I chose to put Single MF just as Eclipse runs out because it's usually when IS goes up. I could change it to just before ICD finishes, but there's a couple issues. First, you're going to have a long period of time when you technically could cast MF, and you aren't. It's gonna be odd-feeling in practice. Second, it gets messy as to whether MF will be up for the whole Eclipse duration, so accuracy goes down. I'm still debating changing it.

I understand why you opted to code it as you did -- Im checking that option as dot clipping is something im very much against. Let me explain how I apply MF.

- I'll never clip (unless im out of other cast options & moving)
- I'll never cast during Eclipse if I can still realistically land 1 more SF.
- Outside of Eclipce & during hidden CD ill refresh just after it drops off (Taking prio over IS if both drop at the same time).
- If it drops towards the end of hidden CD ill use it together with IS to easy the phase transition from SF into W spam.

- If it drops during W spam, my cycle will look like this :
a) W repeat.
b) W cast (SCT shows Nature's Grace)
c) W cast (allready queue'd)
d) MF (Too soon to know if the crit procced elipce or not)
e) Usually you know if Eclipse procced or not while waiting for GCD to clear -- If not theres two options, if c also resulted in a crit then I go SF, if not I go W again.

Using the above steps as an alternative to clipping towards the end of eclipse hidden CD will lower your eclipse proctime slightly on average, but has positive side effects aswell (Activating potion is easier during a GCD than after a W cast before SF // small position adjustment possible just before 15seconds of forced standing still).


Not really expecting you to model this, just my 2 cents to why Im against clipping (Even though the sheet shows me a slight dps gain for doing so). That the two options are so close dps-wise gives me a pretty good indication that the non-clip style is a better ingame approach, a side note is that it has better dpm aswell!

Not looked at the new WC in too much detail yet, but at first glance upon putting my stats in, I noticed that marginal DPS value reported for Int is very high (0.73 DPS/Int). Something might be up here.

Downloading 1.2, and just changing character-sheet Int by 1, I see the value of int jumps up and down substantially.

In Basic Calcs, C14 (Total Int), the formula is

"=FLOOR(big_long_expression,1)"

changing that to just

"=big_long_expression"

makes things smoother and reasonable (0.39 DPS/Int).


Spirit has a similar issue, same fix in Basic Calcs, C13.

Every other use of FLOOR in the spreadsheet seems to be associated with Mana, not Damage. I don't know if there may be similar problems with mana. I believe it is appropriate to measure spell costs with

FLOOR(pct_cost*base_mana*moonglow_reduction)

but regen talents (Intensity, etc) should probably use a floating-point result (and it appears they do).

Few things:

1) I saw you mentioned the lack of proper IFF support further up. This would be really useful - I'm in a raiding guild that insists that I keep Faerie Fire up on every boss. (Heck, the guild leader, who is currently inactive, would yell in vent if he saw it had fallen off.) This was no problem when I was tanking, but as boomkin it hurts. I'd like to make an argument against doing so, at least until 3.1 with the 5 minute FF hits, but I can't tell how bad it nerfs my DPS - if I wanted to add the affect of it in, where should I start?

2) While 99% of the time, single target DPS is all we care about, there are times when measuring AoE is useful. I see that Starfall damage is mixed into the rotation (damage changes as the talent is selected/deselected in 1.2, and the cooldown tab shows this), but it'd be interesting to add support for Hurricane. I can see this being useful for guilds struggling with Sarth+3, as well as any Ulduar AoE fights (I've not followed the Ulduar info closely enough). I may try to add this based off of your Starfall calculations.

3) I saw a comment earlier about you wanting to avoid entering pre-LG spellpower. That raises a question - do we enter humanoid or boomkin SP? Crit and haste are modified on a percentage basis, so I don't think that they would matter. I would have guessed that it would be boomkin SP, but I'm not sure. What I do find odd is that the LG field is modifying DPS, so it almost has to be pre-LG. The difference is much smaller than I'd have expected - about 7 DPS per point in the talent.

4) In OO.org 2.4 and 3.0, I get Err:522 in almost all DPS/MP5 calculation fields on the main page. This refers to circular references, and OO.org recommends turning on iteration options - doing so, however, just makes all DPS/MP5 values in this be zero. Excel Viewer 2003, however, is giving valid values and MUCH higher overall DPS. Looking at Erdluf's comment explains why. The bad news, however, is that even editing the spreadsheet in OO.org breaks it, causing Excel Viewer 2003 to spit out #VALUE? errors in these same fields and give the same low numbers that OO.org gives. Has anyone found a workaround for this, or am I going to just need to get Excel 2003 (at least to do the edits)? The document is in one of the older binary office formats, so I can't edit these values manually, either.

iFF is one GCD every 40s or so, ~3.75% DPS loss just in time spent casting before Haste is factored in. If you wanted to be more specific with Eclipse rotations, you'd probably be putting it up in the post-Eclipse phase, so just subtract GCD+Latency from the post-Eclipse time field. I'll probably add in a FF field in the next version, with options for whether you, someone else, or no one is keeping up FF.

