ilkori

I've been digging a bit more into the nitty gritty of the spreadsheet. I see progress in addressing some of the initial concerns posted at the beginning of the thread, but there is still plenty of room for improvement. (By no means do I expect to find a perfect tool, though having something is better than nothing by a long ways - and the progress on your sheet means I don't have to make my own.) I have a few more suggestions, which I hope do not cause too much headache. (Suggestions numbered for ease of reference.)


1) A note about the haste modification would be nice in either first post documentation, or in the spreadsheet itself. I was a bit shocked on this, personally. The value of 1/2 seems rather arbitrary, and would vary quite a bit based on casting style and situation. While I could use custom weights to give a personal feel. (I would guess around 2/3 effectiveness, though this is a purely "feel" thing and I haven't played with much more than Bloodlust in game.) Haste is a tricky bugger. While you jump directly to a # casts to OOM calculation, the intermediate value of MPS casting may be worthwhile to balance with HPS and mana regen. But, that's about as far as I get there... Ooh, and haste increases the chance for ticks to be outside the 5sr since casts end more quickly when chain casting (though I do not know if this effect is significant enough to calculate into a spreadsheet).

2) Another alternative for evaluating stat weightings would be one based on a ratio of specific spells for a given fight. For example, if I find that I cast 2 GH1, 1 GH5, 1 FH9 and 1x Renew, then those ratios could be used to weight the values for various spells and abilities. (A FH1 option would be handy for me, though I doubt many other people use the spell.) Combined with a %oo5sr, this would give a general ballpark for time to OOM in a realistic fight (ignoring Clearcast, IF, etc.). This would also provide an approximation for average casting time if combined with another stat (such as % time casting, or % casts per minute(second) or something of that nature).

3) Trinkets are an odd case because of how specific they can be. For example, DMC: Blue Dragon could really use its own toggle option near the oo5sr% cell since it affects regen even more than the PMC bonus. There are a couple flaws with the way it is calculated as well. This is nearly identical to the Eye of Gruul calculation. Cooldown = 1 - (1 - .02) ^ N, where N is the number of casts in the last 15 seconds. Further, the effective uptime is reduced by the time already spent oo5sr (i.e. 200% of normal regen is not possible). Also, it should be reduced by both Meditation and PMC regen (along with Bangle of Endless Blessings if one has that for a 2nd trinket). Due to the way it is calculated, it is much higher than it should be in terms of calculating regen, but possibly conservative in terms of casting time. If a change is added with average #casts/minute or #casts/second (possibly incorporating the second suggestion), that would give a much more realistic valuation of the trinket.

4) [e3 - found where meta was enabled on gem sockets] Given the recent changes to regen, [Ember Skyfire Diamond] should be included. This meta outdoes [Bracing Earthstorm Diamond] by your spreadsheet for any priest who has at least 600 raid buffed int and an int/heal weight greater than 1. (Noting that threat reduction is near impossible to quantify as it is a function of your tanks more than anything else.)

5) [Adept's Elixir] could be added as a Battle Elixer since it stacks very close to [Elixir of Healing Power] on CoH healing and those of us on Tank duty may value it more highly due to Inspiration uptime.

6) I'm still thinking about the Stamina proposal. Is there any way to adjust for other factors, such as not wearing best-in-slot for given gear level?

7) Speaking of gear level, is there any chance that BoE equipment (or crafted BoP pieces from BoE recipes) of higher tiers be set at Tier 0? (Lootzoor has a handy way to check/uncheck the crafted patterns, and I don't know if this is something beyond the scope of what you had in mind for the spreadsheet.)

[e2]
8) Something seems broken with talents and stats... I'm still trying to to figure out what is going on. When Spirit of Redemption is toggled off, there is no change to mp5 values. There is other funkiness as well with stats and talents.

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

 

gasilia

Very interesting your stats, yeah possible that renovation and PoM change a lot the stats in favor of the card.
as the cost for CoH is exactely the mana you gain by the eye it could theoretically allow so leave 5SR, do you really have the leisure to stop casting for a few seconds to get one tick of regen O5SR?

Furthermore looking at your stats for the fight of Naj'entus and bloodboil there are proc rates of 1/10 for bloodboil and 4/10 for Najentus
It is obvious that for bloodboil Gruuls Eye outweights by large amount the card, but for Najentus I get (if I am not mistaken)

10* proc of Eye of Gruul = 10*450 mana = 4500 mana - plus perhaps some tick O5SR - is it relayable to say 5 ticks of regen O5SR? so 4000 mana = 8500 mana
4* proc of Blue dragon card = 4*15 sec full regen - I am not a manawhore but I am getting easily 800 MP% raid buffed, this means a proc of the Blue dragon card is worth up to 9600 mana

So unless CoH does not more than 70% of your healing the blue dragon card seems to me more appealing.

would you agree to this?

