So here is a question I would like to ask, the tier 9 gear has significantly less spirit than the tier 8 gear which means that the benefit molten armor gives will be diminished. Could I make the case that these changes no longer make molten armor the best buff to have? Would mage armor be preffered as it will would help make up for the loss of spirit mana regen needed in longer raid fights? Perhapse with the 4 piece bonus that would free a couple of points off of arcane meditation to be used in something else in the talent tree.
Something entirely different but I think I'm allowed to ask ;-)
I have been trying to figure out the best rotation to use from asking different mages ingame but they all tell me different. The point is, I'm not sure whether to use ABx3AM regardless of the proc, or use ABarr without a proc ('cause I seem to remember that's what it was always like). Up until this point I believed that using ABx3Abarr is better than using an unprocced AM after ABx3.
So this still leaves me with my question: I cast 2xAB, and there is no proc yet. Then I can do 3 things:
1. I cast the third AB, MB procs, but I have to wait 1 second to see if it procced in my interface (reaction speed), and then I obviously cast AM. I lose 1 second ish of DPSing here
2. I cast the third AB, MB does not proc, but I have to wait 1 second to see that it didn't proc (reaction speed), and then I obviously cast ABarr (because according to the above poster, that's more DPS) I lose 1 second ish of DPSing here
3. I cast the third AB, and regardless of the proc, I just cast AM instantly. Reaction speed = zero, and I don't lose a second of DPSing here.
If an unprocced AM is better than ABarr (aka. there is no use for AB at all except when moving), then the answer is easy: Always use situation 3. But since I don't know what results in better DPS, which do I pick? Is it still better to wait a second and then use ABarr even if that second is a loss of DPS?
Please enlighten me :-)
Thanks.
I'm wondering the same thing.
also would "/cast [nochanneling] arcane missiles" reduce dps at all due to latency?
Waiting is always a dps loss. With the new patch, ABspamAM is the way to go.
after 4 AB, the chance of no proc is 13%, after 5 AB 7,8%, after 6 AB 4,7% and after 7 AB 2.8%.
Use rawr and see in the Arcane cycles window what to use and when. Rawr actually includes the "buff not visible" option.
Waiting is always a dps loss. With the new patch, ABspamAM is the way to go.
after 4 AB, the chance of no proc is 13%, after 5 AB 7,8%, after 6 AB 4,7% and after 7 AB 2.8%.
Use rawr and see in the Arcane cycles window what to use and when. Rawr actually includes the "buff not visible" option.
If so, there is no point of getting ABarr talent at all... !?
ABarr is used for when you have to move for an extended period, such that your 6 second AB stack would fall off and be wasted. A small example is when you get targeted for Legion Flames on Jaraxxus while at 4 stacks. Your going to have to move for ~6 seconds anyway, so you'd throw off a ABarr instead.
Its not the most exciting use, but there are a number of situations where situations like the aforementioned example will crop up.
And another 2 little nub questions..
1. Is the best use of PoM is just to force AB stack?
2. What's the best timing to pop IV? I'm using it at the very start of the encounter, to use maximum of it's CDs or maybe its better to go ABspam on its active period, go OOM and Evocate fast.
So here is a question I would like to ask, the tier 9 gear has significantly less spirit than the tier 8 gear which means that the benefit molten armor gives will be diminished. Could I make the case that these changes no longer make molten armor the best buff to have? Would mage armor be preffered as it will would help make up for the loss of spirit mana regen needed in longer raid fights? Perhapse with the 4 piece bonus that would free a couple of points off of arcane meditation to be used in something else in the talent tree.
I have very little spirit on my gear and in a raid environment I still get over 10% crit from Molten Armour. It has scaled ridiculously well since the changes that tied it to spirit, and frankly, is borderline overpowered and likely to get a nerf at some point in the future. In short, 50% regen from Mage Armour isn't even close. You should never be running oom and so you should never be using Mage Armour.
also would "/cast [nochanneling] arcane missiles" reduce dps at all due to latency?
If you're using a rotation that does not include using MBAM at 0 stacks of Arcane Blast, a nochanneling macro for Arcane Missiles is useful to stop yourself from starting to cast a 2nd Arcane Missiles in the middle of your previous one (which can happen if you're spamming the key). However, since most of the rotations now include using MBAM at 0 stacks you should not use a macro like this as it will needlessly delay the casting of your second Arcane Missiles in a row.
