[Mage] Raiding as Arcane mage post 2.4
This is a thread that hopes to have serious discussions about raiding as arcane post patch 2.4.
For starters, we aren’t here to argue about whether arcane, frost or fire is better or whether you are top in damage meters in your raid. Discussion on this thread should be limited primarily to raiding as arcane. About how best to raid as arcane, how to gear up for arcane, and any game or item mechanics changes that affect arcane raiding.
Part 1: How to play Arcane post 2.4
Patch 2.4 had some subtle but key effects on arcane raiding.
1) The mana regen from spirit goes up as your int goes up. So, now, the higher your int, the more mana regen you have, and as we all know, mana = damage for arcane mages. This has been a very key change that has enabled arcane mages to spam arcane blast for the majority of their fights. (exact formula per Tinybronco: "5 * 0.0093271 * Spirit * Square_root ( Intellect ) is the formula relating mp5 with int/spirit")
2) Haste can now bring your global cooldown below 1.5 seconds. The lowest is 1 second. Another key change given how important an impact bloodlust and heroism have on end game raiding. Plus, this enables arcane mages to add some haste to their gear without being overly penalized for it. Haste is still not the best stat for arcane mages, more discussion would be good. But at least it is not an entirely wasted stat as it was before this change.
Arcane raiding post 2.4 is made up of just one thing. Spam arcane blast as much as you can for as long as you can. Everything else you do is to support that. The theoretical perfect arcane raid sequence (putting aside cooldown abilities for now, and bloodlust), is one where the arcane mage managed to cast 1.5 second arcane blasts 100% of the time for the entire raid fight. Where this is not possible, then everything else that is being done when raiding as arcane is to get as close to this “nirvana” as possible. Hence, the importance of managing your mana, gearing properly, whether to gem for int, and when to use Arcane power.
The core of arcane raiding now, is to spam 1.5 second arcane blasts as much as feasibly possible. If you have received 2 druid innervates that enable you to do it, then all the power to you. Even without such raid support, the int/spirit change that was mentioned has been such that arcane mages can now still spam arcane blast most of the time, for the majority of raid boss fights.
So, the three key ingredients to be able to raid successfully as arcane.
1) Have two piece T5 (for the boost in damage to your arcane blast)
2) Have good mana support (either SP or shammy in group, gem, pot, etc)
3) The willingness to push AB spam to the max. Start off with pushing AB until you are really and truly close to dry. Then work backwards as you gain experience about the true length of the raid boss fight.
Assumptions: This assumes at least T5 level of gear. Given the ease of getting T5/T6 equivalent gear through ZA, badge and T5/T6 dungeons. Its a reasonable assumption. This kind of AB spam style arcane raiding is not being suggested for people with T4 or worse gear just starting to raid. (unless your group support is that good). And people with spellfire, spellstrike in their setup also need to be careful. The stats on these are so low it is a lot more challenging to keep up AB spam for any length of time if you are wearing them.
This is all only the basic core concepts of arcane raiding. It doesn’t address more advanced arcane raiding, or the many issues which there is no standardized agree “best” way of doing to.
Part 2: Advanced Arcane Raiding concepts
Part 2A: Why and what type of filler spells
More advanced portion. (some open to debate).
What to use as filler spells when you can’t spam AB 100% of the fight. Depending on the length of the boss fight, this part could be anything from 0% to 30% of your total damage. Though if its 30% of your total damage, it would have to be a really really long fight. The objective of filler spell, is that when you are casting this spell or spells, you are still regenerating mana (due to the awesome abilities of arcane mages to regen mana with proper raid support). But at the same time, you are also still doing a decent level of damage.
Have no illusions, any amount of time casting filler spells is time where you are doing below average DPS. Because it is time you are not spamming AB. And no amount of complex or simple spell sequence of filler spells will match up to AB spam.
Part 2B: Introducing the three arcane variant specs
This is where the three arcane specs come in. Often, the spec of choice will influence the filler spells and filler spell sequence used. There could be more, but I will stick to the three I know.
Spec 1 – Arcane/fire
At least 40 points in arcane, don’t take any talents that improve AM, take 3 points in elemental precision, and the rest into fire. (get talents like improved fireball, ignite, pyroblast, burning soul, etc).
