I see. That means the optimized ratio reported earlier will only be valid for gemming, and not for gear choices. Blizzard punishes large, single stats with your formula ( or with something close ) on item budgets. So, given a gear set, optimizing mana return means gemming it such that you will have a total Intellect equal to half your total Spirit ( as close as possible ).
Can anyone confirm whether I am properly rounding down the MP5?
-----------------------------------------------------------------------------------------------
I found some issues with your formula's so I rewrote some of the formula's as well as I added some more controls for buffs, more druid support etc. I put an adjustable percentage for talents...as well as Primal Mooncloth set bonus affects the mana regen while casting.
This looks good to me. However, can you change the modifier on MOTW to 18.9 instead (for improved MOTW), or give people the option of selecting MOTW vs. Improved MOTW?
Done. You can now choose between MOTW and Imp MOTW. Also added support for Mages, and relevant Mage/Druid/Priest talents. Added total mana gained from an OO5 innervate. Fixed Human class buff, and added Gnome class buff.
Hello, I'm the author of HealCrafter. I can't seem to get a toon transferred over to the PTR so I was hoping you guys could help validate the 2.4 regen changes I made.
Healcrafter uses the armory and allows you to download your character and equip any gear that is on the armory on to your character to see how much it improves your stats. It does MUCH MUCH more like effective MP5, etc.... Anyhow, I added check boxes to allow you to change change mana regen to the new equation you guys came up with.
It would be awesome if some of you could try it out and give me some feedback on v2.4 accuracy, etc...
Keep in mind that it does not use buffs, so the numbers wont match armory if you logged off with buffs. A WoW Priest Application - Home
Here is an example where two different characters are compared using the new numbers (you can just as easily compare the same character with/without the patch). Using the checkboxes, you can look at the old numbers pretty easily.
Before:
After:
Thanks Ssjsigma! I added that to the gems.xml file and posted the new version 1.1.0.1.
I did a little experiment....
Royal Nightseye vs Purified Shadow Pearl
I purposely broke my Primal Mooncloth set bonus by using this belt. One belt has a Royal Nightseye in both slots and the other belt has Royal Nightseye in one slot and a Purified Shadow Pearl in the other slot. They both match the set bonus, so this is a direct comparison of 2 MP5 vs 4 spirit assuming you have around 513 intellect. The blue, "New Character" is the one with the Purified Shadow Pearl.
Looks like the shadow pearls may be going up in price soon...
Done. You can now choose between MOTW and Imp MOTW. Also added support for Mages, and relevant Mage/Druid/Priest talents. Added total mana gained from an OO5 innervate. Fixed Human class buff, and added Gnome class buff.
Well, your methods certainly produced a very accurate formula! I look forward to the 2.4 release of RatingBuster.
RatingBuster is all updated with the 2.4 spirit regen change and ready for beta testing on test servers(lastest beta here)
It has wow version detection so that it uses 2.3.3 code in 2.3.3 and 2.4.0 code in 2.4.0 and up.
This beta also has a new feature that auto fills empty sockets with gems of your choice(more info here)
RatingBuster is all updated with the 2.4 spirit regen change and ready for beta testing on test servers(lastest beta here)
It has wow version detection so that it uses 2.3.3 code in 2.3.3 and 2.4.0 code in 2.4.0 and up.
This beta also has a new feature that auto fills empty sockets with gems of your choice(more info here)
i love you. it was always annoying to compare items with slots.
Healcrafter uses the armory and allows you to download your character and equip any gear that is on the armory on to your character to see how much it improves your stats.
Kevlar, your application looks very interesting, but it seems it only works for the US-armory and not for European realms? Any possibility of adding this, or am I simply overlooking something?
I see. That means the optimized ratio reported earlier will only be valid for gemming, and not for gear choices. Blizzard punishes large, single stats with your formula ( or with something close ) on item budgets. So, given a gear set, optimizing mana return means gemming it such that you will have a total Intellect equal to half your total Spirit ( as close as possible ).
Forgive the newbie his naivete, but wouldn't changing the formula for the item budget change the location of the peak ratio? Since your analysis was based on the assumption that spi + int = constant, I would hazard a guess that changing that formula would also change the analysis. I did some tentative theorycrafting yesterday and came up with a 3:2 ratio of spirit to intellect for (spi^exp + int^exp)^(1/exp) = constant, where exp is the previously stated value of ln(2)/ln(1.5). It can be shown that for exp = 1, that is, the simplified case of spi + int = constant, the ratio does equate to 2:1. I've uploaded a spreadsheet with my theorycrafting here.
[e]I'm silly and forgot to specify the value of exp that got me the 3:2 ratio.
Done. You can now choose between MOTW and Imp MOTW. Also added support for Mages, and relevant Mage/Druid/Priest talents. Added total mana gained from an OO5 innervate. Fixed Human class buff, and added Gnome class buff.
-----------------------------------------------------------------------------------------------
Havenwood,
I still see problems with your formulas, especially because you are applying the human bonuses ontop of numbers that have already calculated that into the base spirit. For example, you look at the character sheet, you are going to say you have 540 spi unbuffed, well you would only apply kings to that total, they way you have your formula setup, it is double dipping. Take a look at my corrections as well as additions to your spreadsheet.
Kevlar's application...Dont want to risk running it on my home computer. Exe files? eep! Looks sweet though.
Yeah, I had a feeling this would be a problem for a lot of people. I'm not sure how to overcome this fear. Maybe if it gets really popular I can get a reputable site to host it for me.
Originally Posted by Chtoll
Kevlar, your application looks very interesting, but it seems it only works for the US-armory and not for European realms? Any possibility of adding this, or am I simply overlooking something?
