Jayde

Have been hibernating a bit with a busy raid and work schedule, but wanted to bump this to mention that I have updated this sheet (hopefully) fully with the upcoming 2.4 calculations, in addition to cleaning up the way healing pool, time to OOM, and OO5SR% is calculated.

Just to reiterate to avoid some confusion from before, there are two elements of this sheet: 1 is weightings, the other is calculations. You may want to use the sheet for both, you may want to use the sheet for one or the other. Use it how you like. I would would reiterate that this thread shouldn't be for debating the downrank ratio calculation method, as it's only a minor subset of the sheet and totally optional. (There is another thread for that.)

I believe that with the updated formulas, the healing pool to OOM should be resonably accurate given the way I am calculating OO5SR% contributions to regen and average casting time. Based on some testing in-game, I've found the results to match up fairly well.

That said, it's a complex sheet and I'm sure I've messed up somewhere or another... so if you find any bugs or suggestions on how better to calculate things, please let me know and I'll be happy to fix them ASAP.

 

Irise

Jayde, would it possible to change the spread sheet so you don't need to calculate the int and spirit from gear etc. What I am thinking off is to input stats from the char sheet without any buffs and the spreadsheet calculates the correct values after taking talents into account.

I created a spreadsheet for my own use as well. Its very crude and unpolished but I think it calculates everything correctly.

Priest Healing

 

Jayde

Hmm, yeah.. I was thinking about that myself.

When I first did it, I was mostly going off of Vontre's great Mage sheet, where you have to downward adjust some of the talents manually. In putting in my values to test it recently, though, I'm definitely leaning towards your suggestion. It would take a minor rewrite of the multipliers, since right now I combine talent and buff multipliers into one named range, but it's not a huge amount of work to change it.

Reversing the Mp5 figure it a little trickier, since you have to do the whole trickle-down regen calculation backwards, but I still think it would be a fairly minor task. Will see if I can do it tonight or tomorrow.

I also would like to test on PTR to make sure the calculations are correct, however EU PTR transfers are still quite borked, so I've been unable to get my Priest copied over.

Edit: Was easier than I thought to convert it... I have a version of the sheet that should do it properly, but since I've converted the Regen formulas to be 2.4, it's hard to test to see if it's working correctly. If someone who has a character on PTR would be willing to test it, please PM me.

 

Starfire

I saw you added some pieces of T6 there, could you add other new items to the list?
[Powerheal 9000 Lens]
[Brooch of the Highborne]
[Vindicator's Pendant of Reprieve]
[Wand of Cleansing Light]
[Robe of Eternal Light]
[Hands of Eternal Light]
[Rod of the Blazing Light]
[Band of Lucent Beams]

 

Havoc12

This is not correct the difference actually increases as +heal increases due to the effect of empowered healing in overcompensating the downranking penalty.

Lets have a look at numbers:

2300 healing 5/5 empowered healing, 3/3 imporved healing 5/5 spiritual healing

With 2000 healing

with 1000 healing


As you can see the improvement in mana efficiency increases it does not decrase. Lets look at this mathematically to derive a formula

(1) Improvement over rank 7 = (HP/m for rank x) / (HP/m for rank7)

(2) HP/m for rank x = (Healing for rank x) / (mana cost of rank x)

(3) Healing for rank x = [(Base healing rank x) + (bonus healing factor)]*1.1

(4) bonus healing factor = (+healing)*(3/3.5+0.2)*(downranking multiplier))

substituting (4) into (3) we get

(5)healing for rank x = (Base healing rank x)*1.1 + (+healing)*[1.057*(downranking multipler)]*1.1

Substituting (5) in (2):

(6) HP/M for rank x = (Base healing rank x)*1.1/(mana cost of rank x)
+ (+healing)*[1.057*(downranking multipler)]*1.1/(mana cost of rank x)

to simply:
(base healing rank x) = BHx
(mana cost rank x) = MCx
(downranking multiplier for rank x) = DMx
+healing = H