This arcane spec is still about spamming arcane. For filler spells, it can cast fireball, cast scorch, and it can use AP, pom pyroblast. The spec is unlikely to use AM as a filler because spamming fireballs as almost as good, and better damage. Your fireball hits are only marginally below a full fledge fire mage, and your crits actually hit as hard as his because you have spellpower boosting them. You will unlikely to go out of mana spamming fireballs thus making it a good filler spell, and if you want to regen even more mana, then just spam scorch instead.
Spec 2 – Arcane/frost
40 points in arcane. don’t take any talents that improve AM, take 21 points in frost. Improve frostbolt as much as you can. Take icy veins and cold snap. This spec is the theoretically highest DPS arcane spec.
To clarify. In a situation where arcane blast is 90 to 100% of your damage, the damage difference between any arcane spec variation is minor. What pushes this spec to highest max DPS theoretical arcane spec is due to two key things.
A talented frostbolt cast by such a spec hits extremely hard, and crits even harder. While casting it is incredibly mana efficient. Plus, the kicker – the ability to cast icy veins, cold snap, and then do it again. Frostbolts are the main filler spell for this arcane spec.
Spec 3 – Deep deep arcane
More than 45 points into arcane, take all talents that improve AM, have only 11 points thereabouts remaining placed into either frost, or fire.
This spec uses arcane missiles as part of its filler spells. Hence, it is talented for it. It could be a variety of sequence of spells, but there would be an AM cast somewhere in it. Take note, if you can spam AB 100%, you still do. AM and AM sequences come in only in situations when you can’t spam AB 100% of the time.
Some worth while debating issues on filler spell sequences.
Part 2C: Issues and discussions on filler spell sequences for each Arcane Variant
Arcane fire – is fireball spam the best way to go? Or some spell sequence like 3xAB, 2xFB, scorch. Etc etc. Now, latency comes into play as well. Take me, I play from Asia, so it is impossible for me to squeeze in a scorch at the end. The best I can do, is 3xAB, 2x FB. And in the first place, it might be easier to just fireball spam until you feel ready to do back to AB spam. Personally, I am not convinced the complexities of have a 3xAB spell sequence is worth it when the alternative or fireball spam as a filler is quite competitive. But le the debates roll, I am all for it. :D
Arcane frost – Is it even worth trying some 3xAb, 3x frostbolt kind of sequence, as compared to just frost bolting until you are ready to go back to AB spam again. Again, this is debatable. Tinybronco also suggested using AB, frostbolt frostbolt as a spell sequence for the filler sequence.
Full arcane – how to weave in AM. Keep on going AB, AM, AB , AM so that you will benefit max from a clearcast? Again, some complex ABx3, AM, scorch sequence? Again debatable. So let the discussions flow!
I would stress again though. The whole idea of filler spell sequences is up for debate in itself. Some people argue that you shouldn’t have a “low mana” time where you do a mana efficient spell sequence. They say. Keep the AB debuff up at max, but weave in the required number of fillers spells such that you run to zero mana at the end of the fight. This is extremely challenging, but possible if you are so used to the fight length.
Example – say you know that for this boss fight X, you will end up having to cast exactly 10 frostbolts as filler, and the rest of the time, its all AB. Instead of waiting until Time 3.20 into the fight where you are low and spamming that 10 frostbolts. You “weave” it into the whole fight. So, every now and then, while you are spamming AB, you throw in one single frostbolt, then go back to spamming AB. This requires very high degree of knowledge of exactly how many filler frostbolts you need to cast. Hence, in this fight, say you will end up spamming 50 ABs. instead of going spam 40 AB, spam 10 fostbolt, then spam the last 10 AB. You will go 5 AB, 1 frostbolt, 5 AB, 1 frostbolt until the end of the fight.
Both ways are up for debate.
Part 2D: About about cooldowns and what to cast during bloodlust and AP?
All this goes back to arcane wanting to spam AB the whole fight. While arcane power is extremely good, it sucks up even more mana if you are spamming ABs during AP. And throw in bloodlust, and the mana consumption goes up even more.
The first AP is usually not an issue. This usually happens before you need to evocate, so you will have no problems timing it with that first bloodlust and spamming AB. I believe the main issue comes if the fight is long enough for the AP cooldown to be up, and for a second AP to come in. Similarly, when the bloodlust cooldown is up, and cast a second time.
