Elemental, Shaman, Spreadsheet, Theorycraft, WoW

Flame Shock in 3.3.3

In patch 3.3.3, so sayeth the PTR patch notes, we can look forward to some changes regarding our core Flame Shock spell.  Just to recap what those changes are (I’m sure everyone reading is already well familiar with them but hey!  bullet points!!):

  • Flame Shock’s periodic damage (its “damage over time” component) will now be able to be critical strikes innately (replacing the present function of the Glyph of Flame Shock – essentially we get a “free” glyph slot, just as in patch 3.2.2)
  • Flame Shock’s periodic damage will also be affected by the player’s haste
  • The Glyph of Flame Shock will change to provide an extra 60% bonus to periodic critical damage
  • 2-piece tier 8 will change to increase damage done by Flame Shock ticks by 20%
  • 2-piece tier 9 and 4-piece tier 10 will also change to accommodate the new Flame Shock mechanics.

Flame Shock is sort of the key to our whole rotation, so these changes are going to have a pretty big effect.  My aim here is to evaluate exactly what the effect will be.  Specifically:

How it all works: how haste actually affects the Flame Shock DoT, and how to work it out.

Verdict on the Glyph: Is the new Glyph any good?  What does it mean for our 3.3.3 glyph choices?

Set bonus changes: yes, this would be about them set bonus changes then.  More specifically, how the changes to tier 9 and 10 will impact the mechanics and DPS values of those sets.

Rotation effect: closing thoughts on how our Flame Shock use will change and what the considerations will be.

How it all works

Nothing complicated here, haste affects Flame Shock’s periodic damage pretty much exactly how you’d expect – it increases the number of ticks that will take place in any given period of time.  You still get the same number of DoT ticks per cast, but now they happen faster, leading to a shorter Flame Shock duration.

Also, the haste value used for Flame Shock ticks is set in stone the moment you cast Flame Shock – if your haste value changes afterwards Flame Shock’s DoT will keep using the haste value you had when you cast the spell.  This is true even when the DoT is “refreshed” by 4-piece tier 10 (more on that later).

So the formula for time between ticks is:

3 seconds (base tick time, like base cast time for a normal spell) divided by total haste value

Remembering the other key factors for calculating haste:

  • We start off at 100% spellcasting speed
  • Haste rating is all lumped into one big … lump and then converted to a percentage
  • All other %age effects like Wrath of Air, Bloodlust etc. are then applied as separate multipliers

Then if I have 1000 haste rating, Wrath of Air Totem and a Moonkin present, my Flame Shock ticks would look like this:

3 [number of seconds for base tick time] / ( 1 [100% base value, remember!]+ 1000 [total haste rating] / 32.79 [haste rating > percentage conversion rate] / 100 [converting it to a decimal] ) * ( 1 + 0.05 [decimal haste from Wrath of Air] ) * ( 1 + 0.03 [decimal haste from Imp. Moonkin Form] ) = 2.13 seconds

Yeah so that probably looks more complicated than it is.  It’s really just base tick time divided by total haste.  Most of the sum above is concerned with working out total haste, which in this case is 41%.  So it’d be 3 / 1.41.  Simple!

Number of ticks

Flame Shock’s total duration is best defined now as number of ticks rather than as a total time.  The total time is easy enough to work out – it’d just be e.g. 18 seconds / total haste, just like the tick time calculation – but because we have two set bonuses that affect the number of ticks rather than the total time, it’s better to think of the duration this way:

Number of ticks times by time between each tick

For our example above, that’d be 6 * 2.13 = 12.75 seconds. (Yeah, that looks odd, but I didn’t round the tick timer in the maths.  Bear in mind when I do maths stuff I’m just sitting with an open spreadsheet typing the sums into blank cells and having it mindlessly spew forth hyper accurate results that require no work 😀 )

So instead of thinking of Flame Shock as having a base duration of 18 seconds, think of it as having a base duration of 6 ticks.

(The flame is a SHOCK, get it?)

