item Y, but that doesn't make sense to me. Why does your spreadsheet lie?
A: The cake is a lie, the spreadsheet is an approximating & averaging algorithm. If you don't like the suggestions you're getting, no one is making you take them. And while the spreadsheet can easily be off by a few percent, be aware that it's cumulative 1-2% increases in DPS that got you where you are now from where you were when you were wearing AH & quest reward blues/greens.

Related Questions (moonkin specific, but not about spreadsheet)
Q: Is Improved Faerie Fire a good investment of points?
A: Depends on your raid make-up. It will definitely increase the threat generated by your main tank. How much that will help your raid, and the direct effect it will have on your other melee DPS depends on how many of them dual-wield, their gear level & how close/far they are from the hit cap.

Q: Is Force of Nature better than an additional point of threat reduction?
A: If you're getting Blessing of Salvation, if your raid is comfortable with allowing you to cast them, and if you remember to cast them, yes, it's better than an addt'l point in Subtlety.

Q: What's the ideal group make up for my Moonkin?
A: Are you trying to maximize the moonkin, or the raid as a whole? Your raid has a unique make-up so there's no general answer for this. We do, however, go great with Shadow Priests (at least, they help us alot), Elem/Resto Shaman, Dest Locks, Mages (regardless of spec) and to a lesser extent BM Hunters, Holy Pallys and Holy/Disc Priests.

Q: Should I use idol swapping macros?
A: Not in 2.4.3 or later.

Previous releases:
0.72c (notes)
0.71b (notes)
0.70e (notes)
0.69 (notes)
0.68 (notes)
0.67 (notes)
0.66 (notes)
0.65 (notes)
0.61 (notes)
0.60 (notes)
0.50

Other (simplified) spreadsheets by others:
Elitist Jerks - View Single Post - Moonkin DPS Numbers/Comparisons
Elitist Jerks - View Single Post - Moonkin DPS Numbers/Comparisons

Attached Files



#2 Tyrnall

Tyrnall

    Glass Joe

  • Members
  • 6 posts

Posted 02 May 2007 - 09:24 AM

I feel like I'm turning into a bad internet stalker for you, ef, but you seem to have the math for the moonkin, so here's another question for you 'bout that spreadsheed:
In another post, on the the WoW Class boards, you've got a post going saying that wrath should out DPS Starfire, and by my simple math (literally as simple as "if I'm hitting for 1200 every 1.5 seconds, it's more than hitting for 2k every 3 seconds...so I use the term 'math' liberally) it should be.

However - be it in my full gear or totally unbuffed and naked, I can't get the spreadsheet to confirm this - absolutely every configuration I've tried has Starfire spam coming out on top of wrath-spam - even assuming 0 latency. My fault somewhere here or is there a flaw in the sheet?

Edit: Figured out the problem - forgot to account for running oom too fast...oy I feel like a noob.
Tyrnall - level 70 druid
Druid Class Lead of <Unpossible>
Zul'jin

#3 Efejel

Efejel

    Piston Honda

  • Members
  • 246 posts

Posted 02 May 2007 - 11:34 AM

Try the following:

Go with your "normal"--not boss--gear, but add ~150 hit rating (I added a "Target level" setting in the next version, but it's currently undergoing construction so I'm not really able to post it.)

Turn most buffs off. I'd say leave MotW (talented if appropriate), Superior Wizard Oil, and Blackened Basilisk--this is about what I normally use in a 5-man.

Set fight duration to 2 minutes, turn all the mana restore potions/consumables to "No" and Innervate to 1 @ 1.5 min.

Select the Wrath idol.

Set Curse of Shadows & Misery to 0. (Or just turn Curse of Shadows off).

Make sure Mystical Skyfire Diamond is "No".

Hopefully it nets you something like this:

Posted Image

EDIT: P.S.-I implemented the Wrath idol as an additional 25 +dmg for Wrath only. So, it's only adding ~ 25 * .78 = 19.5 dmg to Wrath. This is the conservative estimate--I haven't tested it yet (it's on the "To do" list) but it could be as much as 27.5 dmg, which will skew things a bit more.

#4 Efejel

Efejel

    Piston Honda

  • Members
  • 246 posts

Posted 02 May 2007 - 10:48 PM

Uploaded version "0.6".

Notes:
Added target level pulldown
Partially-implemented gear selection
Implemented most of the previously greyed out World buffs.

Spreadsheet was set to demonstrate a fight vs. level 72 mob but "6% resist" was inadvertently set to "Yes". Set that to "No" for more accurate results.

#5 Efejel

Efejel

    Piston Honda

  • Members
  • 246 posts

Posted 02 May 2007 - 11:25 PM

Found & fixed a couple of errors where it was still assuming 17% boss miss and not taking into account level. Also found & fixed some order of operations errors with partial resist, while I was at it.

