Elemental Shammie DPS Optimization and Recount
facema...@yahoo.com
all too common in raids these days (e.g. camping the boss, unnecessary
AOEing mobs instead of DPS the boss).
I have had the sense that my DPS was lower than it show relative to
comparably equipped people. Today, ran the weekly (Naxx,
Plaguebringer) with two other shammies. We knocked the boss down
before he had a chance to teleport.
The recount results were puzzling.
Me: GS over 2900
Sham2: GS: 2478
Recount reports:
Me: DPS 6795, Damage 560K
Sham2: DPS: 7764, Damage 613K
(Opposite of how it reported the totals for the fight--completely
different values)
So I am wondering what I am doing wrong. So I copied all the recount
data for the fight to a spreadsheet. When I add up all the damage I
get:
Me: 509K
Sham2: 479K
Adding the damage from each spell comes to a different sum than what
Recount reports for the entire battle.
** Why might Recount be reporting the figures differently? **
Even if these figures are correct, it seems there ought to be a much
greater disparity in damage.
These are the base stats:
Sham2 Me
SP 2434 2710
HIT 212 342
Crit 15.04 19.32
Haste 556 686
Both are identically spec 57/14/0
I have Glyphs of Lava, Flameshock and Lightning Bolt
He has Glyphs totem of wrath, flametongue weapon and lighning bolt
Oddities that stood out:
Counts of spells in the battle:
Lightning 45 42
CL 34 32
LaveB 6 7
ThunderS 5 0
Fire Nova 6 0
FS 3 5
I did not use Thunderstorm and Fire Nova because I was away from the
pack.
That said, I was casting as fast as I could and had no delay getting
into the fight. However, the other Shammie has gotten of 13 more
spells. 9 of those are instant cast---but they still take time.
** How could there be such a disparity in the number of casts? **
My LavaB Crit rate was 100%, his was 16.7%
My My Lightning crit rate was 40.5% his was 35.6%
However my CL crit rate was 28.1% and his was 50%
Obviously he was not keeping his FS up. However, if you look at just
chain lightning and lightning bolt, he was criting 41% of the time. I
was criting 35% of the time, in spite of my having a 28% higher crit
rating.
**What would account for this inversion of crit rates between LBolt
and CL? **
The flight lasted less than 110 secs (Noth was dead before getting to
Phase 2).
Recount lists him throwing off 5 thunderstorms in a 2 minute fight.
There's a 45 sec. CD on thunderstorm. It should take a minimum of 180
seconds to get that many off.
My rotation in a multi-target fight like this is:
Keep Flame Shock Up.
Use Chain Lightning whenever available
Use Lava Burst whenever available.
Use Lightning bolt to fill in the gaps.
So, are there any elemental shammie experts who might explain this
all?
(Can supply more data if necessary.)
John Gordon
> Recount lists him throwing off 5 thunderstorms in a 2 minute fight.
> There's a 45 sec. CD on thunderstorm. It should take a minimum of 180
> seconds to get that many off.
Was Recount reporting just for that fight, or for the preceding trash
as well?
> So, are there any elemental shammie experts who might explain this
> all?
What totems were you both dropping? It would be silly for you both
to drop Wrath. Did he drop searing/magma/fire elemental? That might
account for some extra damage.
--
John Gordon A is for Amy, who fell down the stairs
gor...@panix.com B is for Basil, assaulted by bears
-- Edward Gorey, "The Gashlycrumb Tinies"
facema...@yahoo.com
> In <98eb7f63-3b52-4c02-88b6-d5b670e9a...@q16g2000yqq.googlegroups.com> "faceman28...@yahoo.com" <faceman28...@yahoo.com> writes:
>
> > Recount lists him throwing off 5 thunderstorms in a 2 minute fight.
> > There's a 45 sec. CD on thunderstorm. It should take a minimum of 180
> > seconds to get that many off.
>
> Was Recount reporting just for that fight, or for the preceding trash
> as well?