Verdict on the Glyph

The present Flame Shock glyph makes Flame Shock’s DoT able to crit, but in 3.3.3 this function will be “baked in” to the base spell and the current form of the glyph would be pointless.  So instead it’s being changed to an extra 60% damage on Flame Shock’s periodic tick crits.

This slots into the crit damage formula as follows:

( base modifier * chaotic skyflare bonus ) * ( elemental fury + new glyph )

This results in a total crit damage figure of 209% pre-glyph (as it is now) or 241.7% with the new glyph.


Not really, sorry 😦  Even with the ticks occurring faster than ever, the table of glyph dominance by DPS value looks like this (values courtesy of ZAP! v1.2.5):

  1. Glyph of Lightning Bolt: 260 DPS
  2. Glyph of Totem of Wrath: 170 DPS
  3. Glyph of Flametongue Weapon / Glyph of Lava: 120 DPS
  4. Glyph of Flame Shock: 80 DPS

Yes, you read that right – Glyph of Flametongue Weapon is looking like it might make a comeback!  At any rate, Glyph of Flame Shock is wallowing at a rather mournful fifth place in the league table, well out of contention for one of those coveted three glyph slots.

Update: it does seem that at very high – that is, best in slot – gear levels, Glyph of Flame Shock may close the gap a little.  See this post for more information.

Set bonus changes

2-piece tier 8: Well it’ll no longer give us a free glyph slot, but is anyone still using this what with triumph emblems flowing so freely and all?  DPS value of the change is… pretty low, assuming it works how I think (I’ve not tested it).  Probably in the region of 80 DPS at ilevel 232 gear levels.  Which is not bad, really, as it’s 80 free extra DPS to anyone still using the set.

2-piece tier 9: This is a bit more interesting.  It present adds 9 seconds to the Flame Shock DoT, but in 3.3.3 it’ll add 9/totalhaste seconds… it’s simpler just to think of it adding 3 ticks, really.  At time of writing the tooltip hasn’t been updated but playtesting has shown 3 ticks being added as expected.  So for the example above with 41% haste it’d add 3*2.13 = 6.38 seconds, for a total duration of 19.13 seconds.  Astute observers will note that it still increases the length of Flame Shock by 50%.  So the set stays about as valuable as it was before, being worth around 100 DPS from saved Flame Shock casts.

4-piece tier 10: This is slightly more confusing.  Presently, 4t10 adds precisely 6 seconds to the Flame Shock DoT timer every time we cast Lava Burst.  In 3.3.3, it will instead add however many ticks give an extra duration closest to 6 seconds (source).  You would work this out by dividing 6 by the tick timer, e.g. for our example above:

6 / 2.13 = 2.82

Then round the result to the nearest integer, which in our case is 3.  So 3 ticks would be added for a total of 3 * 2.13 seconds = 6.38 seconds total per LvB.

The DPS value of this set bonus will actually increase quite a bit in 3.3.3 – it’s worth maybe 200 DPS at present, but in 3.3.3 it could be worth anywhere between 250 and 350 DPS.  What’s more, the set value really skyrockets with high haste, so it’s pretty future-proof.  This is because as haste gets higher and higher, more and more ticks fit into that ~6 second extension, and so the DPET of Flame Shock goes nuts.  We can end up getting ridiculous numbers of total ticks even though the DoT duration remains fairly short.

Rotation effect

Actually though, without 4t10 the DPET (damage per execute time, aka DPSC, DPCT, etc.) of Flame Shock isn’t affected by this change – it still does exactly as much damage per cast as before.  Only now, the damage occurs faster.  So that means more DPS!  Hooray!

But it also means we have to cast Flame Shock more, which means casting Lightning Bolt slightly less, which means a slight DPS loss.  But the gain from hasted DoT ticks is greater than the loss from Lightning casts.  So it really does mean more DPS!  Hooray!!

So casting Flame Shock more is basically the main effect of this change.  However it does bring up two other things to think about:

1. Firstly, the interaction with Lava BurstFlame Shock’s DoT is necessary for Lava Burst to crit, which constitutes a large portion (~17%) of our damage.  The DoT has to be present at the moment Lava Burst finishes casting.  This change will make Flame Shock’s timer vary depending on our haste, which will vary during a fight (EM, relic buffs, Heroism, etc).

