Let’s make an additional assumption as an extension of #1 above:
1a. Illusion generation is not a factor. Obviously it is, but you can get it down pretty well with Deceptive Evasion. We can come back to it later if we need to.Traited cooldown for Mind Wrack is 10.2s, CoF 21.25, Diversion 12.75, Distortion 16.15.
Duration for torment is 12s, Confusion is 6s, by your numbers.
If you can keep all four shatters on cooldown, never shattering at less than 3 illusions:
Mind wrack: 4.67 stacks of torment up with 100% uptime, 4 stacks of confusion with 58.8% uptime.
CoF: 4 torment with 56.5% uptime, 8 confusion with 28.2% uptime
Diversion: 4 torment with 94.1% uptime, 4 confusion with 47.1% uptime
Distortion: 4 torment with 74.3% uptime, 4 confusion with 37.2% uptimeNormalizing these numbers,
Average torment stacks per second: 13.67
Average confusion stacks per second: 7.98At the per-tick damage numbers I quoted above, average per-second damage is
No triggers: 1995.9
Torment but no confusion triggers: 3381.3
Confusion but not torment triggers: 2443.4
Confusion and torment triggers: 3838.8So an always-moving enemy that makes no attacks will be close but not quite to our bleed numbers.
It’s not until you add some skill usages on an enemy that moves a lot (or an enemy that moves a little but attacks like a thief) that you pass up the bleeds respectably.How many more bleeds would we need to catch up to full triggers?
3828.8-3400 = 428.8. 428.8/136 = 3.2 stacks of bleed. So 28.2 stacks of bleed is evenly matched to the shatter build with a constantly-moving enemy that makes 2 attacks every 6 seconds.
What if the enemy makes more attacks than that? Well, we can calculate how many stacks of bleeds we’d need to make up for one extra attack every 6 seconds.
At 168.25 damage per attack, one extra attack every 6 seconds gives 28.04 extra damage per second on average per stack. With average stacks at 7.98, that’s 1342.6 extra damage per extra attack in 6 seconds. That means you need 9.87 stacks of bleed to make up for it.
So if there’s anywhere for the confusion/torment build to gain the most ground, it’s here. Even a single extra attack in 6 seconds is almost as much extra dps as the entire torment moving-penalty (3828-2443=1385).So against enemies that use a skill 4 or more times per 6 seconds, shatter wins.
Against enemies that attack 3 or more times per 6 seconds, and move a fair bit, shatter wins. Against anything else, bleed wins.
Are there other factors that can change our math?
^ Your calculations cant compare to 100% of bleeds because -
while the 100% bleeds your calculating is with everything included, your auto attacks, clone and phantasm attacks included. you probably apply 3-5 stacks of burning at max while sitting in staff, cause there’s no way you can play bleeds in scepter, correct me if i’m wrong here.
what your calculating is only the shatters damage without any application of exterior skills, namely confusing images, lets not add illusionary counter. but there’s still scepter auto attack granting you 4-5 stacks of torment, not to mention the bleeds that your illusion crits.
Your calculations above are for example – 20 seconds of bleed build all skill usage vs 20 seconds of shatter skills only*** (this is only an example, but i hope you get where i’m coming from)
First are our assumptions.
Loosening assumption 1 to assumption 1a still leaves us wondering if you can keep up the kind of illusion generation needed for constant shattering. If this is not true, the shatter build loses even more average dps. I’ll say this could be reasonable, though. I really don’t know, I don’t use f3/f4 offensively that much. I know I can keep f1/f2 on cooldown if I want to, using Deceptive Evasion for clone generationWe already tackled Assumption 2 by considering how much faster an enemy would need to attack.
Weapon skill rotations is too complicated, imo. At the point you start to account for that, I think you’d need to start doing real-kill comparison trials, as things start to get really complicated really fast. I think the basic assumption is theoretically sound, however, given that both builds appear to be using the same weaponry. There is some consideration to be given to the shatter build’s focus on confusion/torment durations over bleeds, as they’ll get more mileage out of the scepter weapon skills.
Stacking bleeds up to 25 or higher? You’ve already admitted you think the norm is about 30, so we’re agreed that my estimate of 25 was conservative.
^ Lets not change the assumptions now for keeping it strictly within bounds of your calculations
Is there anything else?
Yes, actually.
(edited by Telekinesis.8312)