Hence HP/M for rank x = [BHx + H*(1.057*DMx)]*1.1/MCx

So (1) becomes

Improvement over rank 7 = {[BHx + H*(1.057*DMx)]*1.1/MCx} / [(BH7 + H*1.057)*1.1/MC7]

[top]> Improvement over rank 7

[BHx + H*(1.057*DMx)]*(MC7/MCx)/(BH7 + H*1.057)

[top]> Improvement over rank 7

[BHx + H*(1.057*DMx)]*(MC7/MCx)/(BH7 + H*1.057)

Hence the part of the equation that decides whether improvement over rank 7 increases or decreases with increasing H is

[BHx + H*(1.057*DMx)]/(BH7 + H*1.057) = f(H) for a given x

Lets find the 1st derivative of f(H) with respect to H

df/dH = [(1.057*DMx)*(BH7+H*1.057) - (1.057*BHx + H*(1.057*DMx)*1.057)]/(BH7 + H*1.057)^2

[top]> df/dH

[(1.057*DMx)BH7 - 1.057BHx]/(BH7+H*1.067)^2

It is obvious for for df/dH < 0 you must have (1.057*DMx)BH7 < 1.057BHx. For GHeal this is in fact never true. Lets just use rank 1 as an example. DMx for rank 1 is 0.769 but the BH1 is 2.6 times less than BH7. So its 0.813*2.6*BH1, which is obviously greater than 1.057*BH1. Hence df/dH never becomes 0. No matter how much +heal you stack the efficiency of rank x compared to rank 7 keeps increasing asymptotically up to a maximum value reachable only at infinite H. In fact due to empowered healing and the nature of the downranking penalty the efficiency of rank x over x+1 always increases as well.

To the original poster: Consider using an alternate means for calculating the relative value of healing and mp5. Instead of relying the decreased mana cost for downranking simply determine the increase in overal mana efficiency with higher +heal. This way you can directly convert +heal to mana regen.

 

Kinien

This might help people make better gear decisions for 2.4

MP5-modified.xls

 

Cauthon

Does this mean we should enter our stats exactly as they appear on the character sheet now and the spreadsheet reverse engineers the original values? I'm assuming based upon the default values that one should be wearing their current gear and the spreadsheet will suggest upgrades from there, correct?

 

Jayde

Yes, you should now be able to enter your stats in the yellow fields as they appear on your char sheet, although you should be also unbuffed when doing this and have the same talents set up as you do at the time. I will be double checking this shortly as I have finally managed to get my character copied to the EU PTR.

As a note, I will be posting an update tonight with some suggested item additions and fixes. Will also probably flag T6 drops so that toggling the option that limits to epic gems will also limit to <=T5 vs. >=T6 items.

 

Cauthon

Post deleted due to answering my own question.

 

Jayde

Updated to v 1.7 now, which contains the changes allowing one to just input their unbuffed char sheet stats in addition to the other notes: added ability to filter gear-based tiers, added all/most of the Sunwell healing gear I could find, updated some inproper stats on the Etherneum Life-Staff, fixed an error where meta sockets were not being considered when comparing head slot items.

If there are any items I inproperly flagged for tiers or missed out of the known Sunwell loot, just let me know and I will add ASAP.

In regard to haste, I'm having a heck of a time trying to factor how I should weight it. For downrank ratio weighting, I'm considering just calculating its HPS increase for heals when comparing to +healing, but I'm not 100% satisfied with that. For healing pool rankings, I honestly can't see of any way to apply it as it has basically no impact on healing pool whatsoever.

If anyone has any suggestions on how to mathematically rate haste vs. healing, I'm all ears.

 

Exinfernis

Jayde, having perused both your spreadsheet as well as Kineen's, I have a few observations to make:

Your calculations ascribe insignificant weight on the value of raw mp5. While it is widely accepted that with the advent of 2.4 the net value of spirit shall receive a tremendous boost, one cannot downplay the contribution of raw mp5 towards regeneration, especially in fights where stepping out of the 5sr in rewarding chunks is largely unfeasable or even ill-advised. I will cite the Najentus & Gurtogg fights, as my Guild has not progressed further.