Good point. Until now, I've never released HealCrafter outside of my own guild, so this is the first time anyone has tried to use it outside of the US. I'll try to add a checkbox or something for EU Realms. I really appreciate the feedback.
Originally Posted by Ackis
Need to add:
<gem id="32202">
<Name>Lustrous Empyrean Sapphire</Name>
<Color1>blue</Color1>
<Color2>none</Color2>
<MP5>4</MP5>
</gem>
Forgive the newbie his naivete, but wouldn't changing the formula for the item budget change the location of the peak ratio? Since your analysis was based on the assumption that spi + int = constant, I would hazard a guess that changing that formula would also change the analysis. I did some tentative theorycrafting yesterday and came up with a 3:2 ratio of spirit to intellect for (spi^exp + int^exp)^(1/exp) = constant, where exp is the previously stated value of ln(2)/ln(1.5). It can be shown that for exp = 1, that is, the simplified case of spi + int = constant, the ratio does equate to 2:1. I've uploaded a spreadsheet with my theorycrafting here.
[e]I'm silly and forgot to specify the value of exp that got me the 3:2 ratio.
It turns out, if you optimize the mana return formula,
(0.001 + 0.009327 * S * sqrt(I) ) * 5 = mp5
with the constraint,
(S^x + I^x)^(1/x) = Constant
where S is spirit, I is int, and x is an ( unknown ) exponent,
the maximum return occurs when,
S = 2^(1/x) * I.
For x = ln(2)/ln(1.5), this means your total Int should equal 2/3 your total Spirit.
It turns out, if you optimize the mana return formula,
(0.001 + 0.009327 * S * sqrt(I) ) * 5 = mp5
with the constraint,
(S^x + I^x)^(1/x) = Constant
where S is spirit, I is int, and x is an ( unknown ) exponent,
the maximum return occurs when,
S = 2^(1/x) * I.
For x = ln(2)/ln(1.5), this means your total Int should equal 2/3 your total Spirit.
So all that remains, then, is to determine what the proper value of x is in the equation, at which point we can provide an "ideal" gearing ratio for int/spirit. I've heard 1.5 and ln(2)/ln(1.5) bandied about; any ideas how one could verify the itemization exponent experimentally?
-----------------------------------------------------------------------------------------------
Havenwood,
I still see problems with your formulas, especially because you are applying the human bonuses ontop of numbers that have already calculated that into the base spirit. For example, you look at the character sheet, you are going to say you have 540 spi unbuffed, well you would only apply kings to that total, they way you have your formula setup, it is double dipping. Take a look at my corrections as well as additions to your spreadsheet.
Hmm, yeah, since most people are going to be plugging in spirit from the spirit total in character info, your way is better. I was thinking spirit totaled from gear for some reason, which doesn't include the human/gnome bonus yet.
Hmm, yeah, since most people are going to be plugging in spirit from the spirit total in character info, your way is better. I was thinking spirit totaled from gear for some reason, which doesn't include the human/gnome bonus yet.
No biggie, just improving on your original. Feel free to use my changes, collaboration is good. Especially considering I pretty much reworked your original idea.
As per request, updated the MP5 spreadsheet to include mage talents and self-buff. Built this one off of Kinien's revision, so formula fixes are in place and Kinien's Non-Spirit MP5 stuff is added. Old stacking buff problem was entirely my doing, as I broke Bendyr's original work on buffs, which didn't have the problem. Oops!
I know that mages tend to have a lot less total spirit, but when mage armor is taken into account this looks like quite a buff to mage mana regen!
Upon hearing about the v2.4 change to mana regen, I was intensely curious to see what would be the best stat to stack and gem for. It accounts for all 4 healing classes and Mages, all talents that affect mana pool and mana regeneration, and almost all possible raid buffs and consumables.
There are three ways to input your character's statistics:
Pre-buff stats: These stats are modified by talents, racial, and buffs to give the paperdoll SPI, INT, and MP5 values.
Post-buff v2.3 stats: Input your SPI, INT, and MP5 directly from v2.3 live or Armory paperdoll screen. Automatically converts to v2.4 mana regen.
Post-buff v2.4 stats: Input your SPI, INT, and MP5 directly from v2.4 PTR paperdoll.
The spreadsheet calculates the following values:
Total mana
Average MP5 (using a FSR% which you can input)
2, 5, and 10 minute Effective Mana
Estimated endurance while spam casting (based on your maximum mana expenditure)
Estimated increase to mana, MP5, and effective mana for +1 INT, +1 SPI, or +0.4 MP5.
What is effective mana? Effective mana is an estimate for the most mana you can possibly spend over a 2, 5, or 10 minute time period, given all buffs and raid consumables. Effective mana is capped by your maximum mana consumption as it is impossible to spend more mana.
Intellect is definitely a powerful stat with the new v2.4 formula. Unlike SPI or MP5, Intel boosts your spell crit chance. Compared to a point of Spirit, a point of Intel generally gives a much higher 2-minute effective mana and similar 5-minute effective mana, and only really falls behind at the 10-minute mark. Of course, you could argue that for most fights you aren't going to run out of mana before 5 minutes anyways, so Spirit still comes out ahead in the mana department.
And as everyone knows, Spirit is overwhelmingly superior to MP5 in v2.4. The crazy thing is, MP5 hasn't changed at all, it is still strong. It's just that INT and SPI are super strong.
Also, the change makes Flasks of Distilled Wisdom (65 INT) considerably stronger than Flasks of Mighty Restoration (25 MP5) for most healers.
02/12/08, 2:32 PM
Similar Threads
