Elemental, Opinion, Shaman, WoW

Lava Burst: looking back, looking forward

Despite the somewhat fancy title, what follows is a slightly edited repost of a piece I originally posted in the Cataclysm alpha forums.  I had thought that the forums would be wiped or at least remain invisible after the beta started, but as this was not the case the original post is still there – but I wanted to post it here anyway both for my own records and to allow folks to comment, disagree, discuss etc.

I was trying to look back over the impact of Lava Burst on elemental’s playstyle, itemisation, scaling and stuff, and also look forward to what might be planned for it in Cataclysm.  From the time of posting until now, the only difference between Lava Burst (LK) and Lava Burst (Cataclysm) is the Lava Surge talent and the ability to Overload – we’re still awaiting the new 31-point talent trees and possible further revisions to the way elemental will play (Edit: 31-point trees are now up but not much has changed, will comment on them another time).  Unleash Elements was also still in its very earliest stages and wasn’t fully functional for Flametongue Weapon.

I was trying very hard to sound positive and constructive without just writing a stupid list of my own bad ideas or a whine session featuring my favourite pet peeves, and in the process I fear I may have not said what I wanted to say.  Nobody replied to it on the forums (I get the impression there weren’t (m)any elemental shamans in the test at that time) so I don’t know what folk thought of it.  Anyway, here’s what I wrote, and I’d be very interested in points of view on the topic: has Lava Burst been good? bad? both? What changes does it need or is it OK as-is?  What do you like or dislike about the spell and mechanic?

Lava Burst feedback

I want to briefly talk about some of the ups and downs of Lava Burst’s present mechanic, at least as it’s been perceived by members of the shaman community as expressed through message boards and blogs.  Historically Lava Burst has only been a concern (in PvE, at least) for elemental, but if it is added to Maelstrom Weapon then I suppose it will become an issue for Enhancement too.

As we all know, Lava Burst is a guaranteed critical strike if Flame Shock is active on its target when the spellcast finishes.  The spell receives extra bonuses to its critical damage which makes it much more attractive for it to crit than otherwise (though its DPSC is still the highest for any elemental spell even if it fails to crit).

Positive points of view

These are points of view that players might see as positive, though I’m aware not all of them may be considered a good thing from a developer point of view.

Fun

I think most folk agree that the present mechanic is, at least in the absence of external factors, a lot of fun!  It’s fun to smack Flame Shock then pull off that fast Lava Burst and see a big number flash up, listen to the Clearcasting proc sound trigger and know that our next two spells will be benefiting from that extra 10% damage.  It’s just a very pleasing package on its own.

Consistency

While Lava Burst’s base damage varies by a whopping 326 (compared to Chain Lightning at 138 and Lightning Bolt at just 100) – which, especially combined with Clearcasting/Elemental Oath interaction, can make Lava Burst crits vary dramatically in magnitude – it is still a pretty solid rock on which to build the elemental ‘rotation’.

Despite the variation in crit damage, there’s still far less variation in a Lava Burst crit than exists for Lightning Bolt normal/crit damage, especially when Overloads are factored in.  Thus Lava Burst is a regular, reliable spike of damage with a very narrow damage range compared to other spells.

Because it procs Clearcasting, it reliably adds the 10% damage bonus to the next two spells cast.  Under normal conditions an elemental shaman will only fire off 5-6 spells between Lava Bursts, which means that Clearcasting is guaranteed to be active for at least a third of those “filler” spells.

Its fast cast time combined with this reliable damage also makes it the one spell an elemental shaman will attempt to never delay even during periods of intense movement (e.g. the old Heigan “dance”), further adding to its perceived consistency.

Ramp-up time

One of the great strengths of Elemental is that we have almost no ramp-up time between zero DPS and maximum DPS.  All it takes is one Global Cooldown to cast Flame Shock and then we can hit with our hardest-hitting spell (yes, Lightning Bolt can do more damage in a single cast with crits and Overloads, but this is very random and it also takes longer to cast).  The rest of our ‘rotation’ time is essentially just filler until Lava Burst is ready again, and as there is no synergy between Lightning Bolt/Chain Lightning and any other spell, we can just “spam” them freely.

(Incidentally this seems to have had unfortunate ramifications for the spec in PvP, though I can’t really comment on this as I’ve barely tried PvP as elemental.)

Randomness

I wanted to keep this separate from the ‘consistency’ point even though they’re very closely related.  There is really no randomness to the elemental ‘rotation’.  I actually very much like this, because I find random procs very frustrating especially when they have a significant effect on DPS.  Although Clearcasting procs are, to an extent, random, with elemental critical strike chances as they are presently combined with the guaranteed uptime from Lava Burst they are very reliably random.  And there is nothing the elemental shaman can do or fail to do to cause procs.