As new gear enters my outfit, I take pains to boost my Spirit but in a logical (to me) fashion. + 10 Spirit gems go for every blue socket, but I will not consider the same gem for a red socket. A prime example of this are the [Adorned Supernal Legwraps], in which the red socket is primed for a + 22 gem. With this in mind, two of the most glaring examples of the difference in weighting raw mp5 between your calculations is the afore-mentioned vs. [Leggings of Eternity] and [Robe of Eternal Light] vs. [Robes of Faltered Light].

Therefore, the valuation rating given to the above heavily hinges upon the particular emphasis each of you place on 16 mp5. Personally, I have tried not go below 90 mp5 (from items) unbuffed. At the moment, I am at 460 oot5sr, 210 it5sr, 522 Spirit, 522 Int & 2162 + to healing (unbuffed).

I did not have the time to visit the PTR servers, thus I am unsure whether the incoming Spirit boost will render my preconceptions obsolete - may well do so.

 

Jayde

Not sure which comparison you are talking about specifically... but I assume Spirit vs. Mp5?

I do have an input for OO5SR%, which will dynamically adjust the value of Spirit based on that figure in addition to talents and potentially the PMC bonus (if enabled.) Spirit is also being multiplied by SoR, Human, and BoK bonuses (if enabled) when comparing weights. Currently Mp5 is a rather static value as none of the buff/talent values really affect it at all.

With 0% OO5SR, no buffs, and no Human setting, Leggings of Eternity does indeed seem to pull ahead. With 0% OO5SR, Human and BoK turned on they seem roughly the same. With any % OO5SR, the Adorned Supernal Legwraps seem to pull ahead. As most fights will have, at a minimum, 10-15% OO5SR (around 30% for many fights though) the Legwraps seem to start pulling ahead.

So, I'd recommend just playing with the settings to simulate different encounter situations.

 

Exinfernis

Indeed Jayde, Spirit vs. raw mp5 is the focus.

And to underscore my musings, I am eagerly awaiting to see whether the loss of 16 mp5 on LoE and RotEL shall be offset by 22 Spirit (net gain by gemming + 10 Spirit & + 22 to heal gems) for the former and 34 Spirit (net gain by gemming + 10 spirit *2 plus 11 to heal/+ 5 Intellect) for the latter. Assuming of course + 10 Spirit *3 for both of the contenders.

I did play around with the settings, however I find the tradeoff to be too steep in the case of the legs, whereas the chest comparison veers clearly in favour of [Robes of Faltered Light].

 

Crosshairs

I was wondering why this gem, [Imperial Tanzanite] isn't included on your spreadsheet.

 

Jayde

I have uploaded version 1.8:
priest_healing.xls - FileFront.com

Updates are:
-Initial modeling/ranking of Trinkets
-Imperial Tanzanite to the gem list
-Modeling for IED and support for selecting which meta gem to use in helm slots
-Initial modeling and support for Haste weighting
-A few missing Sunwell patch items
-Support for showing healing data for Prayer of Mending (x1 and x5), Prayer of Healing (x1 and x5), and Binding Heal (x1 and x2) in the Calculations sheet for references purposes

This is a reasonably large update, so a few notes...

Haste weighting:
This was a bit tricky, and I may change how this works... however, I feel that this is a good enough solution for ballparking it for the moment. First, the weight for healing pool is 0, as it has zero effect on healing pool whatsoever. However, for the purposes of downrank ratio-based weighting, it does have a notable impact and is calculated in an equivalency ratio to +healing. Basically I am looking at the HPS increase from +1 Healing and comparing it to +1 Haste for the selected downrank. Haste is actually much more efficient in this regard, so the base ratio is something around 2.5:1 with my test settings. However, I felt this value was too high simply due to the fact that +Healing also raises healing pool by the same percentage as HPS, whereas +Haste does not raise healing pool at all. Therefore, the ratio is thus halved, giving haste a weight of around 1.3, depending on your selected downrank. (Range is roughly between 1.15 and 1.45)

In looking at the items, I have the feeling that this weight is reasonably close, however I'm open to feedback on how better to calculate this. Realistically, most of the items with +haste did not change ranking position even with weighting +haste, with a few exceptions--mostly rings.

