This site makes extensive use of JavaScript.
Please enable JavaScript in your browser.
Classic Theme
Thottbot Theme
[Paladin] Crit vs MP5 for Flash of Light Spamming
Post Reply
Return to board index
Post by
tecnobrat
Here is some theorycraft math for some of you who are comparing MP5 vs Crit and heal based on spamming FoL
This math is based on 300 seconds (5 minutes):
Number of crits = Crit % * ( 300 secs / cast time )
Mana returned from crit = Mana Cost * 0.6
Total mana returned in 5 mins = Number of crits * Mana returned from crit
Spamming Flash of Light
1% crit rate using 1.5 sec Flast of Light costing 180 mana, you get:
1% * (300/1.5) = 2 crits
180 * 0.6 = 108 mana
Total mana returned = 216 mana
In 300 seconds, you have sixty 5 second periods, which equates this out to 3.6 mp5.
Therefore 1% = 3.6 mp5 spamming FoL
Investigating what is the best way to spend item points
Each 1 spell crit rating = 1 item stat point.
Each 1 mp5 = 2.5 item stat points
It takes 22.1 spell crit rating to gain 1% crit or 22.1 item stat points.
It takes 3.6 mp5 to equal 1% crit, or 9 stat points, spamming FoL
Healing is 0.455 stat points per point of healing.
Which means, for every point of crit, you can get the equivelent in mp5, and 28.8 healing.
But, but but ... crit add healing because you crit for 1.5 times your healing!
Okay, lets do the math. Lets say you have 2k healing, and have spent every stat point available into crit, you have 50% holy crit from having a base of 10% crit (from int, etc) and 5% from talents, the remaining you gained from 774 crit rating, and absolutely no mp5.
Lets calculate the base heal
448-502 base FoL
+185 healing from BoL
+105 healing from
Libram of Souls Redeemed
+857 healing from your +healing (2000 * (1.5 sec cast / 3.5 sec cast base))
+ 12% healing from Healing Light talent
= 1786 - 1847 healing
= 1816.5 avg heal
Now in 300 seconds, you will cast 200 FoL, and 50% of them will be crits. So that is 100 crit, and 100 non-crit.
That is 100*1816.5 + 100*(1816.5*1.5) = 454075.5 healing.
Okay, and what about mp5, surely it heals for less
Now, lets assume the same for 2000 healing + 126 mp5 (equiv for 35% crit) + 10.00% base crit + 5% crit from talents + 1009 healing, which is what you saved in stat points from using mp5 instead of crit.
Lets calculate the base heal
448-502 base FoL
+185 healing from BoL
+105 healing from
Libram of Souls Redeemed
+1290 healing from your +healing (3009 * (1.5 sec cast / 3.5 sec cast base))
+ 12% healing from Healing Light talent
= 2271 - 2332 healing
= 2301.5 avg heal
Now over a 5 min fight, you cast 170 non-crit heals and 30 crit heals.
That is 170*2301.5 + 30*(2301.5*1.5) = 474822 healing.
Its actually MORE healing, because you are getting cheaper stat points, and can reallocate those stats as healing instead of crit.
If anyone finds any flaws in my logic / math, let me know.
Post by
tecnobrat
Also worth noting: MP5 does not follow the 5 sec rule. MP5 regens all of the time, even when casting.
And yes, it does depend on your play style. Really the difference is FoL vs HL. Me and another Paladin in my guild heal wildly different. His paladin is built for HL crits with 4/5 tier 5, and almost requires a shadow priest.
However mine is built for longterm healing with FoL, topping off a tank or keeping the raid at 100%, while not needing to use pots or a shadow priest at all.
Post by
profilerv2
Good info :D
One thing I'd like to thrown in there is the fact that MP5/heal is a constant and guaranteed amount of mana and healing, whereas +crit is rolling an RNG for the mana return, which can be good or bad. In some lucky cases, crit may prove to be a greater improvement or significantly worse, while mp5/heal is a constant, non-spikey result.
I'd also note that crit adds a bit more fun to the class, thus its higher point cost. :D
Post by
FlatGreenWind
Very useful, I was looking to see the best stats for gemming my holy pally up and this helped me decide +healing and mp5 is better than anything else. Crit is nice but I heal to keep a tank or anyone else for that matter at 100%, so when they are down 1k i heal for 1k, if it crits then i wasted heals. I could wait longer but with a crushable tank it isnt worth the risk of them being crit'd and dying cuz i didnt heal them for that last 1k.
Post by
183060
This post was from a user who has deleted their account.
Post by
9667
This post was from a user who has deleted their account.
Post by
Macanam
Also remember that once you reach sunwell progression and begin to stack haste, the benefit from mp5 per spell is reduced, whereas crit remains the same
i.e. for every flash of light you cast you gain less of your mp5 because you take less time to cast it, however the crit chance and mana regained from the crit will remain the same.