Most arcane mages will have enough mana for that first AP powered AB spam. It’s the second that is the key issue here. If you have the gear, and mana support to be able to AP, AB through the second bloodlust. Then all the more power to you. But what if that would prove difficult? Then, the debate is whether it is better served casting frostbolts, or fireballs, etc during the AP, bloodlust period because you can’t afford the mana to AB spam. Some would even argue that it is so important to time this right, that you have to lapse into a low mana spell sequence BEFORE you reach this stage, intentionally saving up the mana for this part. Again, this is subject to debate and discussion. Given the danger of over compensating and ending up with excess mana. I am personally the believer that if you know you can’t AP,AB spam the second bloodlust without having to go to extreme lengths before that, then go with a filler spell sequence spam instead. Or possibly, forgo the second arcane power instead. It’s a whole discussion in itself, and I am just starting it off.
Part 2E:Mana management and use of consumables.
Again there may be further issues to discuss here. Arcane raiding is simple in concept now, but not so simple when you want to push max dps.
Mana management. On the face of it, its eating gems and pots as soon as possible, once the cooldown is up, then eating them again. If its a short fight, where you can spam AB 100% without evoc, then great. If not, evoc usually comes after the first AP fueled AB spammage.
Also, the importance of getting SP, or a shammy, or both into your group, and the debate on whether druid innervates are justified being cast on arcane mages. And Judgements kept up. (some of this may be out of your control).
Even filler spells and spell sequence I brought up previously is due to the complexities of mana management. Because to an arcane mage, he can always do with more mana.
And sometimes, mana regen can be taken to extremes. For example, better to flask for distilled wisdom than to flask for blinding light? Is that extra more AB you can do going to be worth the damage lost? Better to go draenic wisdom and adepts elixir/ elixir of mastery instead? The average length of the boss fights you are facing is key here, and the raid support and your own gear all comes into it as well.
On one extreme, if an arcane mage has 5 druids lining up to innervate him, and he is in an SP, shammy group to boot. He might as well cast molten armour, eat destruction pots and still spam AB the whole fight. Then just flask for damage, gear for damage.
On the other extreme – zero raid support, no SP, no shammy. Then definitely, mage armour, and perhaps he will need to really gear for int/spirit and flask for mana. He may even have to enchant for int. (Maybe he should just go back to fire if its going to be so rough.).
Most people fall in between, and most raid fights come in between. Some boss fights are longer, some are shorter. But your flask lasts 2 hours and your elixir lasts 1 hour. So, unless you are so rich, you flask for damage for that 2 min boss fight, then flask/exlir for mana 30 min later for that 8 min fight. Most people will have to choose what they can that will give them the most bang for buck though the whole raid.
Theorycraft right now is saying best if use elixirs. Personally I try and use those more. But for the harder fights where possibility of wipeing is there. I use flask of blinding light or supreme power because my guild is mainly farming MH,BT. So, the raid fights are short.
Part 2F: Gearing.
Arcane mages favour T5/T6 gear because they have int and spirit in addition to the damage stats. To an arcane mage, having spirit and lots of int on their gear is not a bad thing. They love it. Damage is still important, but stats like Int are important too and stats like spirit are good to have.
Rawr is a great application to help you choose your gear. I use it, every arcane mage should, its that good. But I have to raise a point here also. Rawr is theorycraft. The assumptions are that the raid boss fight is 5 minutes standard. Most MH,BT fights are nowhere near that long. When you tweak the length of the boss fights, you will get different results.
Rawr favours +int gems for arcane mages. Its very counter intuitive. And a lot of arcane mages balk at the idea of really re-gemming their entire suit for +int. I myself only have 3 +int gems in my gem slots, and the rest of my gem slots are geared for damage. I feel that ideally, you have substitute gear options. So, you will have a full out +dmg gemmed suit of gear that you will use for all short fights. And you will switch out a few selected slots with +int gems for the few long fights you might have to face.
Kavan might disagree. Maybe it is better to still gem for full int, and then use destruction pots and don’t cast evocate if you have so much mana from your gear. However, this is a really open topic and it goes against a lot of conventional gearing wisdom. Personally, I have yet to see many even true arcane mages gem themselves fully with +int.
This idea of gemming for int is still one that I believe is not so mainstream despite what the theorycraft supporting it. And in the end, it may be due more to practicality than just pushing DPS. Most mages balk at making their gear so one sided its only suitable for raiding and sub par for everything else.
Part 2G: Haste Rating. How important is haste to Arcane?
Haste rating in gear. Another big topic in itself. We at least benefit from haste now. Hence, we benefit from bloodlust and icy veins. However, its not the best stat for arcane mages in general. After pages of discussion, here's what came out with regards to haste.