Critical points of view (no pun intended!)

These are perspectives that might be considered a problem for players.

Value of critical strike rating

Lava Burst has been blamed for the poor scaling of elemental’s DPS with critical strike rating as compared to haste rating and spellpower.  There is some merit in this view as clearly having a spell which accounts for 15-25% of a shaman’s damage unaffected by crit is going to have an effect.  However, mathematical models adjusted to render Lava Burst subject to the same critical strike chance as other elemental spells show an insignificant rise in the value of critical strike rating – in one example, from about 70% of the value of a point of spellpower to 80% of that value (compared to haste at >100% of the value of a point of spellpower).  This isn’t significant because, even with Lava Burst’s huge crit modifiers now positively affecting the value of critical strike rating in this example, the elemental stat priorities are not even close to changing.  There would still be no normal situation where critical strike rating was preferable to haste rating.

A rather bigger culprit of elemental’s poor scaling with critical strike rating (as compared to haste rating) is the generally high crit rates in the present raid game.  This is partly symptomatic of item inflation through hard mode gearing, but was present even at the release of WotLK through elemental’s inherently very high crit chances.  This seems likely to remain a problem in Cataclysm especially with elemental’s mastery providing an innate bonus to critical strike chance.  That is to say, critical strike is much more valuable compared to haste when the shaman has a low critical strike chance.  The total effect of raid buffs, talents and glyphs in WotLK is to add 28% to an elemental shaman’s crit chance before gear and consumables:

  • 2% Glyph of Flametongue Weapon
  • 5% Thundering Strikes
  • 5% Call of Thunder
  • 5% Elemental Oath
  • 3% Totem of Wrath
  • 3% Focus Magic
  • 5% Improved Scorch (Etc.)

(Blessing of Kings adds another percent or so through intellect > crit conversion.)

This means that a naked elemental shaman in a raid, aside from being very cold and a little embarrassed, has already got 28% critical strike chance on his or her main spells.

Therefore it’s the complete package of Lava Burst guaranteed crit + high base crit + haste/crit rating conversion rates which renders critical strike rating an unambiguously third-placed stat for elemental.

While not a “problem” per se, the dramatic inferiority of crit compared to haste has left the elemental community very unhappy with items which have crit but not haste – mail items and caster shields especially.  This perception, combined with the difficulty of itemising mail for restoration as well as elemental while trying to keep some variety in loot tables, has contributed to the general dissatisfaction with elemental itemisation in some parts of WotLK.  If crit and haste were not so dramatically apart in value then elemental shamans would be far more willing to consider items which lacked one or the other stat.

Value of haste rating

Ironically considering the last point, there has been some concern regarding elemental’s scaling with haste rating because of:

  1. Lava Burst’s fixed, 8-second cooldown and
  2. Lava Burst’s 1.5s cast time which equals the global cooldown

(2) is partly an issue because of the very high haste ratings we have seen through itemlevel inflation in WotLK.  It has been relatively easy for shamans raiding cutting-edge content to get to a point where they have enough haste to push Lava Burst’s cast time below the global cooldown hard cap of 1 second, at which point the spell ceases to scale with haste in a meaningful way.  However, as elemental shamans have two different haste cooldowns – Elemental Mastery and Heroism/Bloodlust – we can anticipate that even with dramatically lower haste levels in Cataclysm, this concern will remain at least in part.

(1) has been a serious problem of perception in the elemental community, especially for new players.  The high damage of Lava Burst has encouraged players to try to fit specific numbers of Lightning Bolt casts in the window between Lava Burst cooldowns with no “gap” – and if there is a “gap” (that is, time between the remaining cooldown of Lava Burst and the final Lightning cast that is less than the cast time of another Lightning Bolt), to cast nothing and wait for Lava Burst to come off cooldown.  I think this is very counter-intuitive behaviour to encourage, even accidentally, for a caster spec!

I say “accidentally” because community theorycrafters (myself included!) have shown that in most cases casting nothing while waiting for Lava Burst to come off cooldown is lower DPS than casting something and pushing the Lava Burst cast back a little.  However there are still theoretical haste values at which it is in fact slightly higher damage to wait for Lava Burst’s cooldown and not cast anything.  The haste ranges are very narrow and the DPS gains very small – and the possibility of executing such an exacting timing requirement in-game almost zero – but the very existence of these possibilities is, I think, unfortunate and has had the effect of confusing a lot of players.