#6 Kharlis

Kharlis

    Glass Joe

  • Members
  • 20 posts

Posted 03 May 2007 - 03:29 PM

If i remember correctly rank 8 wrath was massively more mana efficent than rank 10, allowing for further extended, albiet slightly lower dps cycles( this is after factoring the downranking co-efficent).

Following the (Spellevel+4)/character level= spell damage applied co-efficent ,Wrath gets the following bonus damage for its various ranks.

Name / Rank / Level / level Co Efficient / Mana Cost (talented) / % cost of rank10
Wrath / 7 / 46 / 71.43% / 141.05 / 0.61
Wrath / 8 / 54 / 82.86% / 163.8 / 0.71
Wrath / 9 / 61 / 92.86% / 191.1 / 0.82
Wrath / 10 / 69 / 100.00% / 232.05 / 1.00

It is entirely possible that i misunderstood the level based spell damage application, in which case please correct me on this.

Mana is one of the major limiting factors for moonkin dps, i feel exploring downranking to increase our efficency is essential.

this is not as usful for starfire

Name /Rank /Level/ level Co Efficient /% of rank 8
Starfire/ 5/ 50/ 0.77/ 250.25/ 0.74
Starfire/ 6/ 58/ 0.89/ 286.65/ 0.85
Starfire/ 7/ 60/ 0.91/ 309.4/ 0.92
Starfire/ 8/ 67/ 1.00/ 336.7/ 1.00

#7 Torondor

Torondor

    Glass Joe

  • Members
  • 8 posts

Posted 03 May 2007 - 03:52 PM

it seems like the "Approx time till OOM" is too low. I was extremely overestimating the amount of mana I would have/regen, and most cycles still say I would go OOM before the end of a 5min fight (including pots, regen flask, food, innervate, shadowpriest, mana tide, etc).

I tried playing with the time using the 5s spirit regen, and adding a full minute of spirit regen netted me less than an extra second of fight before running oom. I'm guessing either it's not all worked out in that department, or something is off.

also, is nature's grace included?
edit: nvm just saw the talents section

#8 Efejel

Efejel

    Piston Honda

  • Members
  • 246 posts

Posted 03 May 2007 - 07:37 PM

it seems like the "Approx time till OOM" is too low. I was extremely overestimating the amount of mana I would have/regen, and most cycles still say I would go OOM before the end of a 5min fight (including pots, regen flask, food, innervate, shadowpriest, mana tide, etc).

I tried playing with the time using the 5s spirit regen, and adding a full minute of spirit regen netted me less than an extra second of fight before running oom. I'm guessing either it's not all worked out in that department, or something is off.

Can you post a screenshot of the Character Sheet & Buffs page, and let me know if you changed Dreamstate/Moonglow/Intensity at all?

This is what my spreadsheet (v0.61, as posted) looks like when I set things up for about how Gruul goes: Posted Image
Now, take away the Shadow Priest and things turn ugly pretty fast. :)

If i remember correctly rank 8 wrath was massively more mana efficent than rank 10, allowing for further extended, albiet slightly lower dps cycles( this is after factoring the downranking co-efficent).

Following the (Spellevel+4)/character level= spell damage applied co-efficent ,Wrath gets the following bonus damage for its various ranks.

Name / Rank / Level / level Co Efficient / Mana Cost (talented) / % cost of rank10
Wrath / 7 / 46 / 71.43% / 141.05 / 0.61
Wrath / 8 / 54 / 82.86% / 163.8 / 0.71
Wrath / 9 / 61 / 92.86% / 191.1 / 0.82
Wrath / 10 / 69 / 100.00% / 232.05 / 1.00

Name /Rank /Level/ level Co Efficient /% of rank 8
Starfire/ 5/ 50/ 0.77/ 250.25/ 0.74
Starfire/ 6/ 58/ 0.89/ 286.65/ 0.85
Starfire/ 7/ 60/ 0.91/ 309.4/ 0.92
Starfire/ 8/ 67/ 1.00/ 336.7/ 1.00

I'm not sure what coeff. is given based on down-ranking, but I doubt Starfire follows the (level + 4) convention, as many of the gaps between ranks (5@50 vs. 6@58, 7@60 vs. 8@67) are much greater than 4 levels. Similarly, many of the Wrath gaps are > 4 levels by a fair amount. I'm inclined to lean towards the (level + 6) convention, but I'm not even sure that's accurate.

I know personally, I down-rank to Starfire 6 quite frequently and note that my pre-debuffs damage changes by ~100 (1550 vs. 1650 average non-crit) for 50 mana saved per cast. 6% less DPS for 15% less mana.

If I get the time to determine the actual coeff. for downranked Starfire & Wrath, I'll look at adding those options to the sheet.