Due to the many variables, its too difficult to have an accepted ranking of the best stats for arcane mages. But a generic ranking might be as follows (quote Zinaida):
"1) +hit up to 6% (reach cap)
2) +int or +dmg (Rawr usually shows int coming out on top)
3) Haste or crit or spirit in the third tier (depending on gear/raid setup)."
Ultimately, there may be lots of situations where this doesn't apply. So, I quote Wizeowl "The best advice about haste is to work it out for yourself, depending on your gear and the fight you are doing". So, use Rawr to see how valuable the various stats are based on the parameters you input in, and see how good each piece of gear is relative to each other for upgrades.
For example. When Roywyn plugs in two sets of numbers, one for BT gear, one for sunwell gear, the stats relative value to each other changes based on Rawr.
"Black Temple/Hyjal gear (or: gear with little haste)
1) +12 dmg ~ 10.6 DPS
2) +10 int ~ 10.3 DPS
3) +10 haste ~ 8.4 DPS
4) +10 crit ~ 6.9 DPS
5) +10 spirit ~ 6.5 DPS
6) +4 mp5 ~ 2.0 DP
Sunwell gear (or: gear with a lot of haste)
1) +12 dmg ~ 11.9 DPS
2) +10 int ~ 10.4 DPS
3) +10 haste ~ 9.8 DPS
4) +10 crit ~ 8.8 DPS
5) +10 spirit ~ 6.5 DPS
6) +4 mp5 ~ 2.0 DP"
So, things can change quite dramatically depending on your gear, raid setup, and even the raid boss fight itself. So, use Rawr and find out yourself exactly how valuable the various stats are to you. The above are all very generic and may not apply at all to your individual situation! You can find lots of discussion specifically on haste on page 11 to 13 of this thread. :)
Part 2H:How to make up for some of arcane’s disadvantages. Spell pushback.
Some people spec arcane frost and just use icy as much as they can, for those few fights that have tons of spell pushback.
My method, I am arcane fire. And I take burning soul. And for that one or two boss fights with that much spell pushback. I just spam fireball instead of AB and play like a fire mage. My rationale – why fight against it when the cards are stacked so heavily against you.
Third method – radical worth discussing. Have a fully tweaked out haste set. Spec into deep deep arcane. Wear the BT mage trinket, wear lightning capacitor and spam AM. You have 100% pushback immunity compared to even fire’s 70%. Maybe, it will be just as competitive.
This is only specifically for those very few boss fights where spell pushback is so significant it matters.
Part 2I: Trinket discussion.
Another big topic by itself. I am currently using serpant coil braid and Sorceror's Alchemist Stone. Others may have their own preference on which are the best trinkets an arcane raiding mage should be using. But mine are these two. I am not sure about the way Rawr ranks trinkets, and whether the intangiblee benefits of alchemist stone and serpant coil braid on mana are factored into the value of such trinkets. But certainly, the extra mana gain is extremely valuable because they allow that much more AB spam. And the longer the fight, which allows for more potting and eating of gem, then the greater the benefit. Certainly worth further discussion.
Part 2J: resources
I personally love using RawrRawr - Release: Rawr b14.1. Its easy to use and allows you to swap gear and see the immediate results. Kavan knows exactly how Rawr works.
Another resource is magegraf by Vontre. It shows your spell sequences and damage in graphical form.
A quick ballpark easy to use way to judge your DPS is to have a damage meter and look at how much your average arcane blast hits for in raids. Let's say it averages 2400 per arcane blast normal hit. This means that it will crit for 4200 (2400 x 1.75). Then look at your crit rate. Let's say its 30%.
This means that on average, your arcane blast spam will give you [30% x 4200 + 70% x 2400] / 1.5 = 2940 / 1.5 = 1960 DPS.
Arcane power, trinkets, bloodlust, can push you above this figure, but on the flip side, if you can't cast AB 100% of the time, if you have to move and bave to reapply the arcane debuff, etc etc, can all cause it to be lower than this number. And of course, Rawr is much better and much more precise. This is just a quick and easy way to get an average ballpark DPS figure you can aim for if you achieve something close to 100% AB spam.
Wow, this has been a long post. Thanks for reading so far. I hope this helps at least set the right track on further in depth discussions about arcane raiding and arcane gearing in this thread.
Let’s keep this a productive thread aimed at educating all mages on arcane raiding, and pushing arcane DPS ever further and further to its theoretical best. So, feel free to contribute any of your comments, disagree with mine, bring forth your theories. But most importantly, keep the topic of discussion on arcane.