>
> > So, are there any elemental shammie experts who might explain this
> > all?
>
> What totems were you both dropping? It would be silly for you both
> to drop Wrath. Did he drop searing/magma/fire elemental? That might
> account for some extra damage.
I use agility (groups usually want unless fear totem needed), wrath
(SP), mana, wrath (haste).
If he used a fire totem of some sort, I did not see it and it does not
appear in recount. I do not know if it would.
In any event, he got in a lot more casts than I did.
John Gordon
> In any event, he got in a lot more casts than I did.
That would suggest he has more haste than you. However, in your original
post, you list your haste rating as about 100 points higher than his.
Maybe he has some gear with activated haste or haste procs, for example
Shard of the Crystal Heart or Totem of Electrifying Wind?
Maybe he was getting the Power Infusion buff from a friendly Priest?
Is he a Troll, which would give the Berserking racial?
Catriona R
>In <b9f33c0a-3b64-4661...@j1g2000vbl.googlegroups.com> "facema...@yahoo.com" <facema...@yahoo.com> writes:
>
>> In any event, he got in a lot more casts than I did.
>
>That would suggest he has more haste than you. However, in your original
>post, you list your haste rating as about 100 points higher than his.
>
>Maybe he has some gear with activated haste or haste procs, for example
>Shard of the Crystal Heart or Totem of Electrifying Wind?
>
>Maybe he was getting the Power Infusion buff from a friendly Priest?
>
>Is he a Troll, which would give the Berserking racial?
And were you using Elemental Mastery (which I always forget about)?
Sounds like he was, maybe he also used a +haste buff food.
--
EU-Draenor:
Sagart (80 Undead Priest)
Tairbh (80 Tauren Druid)
Buinne (80 Troll Shaman)
Balgair (73 Human Rogue)
Eilnich (70 Blood Elf Warlock)
Ruire (70 Blood Elf Paladin)
notsmart
<facema...@yahoo.com> wrote in message
news:b9f33c0a-3b64-4661...@j1g2000vbl.googlegroups.com...
I thought you had to change a setting to have recount display totems or pets
seperate, otherwise their damage was included with the character's damage?
Shiflet
"notsmart" <nots...@aol.com> wrote in message
news:hlhsdp$qt6$1...@news.eternal-september.org...
> I thought you had to change a setting to have recount display totems or
> pets seperate, otherwise their damage was included with the character's
> damage?
Correct. If you don't have it set to display pets, all your pet's damage
will be shown with your character(and if you don't have that option checked
and mouseover a toon's name in recount, it will only show the pet that did
the highest damage for your character, which means if the character used
more than one, say fire elemental AND magma totem, only one will be shown).
ti...@thsu.org
wrote:
> I use recount as a diagnostic tool and avoid the DPS games that are
> all too common in raids these days (e.g. camping the boss, unnecessary
> AOEing mobs instead of DPS the boss).
You really should use WorldOfLogs.com to do post-fight analysis.
Recount just doesn't save the right data when it comes to a
postmortem. Plus, you can post a WoL, and we can all peruse it, to
help you out.
Lots of things can affect the dps, which you need to know about.
Elemental Mastery, 2xT10, haste food, haste potions, Black Magic
enchant, Hyperspeed Accelerator, trinket procs, etc. In addition, how
much was he actually dpsing the boss vs the adds?
But my guess is that he's using more cooldowns than you are, and that
he probably used consumables.
--
// T.Hsu
Shiflet
<ti...@thsu.org> wrote in message
news:d1c256a6-5fde-42f7...@k41g2000yqm.googlegroups.com...
> But my guess is that he's using more cooldowns than you are, and that
he probably used consumables.
Aye, seems likely. Someone else mentioned the fire elemental, even just
dropping that can be a nice DPS boost. I dunno for ele, but I know my
enhance shammie's elemental melees for around 1.5k average non-crit,
fireblasts for about 1.2k non crit, and his firenova usually hits for 1.7k
non-crit(and he does a whopping 150-200 damage to anything that hits him, if
you wanna get technical). So even though he's only around for 2 minutes, he
can put out a nice chunk of damage, especially if there's adds or multiple
mobs so his fire nova can hit several enemies at once.