#9 Kharlis

Kharlis

    Glass Joe

  • Members
  • 20 posts

Posted 03 May 2007 - 10:10 PM

According to http://forums.worldo...&sid=1&pageNo=1

its (spell level+6)/character level.
this changes the values above to -

Wrath % SD % Mana Effectiveness
7 74.29% 61.00% 103.26%
8 85.71% 71.00% 110.57%
9 95.71% 82.00% 112.94%
10 100.00% 100.00% 100.00%
Starfire
5 80.00% 74.00% 100.80%
6 91.43% 85.00% 105.14%
7 94.29% 92.00% 101.83%
8 100.00% 100.00% 100.00%

%SD = % of spell damage applied due to "downranking"
% Mana = % of mana that this rank costs compared to the highest currently available rank.

Effectiveness calculated as (%SD+(1-%Mana)x%SD) or more simply , how much damage you get from the same amount of mana as a top ranked cast.

Looking at it this way it looks like rank 8 & 9 of wrath will give you most over a static mana pool.

Digression --- with mana being so important, what relative value would you assign to something like an alchemist's stone(using super mana pots its an additonal 40 mp5, with fel mana it becomes closer to 53 mp5)?


I really like what you have done so far though. nice job.

#10 Efejel

Efejel

    Piston Honda

  • Members
  • 246 posts

Posted 03 May 2007 - 10:42 PM

Digression --- with mana being so important, what relative value would you assign to something like an alchemist's stone(using super mana pots its an additonal 40 mp5, with fel mana it becomes closer to 53 mp5)?

I really like what you have done so far though. nice job.

Well, the "Alchemist's Stone" option is one I added early because I knew people would be curious, and it's fairly easy to implement. My impressions based on the numbers & my experience without it are: it's good, but I wouldn't level Alchemy solely to acquire one.

My take on professions right now, for raiding Balance Druids:
Enchanting ~= Alchemy > Tailoring > Jewel Crafting >> Engineering ~= Leatherworking > Herbalism

Enchanters will always be able to find rings which are upgrades, and then slap +12 damage on them. They also gain easier access to Wizard/Mana oils.

Alchemists will always have the Alchemist stone, but unless we see new potions it's going to continue to supply 40 mp5 / 53 mp5 while base amounts of mp5 from gear (and gear based talents) gets bigger. They also gain easier access to potions/flasks/elixirs.

Tailoring, Jewel Crafting, Engineering & Leatherworking all have a few nice BoP crafted items that make them worth considering. BUT we have every reason to believe, based on what was seen pre-BC, that new items which are more powerful will be added to PvE & PvP progression, while professions get left behind. Could I be wrong? Sure, but the track record is certainly for crafted BoPs to fall woefully behind.

Herbalism = being mostly self sufficient in terms of consumables, and thus being more likely to have them available each fight. You may need to find an enchanter or alchemist to do the combine, but no doubt there's one of each in your raid.

With an eye to the long term, I'm very fortunate to have chosen Herbalist/Enchanter.

And thanks, I'm glad someone is finding this useful. :)

#11 Efejel

Efejel

    Piston Honda

  • Members
  • 246 posts

Posted 04 May 2007 - 08:11 PM

Has anyone found anything wrong with the mechanics in the latest version?

I've been working on cosmetic enhancements & fully implementing the gear selection window, but I'm more interested in the accuracy of the calculations I'm doing.

#12 Efejel

Efejel

    Piston Honda

  • Members
  • 246 posts

Posted 05 May 2007 - 04:19 AM

I just hopped on the PTR & did some quick testing. It seems that the Wrath idol will actually update the in-game tooltip, and is affected by Moonfury. The Starfire idol will NOT update the tooltip, but is similarly affected by Moonfury. Neither appears to be affected by Wrath of Cenarius.

I looked at the Moonfire idol in my bags and.... decided I didn't care that much.

#13 Efejel

Efejel

    Piston Honda

  • Members
  • 246 posts

Posted 07 May 2007 - 09:04 AM

I'm uploading a new version, 0.65 with the following features:

Wrath & Starfire idols should now be factored more accurately.

Conditions are set to mimic those present in this thread: http://forums.worldo...101801597&sid=1
I'm pleased to find that the spreadsheet predicts an IS:SFx4 rotation as doing 280,065 dmg and Horao's screenshots show 280,223 total damage. :) Clearly the fight went 4.25 minutes + 1 IS tick, in reality.

Cosmetic improvements, included automatic highlighting of the highest DPS cycle.

Gear selection page has seen continued additions, but is still not complete.



Additionally, I don't expect to be able to do much more work on this in the near future, but I should be able to handle simple bug fixes. And if anyone else is interested on doing further work on the Gear Select page, you can get in contact with me via PM to ask questions or to give me newer versions to upload & link.