(Fire and frost spec discussions or comparision are not entertained and welcome here). We are die hard arcanists! It is a choice we made, simple as that. So let’s keep to the subject and discuss arcane raiding!
To summarise, for those that don't like to read super long posts. :)
Arcane raiding post 2.4 is all about spamming arcane blasts. Your mana management, gearing, spell sequences are all centered around that.
Issues to discuss:
1) How best to play each of the three main arcane specs?
2) which filler spell or spell sequences to use? which may be spec dependent
3) AM? good or bad? Have a place? have no place?
4) Rotations - have a place? no place? How about "weaving" in your filler spell
5) What to do during the second bloodlust and second Arcane power.
6) Consumables - which is the best flask/elixir to use?
7) Gearing - which is the best way to gear. The great debate or whether to gem for int vs gemming for dmg
8) The great debate on haste rating. Good for arcane? worth adding? worse than crit? How to value haste?
9) How to make up for arcane disadvantages - dealing with spell pushback
10) trinkets - which are best for arcane mages?
11) Whatever else you can come up with or might disagree with me about. :)
This was as concise as I could express it. For finer points of arguments and discussions, by all means read the whole post. And post forth your comments. :D
To be honest, ironically I think you'll have the same problem with this thread because you have too much discussion points. Trying to cover all the finer details of arcane spec will do the reverse of your goal - just continue to confuse players.
I think for all practical means, you should focus on 40/0/21 and I doubt anyone will disagree it isn't the best general-use arcane spec.
edit: also put [mage] in thread title.
By all means, focus on 40/0/21 spec if you so wish to. :)
A lot of the issues brought up still apply to a 40/0/21 spec.
Quoted from Manly with his permission.
Manly said in the other mage TC thread
"Well, the problem is that people are bringing the wrong arguments. There are upsides to every spec, but people trying to vouch for arcane give the arguments that don't matter. Or my favorite people trying to ague about fire spec mana.
With this said, I think this is kind of the unease of talking about arcane is that everything is situational. Everything comes into play. Mana tide, spriest dps, JOW, number of innervates, fight duration will totally change the numbers. Then I don't think many mages will agree that going full int gemming is 'acceptable' in the sense that - while optimal, it is basically 'single-sided' gemming that won't work on any other spec. I think the playstyle of arcane is quite simply all about understanding DPM. The guideline isn't hard at all. I think what kind of is missing is strategies/methodologies about how to plan your DPM ahead of time. This is something I would instinctively do, if I were to start play arcane seriously. But oddly enough it seems like nobody took the time to put more than basic outline about how to play arcane. Or maybe just everyone assumes that understanding how tools like rawr/magegraf is what it takes to know how to play, and maybe that looking at the result is good enough to satisfy most peoples desire to play well? (again, this is pure speculation)."
I am not so sure that it is that easy to plan DPM head of time for each boss. Each boss has a certain total time it takes to finish, which might vary dramatically depending on whether you had one or more people die during the fight. And your own DPM may change depending on which group you were placed in for different fights. So, all this may result in the fight taking longer or shorter than what you expected, and your own mana regeneration being more or less than what you expected.
Thus, I am not sure its that easy to plan DPM ahead of time. I mean, you could estimate that Supremus takes 4 minutes usually. But maybe a few DPS got careless and got smeared early. He still goes down, but instead, it became a 6 minute fight. You can't really plan ahead of the fight for such things.
This is why I prefer to go with the "Burn AB until you are almost out" method for most fights. Then lapse into a low mana filler spell while waiting for mana to regen up if the fight is still not over. If the fight is shorter than you expected. You would have achieved maximum DPS. If the fight turns out to be longer than you expected, you still have achieved a big chunk of your damage and you have leeway to drop down a bit as you go into cruise control until your gem and pot cooldowns are ready for you to ramp up again.
For example. We had a Rage Wintechill bug and evade on us the other day at around 50% back to 100%. Plus in the process of evading, Rage aggroed and killed some DPS. So, the fight turned out to be much much longer than expected since we had to kill Rage one and a half times with less DPSers. But I still spammed AB all the way with the intention that when I did eventually get low, I would spam my filler spell.
It turned out that despite the much longer time it took, my raid support, plus Jaina's awesome arcane brillance was able to keep me spamming AB all the way till the end of the fight even though it lasted much longer than expected. I just about ran dry as rage died. So, in hindsight, it turned out to be a good decision.