Trinkets:
There is now a sheet for trinkets, with basic modeling of most trinkets I could think of. There may be some issues here if I got the hidden cooldowns wrong, or modeled things incorrectly. This was done pretty quick, so if you have any suggestions for how to model some of the trinkets better, please feel free. Some of them are modeled in an odd way to make sure they are all on the same duration/cooldown/uptime model, but I believe they are all reasonably accurate at the moment.

(Edit: I realize I forgot to put in the Alchemist Stone trinkets.. sorry about that, have uploaded a fixed version.)

 

Muck

There is a minor bug with the rank sorting if you change the # of T5 value.
Also it would be nice if you add the draenei racial "Symbol of Hope".

 

Bendyr

Jayde I love this spreadsheet. You've just given me a new toy. One thing I've noticed is your gem selection doesn't seem to include [Seer's Chrysoprase], which by my limited calculations is the best Yellow Gem available after 2.4. Or am I a noob and just missing something huge?

 

Jayde

Thanks for the note, will add it in the next update. I had left out the heroic ones previously due to their unique nature, but I will add in some of the more interesting ones.

 

Pulchellus

Thanks for putting in the hard work on this.

One thing I would like to see is the stamina values for items - and maybe the possiblity to give stamina even a nominal value in calculations. For instance Avian Cloak of Feathers gets over-rated imo, because it has a good amount of regen potential. Whereas Stainless Cloak of the Pure Hearted drops down the list because it looses regen in favour for stamina. Given how stupidly squishy I am it would be nice to see if that extra regen is worth giving up over 200 hp for.

You should also add the whitemend set bonus if you are looking at completeness.

It would be nice is the ability to put in your own gear and compare +/- from there, similar to the elemental shaman spreadsheet.

Last thing would be to have a source column, but I can add that in myself for items of note that I haven't yet remembered where they come from.

 

Arvak-

It's nice to see some objective numbers on all the gear floating around. Are you planning to add some method of inputting manual base stats and adjustments? Given how int scales, that would help a great deal in determining its real upgrade value in relation to current gear.

 

Jayde

Added Version 1.9:
priest_healing.xls - FileFront.com

-Added much better T5/T6 set bonus support
-Added some new heroic gems
-Added the new SSO exalted neck items
-Fixed the modeling for average adjusted OO5SR% cast times (was off by around .1s for GH, some faulty logic before... other heals were waaaay off, even though it didn't affect any calculations, should be much more accurate averages now)
-Fixed the modeling on Hyjal exalted ring proc
-Fixed some trinket modeling

Next on my list will be to add support for Stamina data (quite a bit of manual work to add it to the equipment data for all the items, though) and some equip/upgrade comparison stuff. Don't think there is much more to add beyond that at the moment.

 

Yellowss

Hey Jayde
First of all: Fantastic sheet!
There is just a little thing there is annoying me... do you include the +heal/spi/whatever you get from the gems you suggest in the "Total Mp5"-column and in the "Total Heal"-column beside every item. (This is for instance interesting if you compare "Leggings of Eternity" and "Adorned Supernal Legwraps", because there is a different number of gem slots in them) If could be a very interesting feature… don’t you think?

Bedst regards
Media

 

Jayde

Nope, currently I don't due to me being a little lazy when I hacked in those "total" columns. I only added them after the 2.4 version to see the impact to raw stats with all the changing formulas. However, I do agree that it's nice info and I will try to work it to be proper into my next version.

 

ilkori

For the next version, could you add in the effects of spell crit? It may not be the best stat to stack, but it would be worthwhile for Circle of Healing calculations, at the very least. (10% crit is 5% higher throughput than 0% crit as calculated now.)

Great spreadsheet, btw.

"It's kind of fun to do the impossible." - Walt Disney

 

Verad

I have the vengeful gladiator wand, and was curious as to why that is not on the list.