The blastweave chained with the flamestrike getting the crit probably has something to do with the fact that "order of operations" in the game isn't compeltely straight-forward. Just like if you look at WWS/combat log there are just some things that you know cannot happen in that order and should've happened in reverse according to common sense, and while most of the time you could excuse it as "lag", sometimes it looks like there shouldn't be any way 1 event happened before the other.
Just like 2 mages can sheep eachother in a duel and both sheep will go off, or you can even kill someone and have him kill you without anyone using damage over time spells - if the killing blow is done by both sides in close enough intervals, unless I was seeing things wrong as I didn't make repeated tests about it, both will die, even though logically it should be impossible, just like the sheeping.
Combustion takes a short half a second to compute consumed charges - which is why mages with low crit chance should make a macro to /stopcasting /cast Fire blast /stopcasting /cast Combustion (at least pre-patch), at the tail of a fireball, so neither crits, and the Combustion essentially starts at 3 stacks. This allows it to refresh quicker, and if they do crit, ideally they'd have crit with combustion anyways so you save time cooling your Combustion down faster. This also works with Flamestrike or Blast wave (macro above except instead of fire blast use blast wave). This allows you to aoe with the spell at the 0% bonus crit rate, and assuming you don't get 3 crits, you'll immediatly be stacked up at 6-10 charges of combustion.
Any mage regardless of his crit should try to fire blast with his assumedly last fireball, i believe it's more beneficial than hoping it doesn't crit and wasting it on fire blast in most cases.
Btw, in regards to napkin math AP Fireball and getting that last fireball: Assuming you wish to do 6x fireball 1x PoM pyro (which does get the benefit at the time of the cast, the naysayers who said otherwise should be shot). You realistically need to get the first 6 fireballs off 14.80 seconds into it (varies by lag, this is hopefully a conservative estimate). Which means a fireball every ~2.4667, which using the 2.4667=3/(1+x) formula, x is ~21.6% spell haste. Which is ~340 spell haste, for the entire length of the 5(6) fireballs. Assuming you pop skull of gul'dan, you need ~165 passive spell haste outside of skull. Good luck.
P.S. If you are mage tanking and can't live one full blast at the start of the encounter, you don't have enough hp/shielding on you when you pull.
Searix your post is slightly dense to understand and I don't mean dense as in dumb, I mean a lot of information with not much expletive: I don't clearly get what you're proposing in the first paragraph and why: Firstly, I don't think a macro which sequences anything with a GCD and -afterwards- sequences a non-GCD spell will work: It'll simply cast fireblast and return an error with the request of Combustion. Secondly, unless I got it wrong (which I suspect I have) you suggest switching combustion on or at the same time as the previous ball and current blast hit so as to tag +2charges to it? Do you have data backing that this won't consume Combustion charges? For the fireball that left before, I can within reason see why it shouldn't consume 1/3 combustion, but I'm quite sceptical about the fireblast you mention in your macro.
Secondly, assuming this does work, it's an awful lot of work to cause the combustion's 3 crits to happen a tiny bit earlier, what's the point? We already know that Combustion works better and gives more value to mages with lower crit, so devising a mechanic that will force it to accelerate it's self is only in a tiny way beneficial as it'll get the Cool-down to start a few seconds earlier (as Comb CD countdown starts when it's run out).
For the fireball that left before, I can within reason see why it shouldn't consume 1/3 combustion
It does though, shoot fireball and click combustion while it's in the air, if it crits it takes 1/3 of combustion, however I'm not sure it get's the 10% extra crit. I always feel cheated when this happens.
I'm almost certain the same happens, though I'm willing to accept Searix's version given proof. Usually I chill for a quarter seccond or so untill the ball hits before hitting combustion; having to down a manastone (for serpent coil braid) and click a trinket makes it a click and a key most of the time, but if in doubt, i'll fireblast to instigate at least a GCD worth of time before Combustion on.
It does though, shoot fireball and click combustion while it's in the air, if it crits it takes 1/3 of combustion, however I'm not sure it get's the 10% extra crit. I always feel cheated when this happens.
Pretty sure it does.
Optimaly combustion should be clicked in air when you cast a fireball, and you should not fireblast nor scorch until you think a third fireball will crit.