So in other words, we’ll need to pay very close attention to just how long is left on our Flame Shock timer depending on when we cast it to ensure that it doesn’t fall off when we’re not ready for it!

** Remember that it’s always higher DPS to “clip” the last tick of Flame Shock to make sure that Lava Burst will crit than to miss a critical Lava Burst because Flame Shock fell off mid-cast. **

“Clipping” (using Flame Shock again while the DoT is still on the target) is already something we’re familiar with thanks to the current incarnation of 4t10, but just be aware it may be necessary more often even without 4t10 in patch 3.3.3.

2. Secondly, consider the precise value of temporary haste effects (like EM, Bloodlust etc.) being used on Flame Shock instead of other spells.  Is it worth it to make sure your Flame Shock benefits from the extra temporary haste?  Nobody’s done any conclusive, serious evaluation of this yet, but it looks like it’ll be a pretty close call – perhaps close enough that it doesn’t really matter.  I’d tentatively say that the main issue of using haste effects with Flame Shock casts is that we have to be more aware of the much shorter DoT timer, but we shouldn’t worry too much about the DPS value.  In general we should probably aim to maximise our haste effect uptime – that means using them as early and often as possible (sensibly) – rather than to trying to time them specifically for a Flame Shock.

What about just clipping Flame Shock when your haste gets temporarily higher?  I dunno, sorry.  With 4t10 and a big difference in the two haste values it could well be a good idea to manually refresh Flame Shock before the big haste effect expires, but that’s just a guess based on limited evidence.  I think, for example, if there was 1 second left on Elemental Mastery and your Flame Shock that had been cast under normal haste was nearly over anyway, it’d probably be worth more to cast a hasted Flame Shock before the old one expired than cast a Lightning Bolt.

Two other things to note that area already true and should remain that way:

  • While moving, it’s better to use Frost Shock than Flame Shock if you have 4t10 and a reasonably long duration Flame Shock already ticking on the target.
  • Again with 4t10, the best time to recast Flame Shock in terms of DPET is right before a Lava Burst, as that way you get the maximum number of DoT ticks possible.  However this may not always equate to more overall damage so don’t fret about it too much.

So that’s what we have to look forward to.

Why do I feel like this post is a big complicated wordy mess?  I think maybe I focus too much on detail at the expense of clarity.  Ah well, maybe Zing or pewter will lay it all out in a more easy to understand and concise way!

Also, self-pluggage: there’s a new version of the spreadsheet out which is up to date for patch 3.3.3.  Just remember not to use it with 3.3.2.



32 thoughts on “Flame Shock in 3.3.3

  1. Hmm, odd. Is this reflected in the current version of ZAP? I was playing around with the ‘new flame shock glyph’ option, and the DEP of the Flame Shock Glyph increased and was still far and above that of Flametongue

    Posted by pewter | March 13, 2010, 5:14 pm
    • Also WTB edit button. Checking the new new version of the spreadsheet now 😉

      Posted by pewter | March 13, 2010, 5:14 pm
      • GAH.

        And on closer reading: resently, 4t10 adds precisely 6 seconds to the Flame Shock DoT timer every time we cast Lava Burst. <– this has now been changed to 2 static extra ticks, afaik? Or did they change it back?

        Posted by pewter | March 13, 2010, 5:16 pm
  2. Right, aside from my momentary panic, I’m glad they seem to have left a scaling 4pc bonus on the PTR, despite totemz recent post. (I’ve looked and looked Zing, but I can’t find a reference for it in blue posts.)

    Posted by pewter | March 13, 2010, 5:35 pm
    • The tooltip on 4t10 on the PTR is misleading – it just says “adds at least 2 ticks to your Flame Shock”. However through extensive playtesting, we’ve confirmed that it works by adding however many ticks gives a time closest to 6 extra seconds (but still at least two ticks). So the precise number of ticks will vary depending on your haste, and more haste doesn’t necessarily mean a longer extension.