Interaction with 4t10

It’s worth mentioning that the 4-piece tier 10 elemental bonus has made trying to cast Lava Bursts as “tightly” as possible even more important.  Unfortunately the mechanics of Flame Shock’s interaction with Lava Burst and the set bonus feels a little awkward, because whether Flame Shock falls off or is refreshed seems to depend as much on lag as on player skill and timing, and has led to a situation where it is optimal to overwrite Flame Shock rather than simply re-cast after it has fallen off.

Sameness

While I made a big deal of how consistent elemental DPS is as a positive point, there are those who see it as a negative – that elemental DPS is “too easy” and “boring”.  I’m not one of those people; I think that with totems to manage, the raid depending on us for Heroism/Bloodlust, the possibility of extra raid roles like interrupting, dispelling, kiting, offhealing and so on and so forth, there is plenty for an elemental shaman to be thinking about while executing our fairly simple “rotation”.  It’s also my opinion that the rotation is not really as simple as it seems as trying to minimize the time between Lava Burst casts and maximise DPS during movement is often very challenging and a lot of fun.

But nevertheless elemental can be perceived – rightly or not – as a sort of a two-button spec, just mashing Lightning Bolt and Lava Burst, with little room for finesse or variation.

The new Lava Surge talent does potentially offset this in a very interesting way – flame shock ticks having a chance to reset the cooldown on Lava Burst – but I fear it doesn’t fit very well to the WotLK incarnation of Lava Burst being a massive damage nuke on a short cooldown with a fast cast time.  There’s not a lot of time in the 8-second cooldown window for those procs to happen in a way that the player can react to it meaningfully, especially if haste levels end up lower than they have been in WotLK.

I love the idea of Lava Surge and I think it would go a fair way to offsetting the above complaint.  However I also greatly fear the impact of adding a random proc element to our rotation – I value our present consistency and reliable DPS output – and I wonder if it’s a bit similar to other DPS talents like Frost Death Knights’ Rime.

I should also mention that Unleash Elements (or Unleash Weapon, whatever it’s going to end up being called!) has a lot of potential to be a really fun addition to the elemental toolkit, but obviously I can’t really comment much on it until there are some more solid numbers in place for it and our other spells at level 85.  I think it’d be great to have another spell to cast, especially one that promotes synergy with our other spells, talents and glyphs – but we need to have a reason to cast it.  As Lava Burst is one of only two elemental spells it presently affects, I think the direction Lava Burst takes in this expansion will be a big factor in the use of this new spell.

Concluding thoughts

As I’ve said, I know that what the players enjoy about a spell may not necessarily be things that the designers intended or support.  I particularly wonder about the lack of ramp-up time and consistency of elemental damage as “good”, and the crit/haste issues as “bad” – perhaps the former are unintentional artefacts of Lava Burst’s design, or the latter intentional hurdles placed in front of elemental shamans in an attempt to make gearing more interesting?  (though I hope not!)

While I’ve really enjoyed elemental in WotLK, it has definitely not been without its frustrations and I feel that many of these can be traced back, directly or otherwise, to Lava Burst.  I would love to see the spell get some serious attention in the Cataclysm development process even if it ends up not changing much, because it has become such a foundational part of the elemental ‘lifestyle’ and seems set to remain a defining part of our gameplay for the foreseeable future.

Discussion

17 thoughts on “Lava Burst: looking back, looking forward

  1. There is a lot here that I agree with.
    I am very much a fan of the more static rotation that we currently have as elemental shaman. The addition of random and unreliable procs doesn’t, to me, necissarily constitute a more fun rotation. Perhaps more challenging, but with all of our other things to pay attention to (as you said, with totem management, and general raid awareness) I don’t enjoy erratic procs.
    This is one of the reasons I started playing my shaman over my Mage during LK. Mages have procs which contribute to their damage, and that to me makes them unique in their own Mage-ish way. But they have no heals or totems, only really a rare decurse or counterspell duty.
    Ele Shaman have a consistant rotation that still keeps you on your toes.
    The consistancy allows me to pay closer attention to what is going on around me in a raid environment, and it doesn’t make the spec too easy. A good player will still shine through, we don’t need ten buttons to push with random procs.

    Posted by Tammerson | July 14, 2010, 6:56 am
    • The problem with a static rotation for a DPSer is that the skill gap cannot exist, and players cannot grow in personal capability beyond a certain point. That brings stagnation into the mix as an issue.

      Random procs are one way to allow the spec to grow in skill requirement. Hopefully we’ll see more of them.

      Posted by Shuuwki | July 14, 2010, 8:24 am
      • I disagree that a static rotation does not allow for a skill gap. Skill comes from understanding.