#14 Bitolan

Bitolan

    Glass Joe

  • Members
  • 3 posts

Posted 07 May 2007 - 02:52 PM

Excellent work Efejel! Some of my non-moonkin friends were checking out this spreadsheet and were jealous at how clean and easy to use it is.

#15 Boevis

Boevis

    Bald Bull

  • Members
  • 1,284 posts

Posted 08 May 2007 - 03:50 PM

I assume you're not including Moonfire crit in the calculations for Natures Grace procs.
Damage_Calcs W2: =(((3.5-(0.1*Starlight_W))+Latency))*(Sf_p)*(Sf_p)

#16 Lambach

Lambach

    Don Flamenco

  • Members
  • 258 posts

Posted 08 May 2007 - 08:16 PM

None of the Dmg trinks(quags eye, nexus horn) seem to have any effect on the calculated dps.

Also, the dps calcutated agrees with my old theory about spell choice which the other forum on moonkin dps debunked. Have you read that thread at all, and have any idea why the diffencences between your math and theirs exist? Did you use different equations in determining the dps?

#17 Efejel

Efejel

    Piston Honda

  • Members
  • 246 posts

Posted 10 May 2007 - 10:59 PM

I assume you're not including Moonfire crit in the calculations for Natures Grace procs.
Damage_Calcs W2: =(((3.5-(0.1*Starlight_W))+Latency))*(Sf_p)*(Sf_p)

Yeah, I'm not. I'll definitely consider looking at that before I get to a version I'm willing to call 1.x. It's going to be a very very slight boost to DPS of the two Starfire + Moonfire cycles.

None of the Dmg trinks(quags eye, nexus horn) seem to have any effect on the calculated dps.

Also, the dps calcutated agrees with my old theory about spell choice which the other forum on moonkin dps debunked. Have you read that thread at all, and have any idea why the diffencences between your math and theirs exist? Did you use different equations in determining the dps?

They greyed out selections are features I hope to implement later, or would love for others to help me implement. This includes some set bonuses & trinkets, etc.

I'll try and find what you're talking about as far as "old theory about spell choice which the other forum ... debunked" but if you mean Wrath vs. Starfire vs. complex rotations, the simple answer is that fight duration, raid & consumable buffs, and exact gear selection make appreciable differences there. That was one of the things I hoped this tool would demonstrate, is that selecting the correct spell rotation can lead to noticeable DPS differences.

EDIT: I'd bet you're talking about this: http://elitistjerks....1061#post331061
And, yeah, the spreadsheet will show that Wrath spam + Moonfire is excellent DPS in certain situations. Those situations just don't commonly occur in 25-man raid boss fights. No arcane dmg specific gear, essentially unlimited mana, frequent Stormstrike debuff application, weak and/or lacking Curse of Shadow, Misery, JoW (which I need to model better) on the target, high latency... not impossible, but highly specific. Change any one~two~three of those factors slightly and the picture can change pretty greatly as well.

Balance Druid DPS, when done "correctly", isn't actually all that boring--at least not to me.

#18 Efejel

Efejel

    Piston Honda

  • Members
  • 246 posts

Posted 15 May 2007 - 05:01 AM

Version 0.66 notes:

SHOULD BE COMPATIBLE WITH OOo (Open Office.org) calc
8 "primary" slots have gear selection implemented. Cloaks, jewelry, weapons coming soon.
Tier 4 4pc & Gladiator 4 pc implemented
Corrected error that was causing Innervate to penalize ALL mp5 instead of Intensity based mp5
Added support for 1/2 Improved DS Priest buff------seriously. Not sure this was worth the 30 seconds it took, but I wanted to get rid of one box of red text

P.S.-Things that I would find helpful:
1) Suggestions on modeling Quag's Eye, Shiffar's Nexus Horn (possibly tied in with #2 below...)
2) Assistance in implementing a spell-cycle similar to the one found in the Shadow Priest DPS spreadsheet
3) Suggestions/assistance modeling JoW (which will probably extend nicely to the T4 2 pc. bonus)

#19 Lorewanderer

Lorewanderer

    Moof.

  •  Patrons
  • 2,360 posts

Posted 16 May 2007 - 12:51 AM

For those of us with nasty latency or other considerations (always keeping IS up, for one), would you be willing to include an IS, SFx3 rotation as well?

I find myself using that, sometimes with down ranked SF, on fights where I am needed on healing for a portion of the fight but want to do something more than just stand there for the other portions.

It might also help people whose gear is particularly bad (or those in more heal-oriented gear).

#20 Efejel

Efejel

    Piston Honda

  • Members
  • 246 posts

Posted 16 May 2007 - 03:57 AM

Sure... do you overwrite the last tick(s) of IS in order to ensure it never drops off, or is your latency bad enough that's not usually a problem? (If so... :-o )




0 user(s) are reading this topic

0 members, 0 guests, 0 anonymous users