I went fire after 2.3, and immediately went back to arcane after one raid.
Something that i haven't seen mentioned(could be blind) is for the burst damage call, "ALL OUT"...if you have managed your mana or have evoc up, arcane blast out dps's everything with the T5 2 piece bonus while in SSC and TK.
I went fire after 2.3, and immediately went back to arcane after one raid.
Something that i haven't seen mentioned(could be blind) is for the burst damage call, "ALL OUT"...if you have managed your mana or have evoc up, arcane blast out dps's everything with the T5 2 piece bonus while in SSC and TK.
Not to sound like a jerk, but a guy with three posts coming into a thread and posting a outright contradiction to all the painstaking math and theorycrafting in this thread without backing it up is not someone who is going to be taken seriously.
Unless you'd like to backup your claim with some substantial theorycrafting, then I suggest you reread the highlights of this thread. In addition, I would suggest that being fire for one raid is not enough to get into the groove of how to maximize your DPS.
I'm in college for an engineer major, definately not english
Macro i meant to post is /combustion /fire blast, and it brings a few dps boosting benefits:
-Your combustion refreshes faster
-If you sync your trinkets/pots with Combustion (which you should), you can guarentee getting the full benefit through all of your combustions, if you don't, and charge up from 1, there's a chance of a resist, or a non crit string etc. that you end up hitting your last 1-2 fireballs without your trinket/destruction pot
-More dependable combustion, ignoring the incinerate talent, assuming you don't crit your first 2 (which often happens if you use the macro, as the fireball you cast in mid-air and the fire blast don't benefit from the combustion), you start with 3 charges and no crits
Btw pint, not macroing your trinket/cools and clicking them can and often does outweigh the benefit of even using them at all.
Also, if you want evidence of the "in-air" spell not getting benefit of combustion if activated while in the air/cast, an easy test is to cast flamestrike on a group of mobs and immediatly hit combustion. What happens a good portion of the time (depending on your crit chance) is that you get 1-2 crits, and your Combustion stacks up to 10.
Edit: I'm not sure if using Combustion after the fire blast works, it might, but i'm pretty certain that if you macro combustion into fire blast, that the fire blast won't get the combustion charge but will consume one
It does though, shoot fireball and click combustion while it's in the air, if it crits it takes 1/3 of combustion, however I'm not sure it get's the 10% extra crit. I always feel cheated when this happens.
There is an easy solution and thats to pop combustion after a scorch and after you activate your trinkets. You can also wait a fraction of a second after the scorch and do combustion then trinkets if you prefer. It is less of a wait since scorch = 0 flight time. This also reduces the chance of getting a mass of fireballs in a row without crits forcing you to refresh scorch while still having combustion (or at least reduces the chance).
Unless I am missing something, frostbite appears to have a substantially lower proc rate.
I was getting ~2.5% chance to proc off max rank frostbolt whilst farming fire/air elementals (level 70/71).
PM convo with the useless GM - I haven't posted in WoW suggestion forum or anything like that yet, just checking I haven't missed something obvious.
Max rank frostbolts.
~900 casts (large sample size).
Level 70/71 mobs that are not immune.
Could this be related to the what i could have sworn was the "frosbite breaking sound" occuring often with my frostbolt casts? Meaning maybe frostbite's occuring the same second frostbolt hits, consuming it instantly?
This business with keeping CDs until combustion is up... I'm not entirely sure I get the point;
Let's look at the data: Icon, has a 2m CD, thus we model it as 43+ [(20/120)*155] which is 78. If we model it, however, as "on every 3m" it becomes a mere +60 dmg trink. Taking into account that for a 20sec duration trink, we're only going to get a maximum of 18sec bonus from it (given 0 haste and 0 latency) it drops down to a pathetic +58. I know this is not an optimal model and I know it's theoretical and inacurate, it does however show a deffinite decline given increased down-time as an illustration.
Then let's look at Combustion. (Note: I'm not examining combustion in Molten Fury range. Given close proximity to 20% I agree that delaying trink CD is optimal because the MF benefit outweighs the "down time" disadvantage. I'm reffering to the 3m rotation of Combustion). Average fire mage crit rate is over 35%. So technically, the 6 fireballs we'd cast under Icon's power, in all likelihood, would have 2 crits in them whether or not we combusted. Given combustion, the chance we get at least three (possibly more if combustion burns off quickly) is higher.