 

Sarjin

You missed a couple of things, which lessen Blue Dragon effectiveness, as outlined in previous posts:
-Regen with Blue Dragon will never top 100%; since you got 30% Meditation and potentially PMC bonus, that's 70 or 65% increased regen.
-It only increases Spirit based regeneration; your MP5 gear will always tick anyways.
-Since Regen with Blue Dragon active does not go above 100%, any time spent outside 5SR is actually wasted proc time. It's obviously hard to quantify this, but it means effectiveness of the trinket will be below the theoretically calculated.

Applying the first two notions while having to just keep in mind the last one, revisiting your calculations:

-4 procs of Blue Dragon, at 15 seconds each
-70% Spirit based regeneration bonus; I'll assume about 60 Mp5 from enchants, gear and such (I personally go for spirit like most these days obviously, but 'avoiding' MP5 completely seems hardly possible)

Let's subtract the 60 MP5 from the ~800 MP5 you estimated (very rough estimate, and indeed quite easy to reach, but just to illustrate difference in effectiveness based on these issues) gives 740 Spirit based MP5, during procs you get a 70% bonus to Spirit based regen. With 60 seconds of uptime during the fight (assuming not dropping out of 5SR during this), this gives:

0.7*740*12 = 6216 mana, which is about 2/3rds of what you calculated previously.

 

gasilia

oh right, you pointed out an error - i missed indeed to subtract the I5SR regen. Which makes it less strong.


I am trying to find a cap on par exemple how many COH you use, when using Gruuls eye will be more interesting than the blue dragon card.
With you remark indeed the threshold will be lower. But the blue dragon card will give an advantage as long as you are not sure how much you will be able to heal with CoH or not (unknowd or less known encounters)

 

Bendyr

It's impossible to model, but it's also worth noting that if the blue dragon procs while you're under the effect of an Innervate, the effect is wasted as well. If you have a ton of druids, and you get a lot of innervates, it's worth mentioning I suppose.

 

ilkori

It isn't impossible to model, but until we get a simulator like Yo!'s Enhancement Shammy simulator it will be difficult to calculate. (Not impossible, mind you. It's just that the number of cases where certain assumptions are valid/invalid is rather staggering.)

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

 

Flinkindolin

I feel a bit like a boob, but I'm attempting to use the sheet and I've noticed some outrageously high values placed on the t6 pieces because of the 4pc bonus.

I've traced it back to the Calculations sheet spitting out some weird numbers for low rank Gheal casts. The numbers, for whatever reason, are extremely large or negative... which is completely messing with the +1 heal/cast average calculated in Calculations!V20.

Gheal 3 (S6) = 8689
Gheal 2 (S7) = -174
Gheal 1 (S8) = -88.6

I think I've narrowed down my error to user-based.

My Stats, as entered are
Int 549, Spirit 607, Healing 2189, mp5 92

Settings:
OO5SR% 20%; Duration 360; Human No; Buffs Yes; Cons Yes; Mana Pots Yes; Shadowfiend Yes; Shadowpriest Yes; Epic Gems Y; Meta IED; PMC No; Max 7

Tier Sets:
All T5 no, Head Chest Shoulder and Gloves T6 Y

Talents (Down the list)
3/3, 4/5, 5/5, 3/3, 1/2, 1/1, 5/5, 5/5, 3/5, 3/3, 5/5

On the buffs page:
Yes to all Buffs
Food: Healing/Spirit
No Flask
Guardian: Draenic Wis
Battle: Mastery
Weapon Oil: Brilliant Mana

I was operating under the assumption that mp5 was a manual addition of the mp5 value present on gear. Given that the sheet provides an initial value (on download) of 312, my assumption was likely in error.

I've provided all my data entry in hopes of giving you the ability to replicate the error.

Questions

• What is the intended value to be entered as "Mp5"? And can this be clarified on the sheet?
• Why, under any situation, is the Calculations sheet providing a negative number of casts?
• And, finally, why isn't the averaging function in Calculations!V20 referenced to Column T (Casts (Max)) in the first place?

And, on an unrelated note, a toggle for Alchemists Stones that alters the mp5 assigned to Mana Potions would be a great addition!

 

Jayde

Actually, in most cases the sheet is working correctly in that case... Negative values in Calculations!S are infinite recast, which is why the Calculations!T, "Casts (Max)" column exists to cap/adjust.

However, in looking at what you're pointing out it seems that my +1 Heal average cell is referencing Casts instead of Casts (Max), which is the error. I've corrected this for the next release. Thanks for pointing it out.

To answer your questions specifically:
a) Mp5 in the yellow cell on the main sheet should be your unbuffed "mana regenerated every 5 seconds while casting" from your character sheet, will make a note on it in the sheet
b) As above, negative is for infinite sustainability which is why it's capped in the next column
c) That's the error, fixed!

In regard to Alchemist Stones, I could possibly do that... right now, though, I am attributing the Mp5 bonus from mana potions to the Alchemist Stones in the trinket sheet if you have the mana potions setting toggled on. I see what you mean in terms of trying to sort out healing pool issues, though, so I'll take a look at if I can do that easily.

 

Eloc

Thanks for the spreadsheet! It's really usefull, as I am a T4/T5 raiding priest trying to gear up.
Anyways, my question is: could you include an option for no metagem? I am currently using [Whitemend Hood] which does not have a meta slot. I realize it might not affect most people, but a lot of priests make the Whitmend set when they reach 70.

