Shadow: 4.01 Checklist - (DPS questions? Read me first)

[BDF]

I did a quick search of the wow forums and there, but couldn't find blizz's mention of dot changes.

My question is...so we still need to refresh our Dots after heroism is popped for them to benefit from the haste? or what about spellpower from things like DFO?

[Kelesti]

All Damage over Time calculations are done when they are cast (or in Pain's case, refreshed).

If you have Vampiric Touch halfway through its duration, and someone uses Bloodlust, that DoT doesn't speed up, you would have to recast it. Spell Power and Crit are treated the same.

So you would have to recast every DoT to take advantage of it, in which case that's lost Mind Flays/Shadow Orbs that you're not getting.

General idea: Refresh when you have to, not because you gain something.

[Limstella]

Do you think SWD is worth casting now at 25%?

[Worshaka]

36.5K crits at execute range pretty much difinitivly says to me it is... with a decent crit rate your DPeT for SW (with the ICC buff of course) is something around 22K, it's slightly ahead of a 3 orb MB (assuming 0 mastery of course).

[Jigawatts]

Yes the GCD used for SW: Death is definitly worth it in the sub 25% health range. Spam it when you can!!

[Worshaka]

Before I get to the point I want to make I'm going to explain my understanding of how DoT haste scaling is currently implemented (which may be incorrect), so if I am could someone please correct me.

As I understand it, haste no longer reduces the duration of a DoT but instead offers the ability for additional ticks during the duration. At 0 haste VT will provide 5 ticks in the 15 sec duration. In order to get 6 ticks, you need an amount of haste that will reduce tick intervals to 15/6.

Taking this concept further (which leads me to my point) at certain levels of haste you can have a situation where the last tick of a DoT is applied significantly before the duration of the DoT ends. Take the situation where you have a level of haste that is approaching the 15/6 tick interval but not there yet. That means the 5th tick of VT will occur sub 15 sec but the duration keeps going for 2 or so seconds.

That being the case if we refresh DoTs timed on the moment they expire are we introducing a delay? Meaning it would be benefical to reapply a DoT slightly early in order to minimise any sort of delay effect?

Again i'll reitirate my understanding of current DoT haste scaling could be incorrect, in which case could someone please educate me.

---------- Post added 2010-10-21 at 08:46 AM ----------

I did a quick search of the wow forums and there, but couldn't find blizz's mention of dot changes.

My question is...so we still need to refresh our Dots after heroism is popped for them to benefit from the haste? or what about spellpower from things like DFO?

I really think it's a case by case basis in which you need to make those decisions.

Firstly the remaining duration of the current DoT is going to skew the result of overwriting the DoT significantly. Take a situation in which you land a DoT moments before BL/Heroism. I suspect that reapplying that DoT would be worthwhile as opposed to a DoT that is about to expire very shortly.

The reason for that is if we take DP as an example, you will lose 24 seconds of 40% more ticks on DP, I suspect that 40% more DP damage is going to be greater than any damage you can do (perhaps excluding 3 orb MB or execute SWD).

So you're going to end up with a complex decision matrix... I tend to overwrite DoTs based on information presented to me based on the situation and the available information. For example if I know phylacatery is comming off ICD and I'm presented with a choice to reapply a DoT slightly early or slightly late I choose the later because it's likely to pick up the proc on application. If I'm dotting multiple targets I refresh SW:P on a target once with SW:P to lock in the proc on a target just before the proc expires and try not to refresh that SW:P until late in the SW:P duration.

All these sorts of things are to achieve relatively minor gains. So at the end of the day they aren't really necessary unless you're a bit OCD like me and enjoy squeezing every last % of performance out.

[Etheros81]

i cant find the changes on DOTs within the forum, but did they remove the haste on DOTs? and is that why crit is a better stat than haste for spriests?

[Thrayne]

i cant find the changes on DOTs within the forum, but did they remove the haste on DOTs? and is that why crit is a better stat than haste for spriests?

a) No.
b). It isn't is is almost the same value by most estimates. (haste still being slightly better)

[Rocketsurgeon]

Haste is supposed to shorten the duration of a dot/hot until it can add a tick at the original duration length.
I've been trying to dance around the haste plateau between 701 and 1016 rating shown on Elitist Jerks. I don't have the gear to hit four digit haste.

[Worshaka]

a) No.
b). It isn't is is almost the same value by most estimates. (haste still being slightly better)

I'm highly skeptical that crit & haste are smiilar with haste only being slightly ahead. Of course it depends on your definition of slightly.

Haste scaling decays at less of a rate than crit... the reason being is crit can only increase the damage of a spell by 100% where as haste allows you to cast 1 extra spell and then that spell can crit meaning haste increase the damage of a spell by 100% + your crit rate.

As you stack more and more haste it slowly decays and can theoretically reach a level of your crit. However for that to occur you'd need almost double the haste to crit and in my gear which is half 277 my crit and haste are at very similar levels.

Now there is this concept of soft caps, however I've always been highly skeptical of those as well. While it's true you can reach the hard cap GCD (which any well geared spriest would have) you still continue to scale DoTs in a linear fashion as well as MF. DoTs and MF make up a very large proportion of our damage (in the realms of 70% to 80%). So while this will increase the decay factor I don't think it can reliably push it to a point where crit becomes better.

I guess from a practical point of view it is perhaps possible to reach a level of haste that goes beyond your means to react to procs or spell queueing. That will be a very individual measure and with things at the 1 sec hard cap you can certainly be pressed to avoid reaction lag. Particular with spell queuing, you may need to change your mind on your next cast based on a trinket proc, orb proc, add spawn, encounter mechanic etc. To this extent you may wish to gear to a supposed haste soft cap and then look to stack a more passive dps increase like crit.

[Kelesti]

Unstickying this for now. I plan on dropping the clutter and rearranging this, so here's a bump to keep it on the front page while I figure out what to do.

Until then, going out to pick up my phone charger and watching Saw 3D, despite the fact the entire series has been crap since the amazing first one. Mind Game > Torture Porn.

[allaiva]

With the change to 4.0, Blizz has changed all dots to scale with both haste and crit as a baseline ability. Haste does not reduce the duration the spell time exactly, just decreases the time between ticks of the spell. One example of this would be Vampiric Touch. The baseline (0 haste at all) ticks off at 3s intervals for 15s (total of 5 ticks). Adding haste will reduce the time between ticks, while the spell still lasts for 15s total. So if you get enough haste for VT to tick at 2s intervals, you will get a full 7 ticks (and some spare time) of VT before the need to recast occurs.

As far as stat values go, Haste is still the preferred stat, but with the changes to SW: D and reimplementation of MB into the rotation crit has become a much better stat to have.

Another tidbit on DoTs are how they are reapplied during a recast. Instead of a new VT replacing the old, the newly casted VT is added onto what is currently there. So if you had .2s left on VT, as soon as the spell was applied you would have 15.2s remaining in theory based upon Blizzard posts about these changes that I've followed (though I may be wrong so please correct me if you know otherwise).

I'm not a heavy math/theorycrafter, but these are the changes I understand from blue posts and browsing shadowpriest.com