      Regarding Glyph of Flame Shock in ZAP! 1.2.4, the main glyph menu setting controlled whether or not Flame Shock could crit, and that was the DEP value it reported – the DEP value of critical flame shock ticks – even if the “new glyph” was turned on in advanced options. The advanced option to enable the new glyph didn’t affect what the old glyph did, if you see what I mean. It just added the 60% crit damage to the formulator’s output. This is because 1.2.4 was rushed out to fix bugs and I was only half way through the 3.3.3 coding.

      In ZAP! 1.2.5, everything is patch 3.3.3 native and there should be no confusion 🙂

      Posted by Charles | March 13, 2010, 5:49 pm
      • Ah, I see, I misunderstood the posts over at EJ ;). I think the next question is going to be the value of chain lightning again. As I’ve had to use compatibility mode, it doesn’t seem to update on changing whether CL is used or not, but I noticed the disclaimer on the EJ post 😉

        ToW still being a top league glyph just annoys me. It’s a ridiculous, clunky buff -_-

        Posted by pewter | March 13, 2010, 5:55 pm
  3. If one had a lot of crit, wouldn’t that increase the value of Glyph of Flame Shock?

    Posted by Crimsín | March 13, 2010, 8:21 pm
    • Indeed it would, but look at the relative value – Glyphs of LvB and FtW are worth about 50% more DPS! You’d have to inject a lot of pure crit to get them even close. In reality there shouldn’t be any situation where a player has that much crit without corresponding amounts of spellpower and haste which would push up the value of the glyphs it’s competing with even further.

      Posted by Charles | March 13, 2010, 9:55 pm
  4. Would you mind expanding a bit more on the Glyph of Flametongue Weapon / Glyph of Lava section? Are there any points at which the scale favours one over the other?

    If both remain equal, I’m pretty torn as to which to use. I’d rather not have to swap out glyphs for each fight. I’m more than happy with a set of glyphs that will suffice for most fights.

    Posted by Maker | March 15, 2010, 12:42 pm
  5. oh god, not flametongue

    Even if it’s “BiS” , should i assume that in high movement, high latency or at least not “stand and nuke” situations the Glyph of Lava would be slightly ahead ?
    The pure thinking of using a flat crit percentage glyph over an added spellpower mechanic to a wonderful spell makes me sad

    Posted by Hexlol | March 15, 2010, 1:42 pm
  6. At my gear levels, sadly it is Flametongue that draws out ahead. (And that is to say, 3020 spellpower before totems/imbues, 979 haste and 34% crit.) I’d play with the 3.3.3 spreadsheet to find the point where they might be roughly even.

    Posted by pewter | March 15, 2010, 2:36 pm
  7. I agree with the “sigh” sentiment regarding using yet another static bonus glyph that’s barely got any real interaction with our spells or talents. It’s just not very fun, is it?

    Glyph of Lava vs Glyph of Flametongue will most likely be a personal choice/playstyle thing. Some fights might marginally favour one or the other. High lag/low haste will almost certainly favour Lava. High movement might favour Lava. Otherwise Flametongue should come out slightly on top.

    (There are options to fiddle about with movement and lag in the spreadsheet, btw.)

    Posted by Charles | March 15, 2010, 4:21 pm
  8. Interesting. I figured the change to FS would still have the glyph being a choice glyph.

    Flametongue making a comeback…I would’ve figured Lava would pull ahead of it. All of my “figuring”, though, has zero math whatsoever to back it up, heh. Besides, seeing big LvB numbers is fun. 😛

    I’ll punch my stats in on Zap when I can and see what results I get. 🙂

    Posted by Kazgrel | March 15, 2010, 6:17 pm
  9. Ever so slightly off topic.. How long did it take you to get that last screen shot there? 😛

    Posted by Reb | March 15, 2010, 8:05 pm
  10. @Kazgrel: Yes, when I first saw the announcement that’s what I thought too. But when I saw the details of the planned new glyph I realised it was going to be underwhelming though perhaps not quite so far behind the others.