Is there theorycraft to prove that the gain from waiting 50% longer cooldown on a -use- trinket in order to get a "generally higher crit" cycle under Combustion honestly optimal rather than mashing the icon whenever it's cooled, even including a scorch at the end just to maximize the last 2sec of it's power?
Searix: Much clearer now, as for trink-macroing, I have /use 13 on my combustion macro, but I'd rather not script-in an if-gate with [Alt] saying additionally "use: Mana emerald, Mana Jade, Mana citrine, Mana blahblahblah" because then I'd still need to find a work-around for when I want to combust, icon and eat the lvl 23 mana stone, as well as when I want to simply combust before a trash pull just for the sake of it (a function I have already). Too complex, too annoying and too shoddy. Combine with the sorry state that Serpent Coil Braid sets-off Icon's CD. What I do in fact, is hit Combustion/trink macro with index finger (T), fraction of a second later click "use mana stone rotation/last stone" macro and simultaneously thumb-press (C) which is my fireball. It might be slower than one-buttoning but honestly, it's less hassle than making a complex mile-long macro which will run into the 250 letter limit and be buggy when I want to do part of it and not the rest of it. Losing .3sec I doubt will out-weigh using them at all. Three different fingers, three different actions: One of which HAS to have an earlier time-stamp than the other two: hardly much delay over two buttons.
This business with keeping CDs until combustion is up... I'm not entirely sure I get the point;
Let's look at the data: Icon, has a 2m CD, thus we model it as 43+ [(20/120)*155] which is 78. If we model it, however, as "on every 3m" it becomes a mere +60 dmg trink. Taking into account that for a 20sec duration trink, we're only going to get a maximum of 18sec bonus from it (given 0 haste and 0 latency) it drops down to a pathetic +58. I know this is not an optimal model and I know it's theoretical and inacurate, it does however show a deffinite decline given increased down-time as an illustration.
Then let's look at Combustion. (Note: I'm not examining combustion in Molten Fury range. Given close proximity to 20% I agree that delaying trink CD is optimal because the MF benefit outweighs the "down time" disadvantage. I'm reffering to the 3m rotation of Combustion). Average fire mage crit rate is over 35%. So technically, the 6 fireballs we'd cast under Icon's power, in all likelihood, would have 2 crits in them whether or not we combusted. Given combustion, the chance we get at least three (possibly more if combustion burns off quickly) is higher.
Is there theorycraft to prove that the gain from waiting 50% longer cooldown on a -use- trinket in order to get a "generally higher crit" cycle under Combustion honestly optimal rather than mashing the icon whenever it's cooled, even including a scorch at the end just to maximize the last 2sec of it's power?
Searix: Much clearer now, as for trink-macroing, I have /use 13 on my combustion macro, but I'd rather not script-in an if-gate with [Alt] saying additionally "use: Mana emerald, Mana Jade, Mana citrine, Mana blahblahblah" because then I'd still need to find a work-around for when I want to combust, icon and eat the lvl 23 mana stone, as well as when I want to simply combust before a trash pull just for the sake of it (a function I have already). Too complex, too annoying and too shoddy. Combine with the sorry state that Serpent Coil Braid sets-off Icon's CD. What I do in fact, is hit Combustion/trink macro with index finger (T), fraction of a second later click "use mana stone rotation/last stone" macro and simultaneously thumb-press (C) which is my fireball. It might be slower than one-buttoning but honestly, it's less hassle than making a complex mile-long macro which will run into the 250 letter limit and be buggy when I want to do part of it and not the rest of it. Losing .3sec I doubt will out-weigh using them at all. Three different fingers, three different actions: One of which HAS to have an earlier time-stamp than the other two: hardly much delay over two buttons.
You shouldn't be waiting on trinket cooldowns, as you noted waiting and delaying it by a substantial amount kills the dps gain. What you should do is either pair it with the first combustion, then whenever either comes up, or aim for waiting a few seconds and re-pairing it on the 3rd combustion. (Not sure which is more dps, saving your trinkets for combustion or saving those 10 seconds and just popping them).
As for my combustion rotations, i've used the following macros for a while and they've served me just fine.