Lancelet
53, I have a talent that can improve this spell, but don't have the spell.
Shiflet
"Lancelet" <lancele...@merci.invalid> wrote in message
news:Xns9D235D41C93E5la...@217.112.180.250...
>I have a question about LavaB. When do you get this spell? I'm level 52 or
> 53, I have a talent that can improve this spell, but don't have the spell.
Rank 1 is trained at level 75.
Psycho
wrote:
<snip>
> Oddities that stood out:
>
> Counts of spells in the battle:
>
> Lightning 45 42
> CL 34 32
> LaveB 6 7
> ThunderS 5 0
> Fire Nova 6 0
> FS 3 5
> I did not use Thunderstorm and Fire Nova because I was away from the
> pack.
>
> That said, I was casting as fast as I could and had no delay getting
> into the fight. However, the other Shammie has gotten of 13 more
> spells. 9 of those are instant cast---but they still take time.
>
> ** How could there be such a disparity in the number of casts? **
I don't know when you start to cast your spells. If you actually wait
till they are finished "client side" before starting to cast the next
spell you are losing a bit of time with each non-instant spell. That
might be old news for you, but one never knows.
I believe there is even an addon which shows you when you are in fact
able to start the next cast, but it is easily possible to do without
one.
> My LavaB Crit rate was 100%, his was 16.7%
> My My Lightning crit rate was 40.5% his was 35.6%
> However my CL crit rate was 28.1% and his was 50%
>
> Obviously he was not keeping his FS up. However, if you look at just
> chain lightning and lightning bolt, he was criting 41% of the time. I
> was criting 35% of the time, in spite of my having a 28% higher crit
> rating.
>
> **What would account for this inversion of crit rates between LBolt
> and CL? **
That's hard to say, it very well could be just a too small sample to
really say that something special happend. Maybe that's just been the
RNG not being kind to you.
> The flight lasted less than 110 secs (Noth was dead before getting to
> Phase 2).
>
> Recount lists him throwing off 5 thunderstorms in a 2 minute fight.
> There's a 45 sec. CD on thunderstorm. It should take a minimum of 180
> seconds to get that many off.
Hm, no idea here. As I'm not playing a Shaman I don't know how this
would be possible. Are there any talents to shorten that cooldown? Or
some ability which resets all or some cooldowns? I vaguely can
remember a talent which automatically produces a second lightning-
spell under certain circumstances, could there be anything like that
for thunderstorm too?
Psycho
i|||| | | || ||| || |||| 2.0
"Psycho" <ysp...@gmail.com> wrote in message
news:2cafb379-3249-4a94...@15g2000yqi.googlegroups.com...
> That said, I was casting as fast as I could and had no delay getting
> into the fight. However, the other Shammie has gotten of 13 more
> spells. 9 of those are instant cast---but they still take time.
>
> ** How could there be such a disparity in the number of casts? **
I don't know when you start to cast your spells. If you actually wait
till they are finished "client side" before starting to cast the next
spell you are losing a bit of time with each non-instant spell. That
might be old news for you, but one never knows.
I believe there is even an addon which shows you when you are in fact
able to start the next cast, but it is easily possible to do without
one.
------
quartz addon
If you have 500 ms ping and are waiting until the interface says you can
cast, or using mouse, you are losing 30%-50% dps.
> My LavaB Crit rate was 100%, his was 16.7%
> My My Lightning crit rate was 40.5% his was 35.6%
> However my CL crit rate was 28.1% and his was 50%
>
> Obviously he was not keeping his FS up. However, if you look at just
> chain lightning and lightning bolt, he was criting 41% of the time. I
> was criting 35% of the time, in spite of my having a 28% higher crit
> rating.
>
> **What would account for this inversion of crit rates between LBolt
> and CL? **
That's hard to say, it very well could be just a too small sample to
really say that something special happend. Maybe that's just been the
RNG not being kind to you.
> The flight lasted less than 110 secs (Noth was dead before getting to
> Phase 2).
>
> Recount lists him throwing off 5 thunderstorms in a 2 minute fight.
> There's a 45 sec. CD on thunderstorm. It should take a minimum of 180
> seconds to get that many off.
Hm, no idea here. As I'm not playing a Shaman I don't know how this
would be possible. Are there any talents to shorten that cooldown? Or
some ability which resets all or some cooldowns? I vaguely can
remember a talent which automatically produces a second lightning-
spell under certain circumstances, could there be anything like that
for thunderstorm too?
-----
Probably better to ask the other player, we are just guessing.
Shiflet
"Psycho" <ysp...@gmail.com> wrote in message
news:2cafb379-3249-4a94...@15g2000yqi.googlegroups.com...
> Hm, no idea here. As I'm not playing a Shaman I don't know how this
would be possible. Are there any talents to shorten that cooldown? Or
some ability which resets all or some cooldowns?
No.
> I vaguely can remember a talent which automatically produces a second
> lightning-spell under certain circumstances, could there be anything like
> that
for thunderstorm too?
There isn't. Beyond that, in raids, Thunderstorm isn't used for damage
anyways, it's used to regen mana and nothing more. An ele shaman shouldn't
be close enough for the TStorm to even hit the boss.
John Gordon
> I don't know when you start to cast your spells. If you actually wait
> till they are finished "client side" before starting to cast the next
> spell you are losing a bit of time with each non-instant spell. That
> might be old news for you, but one never knows.
I thought they did away with this? I can't remember what patch it was,
but I swear I remember reading that they did something to client-side
spellcasting such that Quartz and the like were rendered useless.
PV
>I thought they did away with this? I can't remember what patch it was,
>but I swear I remember reading that they did something to client-side
>spellcasting such that Quartz and the like were rendered useless.
Not in the least. Quartz calculates the client-side lag for you and shows
it with a red section of the casting bar. If you cast anything while your
current cast is inside that bar, the next spell starts up immediately.
If you don't use a cast timer like quartz, your DPS will have dead time. *
--
* PV Something like badgers, something like lizards, and something
like corkscrews.
John Gordon
> Not in the least. Quartz calculates the client-side lag for you and shows
> it with a red section of the casting bar. If you cast anything while your
> current cast is inside that bar, the next spell starts up immediately.
But you can't just cast the new spell, right? You have to explicitly
/stopcasting the current one, which will then successfully finish casting
anyway because you're in the "red zone", thus giving you a head start on
the next spell.
I've never used Quartz but that's my understanding of how it works.
Is that right?
Jamie Kahn Genet
Any good castbar and unit frame addon will do this. I use Pitbull to
show when I can actually start casting the next spell.
--
If you're not part of the solution, you're part of the precipitate.
Jamie Kahn Genet
> In <vbCdnRjRKpheV-DW...@supernews.com> pv+u...@pobox.com
> (PV) writes:
>
> > Not in the least. Quartz calculates the client-side lag for you and shows
> > it with a red section of the casting bar. If you cast anything while your
> > current cast is inside that bar, the next spell starts up immediately.
>
> But you can't just cast the new spell, right? You have to explicitly
> /stopcasting the current one, which will then successfully finish casting
> anyway because you're in the "red zone", thus giving you a head start on
> the next spell.
>
> I've never used Quartz but that's my understanding of how it works.
> Is that right?
No, you just cast and it will work if server-side the previous spell has
already finished casting/CD. No need to stop a cast.
Steve Kaye
> In<vbCdnRjRKpheV-DW...@supernews.com> pv+u...@pobox.com (PV) writes:
>
>> Not in the least. Quartz calculates the client-side lag for you and shows
>> it with a red section of the casting bar. If you cast anything while your
>> current cast is inside that bar, the next spell starts up immediately.
>
> But you can't just cast the new spell, right? You have to explicitly
> /stopcasting the current one, which will then successfully finish casting
> anyway because you're in the "red zone", thus giving you a head start on
> the next spell.
>
> I've never used Quartz but that's my understanding of how it works.
> Is that right?
It was, but it isn't anymore. IIRC, the change that you half remembered
queued one spell if it was cast close to the end of the currently
casting spell's cast. This made the /stopcasting redundant.
steve.kaye
DarkRose
wrote:
> John Gordon <gor...@panix.com> wrote:
> > In <vbCdnRjRKpheV-DWnZ2dnUVZ_hJi4...@supernews.com> pv+use...@pobox.com
Right, I've used Quartz for quite a while now. Once your spell
"progress" indicator is in the red, click your next spell (or hit your
key, whichever, I'm a clicker) and it will cast immediately after the
current one with no downtime in between. In lower level instances
people seem amazed at my non-stop casting. Also, on the bar (lower
right I believe) it shows you your actual server lag, as your lag gets
worse, the red bar gets bigger to compensate.
John Gordon
> > But you can't just cast the new spell, right? You have to explicitly
> > /stopcasting the current one, which will then successfully finish casting
> > anyway because you're in the "red zone", thus giving you a head start on
> > the next spell.
> No, you just cast and it will work if server-side the previous spell has
> already finished casting/CD. No need to stop a cast.
If you can cast the next spell normally, how does Quartz provide any
benefit over just spamming the spell key?
Steve Kaye
>
>>> But you can't just cast the new spell, right? You have to explicitly
>>> /stopcasting the current one, which will then successfully finish casting
>>> anyway because you're in the "red zone", thus giving you a head start on
>>> the next spell.
>
>> No, you just cast and it will work if server-side the previous spell has
>> already finished casting/CD. No need to stop a cast.
>
> If you can cast the next spell normally, how does Quartz provide any
> benefit over just spamming the spell key?
Unfortunately, the WoW.com page* that I read this on refers to another
page for the technical details but that page has gone so I can't confirm
if I remembered the details correctly.
I think that the theory was as follows:
If you press a key too early then it goes to the server, the server sees
that you can't cast it because you're already casting and sends you the
fail message. It won't send another message to activate the ability
until this fail message comes back. If you press it a fraction of a
second too early then you have to wait for the message to go to the
server and back before you can try to use the ability again. This could
be a long time if you have high latency.
* http://www.wow.com/2007/11/05/dont-mash-in-2-3/
steve.kaye
PV
>But you can't just cast the new spell, right? You have to explicitly
>/stopcasting the current one, which will then successfully finish casting
>anyway because you're in the "red zone", thus giving you a head start on
>the next spell.
You are way behind the times! /stopcasting doesn't even work anymore for
this purpose.
>I've never used Quartz but that's my understanding of how it works.
>Is that right?
No. If you press a cast button while you are "in the red", the client will
stack the spell on top of the one you're current casting so that it will
begin the moment the current cast finishes. Zero downtime. *
PV
>> No, you just cast and it will work if server-side the previous spell has
>> already finished casting/CD. No need to stop a cast.
>
>If you can cast the next spell normally, how does Quartz provide any
>benefit over just spamming the spell key?
Spamming a key is another lag-causing event, and it also takes measurable
time to complete. Not a good idea at all. *
John Gordon
> No, you just cast and it will work if server-side the previous spell has
> already finished casting/CD. No need to stop a cast.
I've been playing with this and I have a question:
If I am casting a regular spell (such as Frostbolt), I can press the spell
key when the castbar is at about 90% complete and it will transition
smoothly into casting the next Frostbolt with zero lag between the two
casts.
But if the spell key is actually a /castsequence macro, it seems not to
work. I have to wait for the client-side cast bar to completely finish
before pressing the key again.
Do castsequence macros interfere with serverside spellcast prediction?