Our Lvl 90 Talents - Why Incanters Ward is Good!

[Malfecto]

I hate to say this, but there are times, like on Horridon killing adds, that I feel I don't have the time to stop and channel, even for 2.5 seconds. I might give this a try next week and put off trying frost so I can have same spec numbers to compare.

[Deleted]

On Horridon, I force myself to Invoc and the result is here.
A good timing :
- Evoc before each door opening, even if you have 15s remaining.
- Evoc just after clicking the orb, if needed.

With this, the buff is up when you need it and you refresh Invocation on "down time".

This fight is a good example of how scraped this talent are.

[Moras]

First of all, I appreciate all the discussion
Second, I'm not really sure I understand/agree with math.

we have for the second formula :

• RoP : 60s/(60s+1.5s) = 97.56%
• Invoc : 60s/(60s+3s) = 95.24%
• IW : 15s/(25s+1.5s) = 56.6%
• IW (passive) : 100%

It's the % of the time where the buff is active.

Ok, that I understand and I basically use the same formula. (Time of the buff) / (Time of the cycle)

We can now found the effective bonus : up-time * bonus. We can keep % as it don't change anything.

• RoP : 97.56% * 15% = 14.63%
• Invoc : 95.24% * 15% = 14.29%
• IW : 56.6% * 30% = 16.98%
• IW (passive) : 6%

And here is where I'm lost.
The way I count is, I basically assumed I can do certain amount of damage over time. Let say I do 100 dps. So without any buff over 61.5 second I can do 6150 damage.
Now with 15% buff, I will do 115 damage every second. So over 60 seconds (length of the buff) I do 6900 damage. And over 1.5 s (RoP casting time) I do 0 damage. Overall I did 6900 damage over 61.5 seconds which gives me 112.1951 dps. Now damage increase is 12.1951% for me here.

Now what you did here is:
You count uptime (which is +15% damage over 97.56% time), but the rest of the time (2.44%) you treat it like it's 100 dps. But you are not doing any DPS during this 2.44% (1.5s) because you are casting RoP. Which deals 0 damage.

That's why my formulas differ. Now question again - do you think, after this explanation, it's accurate or my logic is still flawed?

[oldnmean]

I think the issue is because Nathiel is multiplying % by percentage...should be % by damage multiplier. As example for ROP...

.9756*1.15=1.1219 which agrees with your 100dps example

[Polarthief]

The problem is, this doesn't take into account for Fire's scaling with Damage Done/Damage Taken buffs/debuffs. Fire scales exponentially with these increases because of how Ignite and Combustion work.

[Deleted]

@ Moras

When you don´t do any damage during cast (refresh) you doing it wrong.

[Moras]

When you don´t do any damage during cast (refresh) you doing it wrong.

Ok, you are right. I actually do some damage while refreshing - about 14% of my damage comes from Bombs, abour 15% of my damage is elemental. So indeed, damage buff is slightly bigger.

So for my previous calculations it will be (for Frost Mage) RoP:
6900 damage over 60 seconds + 30%*1.5s*100dp/s = 6945. 6945/112.92. So it's 12.92%.
I guess for Fire Mage this number might be slightly bigger (bomb+DoTs), for Arcane probably lower (only bomb DPS counts).

[Deleted]

Ok, you are right. I actually do some damage while refreshing - about 14% of my damage comes from Bombs, abour 15% of my damage is elemental. So indeed, damage buff is slightly bigger.

So for my previous calculations it will be (for Frost Mage) RoP:
6900 damage over 60 seconds + 30%*1.5s*100dp/s = 6945. 6945/112.92. So it's 12.92%.
I guess for Fire Mage this number might be slightly bigger (bomb+DoTs), for Arcane probably lower (only bomb DPS counts).

Do not forget you don´t lose any damage from Mirrors, Frozen Orb, FoF IL, BF FFB. As already stated its more like:

(115% x 60) + (83% x 2.5) / 62.5 = 113.72%

[oldnmean]

FoF IL and BF FFB would not count toward damage in the buff refresh window. They are cast using GCDs during the 60 sec of buff.

Anything DoT like MAY end up contrbuting damage during the buff refresh window. Some of them depend on lining up well with refreshing the buff (like Mirrors, Frozen Orb). 83% feels like a really high number especially for Frost/Arcane but I would have to analyze some logs to defnitely say yes or no.

[Moras]

Lining up Mirrors during refresh in a heat of the fight, I call that challenging at least, would love to see your logs Pete showing you doing it. Timing just isn't right if you don't want to waste DPS by holding off mirror image when you have precast Invocation at the fight's start.
Casting Frozen Orb and then channeling Invocation - depending on number of enemies you actually risk loosing FoF, I wouldn't do that unless I know i'm very far from enemy. And also in this case timers don't line up correctly.

Anyway, is there something you want to prove with this speculation? I admit I wasn't very accurate, but my calculations are still very close to actual numbers.

In summary difference between RoP and Invocation is very small and is more of a matter of personal preference. IW is definitely worth it if you can time it right during encounters. But saying that IW passive is better then RoP or Inv, even if they are used only with moderate skill, is just plain false and incorrect. You actually need to use IW at least once a minute to beat well used RoP/Inv. And IW in my opinion might promote bad gameplay if someone intentionally gets damaged because he wants to maximize DPS output.

[Deleted]

This is why I have only calculate the uptime of the buff. The importance here is to compare how many time the buff is up. The longer, the better.

In my calculation, I have say that using the buff as 15% or 1.15 dont have change a things.
All of this was to prove that it need more than 30s without the 15% buff to equal IW's passive.

After, it easy if you have uptime. With 95% at 15% damage, so you did 115 for 95% of your time and 100 for the 5% remaining. But in reality, you will Evocating, refreshing IW/Rune in those time. So the dps will only be rolling DoT, Water Elem, Image, etc.

So for Invocation/Rune, the down-time is negligible in comparison to the up-time.
But for Incanter's Ward, the down-time represent 45% of the DPS time : it have to be taken into account.
So you have a global formula egual to this x*1.3*25/26.5+x*1*10/26.5+y*1*1.5/26.5 with :
x = normal damage
y = rolling damage (dot, etc) while you are under the GCD (negligible)

Conclusion : what ever the bomb you choose, use CD/potion/trinket when the buff is up.

[Akraen]

I don't know IW could possibly be calculated with any semblance of accuracy. +30% damage when Wushoolay's is gaining stacks of intellect is going to add an excessive amount of RNG swing to any models. It requires math beyond my education to come up with a true added DPS number of IW.

On Durumu last night with Wushoolay's I was able to proc IW off a lingering pool. I ended up being #1 in damage around 120k that attempt. The next attempt I couldn't get anything to line up with IW right and I was at 72k dps.

Maybe some math boffins in this thread can help explain a way to model the consistency factor of RPPM trinkets layering with the constant uptime of Invoc?

[Soulstrike]

cheers zomgdps your wonderful work is truly amazing towards mage community

yea i was wondering if i should use IW(not that im lazy >_> <_<) just seems more attractive for alot of fights like last 4 bosses(constant raid damage pretty much) more before that but cbf to name them all lol

[LazyWithMath]

I did a spreadsheet and calculation comparing IW and Invocation and agree with zomgDps if you don't account for haste. With haste, I have the equation for the amount of downtime when IW is better than Invocation:

downtime >= (1.15 * (1 + Haste%/100) - 1.06)/1.06 * 60

For no haste, if I spent 5.094 seconds or more over the course of 65.094 seconds not doing damage when I could, IW is better than Invocation.

For 10% haste, I get 11.604 seconds and for 12.5%, I get 13.231. Also because the Evocation is faster, the amount of doing other things increases.

I did a spreadsheet writing out a fireball cast with a 25% crit rate (and Evocation at every 60 seconds for Invocation) for 300 seconds that compares the damage between them and it is close to the equations. The output of Invocaton is slightly less than the theoretical value because of clipping.

I don't account for using IW which complicates the math and the spreadsheet. I might try it later.

I might have made a mistake and this is my first post so I wouldn't trust my number. Please do a spreadsheet yourself. Since I am around 10% haste, I will use Invocation unless I have downtimes greater than 10.56 seconds beyond doing evocation (2.667 seconds for Evocation).

I agree zomgDPS is doing a great job for the mage community and completely predicted the issue with 5.2. I wish Blizzard would have listened to him instead of some MVP on their forums.

[DeLos]

OP, You have some of the math wrong. And a couple bad assumptions.

Going to use Invocation for math. RoP is basically the same, with a VERY slight difference.

Yes there is a point where IW is better then Invocation, but a lot lower then you propose. For the time being, lets deal with the passive. The active changes things up a bit and we can deal with that next. Lets do the equations:

The basic equation:
(DPS) * (Time casting) * (Modifier) = (Damage Done)

We will call X the time in seconds that you get to cast. Now with invocation you will have to spend 3 seconds of that time casting evocate. Haste will reduce that cast and improve invocations performance, so we will do this with zero haste and give IW its best showing. So, lets put the equations down.

