I agree that if you want to consider 40/0/21 you need to consider the DPS of the frost mage buffing him too, as well as the fact fire mages buff eachother via less scorching and more fireballing (assuming they actually sync their scorches - does anyone actually do that with success btw?).
Remember a boss dies when the RAID deals enough damage to him, so the change to total raid dps is what should be considered when choosing a spec/spec setup.
If you're damage capped the only way to increase your damage is to go for spell haste
Actually there are TWO ways to increase your damage.
Etherealz is 100% correct. Critical Strike Rating definitely allows you to go over the AoE cap. It's actually the entire reason Deeper Arcane Builds are the best for AEing since their Critical Strikes with Arcane Explosion hit harder then any other build.
Spell Haste is the second way to break the AoE cap and that method only works with Flamestrike and Blizzard since Haste will have no effect on the GCD meaning it will do nothing for Arcane Explosion.
That's why a combination of Improved Flamestrike and Icy Veins would actually be capable of putting up numbers similar to those attained by a Deep Arcane build with Arcane Explosion. Since Flamestrike can go over the cap through both haste and crit.
Only real issue is the truly pathetic diameter of Flamestrike which greatly limits it's practical applications. Why they limit Blizzard to 8 yards, Flamestrike to 5 yards, Arcane Explosion to 10 yards and give Seed of Corruption 15 yards is beyond me. You would think standardizing all of them at 10 yards would make a lot more sense as well as being a much more balanced mechanic.
I'm still planning on my AE rotation of rank 1 Blizzard (1-2 seconds - just to get them all snared) -> Flamestrike (with it's crits activating Ashtongue Trinket) -> pop Icy Veins and max rank Blizzard -> Flamestrike -> Blizzard. Planning on wearing the Sextant and my Tier 4 helm (which still has the Mystic Skyfire in place) along with the Ashtongue and all the spell haste I own. No AE spell in the game will do more dps then Blizzard with Icy Veins, Ashtongue and a Focus proc from the Mystic Skyfire when you are capped. Even without the Focus proc it should still be very competitive with all other forms of AE.
The full snare of Blizzard (my build is 3/3 permafrost and 3/3 Improved Blizzard - well, it will be when 2.3.2 goes live) also lets Warlocks start seeding earlier meaning everything dies faster and we get purples even quicker. Only real issue is the mana consumption but should be acceptable given the new mana gems.
I agree that if you want to consider 40/0/21 you need to consider the DPS of the frost mage buffing him too, as well as the fact fire mages buff eachother via less scorching and more fireballing (assuming they actually sync their scorches - does anyone actually do that with success btw?).
Remember a boss dies when the RAID deals enough damage to him, so the change to total raid dps is what should be considered when choosing a spec/spec setup.
And there are mages out there that are and always will be pure frost, and if you have one which spec is best for dps
This is a list of choices you can make to maintain your mana.
It's ordered by efficiency, by the amount of mp5 you gain per point of sacrificed damage.
1) 5.27 mp5/dmg - Chaining Mana Potions over Destruction Potions gives you +99.5 mp5 at the cost of -19 damage.
2) 3.70 mp5/dmg - Chaining Mana Gems over Flame Caps gives you +100 mp5 at the cost of -27 damage
3) 2.86 mp5/dmg - Evocation gives you +140 mp5 at the cost of -49 damage.
4) 2.03 mp5/dmg - Arcane Concentration instead of Icy Veins yields +79mp5 at the cost -39 damage.
5) 1.80 mp5/dmg - Arcane Meditation instead of Elemental Precision yields +54 mp5 at the cost of -30 damage.
6) 1.40 mp5/dmg - Serpent-Coil Braid gives you +28 mp5 at the cost of about -20 damage.
7) 1.35 mp5/dmg - Mage Armour gives you +69 mp5 at the cost of -51 damage.
8) 1.36 mp5/dmg - Talented Blessing of Wisdom over Blessing of Kings gives you +25mp5 at the cost of -19 damage.