Alvira, I think you've done a great overview on what is the currently accepted state of Arcane and what is up for discussion. Not to reinvent the wheel I'll copy this port regarding fillers for AB spam so that we have some framework under which to do the analysis.
One of my particular items of interest currently is the following. For deep arcane spec with relatively high amount of haste I'm encountering the following situation. Heroism+AP+IV+AB spam is no longer optimal and it is required to desync AP and IV. Rawr is suggesting that the following should be the optimal way to desync it:
Heroism+Icy Veins+ABAM3FrBScCC: 20.00 sec
Heroism+Arcane Power+Hex Shrunken Head+Arcane Blast: 15.00 sec
Heroism+Hex Shrunken Head+Arcane Blast: 5.00 sec
So using both AP and IV during heroism one after another, AB spam on AP and cycle on IV. The obvious problem where assumptions from theory no longer hold is that that cycle does not fit perfectly into that time span. Theory assumes averages for cycles and that they can be partitioned as much as needed. The particual cycle used there is the normal (AB-AM)x3+FrB except that due to haste it is expanded into (AB-AM)x3+AM+FrB.
So the question now is what subset of spells to use during the IV portion of Heroism that would match that cycle as closely as possible in dps/dpm terms or even exceed it if possible. The numbers are at around 1.73 casting speed, so we're fitting around 35 seconds of unhasted spells worth.
Anyway, just some food for thought.
6) Consumables - which is the best flask/elixir to use?
For fights I think I might have a good chance to OOM, bloodberry elixir and draenic wisdom. Otherwise I would use Adept +draenic wisdom if I have less than ideal support(no chance of getting innervate etc). With full support, I will use flask of binding light.
9) How to make up for arcane disadvantages - dealing with spell pushback
The usual way I do is to be asked to be grouped with a Paladin with imp Conc Aura or ask for an earthshield.
10) trinkets - which are best for arcane mages?
Currently trying to decide between
1)SCB + Hex.
2)Mark of the Champion + Hex.(for UD/Demon fights)
I applaud your initiative in making this thread Alvira, lets hope it turns into fruitful discussion regarding arcane and arcane DPS.
5 * 0.0093271 * Spirit * Square_root ( Intellect ) is the formula relating mp5 with int/spirit.
As arcane you should be looking for the stam/int/spirit gear available(along with spriests) instead of the fire mage/destro lock stam/int pieces available. These stats are inherent in our tiered loot however in choices such as velvet boots of the guardian from lurker or boot of blasting (ssc/tk crafted) assuming your are hit capped the lurker boots would be more beneficial towards mp5 and AB spam uptime.
I am a 40/0/21 mage (4/5 hyjal 3/9 BT) and recently I have invoked the 2 cycle theory of playing. I use 2 spell cycles;
1) AB spam (high DPS high DPM) pots/gems/evocate
2) AB(1) FrB FrB (or however many FrB you can achieve with spellhaste before you can reapply AB(1) with speed bonus)
The AB with filler FrB I find to give mana back at a very high rate and is easily to manipulate when ranking up AB using the AB-->FrB-->AB(2)-->FrB etc method. I'm interested in hearing other peoples ideas/styles on filler for arcane.
I've seen AB-FrBx2 suggested before as the dpm cycle. Just so that it is easier for people to compare things I'll add it to the spell info in Rawr.Mage.
Expanding the AB-FrBx2 into the full cycle will give you better AB spam tradeoff. For the typical ~T6 level arcane/frost mage I'm getting 3.23 dpm tradeoff for AB-FrBx2 compared to 3.11 dpm on equivalent expanded cycle (AB-FrB)x3+FrB. Adding the 2 extra AB-FrB should always improve your AB spam tradeoff I believe. I came to similar conclusions for deep arcane comparing partial and full AB cycles. The only time where it's optimal to use partial cycles is when using the full cycle for full fight would run you oom without even doing AB spam.
Good thread ; and stick [Mage] in the title as Manly said ; so it looks official and gets more attention.
I'll go through your text in a bit Alvira ; thank you for taking the bull by the horns and getting this thread started. If you can help bring to the table some discussion that results in more people being better able to understand the thought processes behind what it is we do... hopefully some number crunching can be done.
Kavan regarding Rawr ; I think trying to bring arcane "cycles" into it is quite hard and less useful than when it used to be the ABx3, AM,scorch ; for example. Although of course, more info is better than no info... And I don't want to knock anyone involved with Rawr because its been a revalation.
However, while you're here and open to suggestions - when you say you incorporate an (AB,FrB,FrB) cycle ; thats great but is it possible also to do something like the following :
Double FrB ; custom AB spam length [ ] "Please enter this field the number of AB casts in a row you would like to simulate" ; Rawr will add a double frosbolt to this cycle and calculate the dps"
Single FrB cycle ; custom AB length [ ] "Please enter this field the number of AB casts in a row you would like to simulate" ; Rawr will add a single frosbolt to this cycle and calculate the dps"
Or alternatively ; it is possible for Rawr to run the numbers and tell you whether, for example, 7:2 or 14:1 are better dps using custom fields x and y where y is never greater than 2?
Or ; is the functionality there already?
To be honest - I love Rawr ; but rotations are not really used by arcane as such ; It's totally dependent on how much mana we have coming in - and things such as SP's damage, raid mana (allowing chance for innervate), crits on pots/gems can vary this hugely - so the only real numbers I look at in Rawr are :
1. Total damage
2. Mana regen
In that order ; and then I vary the enchants, gear, buffs and see how they change.
For example, with JOW up ; I can see whether 30 int or 40 spell damage is better for total damage, and how my mana regen is affected in relation to TPS. With and without shammie totems, I can see whether for example, two +12 spell damage gems are better in my belt for overall damage than an int and a spirit gem, and I can visualise in my mind the fights that I'm going to be facing and whether I'm likely to get the totems.
IF (and it may be a big if) I were able to either...
a ) Get Rawr to tell me directly or b) able to use swap-outs to determine
... how many AB I can spam in a row with 2 frostbolts interweaved, and like wise with 1 frostbolt interweaved... I can then visualise whether I want more damage or more regen.
This greatly affects decisions such as : hit vs intellect/spirit regen.
For example ; If I use a piece of gear which as huge damage, but incorporates hit ; then this adds substance to frostbolts (as miss chance falls) and I can "get away with" a bit less regen, because the frostbolt damage overall will be higher, and I won't need to AB spam so much.
Conversely ; If I want to use a flask of supreme power instead of elixirs (lowering my int and spirit) for a wipe-night, I would like to see how many ABs I can ideally use.
The reality is that Rawr is going to struggle when it comes to Arcane ; as the arcane game-mode incredibly situational ; and really only the total damage and mana regen statistics are crucial ; in relation to the raid groups that the various users will be accustomed to. I welcome your thoughts on this, as it would not surprise me in the least if I was missing a few features of the existing "cycles" within Rawr ; which from what I've seen mostly incorporate [AM, scorch], [FrBx3 or more], [AB,Fb,AB,Fb ] none of which are relevant ; and which contributed to me avoiding using either the "DPS" or "Cycle DPS" data.
Many thanks in advance
50/0/11 -- spellfire special
If you are using spellfire, the extra 172 damage on AM causes AM with a 50/0/11 build to outdamage Frostbolt from a 40/0/21 up till the t6 gear level when spellfire itself falls behind. Due to both the spellfire (7% int to damage) bonus and spare OOC regen ticks, this build heavily favors gemming straight +int gems in every socket, which isn't terribly appealing if you're unsure about being an arc mage.
The nuking pattern is AB spam till you get far enough down to pot+gem and fire off your IV/AP/trinket macro, spam until the IV has 1 second left and finish with an AM. From there, rotate AB/AM until your 3 minute cooldowns are almost up, evo, restack AB, and spam out your cooldowns. Why AB/AM? You don't normally want to drop the AB stack and AB/AM ensures your clearcasts go to a 700 mana spell. It takes advantage of the Arcane Potency bonus where an AM that procs clearcasting gets the bonus as well as the spell that consumes the clearcast. This rotation also gives you the option of AM on clearcast without ever losing cast time. Doing an AM on clearcast costs you dps (the time it takes you to restack AB) but is useful for extending your mana pool when you've overnuked with ABs. Finally, rotating AB/AM gets you ticks of OO5SR regen. You are OO5SR from when AM ends until you finish the next AB cast, which is normally enough to get a tick in.
So, the most common objection is that AM is expensive and doesn't scale. I won't debate the second, but AM's efficiency isn't bad if you have JoW on the target. JoW reduces the cost of AM an average of 2.5*74=185 mana and you get a tick of OOC regen if you're not chaining them. Using myself as an example, I'm 600/405 for regen when raid buffed, so that's another 78 mana. This pulls the cost down to 521 mana for the AM compared to the (345*.82-(74*.5))*2= 492 mana for 2 fully talented frostbolts.