This means that your, sort of, mp5 from crit scales positively with haste, whereas mp5 scales negatively.
Perhaps not the best way of wording it but im sure at least some of you will understand
Post by
tecnobrat
Macanam, very good point, this does not take into factor haste in any form.
Post by
9667
This post was from a user who has deleted their account.
Post by
52521
This post was from a user who has deleted their account.
Post by
blademeld
Theistica, the reason that the amount of healing done possible was introduced into this mathematics were to counter any points that claimed that +crit will be more beneficial since it will do more healing.
Post by
Bazar
The thing with spell crit and paladins, is that their spell crit gives near exponential returns, not linear like mp5/+healing.
Testing spell crit at 0% base crit like your workings have, is absolute worst case scenario. I'm going to give a more interesting (and more reasonable) scenario, where the paladin has a flat 50% crit rate
given:
FoL = 180 mana
Crits refund 60% mana
player has 50% crit rate
Simulation Duration = 300 seconds
Base workings show that the paladin will have:
200 spell casts, 100 of which crit
Costing 36,000 mana, refunding 10,800 mana, and giving 250 FoL Throughput (50 bonus throughput = 9000 mana 'saved')
Adding 1% spell Crit gives:
200 spell casts, 102 of which crit
Refunding 11016 (216 additional mana), 1 additional FoL (180 additona mana 'saved')
Total mana refunded: 216
Total mana conserved: 396
<Recursive Headache time>That 216 mana isn't written in stone yet. It still hasn't been spent yet, and when it is, its looking at 52% crit rate, meaning that 31% the bonus mana will regenerate itself, giving 31% of 31%, so on till infinity.
it'd be reasonable to assume that fully spent till you go OOM, the 216 mana will actually work out to about 323 mana. An additional 107 more mana, which as your crit grows, your number of refunds grow, as does their finite refund value)
But i'll put that to the side for now, just know that its another point of contention to why crit is better. This hidden value used to be a lot more powerful when illumination was 100%, but the nerf crippled this recursive aspect</Recursive Headache>
Assuming your workings are correct, in that 3.6mp5 is equal to 1% crit, we will compare the mana generated by 3.6mp5 over 300 seconds.
8.84mp/5 mana generated: 530
So in comparison
:
1% crit total mana refunded: 216
1% crit total mana conserved: 396 (excluding recursive refunds)
8.84mp/5 mana generated: 530
Now you can cry foul saying that no paladin gets 50% crit rate, and you'd be right, but my point still remains. Crit feeds into itself, it grows exponentially. Mesuring it at 0% base crit rate is worst case secenario, working at it with 99% crit base is best case scenario.
So yes, the summary is this
Crits give you more healing throughput and although initially weaker then mp5, scale better end-game. Also those that think crit is spiky, their right, but the more crit you have, the more dependable it becomes.
Final judgement i'll leave to the healer, its a matter of preference.
Post by
blademeld
There's a crucial flaw in your logic, I'm going to wait and see how long it takes you to figure it out.
Post by
Bazar
you'll be waiting a long time then
Post by
blademeld
Adding 1% spell Crit gives:
200 spell casts, 102 of which crit
Refunding 11016 (216 additional mana), 1 additional FoL (180 additona mana 'saved')
Total mana refunded: 216
Total mana conserved: 396
This, there is no "additional FoL"
If you're referring to the fact that 2 crits = same amount as 3 non-crits, there are 2 major flaws in this.
Post by
Bazar
I was referring to the fact that 2 FoL crits = 3 normal FoL.
Post by
blademeld
If you're referring to the fact that 2 crits = same amount as 3 non-crits, there are 2 major flaws in this.
I was referring to the fact that 2 FoL crits = 3 normal FoL.
No need to repeat what I said
>.<
Flaw 1:
This is an experiment to calculate 200 FoL's over 300 seconds in an ideal condition, if you're taking out that 1 extra FoL you "gain" through crits than you're breaking the controlled variable by reducing the total number of FoL's cast by 1, resulting in 199 FoL's instead of 200.
Consequentially, that means that your "extra" FoL cannot and will not take into account your critical strike chance, losing out on (51% * 50%) 25.5% less of the total healing of an FoL, consequentially doing that much less healing by not healing for the cost of 180 mana.
Flaw 2:
Ok, say that you heal for that 180 mana, than you're doing 74.5% of an FoL more, however, that is not accounting for the stat point that you have left over from MP5 set, which is ~13 points, equating 18 + spell power or 5.2 MP5, 18 spell power would be inferior in terms of total healing done because +crits will stack with +base +healing already on the paladin, but 18 spell power will give you 375 extra healing done over the 300 seconds, or the 5.2 MP5 will give you 312 mana, which is 205 more than your recursive headache I might add.
Post by
Bazar
A few points.