    Situations can change of course, but so far all the numbers seem to agree that it’s just not very good, which really does seem a shame 😦 I have a feeling a lot of people will end up using LB, Lava and Flame Shock glyphs anyway because it “seems good”.

    @Reb: I think I stood around for just over a minute before I got that screen shot. Chayah just happened to blink as I snapped it, which meant I had to wait for another minute or two to get the shot I actually *wanted*. Then I realised that I’d accidentally ended up with a bit of an impossible shot to intentionally get, so was quite pleased 😀

    (Discussing the pictures in my posts is never off topic!)

    Posted by Charles | March 16, 2010, 3:32 pm
    • One thing I’d be really curious to see is a comparison between the glyphs of Lava and Flametongue, detailing what situations would make each the “better” glyph, assuming a spreadsheet DEP difference of about 25. For example, are low haste and high movement scenarios the only ones that would favor Lava, and if not what other situations would favor it?

      I’ll likely be sticking with FT as my third glyph, but I think some sort of comparison like that would be a great tool for the community – perhaps in a “How to choose your third glyph” type of post/question!

      Posted by Shkarn | March 16, 2010, 6:02 pm
      • There’s already a similar sort of thing posted by Abakus on the EJ forums:
        Though I should note that the “movement per LvB” setting in ZAP! is pretty extreme and will apply to very few fights (moving every 9 seconds is not very common in ICC).

        Of course in general it’s easy to say what circumstances will “improve” the Glyph of Lava relative to Glyph of FtW (high movement, high lag, low haste), but specifically it’s much harder to conclusively evaluate which is “best”. I’d be nervous of doing such an attempt at a conclusive evaluation at this stage given that the new spreadsheet version hasn’t been tested against any other sources yet. If Irdx ever releases his Elemental_HEP-style log parser then it might get easier to make such conclusions!

        For now I think I’d lean towards saying “depends on your playstyle” and leave it at that, suggesting that folks use the spreadsheet or other tools (like Team Robot’s sim) to get a second opinion and carefully evaluate their own playstyle compared to how the spreadsheet is constructing their rotation.

        …except now I’m wondering whether all the above might make suitable material for a post. Hmmz.

        Posted by Charles | March 16, 2010, 11:47 pm
  11. I wouldn’t really say it has to be a conclusive post – I think it would be a great discussion post, though. I agree that it’s going to wildly vary by boss fights and playstiles, but I think it would make a great blog post discussing the merits of each in as much detail as possible!

    *insert obligatory I kinda told you so comment* 😉

    Posted by Shkarn | March 17, 2010, 1:48 am
  12. Hey Charles, would you say that Glyph of Lava would produce more burst DPS than Glyph of Flametongue? Our burst DPS is incredibly important on Heroic Putricide and Heroic Lich King, so I’m wondering what your thoughts on these glyphs’ burst values are. I assume Glyph of Lava would be better, but I have no numbers to back that up.

    If Lava will produce more burst DPS, then that’s probably the one I’ll be using. It seems more dependable than Flametongue.

    Posted by Bergg | March 19, 2010, 4:36 am
    • I think it’s hard to say, really! On the one hand is 2% more chance for all your spells except Lava Burst to crit and do 209% damage, and on the other is a guaranteed extra ~1k damage on Lava Burst casts.

      Mathematically speaking, if you’re not getting a full rotation but have enough time for two Lava Bursts, LvB wins out. If you only have time for one LvB and a few LBs, then Flametongue wins out. It’s a ratio thing.

      So I dunno, what do you think? Mathematically, we apply crit as a flat damage increase to every spell (normal damage * (crit chance * crit multiplier) ), but obviously in reality it’s an all-or-nothing thing on each spellcast. Is the definite 1k extra damage on LvB which you wouldn’t get without the glyph worth losing a 2% chance for ~5k extra damage on LB that might happen anyway? Put like that it seems so, doesn’t it?

      Posted by Charles | March 19, 2010, 10:44 am
      • So from my experience, if we look at the first part of your reply, the ratio comment, you definately get enough time on Val’kyrs to do a couple LvB. Vile Spirits are up for 30s, though on normal we didn’t even dps these. On Volatile Oozes, I would say most of the time you can do 2 LvB per Ooze (and if not, only because of exessive running needed, rather than it dying in less than 10s).