Having played both a 50/0/11 and 40/0/21, I prefer 50/0/11 for grinding. The damage from AM is fairly predictable and a FrB->AM will drop more or less any mob if you have 1050 +dam or so. If you take full permafrost (no reason not to, you don't shoot frostbolts in a raid), the mob will die before it gets to you. 40/0/21, by contrast, varies wildly depending on whether the frostbolt crits or not, which makes it harder to zone out and just grind. I usually take one hit from a mob with 40/0/21, which means eating every so often.
Alvi, your post is hugely long and makes my eyes hurt. You need to work in some headers, font sizes, perhaps some colours. You need to organize it into chapters and make it easy to read for first-timers. When I get undergraduate coursework to correct that looks like this, I mark it down simply because it doesn't help the reader understand.
Bold the title lines (like Trinket Discussion) and include references including numbers. "mana regen from spirit goes up as int goes up" is not enough to back up a TC thread. Get the numbers and insert, then add a link to where you got it from. This ensures (1) you have your ass covered when someone says "but surely +18int +4spi is better than +7int +19spi for mana returns?" (2) new readers who are not familiar with the issue may look at your references and learn more without clogging your thread with "so how much better is spi than mp5?" or "sorry to ask this again, but how much...."
Also, many of the questions you pose have already been answered. How to weave AB-AM, whether AM on CC is beneficial and under which circumstances is it not have all been discussed. Can arcane-fire work, are elixirs better than flasks.
All in all with some more work to shape-up this first post you may have an informative arcane-spec FAQ which many have been looking for.
Arcane Tip Grab Bag
This is a collection of stuff I've posted elsewhere.
On the increased mana cost of 2t5 and AP...
The increased mana cost on the AB debuff is fixed. Each debuff increases the mana cost by 146 mana and reduces the cast time by .33 seconds. AP and 2t5 increase the base cost but do not affect the mana added by the debuff.
The base cost of AB is 195 mana, so 2t5 and AP add 20% (39 mana) and 35% (68.25 mana) respectively and 62% (120.9 mana) combined. When fully stacked, AB costs 633 base: 672 with 2t5 (106%), 701.25 with AP (110%), and 753.9 with both (119%).
The result is a significantly increased DPM for AB.
The tricky gem slots to fill are always the blues to activate your CSD or hit a +dam bonus. Heroic gems can be extremely nice for this if you're gemming for regen, specifically [Seer's Chrysoprase] (crypts) and [Fluorescent Tanzanite] (BF).
Manly currently has a list of food buffs that stack with other buffs. Most are seasonal. The exception is [Kreeg's Stout Beatdown], which is buyable from Kreeg, in the initial courtyard, after doing a tribute run. In case you've forgotten how to do a tribute run, you'll need a rogue (or seaforium) and the mats for a [Gordok Ogre Suit]. You don't absolutely need them, but it's easier to just do a full tribute when you're 2-manning stuff.
Value of int/spi
Socketing a 10 int gem at 750 int and 320 spi--fairly low stats, if you're stacking int/spi, you'll be in the 870/450 range with raid bufs--gives (assuming BoK):
* 12.65 intellect (non-gnome)
* 189.75 mana
* 3.43 OOCmp5 or 2.06 mp5
* 113.85 mana from evo (151.8 2t6)
* .16% crit
* 3.16 +damage (4.05 +dam spellfire)
Assuming a 6 minute (300 second) fight, the total contribution to your mana is 189.75+113.85+2.06*(300/5) = 436.2 mana, or about 2/3 of a fully stacked AB, which is a significant amount from a single gem, not to mention the +dam/crit roughly equivalent to a [Potent Flame Spessarite].
Stacking int, in turn, makes spirit a better stat. Socketing a 10 spirit gem yields (these are I5SR):
* 450 int: 6.53 mp5
* 550 int: 7.22 mp5
* 650 int: 7.85 mp5
* 750 int: 8.43 mp5
* 850 int: 8.97 mp5
So the two sort of grow together.
For my first post here, i'll copy-paste something i posted elsewehere :)
trying to get some answers by math, etc. If you can help :) (sorry for my bad english :( )
|All times are GMT -4. The time now is 11:42 AM.|
Forum Infrastructure by vBulletin 3.6.12 ©2000-2007, Jelsoft Enterprises Ltd.