9) 0.89 mp5/dmg - Untalented Blessing of Wisdom over Blessing of Kings gives you +17mp5 at the cost of -19 damage.
10) 0.63 mp5/dmg - The Alchemist's Stone gives you +47 mp5 at the cost of -75 damage.
Since the PTR characters accidentally had the new spell ranks, I checked the numbers on them, to see how good or bad downranking is, or how much the new ranks are worth it.
Arcane Missiles
The new rank equals +56 dmg, -43 mp5.
Downranking once equals -56 dmg, +53 mp5, twice equals -81 dmg, +81 mp5.
Frostbolt
The new rank equals +34 dmg, -22 mp5.
Downranking once equals -70 dmg, +44 mp5, twice equals +83 dmg, -59 mp5.
Fireball
The new rank equals +70 dmg, -56 mp5.
Downranking once equals -38 dmg, +21 mp5, twice equals +73 dmg, -42 mp5.
So, downranking for frost/fire specs is even worse than using an Alchemist's Stone, and downranking AM is just a tad better.
So, don't do it unless it's the last opportunity you have.
And there are mages out there that are and always will be pure frost, and if you have one which spec is best for dps
That's fine and I really have no issues with the Frost spec whatsoever (take a look at my Armory and you'll see why). The issue is that you are stating that 40/0/21 spec is very competitive but you are not making it clear that it is only when they are receiving WC from another mage. A deep frost mage in isolation will outdps a 40/0/21 mage. A 40/0/21 mage can outdps the Deep Frost mage and possibly a few other mage specs when they gain the benefit of WC from a Deep Frost mage. Fine thats all well and good but pair that 40/0/21 mage with a Fire mage and where are they on the charts.
Argue for the 40/0/21 spec all you want but without taking into account where it is getting the WC buff is not doing anyone in this thread any real service. Give us numbers showing that your 1 Deep Frost mage (for WC) when combined with 40/0/21 mages will as a group outdps similarily equipped Fire mages and you will really have shown everyone something and will probably cause many people to rethink their current prejudices against Frost.
Originally Posted by CHeeSY-CrAfT
Fixing your mage friend is the best option clearly!
Unless you have something real to contribute that statement was nothing but a lot of hot air with no substance. I understand that you are trying to be funny but since the jury is still deliberating on this topic and no judgement is in sight as far as I can tell.
Hop on over to the Mage forums on the WoW general site and take a look at Faxmonkey's thread on Frost vs Fire. It might open your eyes a little bit and help you to keep in mind that in a thread on Theorycrafting, preconceived notions really have no place.
The problem with that thread on wowgen forums is that they worked some theorycraft, but oddly enough none of them are willing to point out the absurd amount of innacuracies they have when they turn up their charts.
Giving a half-truth is not much different than straight up lying. All the TC on that thread does not takes into account a lot of things such as cooldown stacking, bloodlust, trinkets, 4pct6, lag, metagem, pushbacks, (the possible but rare WE dying) and consumables*. Believe it or not, all their charts are graphed at 1200 spell damage. That's great for unbuffed numbers, but then they go and try and derive a conclusion from partial data that they claim to be, in fact, the full deal, which really it isn't.
As much as you point out that people have prejudice in that thread, I find it fairly funny that nobody seems to have noticed the very huge gaps in the TC that I just pointed out above. Yes, they hugely affect numbers. I personally elected to not post into that thread simply because the TC'ers have a pre-established opinion, and as such, it is an uphill battle to try and prove them wrong. Honestly, I don't really care anyway. I know the numbers first-hand, and I know they have a very distorted view of the reality. Yes, they have the general idea correct, that deep-frost gets a huge dps spike with WE out, but other than that its fairly deceptively lacking (their TC, not the deep frost dps).
As for me, I am impartial to whatever spec, but it does infuriate me that people spew that kind of drivel. I spec whichever gives the best DPS. I was the first in line to try AM spam the day 2.2 came out, even though the numbers weren't clearly in its favor. I have worked out the TC for many months now, and the maths points towards firespec, plain and simple, under any kind of high-end realistic scenario.
Last edited by manly : 12/27/07 at 6:53 PM.
<Eej> YOU"RE GONNA PULL
<Eej> IF YOU SQUEEZE OFF ANOTHER ARCANE BLAST
<Spectear> You've obviously never played with Manly.
<Spectear> That's hardly a reason to stop DPS. Very Manly Staff
Manly, I completely agree with you that Fire spec has a bit of an edge. My point was that nothing is really that cut and dry and was more about the sarcastic nature of the post I was responding to.
That thread is a bit lacking in polish and could use a lot more math but it does present an interesting concept and one that surely has some validity.
When Vontre theorycrafted for 2.3 before the MSD nerf, he was showing Deep Frost and Deep Fire with very similar numbers. Well Deep Frost just got a pretty substantial buff, not only with Icy Veins but more with Cold Snap and that should be taken into consideration.
I have no numbers to back it up but I would bet that Deep Frost will beat Deep Fire on certain encounters due to those changes. The real difference is that Frost is much more spikey and a heck of a lot more dependent on it's cooldowns then Fire and when you combine that with the lack of push-back prevention you get a situation where Deep Frost can be better but rarely will be. Fire is reliable and Frost isn't but both are definitely viable for raiding.
Well, I've never debated that frost was viable for raiding. What irks me is the innaccuracies in the given TC. They've got the big picture correct, that frost with WE out is a huge burst, but other than that their numbers are highly skewed because there are severals things that were left off. 1200 dmg in a raid scenario? I'm sorry, but wth. We assume full WC up with COE up, but no luck on buffs? I've made a rather complete list in my previous list just pointing out the obvious innaccuracies, and they really affect the curve a whole deal.
Personally, I prefer to call it theorybending more than theorycrafting. Omitting that many details is borderline inacceptable.
<Eej> YOU"RE GONNA PULL
<Eej> IF YOU SQUEEZE OFF ANOTHER ARCANE BLAST
<Spectear> You've obviously never played with Manly.
<Spectear> That's hardly a reason to stop DPS. Very Manly Staff
The numbers on that thread are not what you'd call perfected, nor are they presented for a t6 mage buffed - however if a thread like that makes some mages think about their class in more than just a 222222223222222223222222223 kind of manner - then I think it served it's purpose.
Edit: And there are also a LOT of stubborn people when it comes to that particular argument. There's still those that will swear up and down arcane > fire always. Hopefully some of the frost die hards reading that thread will realize that fire is going to out perform frost at the very least a majority of the time - thusly it is better.
It mostly depends on how much you want to tunnel down into the theorycrafting. For a quick overview I did for myself a while back (comparing just fireball vs frostbolt/waterbolt), Water Elemental with an uptime of around 50% was close to fire damage. I gave up on the theorycraft at that point because that timeframe didn't work for me and/or I had pushback/AE concerns.
I'll second that 1200 dmg being completely unrealistic - I remember I had exactly 1400 dmg when we were learning lurker... aka 1/4 TK 0/6 SSC. So assuming 1200 dmg in a raid would be way off. Granted from where I was to actually starting hyjal/bt you wouldn't get a lot more since most t5 isn't really more dps than the crafted gear. While I usually say that a small change of damage isn't really going to change which stats/talents are better, 200 dmg is already quite a difference. Same goes for buff stacking, when you evalulate just your use trinket multiplying combustion it's not really that much, but when you put it on top of bloodlust, icy veins and flame cap it starts getting significant.
Of course at the end what really matters is what spec does more dmg in a realistic scenario, and I think vontre's spreadsheet shows pretty well that fire wins hands down for overall damage done in a fight, and the "burst" difference for the few fights where it matters is not nescessarily favoring frost or arcane/frost.
I finally took a look at how Vontre calculates his uptime and to my understanding the numbers he gets are overestimates.
Let d be duration of the effect, c cooldown and t length of fight. Let's analyze worst and best case scenario.
Under worst case the fight will end right before cooldown is ready again. This means that the trinket is activated n*d out of n*c, where n is number of activations. This means in worst case (n*d)/(n*c) = d/c uptime as is generally known.
Under best case the fight will end right after we used the trinket for the last time, so if we used trinket n+1 times it was activated for (n+1)*d out of n*c+d. Let's express this in terms of t.
n*c + d = t
n = (t-d)/c
uptime = (n+1)*d/t = ((t-d)/c + 1)*d/t = (t/c - d/c + 1)*d/t = d/c - d/c * d/t + d/t
Now if you look at Vontre's spreadsheet for example at Icon you'll see the formula used is d/c + d/t (Calculations!F92). From the above you can see that uptime is overestimated as it's higher than it can be in best case scenario.
Let's take an example of a 5 minute fight. Icon is 20 sec with 2 min cooldown. This means that true uptime is somewhere between 16.7% and 22.2% (d/c = 0.167, d/t = 0.067) while the spreadsheet would estimate it as 23.3% (actually the spreadsheet has wrong value for cooldown in the first place so that is off again, same for Hex).
So what is the correct estimate of uptime?
This depends on how exactly you model the fight, whether you take dps uptime into account and what kind of fight length distribution you assume to have. If you're ok with an approximations then you can express best case uptime as d/c + d/t * (1 - d/c). Then "average" uptime would be d/c + 0.5*d/t * (1 - d/c). Note that this estimate is very much arbitrary and is not much more than it is somewhere between best and worst case. That is as much as you can expect to do without settling on a specific fight length distribution.
For any 20 second duration trinket, over the 2 minute cd, you get a 16.7% uptime on the trinket.
Now Supremus/Akama aside, you wont use your trinket first till about 20-30 seconds into the fight, which could possibly lead to aggro pulling, so you cannot include that time into your trinket uptime percentage. Taking your 5 minute fight, you're now at a 4:30-4:40 duration fight, which means only the last 10-20 seconds of the fight are wasted with the trinket uptime. You now get two durations of a 16.7% uptime(20 seconds over 2 min cd), but a third duration of 75-82%(20 seconds over 30-40 second duration on the 2 min cd) uptime. With 3 activations, 60 second duration, and a 270-280 second fight time, thats a 22% uptime.
The main thing to look at is the beginning of the fight. If you blow cooldowns right at the start, you'll pull aggro and die, where as if you wait for established threat, 20-30 seconds into the fight, then blow you cooldowns, you get a much higher uptime. While this is risky for shorter and shorter fights as your raids dps increases, you have to model uptimes differently then too.
To be honest, usually I scorch until 5 are up, then 2-3 fireballs, then I am free to pop all cooldowns (besides lust). I've never really had aggro issues. I can see this changing with IV, but 20-30s sounds really really overstated to me,
<Eej> YOU"RE GONNA PULL
<Eej> IF YOU SQUEEZE OFF ANOTHER ARCANE BLAST
<Spectear> You've obviously never played with Manly.
<Spectear> That's hardly a reason to stop DPS. Very Manly Staff
To be honest, usually I scorch until 5 are up, then 2-3 fireballs, then I am free to pop all cooldowns (besides lust). I've never really had aggro issues. I can see this changing with IV, but 20-30s sounds really really overstated to me,
Can lower it to 10-20 seconds if you want. And then it drops down to 20-21% instead of 22%.
Personally I scorch till 5, Fireball for a bit looking at Omen to see how far I can push myself. If my first 3 fireballs crit, I cannot pop consumables as early, due to threat limitations. Different tanks and different days produce different numbers I guess.
Meh. We get 2 misdirects on pure tanking fights immediatly, so I pop everything after 5 scorches and hope the fight lasts long enough to squeeze in that last cooldown. I can't count the number of times the boss has died as they.re coming up. It would be nice to be able to, without hours of theorycrafting, see how effective stacking your on use trinkets are at each step in the game. 10 seconds in, 70 seconds in, 130 seconds in, 190 seconds in, and so on to say 10 minutes or so. It's really difficult to compare 3 different types of trinkets (on proc, on use, static buff) for a 3 minute fight. I really feel the current spreadsheets fail at this. It's just really complex and situational.
You'll be sad to learn that cooldown stacking has been supported for many months now by vontre' spreadsheet. However, I do readily admit it doesn't allows to set on your own when you want cooldowns popped. But the TC does takes all of that into account.
<Eej> YOU"RE GONNA PULL
<Eej> IF YOU SQUEEZE OFF ANOTHER ARCANE BLAST
<Spectear> You've obviously never played with Manly.
<Spectear> That's hardly a reason to stop DPS. Very Manly Staff
I'll second that 1200 dmg being completely unrealistic - I remember I had exactly 1400 dmg when we were learning lurker... aka 1/4 TK 0/6 SSC.
I am seriously confused here. The mage with the highest fire damage in my guild, and also on the server according to Gankbang.com :: Armory - has only 1150 damage with 5/6 - 2/4. I am rated number 10 on that list with about 1050.
How the heck could you possibly have 1400 damage (to fire) at beginning of Tier 5 content? I would have read 1200 damage as being unrealistically high.
I am seriously confused here. The mage with the highest fire damage in my guild, and also on the server according to Gankbang.com :: Armory - has only 1150 damage with 5/6 - 2/4. I am rated number 10 on that list with about 1050.
How the heck could you possibly have 1400 damage (to fire) at beginning of Tier 5 content? I would have read 1200 damage as being unrealistically high.
The point is that you have to include raid buffs, and if you make your chart for 1200 damage, then you are not including raid buffs.
<Vontre> I removed the cooldown on evo
<sancus> and what happened?
<Vontre> DPS went down rofl
I am seriously confused here. The mage with the highest fire damage in my guild, and also on the server according to Gankbang.com :: Armory - has only 1150 damage with 5/6 - 2/4. I am rated number 10 on that list with about 1050.
How the heck could you possibly have 1400 damage (to fire) at beginning of Tier 5 content? I would have read 1200 damage as being unrealistically high.
Including having a shaman in my group and IMPDS - I'm over 1600 buffed (using darkmoon card)
Also considering the t6 bonus as manly pointed out earlier is another hit to frost scaling.
You'll be sad to learn that cooldown stacking has been supported for many months now by vontre' spreadsheet. However, I do readily admit it doesn't allows to set on your own when you want cooldowns popped. But the TC does takes all of that into account.
I don't think this is correct. As far as I know the latest available version is 2.7.3.1 and I've digged quite deep into its workings. I'd be suprised I missed something like this. Can you point where specifically in the spreadsheet this is handled or which specific cooldowns you believe are handled for stacking? Maybe you're thinking about his simulator but that's not available for public as far as I know.
Including having a shaman in my group and IMPDS - I'm over 1600 buffed (using darkmoon card)
Also considering the t6 bonus as manly pointed out earlier is another hit to frost scaling.
That post seems to imply that Frost is dead, by comparison to Fire, for raiding.
Is it really though?
Sure the 4T6 bonus doesn't affect the Water Elemental but don't forget that the Elemental gets a lot more out of all the shaman raid buffs. Wrath of Air gives you 101 additional damage, it gives the Elemental 101 + 37.88 = 138.88 additional damage. Bloodlust is also amazing for a Frost mage, if timed right, since the Elemental gets it too.
There was this post by Vontre back before Icy Veins and the Cold Snap changes were announced. (10/12/07)
Originally Posted by Vontre
THEORYCRAFT RESULTS WHOA
Hey, the sky isn't falling. I went ahead and assumed a 10% proc rate for our new MSD friend, and 0 ms casting latency thanks to blizzard's fix. Ok the scaling might help that out a lot. Anyway, I went ahead and did some comparisons using my new module. I'll be releasing a spreadsheet update too, but the spreadsheet is still limited to the "Patchwerk scenario", for the most part. Programatic approach tells the whole story. I'll have these calculations visible before too long.
Ok, how did I build this? Basically I included everything, shaman totems, bloodlust, etc. I did NOT include malediction. I DID include Molten Fury, and furthermor assumed that Bloodlust is popped during Molten Fury. I also assumed bloodlust would stack against arcane power and water elemental. Realistic, I think. For arcane damage, I added 72 spell damage, to represent the amount of spell damage you might gain for dropping hit gear. This number was derived by theoretically dropping 8 hit gems (64 hit rating) and replacing them with 8 spell damage gems (at 9 damage each). Guess this would be even better in BT gear? Hmm.
Anyway, I finally got around to making a complete calculation, which means everything included. Not just the extra buffs Bloodlust and Molten Fury, but also options for interruption, pushback, and target switching.
So the theorycraft simulations are not just stand still and spam fireball for 6 minutes, they account for scorching, buff ramping, timer effects, etc. I assumed there was a second fire mage to help with scorching, fairly common. This doesn't mean "scorch bitch" (a concept which is inherently retarded), this means help, as in the debuff goes up twice as fast. Awesome. No such luck for our poor frost simulation, though. He has to stack Winter's Chill on his own.
So how does our "Patchwerk scenario" look? Meaning standing still and doing your spells and scorches for 5 minutes. Pretty much like this:
Yeah, that's a big WTF. Apparently if the water elemental actually stays alive, and you never have to move or deal with pushback, frost fucking owns. Actually isn't a huge shock, considering how often the water elemental actually stays alive... so let's assume he dies about half the time, for starters. But we want to make this look even more realistic. So let's add some more encounter effects.
The first is simple enough: interruption. Not like a counterspell, but something that forces you to cancel a cast and move. Al'ar makes a flame patch and you need to go right fucking now. Sucks that you had 2 seconds on that frostbolt, wasted time there. Or in the case of AM, wasted mana, since you fire bolts at every second. Ok, cool, let's say there's an interruption every... 45 seconds. So to account for interrupt I added 3 seconds of downtime for moving, and half of the spell's casting time to represent the average lost casting time. This hurts fireball the most, because it has the most potential to lose casting time.
Ok, so it's counting the 3 seconds downtime as dps time, so we're looking at a reduction around the board. Fire gets hit the most, arcane gets hit the least.
Now let's throw in pushback. Say you get hit maybe 5 times for the ecounter... or once every 60 seconds. Pushback reduces your channel bar by 1 second, so it has a slightly lesser effect on fast spells like scorch. Frost has no pushback resistance at all, and arcane is flat out unaffected. This is the result:
Ok, last thing is target switching. Say it's add time, you have to kill adds. Great. Fire needs to ramp scorch, frost needs to ramp WC, arcane can just nuke away at full potential. So let's say you switch targets 6 times during the fight. This is how it looks:
Ok, so fire is basically still dominating these numbers. If you count malediction for arcane spec, the nice 2.7% damage increase brings it up to 1568. Still less than frost.
So I was wrong, the sky has, in fact, fallen.
This was taken for around 1200 spell damage, if you drop down to about 1000 it's a little more even and you can justify arcane spec for encounters with a lot of interruption, switching, or pushback. Overall though fire just scales too well. The only thing I haven't considered is mixing in blast spamming, which could be considerably better with zero latency. I doubt a mana dump will make up for it, but you never know. Time to start buying Flamecaps.
The post was made regarding the death of Arcane spec with the changes to the MSD. What I think is very interesting is that the numbers for Fire and Frost are so close to each other. In 2.3.2 Fire gained a couple of percentage points from having Icy Veins available but Frost gained pretty much double that due to the buff to Cold Snap allowing ~5% more uptime with the elemental and 50% more uptime for Icy Veins (using a 10 minute fight duration length as the model).
Those numbers from Vontre also didn't include the current bug with Elemental Precision.
It seems to me that if the 2 specs were that close before 2.3.2 then the buff to Cold Snap (and the bugged EleP) should push Frost that much closer, if not superior to, Fire spec. Pushback and Elemental fragility will, of course, change the numbers dramatically in actual practice but I'm really not seeing where Fire suddenly gained an obvious superiority over Frost, especially just based on the theorycrafting I've seen in this thread.
12/23/07, 7:31 AM
Similar Threads