Firstly it is
not
an 'experiment to calculate 200 FoL's over 300 seconds'. Its simply comparing FoL spamming to MP5
Neither the OP or my post makes such claims or restrictions. It is simply comparing long term mana efficiency. 300 seconds is just an arbitrary value given so a simulation can be given.
Without a time restraint, its reasonable to assume that the paladin will keep casting till his mana expires. In such a situation, then the total mana conserved truly is the total mana saved per point of crit.
Even if we were working within a time constraint of 300 seconds, the additional FoL provided by crit is still there and noteworthy.
Reading your second point, it seems that i misread the itemization when doing my workings
It takes 22.1 spell crit rating to gain 1% crit or 22.1 item stat points.
It takes 3.6 mp5 to equal 1% crit, or 9 stat points, spamming FoL
Which i took to mean 1% crit is as easy to get a 3.6mp5
Using new workings, assuming its along the lines that 1% crit = 8.84 mp5
Which is easy to update the workings.
8.84mp/5 mana generated: 530
So in comparison:
1% crit total mana refunded: 216
1% crit total mana conserved: 396 (excluding recursive refunds)
3.6mp/5 mana generated: 530
mp/5 is now shown to have a much better footing for mana conservation in FoL spamming. (If you take recursive refunds into account at 51%, they'd about tie however)
The thing i'd like to point out however, is that crit will be far more useful when you start using holy light. As the mana returned is a reflection of the cost of the spell.
Post by
blademeld
Even if we were working within a time constraint of 300 seconds, the additional FoL provided by crit is still there and noteworthy.
As I have shown above, the additional FoL does not exist and cannot exist for any reason.
Unless you provide a counterexample or mathematical proof, it will not and cannot exist due to your base 50% critical strike chance.
mp/5 is now shown to have a much better footing for mana conservation in FoL spamming. (If you take recursive refunds into account at 51%, they'd about tie however)
I don't even understand this 31% of 31% of 51% thing you're talking about, but the net difference is 25 mana including your "additional FoL conservation".
The thing i'd like to point out however, is that crit will be far more useful when you start using holy light. As the mana returned is a reflection of the cost of the spell.
Which is the reason we're
strictly
discussing FoL.
Post by
Bazar
Alright, lets try and come to an agreement of minds.
I'm going to quote your old post again, because i think your still working on this assumption somehow
This is an experiment to calculate 200 FoL's over 300 seconds in an ideal condition, if you're taking out that 1 extra FoL you "gain" through crits than you're breaking the controlled variable by reducing the total number of FoL's cast by 1, resulting in 199 FoL's instead of 200.
As i've mentioned before, this is
NOT
an experiment to calculate 200 FoL's.
If your working on that assumption, then your doing something that wasn't requested or stated.
The orginal topic was:
Here is some theorycraft math for some of you who are
comparing MP5 vs Crit
and heal
based on spamming FoL
This math is based on 300 seconds
(5 minutes):
It is
NOT
based on a set number of casts, its
not
even set on a time frame. It is simply comparing mp5 vs crit, with a 300 second window for maths. The 300 seconds makes things simple, rather then working it out till the paladin runs OOM, which requires taking into account total mana,and mana regeneration. Again, and for the last time, there are no restrictions in this theorycraft, other then its restricted to crit, mp5, and fol.
If you still wish to deny the existence of the bonus healing, equal to so many units of FoL . Then we will have an irrevocable difference of opinion.
I don't even understand this 31% of 31% of 51% thing you're talking about, but the net difference is 25 mana including your "additional FoL conservation".
You shouldn't be adding the mana conserved with the mana refunded, as conserved mana already includes what was refunded. I guess i should of been clearer there, either way, the grand total is 396 vs 530, or a difference of 134 mana in mp5's favour
So in comparison:
1% crit total mana refunded: 216
1% crit total mana conserved: 396 (excluding recursive refunds)
8.84mp/5 mana generated: 530
As for the recursive aspect, lets say illumation was 100% again
You had a 10% crit rate
Which ment you would be refunded 10% of what you cast right.
So if you chain-cast till you went OOM, you'd of regained 10% of your mana.
Thus you'd kinda have 110% mana.
But that bonus 10% mana can also crit, 10% of 10% is 1%
So you'd really be working with 111% total mana
That 1% additional mana could crit however, so it'd be 111.1% mana.
And so on.
This example only had 10% crit, but since refunded mana was recursively refunded, it resulted in granting over the 10% bonus.
This effect grows exponentially the higher the crit, as you will crit more often, and get larger recursive refunds from each point in crit.
The nerf to illumination to make it only 60% greatly crippled this aspect of paladin healing. But its there, the maths, although weakened to but a fraction of its original glory, still hides in the background granting an additional cast now and again.
Post Reply
You are not logged in. Please
log in
to post a reply or
register
if you don't already have an account.