I'll also be fleshing out the AoE fields in DPET, I just need more solid data on the coefficients and such for Hurricane and Typhoon.

Enter your stats as you see them from the Armory in caster form. Everything else is taken care of.

I don't, nor do I plan to, support OOO.

Looking at the sheet, why is there no DPS contribution for Owlkin Frenzy? I see a lot of builds that do not have any points into this talent and it thought it would behoove me to put points toward a scalable % rather than dreamstate or +3% haste...

I just lvled to 80 still with my 2xT6 4xT5 and doing decent DPS, I believe, because of the T5 four set bonus. I will be running for T7 four set tonight, should get all pieces because my guild is offspecing them atm... but still just wondering about those three points. Thanks!

Lath

EDIT: Raid geared and buffed do I really need two in BoP (+4% spl hit)? thx again

Well put

I guess I'm living in the past. I remember having to spam Wrath on Essence of Desire because that was the only way with my haste I could punch out another cast before getting tacked for the aura without getting pushback... I guess I'll spec brables and try trenets in my rotation as it seems to be the norm. And with aprox 20k damage per cast it seems to be worth it's weight for now.

I expect 3.1 or later to spec back to OF... it seems like a useful talent that, as stated before, it just not needed... yet.

Thanks much for the quick response, you sir, are a gentleman and a scholar.

-Lath

WCv1.3:

Added split between Reaction Time and Latency. Per Erd's testing in another thread, spellcasts bottom out at 1s+Latency, and Reaction Time is always applied. I haven't built the ISV fields for Reaction Time, but they should be similar to what latency used to be. Notice that the Latency field is now much less important, which should be true in most cases; if you're queueing correctly, Latency will only matter with spells close to the GCD.

Added other 3.1 changes, including the NG change, new IS coefficient, new Starfall cooldown and glyph change, T8-2 bonus (the option exists for T8-4, it isn't implemented yet). Did the minor bump to base damage of spells to coincide with leveling that Erd mentioned in the same thread as before. Mucked minorly with raid buffing (Added Fel Intelligence). I also made a consistency pass through most fields to make sure things were being calculated properly. Mostly, this meant MP5 calculations for most rotations got fixed, thought there were some other minor changes. Changed 1-MF rotation to place MF at the end of the ICD instead of the beginning.

[edit] Just as I finish uploading, GC strikes again! Per MMO-Champion BlueTracker - Vanquish the evil!--More Moonkin stuff, 1.3.1 increases the T8-2 bonus, which makes it pretty much on par with T7-4 (even a little better).

Ah the main thing that's surprised me with the changes for 3.1 is how haste is now valued. it was from nearly being equal to spell power to being vastly weaker (from like 1.2 dps to .86 with my setup). hmm?

-- Love your work, thanks for the quality/excellence

Hey Adoriele,

I just want to start out by saying I'm a big fan of your work. Thank you so much for the awesome spreadsheet and addon.

Anyway, I was wondering if you could clarify the reaction time and latency fields a little more for me. I've realized that these values have huge effects on the dps value of haste, so it is crucial that the spreadsheet accurately reflects my in game experience. Basically, my question is what's the best way to determine my reaction time and latency. However, I have some notes about my attempts to do this on my own:

-Entering in 0 for reaction time and my in game latency (taken from putting my mouse over the little menu button) doesn't seem to give accurate calculations. For example, I spammed wrath on a target dummy for something like 85 seconds was able to get off 72 of them. My tooltip cast time for wrath was 1.24 sec and NG was something like 1.01. My latency was 250 ms. Entering 0 for reaction time and 0.25 for latency led to your "WrathCalcs" section predicting a 1.25 sec avg cast time for all wrath casts (normal, NG'ed and avg). However, 85/72 = ~1.18. What this tells me is that more haste would still be effective in speeding up my wrath casts and should therefore be valued the way it is when I enter 0 for reaction time and 0 for latency.

In short, the way you calculate latency me be incorrect and what's a good method to find my reaction time?

P.S. When haste is "wrath capped", i.e. wrath is no longer benefiting from haste because of latency or reaction time, haste is valued only slightly above crit (0.64 compared to 0.59 dps). When uncapped, it can shoot up anywhere from 0.86 to 1.07 (compared to spell power being ~1.26). So I really need to know if I should treat haste as roughly equal to crit or as vastly superior to it!

Thanks in advance,
Himtaurgar from Tichondrius