        Assuming two identically geared players, why would one perform better than the other? Excluding a lucky crit streak, the only differentiating factor is skill/understanding. Did one drop a fire dps totem instead of ToW since there’s a demo lock in the raid? Did one clip their flame shock in order to not have it fall off mid-cast during Lava Burst? Did one just blindly hammer out a rotation of FS>LvB>CL>LB>LB>LB or did they weave in CL as a filler to avoid delaying LvB?

        With a static rotation, it’s the finer things that differentiate who knows their class and who doesn’t. Random procs mask that to a degree. A good proc streak will mask mistakes a bit better. Conversely a bad proc streak will make you look like you suck. Our rotation will highlight anyone who screws up in the slightest. How many times have you briefly lost focus only to watch your dps drop by a good 500pts?

        How often in VoA do you see another elemental underperform given their gear? Those that underperform always attribute it to luck or that someone vastly outgears them. I assure you that going from full 251 gear to full 264 is not a 3-5k dps gap. It’s a lack of skill and understanding.

        Posted by Maker | July 14, 2010, 1:45 pm
      • I agree with Maker, in that I frequently see elemental shamans with great gear performing poorly or (more rarely) elemental shamans who’ve made poor gear choices performing exceptionally well. Even the choice of gear itself is an expression of skill which differentiates between players. Despite the surface simplicity and consistency of our damage rotation there’s still a very wide gap between players of different levels of skill, experience etc.

        It’s probably more difficult to utterly screw up your rotation as elemental (in the way, say, an arcane mage can end up OOM mid-fight or a feral druid can have no energy and all his bleeds/savage roar falling off at the same time), but I think it’s also more difficult to really scrape out those last few hundred DPS through clever use and timing of FS, CL, totems, movement etc.

        Posted by Charles | July 14, 2010, 2:05 pm
  2. But how can anyone not like the OMG 5 DIGIT CRIT WOOO factor of LvB? 😛

    Of course, pvp enthusiasts have been up in arms for a while over elemental’s burst damage potential. I don’t see this burst being as big an issue in Cataclysm, since health pools are going to increase significantly (least that’s what I read in the past); couple more health with resilence, and folks should be much more durable than they are now.

    As for pve, LvB is the ugly stepchild that can’t really be touched much without making the pvp crowd break out the pitchforks and torches. Having quick on-demand burst has proven to be invaluable in some situations (sup Saurfang Blood Beasts).

    I’m wondering if we will be able to pseudo-tab dot using Flame Shock for great Lava Surge justice…only time will tell. o.o

    Posted by Kazgrel | July 14, 2010, 7:40 pm
  3. What do you think about a Lock and Load type treatment with Lava Surge?

    *Flame shock ticks have a 30% chance to make your next Lava Burst not trigger its cooldown*

    Would definately solve the problem of the 8 second window having meaningless Surge procs.

    Posted by Santux of Dark Iron | July 15, 2010, 1:23 am
    • I think that style of mechanic would work a lot better, yeah, because as you say it would give us time to react and also make sure that the effective cooldown reduction provided is always significant enough to be useful.

      Posted by Charles | July 15, 2010, 4:27 pm
    • I think that Lock ‘N’ Load is an excellent mechanic (having played a survival hunter through Ulduar before switching to my elemental Shaman as my main). It’s very easy to keep an eye on your toolbar to see when the ability’s cooldown is reset, and then hit it again right away.

      You would potentially have to lower the coefficient on LvB with a mechanic like that, because it would allow you to cast two LvBs in one CD period, rather than merely shortening the CD down a second or two.

      Overall, though, I agree than when you only have an 8 second window to cast the reset LvB, you really only have about 8 seconds – 1.5 seconds (castime of LB), because right after a LvB cast, you will immediately begin casting LB or CL. Take that 6.5 seconds remaining, and if the lava surge procs mid-cast, during the middle of your rotation, you’ll have to wait another bit of time to finish your cast, meaning that you’d really get a best-case scenario of casting LvB 6.5 seconds earlier IF lava surge procs during the first LB cast after your normal LvB cast. What seems more likely to me is that you’ll get a proc somewhere in the middle of your rotation, leaving you with only a couple GCDs to react. If Lava Surge procs right before your normal LvB cast, then you get no benefit at all. This introduces huge RNG questions and could potentally give you awesome dps when you get lucky procs and a string of LvB only seconds apart, or unlucky with only procs towards the end of your LvB CD.

      I suppose you could make a conditional macro that casts LvB first as a priority, and LB if LvB is on CD. Is that type of macro still possible? The problem I’d have with that is it would start to become more like DKs macroing in Rune strike to their other strikes, or hunters in TBC mashing one button with their entire rotation spelled out.