 

ilkori

The meta choice only applies to helms with meta slots (see the helm tab of the spreadsheet). Since Whitemend has no meta slot, it does not get the values added in.

I admit it is a little vague, but it does seem to be working.

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

 

Jayde

I presume he's mostly asking for the purposes of calculation adjustments--currently the selected meta is presumed for sorting out your healing stats in the pool calculations, etc. I'll add a "none" option in the next version.

I should be able to push it out tomorrow. All the Stamina additions to the data are taking a bit of time to implement.

 

Eloc

Alright, thanks. I'll be sure to get the update.

 

ramyalexis

If it isn't hard i'd like to see some more info on equipment pages and more info about gems.
Such as sourse of whee can we find them. (vendor, craft, what lvl)
Equipment(quest, craft, drop) or some kind of such info for more easy sorting.
Thx

Edit:In ring section Aveena's Touch should be [Anveena's Touch] according to wowhead.com

 

RubyTuesday

If I understand how things are entered correctly it seems that the stats for the Lightfathom Scepter are not correct on the Equipment_Mainhand tab. The values should be: Int=20 (not 19), Mp5 = 11 (not 10), and healing = 443 (not 431). Making these changes moved the Lightfathom Scepter from 5th to 2nd with my gear/stats.

 

Skabro

While using your sheet, I noticed an abnormality dealing with int. Maybe I am reading or understanding the values incorrectly. Under the Calculator Tab, Regen Box, Regen per Intellect is 267.641. I noticed this because the values given to gems with +int are astronomically higher than gems with out int.

Stats input: 519 int, 575 spi, 2416 heal, 67 mp5(from gear)
Talents: 3, 3, 5, 3, 0, 1, 5, 5, 5, 3, 5.

Settings:
OO5SR% 30%; Duration 360; Human No; Buffs Yes; Cons Yes; Mana Pots Yes; Shadowfiend Yes; Shadowpriest Yes; Epic Gems Y; Meta IED; PMC No; Max 7

So the numbers I am scared of are gem values: Luminous Noble Topaz(9 healing, 4 int), value 1805.16
For comparison a similar gem's value: Purified Shadow Pearl(9 healing, 4 spirit), value 19.20

How on earth does trading 4 spirit for 4 int result in a value 1785.96 higher? I am not sure what if anything is wrong or how to to fix it.

PS The addition of the crafted +11 heal, 5 spi Shadowsong would be nice.

 

Turgid

I put in the values you listed and I get a regen for intellect of .3511. My guess is that either one of the values outside the yellow fields got changed inadvertently, or something else got munged. Double-check all the fields shaded yellow to see if anything is out of whack. But something has defnitely gone awry. My Noble Topaz is valued around 14, and my PSP around 16.

For the Purified Shadowsong Amethyst, I would just replace the Imperial Tanzanite. I added another row for it, but then you have to modify several of the hidden columns as well to make it realize it's an epic gem and get everything else sorted. I've made many other modifications to the file as well (added a section for choosing my gear from the lists, added some conditional formatting, fixed a few typos in item stats, added the Rod of Blazing Light, etc), so my version might be bit a too far from source for most folks. I'm currently working on adding the Stamina for all items in, since Jayde has been MIA a while.

 

Havoc12

I have been thinking about how to come up with a more accurate connection between haste and +healing/regen. My solution was this formula

H>= m*HPStg*Ft/( k*(mana pool + (mana regen)*Ft)(1-O) ) - A/k

H is +healing
m is mana cost of the spell
Ft is fight time
k is % of +healing converted to actual healing for this particular spell
O is overheal
A is base heal for this spell.

Basically it calculates the minimum +healing and regeneration you need to maintain a particular target HPS (HPStg), with a particular spell, using a predefined amout of your mana pool given certain parameters.

Basically by inputing all these parameters one can determine how much of the current +healing can be converted to haste without losing sustainability, thus coming up with an equivalence on haste that is appropriate to the current gear level.

 

Cyraine

Hi, my previous post was deleted because I worded it completely wrong -_-. What I was trying to ask was, How do i select which gems I want the spreadsheet to consider as "default" for different colored sockets? Or do I set them manually for each item? I went through all the pages, sorry if I missed it.
Thanks for all the hard work!

Edit: Ah allright thanks alot for the info it's alot clearer now ^.^

 

Turgid

The spreadsheet automatically picks the gems based on the value of the socket bonus and the value of each gem, based on the values you have selected for +heal, mp5, haste, and crit. It will compare the value of the best gem overall and the value of the best gems that match the colors for the socket bonus combined with the value of the socket bonus. If the socket bonus + matching gems is higher than ignoring the socket bonus, it will choose the best matching gem.

If it's not picking the gem you think should be the default, it's likely that your stat values don't match your perspective on the relative values of the stats, or maybe there's a gem that's better/worse than you thought it was.

 

funkydrummer

Hi Jayde,

Any updates on the new version? Would like to see it, thanks for the effort on this!