Originally Posted by Crowl
If you have to control a robot dinosaur that fires lazers and there's a time when you shouldn't be shooting those lazers then the encounter is clearly flawed beyond hope of fixing.
after 4 AB, the chance of no proc is 13%, after 5 AB 7,8%, after 6 AB 4,7% and after 7 AB 2.8%.
This is a calculation that I don't understand. AB has a 40% chance to proc MB, this proc chance is stateless insofar as the proc rate is unaffected by any number of previous AB casts. I understand that you can calculate some probability figures based on that, but would it be any more accurate than the probability of rolling a six on a die?
The Mage theme song.
<+icesurfer> this is the fucking security industry; if you want ethics, join the Red Cross
Probability is hard to grasp as a concept. But no, the chances don't get any higher on any given "roll" (or AB-cast, for this matter), still, over a certain number of casts, the chance to not have a procc gets very low statistically. Obviously, you could have a freak strain of 8-10 casts without any procc easy enough. Or you could have a whole night of always proccing on the first cast. Its just not very likely. And since TC assumes averages, those numbers show you what you can expect to happen most of the time.
This is a calculation that I don't understand. AB has a 40% chance to proc MB, this proc chance is stateless insofar as the proc rate is unaffected by any number of previous AB casts. I understand that you can calculate some probability figures based on that, but would it be any more accurate than the probability of rolling a six on a die?
That's exactly what he calculated, the probability of not seeing a MB proc. You are correct that the proc chance is a static 40% per cast, and theoretically you could cast an infinite amount of times and never see a proc, but the probability tells you that you should have a MB after 4 casts of AB 87% of the time, 92.2% after 5 casts, 95.3% after 6 casts, 97.2% after 7 casts, etc. The more you cast, the less likely you are to not see a proc.
Waiting is always a dps loss. With the new patch, ABspamAM is the way to go.
after 4 AB, the chance of no proc is 13%, after 5 AB 7,8%, after 6 AB 4,7% and after 7 AB 2.8%.
Use rawr and see in the Arcane cycles window what to use and when. Rawr actually includes the "buff not visible" option.
I can't get the arcane cycles window to open without it crashing
This is a calculation that I don't understand. AB has a 40% chance to proc MB, this proc chance is stateless insofar as the proc rate is unaffected by any number of previous AB casts. I understand that you can calculate some probability figures based on that, but would it be any more accurate than the probability of rolling a six on a die?
Seonid is actually correct here--assuming I am understanding the situation he is describing. To be clear:
In the general analysis, yes, the chances of not procing an MB after 5, 6, or 7 casts of AB is very low. Seonid is (I think) talking about the situation where you have cast 4 AB and have not gotten a proc yet. This changes the analysis. We have prior knowledge about the sequence. The fifth cast does not magically have a higher chance of procing MB just because the the first four missed. I think we all know this.
This does suggest a game play behavior though. If you have cast say six ABs and not gotten a proc and your mana is draining quickly it is perhaps incorrect to keep casting AB fishing for that proc. The chance on your 7th cast is of course only 40%. So it might be better to cut your losses, cast ABar, and restart the sequence.
In playing with Arcane last night, I only had this happen once. So perhaps a minor concern, but the possibility does exist.
Seonid is (I think) talking about the situation where you have cast 4 AB and have not gotten a proc yet.
This was my thinking in looking at that specific situation and rationalising the TC aspect with reality. If you have AB at a 4 stack and no MB proc, you can look at probability and continue to spam AB until you do get an MB proc. The non TC aspect of that is you are hitting a phenomenal MPS figure and at some point, you will either go oom or bail out on the AB spam. Looking at my stats in Rawr, a 4 stack standard AM is slightly lower DPS than a 4 stack AB, but the (4 stack) AM is close to 4 times the DPM. Excluding edge cases such as movement or low target HP, I'm looking at factors for using AM on a 4 stack if no MB proc.
The Mage theme song.
<+icesurfer> this is the fucking security industry; if you want ethics, join the Red Cross
In the general analysis, yes, the chances of not procing an MB after 5, 6, or 7 casts of AB is very low. Seonid is (I think) talking about the situation where you have cast 4 AB and have not gotten a proc yet.
...
This does suggest a game play behavior though. If you have cast say six ABs and not gotten a proc and your mana is draining quickly it is perhaps incorrect to keep casting AB fishing for that proc.
The question I would like to ask, having experienced this occurence a few times last night, is what is the best way to reset the AB stack? i.e. What is the lowest dps loss vs. continuing to spam AB: non-MBed Arcane Missiles, or Arcane Barrage?
This is a calculation that I don't understand. AB has a 40% chance to proc MB, this proc chance is stateless insofar as the proc rate is unaffected by any number of previous AB casts. I understand that you can calculate some probability figures based on that, but would it be any more accurate than the probability of rolling a six on a die?
Every time you cast AB you have a 40% chance to proc MB. That never changes. However, the probability that you will see at least 1 MB proc changes (grows) as you cast more Arcane Blasts.
In general for binomial probabilities like coin flips, events that can have exactly 2 different outcomes, you want to look at three sets of probabilities. The "event" in this case is casting Arcane Blast, the two outcomes are "Proc MB" and "Don't Proc MB".
1) the probability of the individual event (40% chance to proc, 60% chance to not proc) 2)the probability that you will see exactly K "successes" in N "trials" (you will see exactly 1 MB proc after 4 AB casts 34.56% of the time, you will see exactly 2 MB procs after 4 AB casts 34.56% of the time, you will see exactly 3 MB procs after 4 casts 15.36% of the time ) 3)the probability that you will see at least 1 "success" in K "trials" (you will see at least 1 MB proc after 4 AB casts 87% of the time, basically summing up the probabilities from (2)).
So if you just cast your 4th AB and didn't see a proc, your 5th cast still only has a 40% chance to proc, but 92% of the time you'll see a proc after 5 casts. The numbers tell us fishing for that MB proc will almost always result in a proc, and the more you keep at it the better your chances are of seeing it. That doesn't mean you should always fish for procs though. But if you have the mana, go for it.
Since there's an 87% chance of seeing at least one MB proc after 4 casts, it means that stacking 4AB after getting a 4T8 proc is a waste 87% of the time. So it's best to just use AM right away, if you don't you might as well start working on Tier9 set earlier since your basically wasting the 4T8 bonus.
I don't know how rawr handles ab stacks but with the spamming of abilities even if your 4th ab procs mb you will still start and finish a 5th ab before it's used. Basically we are always 1 ab behind in our rotation than you can sort it out on paper. The chances you cast an mbam after the 4th ab is only 78.4% not 87.04%, either you notice mb on the 3rd cast and you use it after 4th is finished or you notice on 4th and use when 5th is finished, you cant interact with it and use on 4th cast if that was the very same ab that procd it. In case its taken into account just ignore my post.
I don't know how rawr handles ab stacks but with the spamming of abilities even if your 4th ab procs mbam you will still start and finish a 5th ab before it's used. Basically we are always 1 ab behind in our rotation than you can sort it out on paper. The chances you cast an mbam after the 4th ab is only 78.4% not 87.04%, either you notice mbam on the 3rd cast and you use it after 4th is finished or you notice on 4th and use when 5th is finished, you cant interact with it and use on 4th cast if that was the very same ab that procd it. In case its taken into account just ignore my post.
I definitely take this into account in all my models, it is an accurate observation.
What about people that religiously cast AM after 4 casts? It doesn't matter if they notice the proc or not, they'll cast AM either way.
Kavan's list of optimal cycles shows pretty clearly that you should never cast AM without a Missile Barrage proc. If you're running low on mana you should clear the stack with Arcane Barrage (the bottom 3 locations).
Originally Posted by Crowl
If you have to control a robot dinosaur that fires lazers and there's a time when you shouldn't be shooting those lazers then the encounter is clearly flawed beyond hope of fixing.
As of last night's version v2.2.19, Rawr does account for the delay in seeing/using MB (ie, 'being one AB behind'). This is also offset by the fact that you get a second MBAM, 40% of the time.
As of last night's version v2.2.19, Rawr does account for the delay in seeing/using MB (ie, 'being one AB behind'). This is also offset by the fact that you get a second MBAM, 40% of the time.
What they were talking about delay in seeing MB was always there. That's just the concept of MB being procced, but you haven't registered it as a player so you can't incorporate that fact in your decision making. What was changed in latest release is the latest bug/feature that the server also doesn't use/consume MB immediately, so the server itself needs time to register the proc, i.e. it actually procs after the next spell already started, not only the user.
09/23/09, 3:08 AM
Similar Threads