      The Lock N Load mechanic, imo, would be much preferable, allowing you to still have your regular LvB casts, but gives you a chance to work in the extras with a more dependable increase in dps.

      Posted by Vanyr | July 15, 2010, 5:46 pm
  4. My biggest problems with elemental are that it just doesn’t scale all that well, searing totem sucks, and we are absolutely god awful while moving.

    The rotation is fine in my opinion, and is just as interesting as almost any other caster rotation out there. It’s fun to see how long you can keep the flame shock up while moving / casting as many Lavabursts as you can. Of course, I played a destro lock for most of TBC, and their rotation was literally CoE, SB, SB, (5 more minutes of SBs), repeat. Anything is a bit more active than that.

    The main problem is the fact that we are essentially a turret who has to stand still (within 20 yards of the boss), and require a demo lock in the raid to do competitive DPS. During the high movement fights, the whole key to DPS is minimizing movement, since you do nothing outside of flameshock dps for that time.

    Also, we just don’t scale all that well. I’m in pretty much all BiS non-heroic 25 / heroic 10 stuff, and my theoretical DPS is still very weak next to a comparably geared / skilled player of damn near any class / spec, excluding ridiculous crap no one plays like frost mages and BM hunters..

    Some of the cata changes seem like they may help this. I’m fairly hopeful. Although I may go back to something else depending on what it looks like once we roll out the new talent trees to live.

    Posted by Enlightning | July 15, 2010, 6:00 pm
  5. This is all pretty spot on for me. Especially the fun factor. Considering our limited arsenal and how often we cast it, it never gets old, especially when it gets a killing blow. I feel like a kid.

    Your comments on haste are interesting. I’m at the stage where I do have to wait a small gap between bolt and burst. I’ll try casting an extra bolt and see how it works out for me.

    Having mad latency doesn’t help.

    Posted by Anaia | July 15, 2010, 11:55 pm
  6. After reading the EJ thread pertaining to Lava Surge, it makes sense that the Lock n Load mechanic is not something we want for Lava Burst.

    Instead, we need to invent ways to make Lava Surge ‘unwastable’ in the form of bad proc timing. Two ideas to start:

    1.) Lava Surge proc increases damage of next LvB by 10% (in addition to resetting cooldown)

    -or-
    2.) Lava Surge proc increases Flame shock duration by 1/2 damage periods (in addition to resetting cooldown)

    -or-

    3.) Lava Bursts cast after a Lava Surge proc debuff target with a “Flame Shock vulnerability” increasing Flame Shock DoT damage much like Haunt does for Warlocks (in addition to resetting cooldown).

    Just some random thoughts I suppose.

    Posted by Sanctux of Dark Iron | July 23, 2010, 5:57 pm
    • Most of the posts I read in that thread were firmly in favour of a lock ‘n load mechanic! The only objection I saw raised was that it would make the talent scale poorly with haste, but Bink’s post on the topic (which I see you’ve replied to also) seems to disprove that by showing that Lock ‘n load would scale better with haste. Unless I’m reading it wrong 🙂

      Posted by Charles | July 23, 2010, 6:53 pm
  7. My objection (to my own suggestion?) is that it is boring and not very skillful.

    Everyone has a different perspective and has different goals I suppose, so it all depends on what your objectives are. If it is purely to increase LvB’s scalar with haste, then LnL looks to be better than current Lava Surge.

    My goal (in my head) is to think of something that both increases scaling of LvB with haste (and maybe even crit if we try hard enough?) as well as increase the skill of the Elemental rotation, and make it even more interesting.

    Also, I can see a LnL treatment being too strong in PvP, especially with the new EM damage bonus.

    Posted by Sanctux of Dark Iron | July 23, 2010, 8:12 pm
    • I think I agree, and would love to see a more creative implementation – but I’d also like to see one which works and isn’t too complicated. I was thinking about this the other day actually, how much I’d like to see something clever and synergistic for elemental vs how much I’d like it to stay simple and “fun”. And I think I prefer simple and fun, but it’s hard to say without trying something new.

      Posted by Charles | July 23, 2010, 8:44 pm

Trackbacks/Pingbacks

  1. Pingback: Elemental Exposé (well it’s a round up without any rope) | The 'mental Shaman - July 14, 2010

  2. Pingback: Lava Surge, Part Two « Blogenstein - July 20, 2010

  3. Pingback: Lava Surge, Part Two - July 25, 2010

Leave a comment

Get emailed about new posts!

Join 27 other subscribers

We write about