1)
/stopcasting
/cast combustion
/stopcasting
/cast Fire Blast
2)
/use 14
/stopcasting
/use Mana Agate
/use Mana Emerald
/use Mana etc.
/stopcasting
/use Destruction Potion
/stopcasting
/cast Fireball.
Trinkets:
Serpent-Braid;
Activatable of choosing
Pop 1 at the tail of a fireball, pop 2 on the heels of that global cooldown. This is in addition to a plain pop mana gems macro (not used in normal pve boss cycle though, more for pvp when i dont have braid on)
This results in the desired effects: Combustion is activated whenever you want (If you want to do it before combat or before sheep), and no dps time is lost hitting your cooldowns.
Also... braid + activatable does work in a macro and always has worked, as long as you pop the activatable first.
I went fire after 2.3, and immediately went back to arcane after one raid.
Something that i haven't seen mentioned(could be blind) is for the burst damage call, "ALL OUT"...if you have managed your mana or have evoc up, arcane blast out dps's everything with the T5 2 piece bonus while in SSC and TK.
To try and be a little helpful to you, you can still burn your excess mana with Arcane Blast even if you have 10/47/3+1 spec. Molten Fury affects arcane spells too.
Could this be related to the what i could have sworn was the "frosbite breaking sound" occuring often with my frostbolt casts? Meaning maybe frostbite's occuring the same second frostbolt hits, consuming it instantly?
If that were correct shouldn't his crit percentage be much higher then that? He had a 28% crit rate from the data he gave to the GM, if the frostbolt that caused the frostbite was also consuming it shouldn't he be getting shatters on those frostbolts increasing his crit rate substantially?
To try and be a little helpful to you, you can still burn your excess mana with Arcane Blast even if you have 10/47/3+1 spec. Molten Fury affects arcane spells too.
While this is correct, the amount of AB you'd have to do to surpass the loss from AB ramp time and make a useful gain over fireball-spam during MF time is substantial. Perhaps in 13.45.3 it might be worth it but let's face it: It's a lot of hassle to gain small DPS gains which are not affected either by Bloodlust (which should be saved for MF range) or by Flamecap, don't gain haste benefit are not interrupt-resistant and have 30y range instead of 41. And cost a horrendous amount of mana.
Even if you've worked your cycle to end OOM, I still don't see why you should: What if 30sec before the end of the fight three DPS die for some freak reason? Or some DPS disco's? No biggie, you simply have 12sec more DPS time... Which you won't be able to do if you worked it out for OOM@0%health.
Searix: I appreciate that a macro that does everything is possible and straightforward; I prefer, however, that I have the option of partial-activation. Hence, to feel comfortable I'd have to build in [modifier=alt] and [modifier=shift] clauses for certain aspects of the macro which then becomes too long-winded and complex for no good reason. Rather than have a button for Trink, one for mana stone, one for Combustion and one for all-three I don't see how losing a quarter second per combustion is such a problem. It's not a case of "either 1-click, or three clicks" it's a case of 3 separate fingers actuating 3 separate buttons almost seamlessly quickly with negligible delay.
Scripting in fireball to the combustion macro is something that I'll do when I take off SCB trink though, why that didn't occur to me before is a mystery, thanks.
There is an easy solution and thats to pop combustion after a scorch and after you activate your trinkets. You can also wait a fraction of a second after the scorch and do combustion then trinkets if you prefer. It is less of a wait since scorch = 0 flight time. This also reduces the chance of getting a mass of fireballs in a row without crits forcing you to refresh scorch while still having combustion (or at least reduces the chance).
Even using scorch I get a combustion "charge" building up if i activate combustion right after the scorch hit! And like Vhad says i too feel "cheated" when it happens.
Since this shouldn't be happening with a spell with 0 cast time i guess it has to do with the latency as well. I do not presume to know how exactly how this works client/server side wise but just to illustrate how i think it might work:
server sends: scorch attempt succesful (crit)
client side: I pop combustion
client recieves: scorch attempt succesful(crit)
client side: -- 1/3 combustion
client side: ++ combustion charge
This has had me sometimes doing an AB instead of scorch right before popping combustion or waiting a bit after any firespell hit before popping it. I don't care for either very much!
While I think it's pretty obvious without calculations to say that turning the icon from 2 min to (more than) 3 min is not worth it for combustion, due to their odd durations/cooldowns and nature of fights (that make you not nescessarily want to mash each ability "on-cooldown"), you may end up with something like the trinket coming up right before combustion or vise-versa.
What would actually be interesting (and useful) is to find a way to figure out, more or less, how much time should I have left on my combustion when my icon comes up in order for me to get more DPS by waiting? Obviously with short time left you should wait and with high time left you shouldn't, but modelling how much is "high" and how much is "low" would be interesting, even if done roughly. Something like "if you have more than 5-10 seconds left don't wait, otherwise wait" would be very useful (as between 5 to 10 seconds the difference between waiting and not waiting would probably be very small anyway in that example).
Even using scorch I get a combustion "charge" building up if i activate combustion right after the scorch hit! And like Vhad says i too feel "cheated" when it happens.
Since this shouldn't be happening with a spell with 0 cast time i guess it has to do with the latency as well. I do not presume to know how exactly how this works client/server side wise but just to illustrate how i think it might work:
server sends: scorch attempt succesful (crit)
client side: I pop combustion
client recieves: scorch attempt succesful(crit)
client side: -- 1/3 combustion
client side: ++ combustion charge
This has had me sometimes doing an AB instead of scorch right before popping combustion or waiting a bit after any firespell hit before popping it. I don't care for either very much!
There might be a slight chance there's something iffy with scorch:
Crusade trinket works on a similar basis of "1 spell hit = 1 charge" but for 1xscorch you gain two charges of the crusade effect.
Perhaps this is a scorch-specific problem: The game is counting Scorch and Improved Scorch as two separate spells that happen at the same time.
Also note, a scorch crit will incraese Crusade by three, though this clearly has no effect on Combustion.
Delaying using trinkets (and consumables for that matter) is only suboptimal if the remaining fight duration is such that you will get to use them more than once. If you somehow knew for certain that there was only one minute, fifty-nine seconds left in a fight, you are only going to get one activation anyways, so it doesn't matter when you use it. The only time it matters when you use it is when there is 2 minutes or more left in the encounter...
Edit: ... or if you know something is going to happen that warps your damage enough to justify the delay (curator evocate, bloodlust, etc.)
Edit: ... or if you know something is going to happen that warps your damage enough to justify the delay (curator evocate, bloodlust, etc.)
which, as I said, is dependant on both how much time you actually need to wait and how good is the buff for which you're waiting for.
For example (assuming you don't know when the fight is going to end in this case and are not going oom to make it simple) if you have 1 minute left before heroism cooldown comes back up, if you wait you lose 1/2 a use, and gain back 45% of a use via the haste. Overall not worth it actually. If you know the fight is going to end before you get to use it again or the 1 minute until heroism is also 1 minute until 20%, it's worth waiting. However if it's up to you when to use the heroism it gets more complicated, but assuming it's worth saving the heroism for 20% (as I assume the sub 20% damage bonus and heroism bonus are both much greater than the trinket bonus), the trinket is worth saving as well in this case as it's multiplied by 1.2 and 1.45 resulting in more than 1/2 a use to make up for the lost 1/2 a use. The benefit from saving it here isn't great, though.
As you can see there are a lot of stuff that can affect wether it's worth saving or not, and I'd like to see some generic way to handle it
Arcane is still viable as long as you have two piece T5. This is what I am finding so far. The buff to arcane meditation means that you have more mana to spam AB. And if you are not quite in a position to get 4 piece T6 yet, then in fights where you get to spam AB a lot, you will still come out pretty well in dmg meters.
Deep arcane's AOE with AP applied is still extremely powerful as well. I think it would take a very well played fire spec with all buffs and destructive potion and such to beat out deep arcane's AOE.
I went with a rotation of 3xAB, 2xFB. To be honest, this is a much easier to keep up rotation than the previous 3xAB, AM, scorch. And actually, I find that I could use AB much more often. So the fireball rotation only comes in very occasionally. So, my sense is that if you are not quite at the stage of having Illidan on farm and still progressing, then deep arcane is competitive to deep fire.
There will be some fights were deep fire is better, and others where deep arcane wins out.
11/17/07, 7:13 AM
Similar Threads