        So, as long as I’m following, even in these high burst required fights, it seems Lava is still a clear winner, due to the fact that the burst situations are long enough to cast LvB at least twice more often than not. (And maybe you could even argue, if the oozes, for example, are dying quicker than you can cast 2, then you guys probably have it under control, and our extra 1k, or whatever, of damage per ooze probably isn’t needed, and thus should go with the glyph that works for the overall highest dps?)

        Looks like little changes with all this, we still want to collect lots of SP, haste, and maybe tweak our EM usage. Maybe I’m crazy, but I often find myself jealous of mages and other classes that get to re-glyph and re-spec for various fights. My contribution to the new challenges is the same as it always has been… Mage1:”Alright, let me spec into x” Warrior3:”K and I’ll glyph for z” Me:”And I’ll…do..what I always do”

        Posted by Acariel | March 22, 2010, 5:11 pm
  13. So now that our flame shock is being affected by haste, and that the new t10 set bonus will even further the value of haste, will haste pass spellpower point for point in either before or after t10 4 set?

    Posted by Octavaria | March 24, 2010, 4:47 am
    • In some formulations haste is indeed coming out very slightly ahead with 4t10, but on an item-budget-point basis it’s still generally behind. Some folk were reporting that Reckless gems looked like they might be superior to Runed gems at near-BiS levels, but the margins are very small and probably not enough to justify a change (haste is a more “fragile” or unpredictable stat than spellpower).

      Posted by Charles | March 24, 2010, 2:15 pm
      • Agreed. I’ve been playing around with simcraft, and with my character’s stats haste only becomes more valuable with t10x4 AND Lava Glyph (on both helterskelter and patchwerk style). Using FTW glyph has haste ever so slightly below spellpower in all scenarios.

        Of course, I am a simcraft noob and am still playing around with it.

        Posted by pewter | March 24, 2010, 2:22 pm
  14. I ran into a strange experience last week.

    I was raiding ICC25, some bossfight (can’t remember which one it was, think it was festergut HM), just mashing my rotation.

    I refreshed flame shock, and saw my timer on 43 seconds left (on the dot!). Turned out I flame shocked while having malible goo on me (silly me not moving fast enough..).

    So this got me thinking. Does this actualy improve my dps? since I dont have to spend a global on flame shock for a long time? Especially since you can throw in a heroism/elemental mastery after it, and mash that LB button like mad!

    I know it increases the time between dot ticks on Flame shock, but would’nt the extra LB’s outweight this loss?

    Any thoughts/math on this? 🙂

    Posted by Tiranimo | April 14, 2010, 10:18 am
    • Considering the 20 seconds you spend debuffed, it really wouldn’t increase your DPS to get hit by it deliberately 🙂 And yeah, off the top of my head (that is, based on calcs I did a while ago for 3.3.3) I’d say that a shorter, hasted flame shock is always going to be more DPS than a longer, slower one with more time for LBs.

      Posted by Charles | April 14, 2010, 12:40 pm
  15. When I originally commented I clicked the Notify me when new comments are added checkbox and now every time a comment is added I get 4 emails using the same comment. Is there any way you may take away me from that service? Thanks! gddffackaefk

    Posted by Johnb613 | May 26, 2014, 9:10 pm


  1. Pingback: Patch 3.3.3 – A Tale of Hasty Theorycrafting « The 'mental Shaman - March 13, 2010

  2. Pingback: Patch 3.3.3 And Elemental Changes « Totemz - March 13, 2010

  3. Pingback: 3.3.3 Hits Live This Week « The 'mental Shaman - March 23, 2010

  4. Pingback: Glyphs: Lava vs. Flametongue « Planet of the Hats - March 24, 2010

  5. Pingback: Elemental Glyphs Post Patch 3.3 « Totemz - March 26, 2010

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

Get emailed about new posts!

Join 27 other followers

We write about

%d bloggers like this: