When to stop gathering haste?

[Fuzzyboy]

Missing a couple pieces of info there - crit also effects magma totems (which haste doesn't I don't believe) as well as other periodic things that can crit. Crit and haste both also effect healing spells - I would imagine this is where haste shines even more though - getting 4x chain heals off 20% faster is a HUGE benefit.

True with the magma totems - I guess it depends mostly on how you use them. I mostly use them on trash tbh. You're very right about the CH though!

[Gares]

Missing a couple pieces of info there - crit also effects magma totems (which haste doesn't I don't believe) as well as other periodic things that can crit. Crit and haste both also effect healing spells - I would imagine this is where haste shines even more though - getting 4x chain heals off 20% faster is a HUGE benefit.

True with the magma totems - I guess it depends mostly on how you use them. I mostly use them on trash tbh. You're very right about the CH though!

I'm pretty much on the boat of not gemming/enchanting for either unless no other option is availble (IE Cloak) other then that its hit->Spellpower as I really don't see any other way to do it to be honest.

[Owltoid]

for the shaman I would think haste and crit are equally important.
haste will increase your DPS by the same rate as a crit chance does.

clear casting and elemental oath aside....
a 15% crit chance means in every 100 casts you will crit for double damage on 15 of them ( equivalent to 115 casts, but the mana cost of only 100 casts )
a 15% haste means for the time it took you to cast 100 casts you can now cast 115 ( again 115 casts but the mana cost of 115 )

as you can see, both give the same amount of DPS increase.

I would personally favor crit over haste because of elemental oath and the clearcasting it causes.

On paper, you're correct. With actual playing lag and enters into play. The more spells you have to cast the more lag affects your DPS (time between casts isn't perfectly timed). In the simple scenario you posted, the haste would be more affected by lag and therefore yield lower DPS.

[Fuzzyboy]

On paper, you're correct. With actual playing lag and enters into play. The more spells you have to cast the more lag affects your DPS (time between casts isn't perfectly timed). In the simple scenario you posted, the haste would be more affected by lag and therefore yield lower DPS.

This is true only if you use a castsequence-macro, not if you use a castrandom macro (or manual button spamming with /cast). In the case of /cast and /castrandom the next spell is qued and isn't affected by lag.

[Alemi]

It's generally considered that haste is stacked until you can reduce your LB cast time to 1.6 seconds thereby squeezing 5 lightning bolts in your lvb rotation - at 400 haste i'm at 1.7 seconds. Every elemental shaman should add +10% crit to their character sheet, as has been mentioned (5% from Call of Thunder, and 5% from EO - Thundering Strikes is counted in your CS). So, most ele shaman sit at ~ 30% crit. You can assume 100% uptime on EO as long as you're not missing your flame shocks/lava bursts. From then, assume that you use Lava Burst every up. So you have an 8 Second spam window for LB - squeeze in 4 lightning bolts. The dps increase from going up 5% on crit vs. stacking haste is as follows, and dps associated (I can break out the math if people would like, I'm that much of a dorkus...) using my baseline hits for LB - crit: 7,513 hit: 3,320

@ 30% crit
0 of 4 13,280 damage - 24.01%
1 of 4 17.473 damage - 41.16%
2 of 4 21,666 damage - 26.46%
3 of 4 25.859 damage - 6.56%
4 of 4 30,052 damage - 0.81%
Net DPS - 2256

@35% crit
0 of 4 - 17.85%
1 of 4 - 38.45%
2 of 4 - 31.05%
3 of 4 - 11.15%
4 of 4 - 1.50%
Net DPS - 2393

@ 40% crit
0 of 4 - 12.96%
1 of 4 - 34.56%
2 of 4 - 34.56%
3 of 4 - 15.36%
4 of 4 - 2.56%
Net DPS - 2498

@30% crit w/ 5 LBs
0 of 5 - 16.81%
1 of 5 - 36.02%
2 of 5 - 30.87%
3 of 5 - 13.23%
4 of 5 - 2.84%
5 of 5 - 0.25%
Net DPS - 2861 dps

Stacking haste is FAR superior to crit.

Edit: added in 40% crit dps to show the horrible scaling of crit vs haste.

[Lyonheart]

JEBUS! i need more haste!

[Kaynin]

Ok, to all you haste junkies, here is why haste sucks:

:P

Consider this, to do most dps in long fights, we have to ideally use Lavaburst ASAP when the cd is up again. We all know this fact to be true. So what we do is fit in as much as possible within the time alotted by Lavaburst cooldowns.

Our rotation is basicly like this

FS->LvB->Lb's->Lvb->Lb's

As an example

Let's say you have X amount of haste and with that X amount of haste you can fs-lvb- 4 lb's-lvb- 3 lb's rotation without any loss on the lvb cooldown.

The amount of haste required to be useful after this point would be based on lvb cooldown, which is 8 seconds. Within 8 seconds you need to be able to fit in another spell, to calculate this:

suppose your cl is... 1.3 casting time (roughly) the amount of haste required to get 1.3 seconds more in an 8 second rotation is 16% haste extra, 16% haste will allow you to be able to fit in another cl into your rotation without loosing lvb cast time. adding quite some extra dps without losing lvb cooldown.

If you're still with me. It means that if you reach an ideal rotation, you will need approximatly 512 haste rating MORE before you will be able to do more dps then you already did.

So, I dunno, let's say you have 250 haste rating and have a solid rotation with lb's, maybe a es or cl or two in there that tightens your rotation the best possible way. If you want to improve your dps, you will either need to stock up on crit, or get 512 more haste rating to fit another cl into your rotation.



Figures are not exact, though pretty close and should give a good idea. I generally believe haste matters not, we are stuck with the LvB timer and messing it up, prolonging the cd too long, will severaly impact our dps. Hence haste is not the way to go, even though some calculators say our dps scales better with haste then with crit. For practical use it does not. Unless all you do is spam Lightning Bolts, then yes, haste is better, but you shouldn't be spamming Lightning Bolts.

[Alemi]

Figures are not exact, though pretty close and should give a good idea. I generally believe haste matters not, we are stuck with the LvB timer and messing it up, prolonging the cd too long, will severaly impact our dps. Hence haste is not the way to go, even though some calculators say our dps scales better with haste then with crit. For practical use it does not. Unless all you do is spam Lightning Bolts, then yes, haste is better, but you shouldn't be spamming Lightning Bolts.

CL is horrible to use in any sense of the word now. I can easily fit in 5 lbs now, and loose exactly 0.5 seconds on the lvb cooldown on fights like patchwerk where i stay put. On fights with movement, I can get my lbs off a lot faster. On fights like Malygos, you can see a HUGE difference in dps, FS/LVB/LB until LVB Is back, rinse repeat.

Regardless, stack what you want. I'm not yet at the 400 haste cap (500 with the totem proc), but practicality wise in both 25 man and 10 man settings, haste IS the way to go. Agree or disagree, that is up to you - crit scales horribly, do the math and you'll see.

Practicality wise, at 0 haste, you can't fit in 4 lbs into your Lvb rotation dealing with minor server lag and human error. Any haste at all and you'll be able to hit it.

[Hachoo]

I utterly disagree that CL is horrible. For boss fights, maybe, but on trash if I don't use CL my recount DPS drops ~500.

I'd also like to keep this MOSTLY on topic for multiboxing. Timers, perfect rotations, etc, are all moot points for 5 boxing IMO - no one is going to get the exact optimal rotation with 4 shamans and a tank unless they're using cast sequences which have their own major latency delay. I think most people here (when 5 boxing) use castrandom or castsequences so talking about needing X haste to make this exact rotation work under 100ms latency is kind of silly. It might be helpful when solo boxing but not 5 boxing.

[Kaynin]

Figures are not exact, though pretty close and should give a good idea. I generally believe haste matters not, we are stuck with the LvB timer and messing it up, prolonging the cd too long, will severaly impact our dps. Hence haste is not the way to go, even though some calculators say our dps scales better with haste then with crit. For practical use it does not. Unless all you do is spam Lightning Bolts, then yes, haste is better, but you shouldn't be spamming Lightning Bolts.

CL is horrible to use in any sense of the word now. I can easily fit in 5 lbs now, and loose exactly 0.5 seconds on the lvb cooldown on fights like patchwerk where i stay put. On fights with movement, I can get my lbs off a lot faster. On fights like Malygos, you can see a HUGE difference in dps, FS/LVB/LB until LVB Is back, rinse repeat.

Regardless, stack what you want. I'm not yet at the 400 haste cap (500 with the totem proc), but practicality wise in both 25 man and 10 man settings, haste IS the way to go. Agree or disagree, that is up to you - crit scales horribly, do the math and you'll see.

Practicality wise, at 0 haste, you can't fit in 4 lbs into your Lvb rotation dealing with minor server lag and human error. Any haste at all and you'll be able to hit it.

five lb's. Oo

I can barely do 4 with 250 haste and loose out on cooldown slightly in doing so. Are your figures bloodlusted or unbloodlusted?



Castrandom and castsequences make my dps drop like a brick btw, I click all spells manually.