Invocations formula: (DPS) * (X - 3) * 1.15 = (Damage Done)
IW Formula: (DPS) * X * 1.06 = (Damage Done)

So, now we set the equations equal to each other to find the break point where IW equals invocation.

(DPS) * (X - 3) * 1.15 = (DPS) * X * 1.06

We work the equation a little: (DPS) cancels on both sides and then we do the math...showing work:

1.15X - 3.45 = 1.06 X

Which becomes:

.09X = 3.45

So, the time spent casting (X) is equal to 31.9 seconds. Any value of X higher then 31.9 makes invocation better. Lets take that out of 63 second period (1 min duration + 3 sec cast of invocation - the 100% up time of invocation). This means that you have to have an up time on the target of less then 38.3 seconds out of 63 seconds, or ~61%, in order for IW passive to win verses invocation.

Using the active on the other hand means 2 GCDs, or 3 seconds every 60 seconds, cast time spent to maintain a 15% buff. Invocation means 3 seconds every 63 seconds spent casting to maintain a 15% buff. The passive goes away when you are using the active. So, it works out to about the same if you can consistently trigger the effect every CD. That is a big 'if' to trust your damage to. Theoretically if you had a fight with the right timing to use the triggered effect at the right moment and use the downtime when you couldn't cast anyway then IW may be able to win. But, that will be somewhat rare.

Conclusion: Invocation wins every reasonable case.

[i]Edited to fix my invocation cast time. Sorry, after all that, I forgot to take haste out of it......doh. The basic principle is correct though)

[LazyWithMath]

My equations mistake is that it is the ratio of the two and not the time.

I have evocation without haste as a 3 second cast time instead of 2.5. With this value, I get 38.3 seconds and the uptime on the target changes to 38.3 seconds out of 63 yielding 60.8%. With haste, it will move the number toward the 51%.

Did I missing sometime that gives a 2.5 second cast to evocation?

I am sticking with Invocation.

[DeLos]

My equations mistake is that it is the ratio of the two and not the time.

I have evocation without haste as a 3 second cast time instead of 2.5. With this value, I get 38.3 seconds and the uptime on the target changes to 38.3 seconds out of 63 yielding 60.8%. With haste, it will move the number toward the 51%.

Did I missing sometime that gives a 2.5 second cast to evocation?

I am sticking with Invocation.

I fixed it. Sorry. But the point is, you have to have less 61% or less time on the target for IW to be the better choice. Some fights are a pain, but I don't know one that has 40% standing around time. So, yes, stick with invocation.

[Agzarah]

Using the active on the other hand means 2 GCDs, or 3 seconds every 60 seconds, cast time spent to maintain a 15% buff. Invocation means 3 seconds every 63 seconds spent casting to maintain a 15% buff. The passive goes away when you are using the active. So, it works out to about the same if you can consistently trigger the effect every CD. That is a big 'if' to trust your damage to. Theoretically if you had a fight with the right timing to use the triggered effect at the right moment and use the downtime when you couldn't cast anyway then IW may be able to win. But, that will be somewhat rare.

Conclusion: Invocation wins every reasonable case.

[i]Edited to fix my invocation cast time. Sorry, after all that, I forgot to take haste out of it......doh. The basic principle is correct though)

my abilities in math aren't great, but I think you are missing something with the active.
correct me if I've missed something, but it looks to me like you have assumed 2 casts of IW gives a full 60sec 'uptime' when its actually only 50s (15 with the buff, 10without x2) so in your reasoning you still need to include the extra 10seconds a minute of 6% passive.

I realise those few seconds aren't much, but doesn't it mean that IW used to its max, every 30s is actually 16%?
[15*1.3]+[10*1.0]+[5*1.06]/30 = 1.16

[schwarzkopf]

I agree zomgDPS is doing a great job for the mage community and completely predicted the issue with 5.2. I wish Blizzard would have listened to him instead of some MVP on their forums.

zomgDPS - is like the broccoli and carrot in your meal.
MVPS - are like the ice cream at the end of the meal.

Blizzard prefers the sweet sounding fatty icecream to the sometimes salty and healthy vegies. It is a problem with much of the world, misinformation spoken sweetly gets through more often that the hard salty truth.

[Deleted]

zomgDPS - is like the broccoli and carrot in your meal.
MVPS - are like the ice cream at the end of the meal.

Blizzard prefers the sweet sounding fatty icecream to the sometimes salty and healthy vegies. It is a problem with much of the world, misinformation spoken sweetly gets through more often that the hard salty truth.

Disagree, i find Lhivera to be a very competent theorycrafter, posting in a forum full of poison. As soon as he dares suggesting something that is not a pure buff to mages the hatred is unleashed upon him by the angry nerds.