killing undead, which weapon set is better

[Manko-ALE]

i have a level 10 finesse rogue. i would like some advice from some of you super smart math guys out there.

which is better for killing standard undead mobs. dual disruptors or an improved curse sperwing and disruptor combo?

advice appreciated.

Thanks

[Agarwaen]

At level 20--fighting against tough undead--you are better off with the Imp. Cursespewer since the DC is higher than on the disruptor. If you are zerging through stuff below your level then the dual disruptors will make quicker work of things.

[Kinerd]

Disruption is DC 14. Improved Cursespewing is DC 20. Hence, the curse will only be useful insofar as making disruption land better in the following range:

The monster has a Will save below 17, otherwise both the curse and disruption will only take effect on a fumbled save, and you may as well have the disruption straight off in that case.
The monster has a Will save above -5, otherwise the disruption would always (except 20) land anyway, making the curse overkill.

Now, in between we have to make a probabilistic judgment on the effective disrupt chance. To do so we have to assume a rate of offhand strikes, as for a rogue that will never be equal to the main hand. This in turn means we have to individually examine the case of having the disruptor in the main hand and the case of having the disruptor in the off hand. Let us take an undead with Will save 5 with your situation (assuming ITWF) and see how it shakes out:

With the disruptor in the main hand, each attack has a 40% chance to disrupt. Each attack also has a 70% * 60% chance to curse, which increases the chance to disrupt on further attacks to 60%. Hence:

Attack 1: 40% of disrupting straightaway, 42% curse
Attack 2: 58% chance not cursed, another 40% shot at disrupting; 42% chance cursed, another 60% shot at disrupting; only (1 - 42%) of the 58% to remain uncursed for the future. This is a total of .58 * .4 + .42 * .6 = 48.4% chance at disrupting on this attack, and a 66.36% total chance of being cursed in the future.

Algorithmically, we can express this for any attack N as:

Disrupt chance = (Cursed on attack N - 1) * (disrupt + 20%) + (Not cursed on attack N - 1) * disrupt
Uncursed chance = (Not cursed on attack N - 1) * (curse)

And the chance of no disruptions is the product of (1 - disrupt chance on attack N) for all N. We can do the same thing with the 60% applied to the disrupt chance instead of the curse chance if we have the cursespewer in the main hand and the disruptor in the off hand. Two disruptors is easier: we have a 40% and a 40% * 60% to disrupt on each strike, or a (1 - 40%) * (1 - 40% * 60%) of not disrupting on each strike, and we can exponentiate that directly.

In the absence of graphs (although I'll probably make some later), we can compare how many attacks it takes for the various set-ups to reach at least 99% chance of disrupting.

Cursespewer main hand, Disruptor off hand: 9.
Disruptor main hand, Cursespewer off hand: 7.
Both Disruptor: 6.

Again, these were all for Will save 5. Presumably the Cursespewer will have a greater impact with a higher Will save, so let's see how those numbers look, for instance with Will save 10:

Cursespewer main hand, Disruptor off hand: 16.
Disruptor main hand, Cursespewer off hand: 14.
Both Disruptor: 18.

So there you go. I'm not sure yet how to graph everything without it being incredibly busy, but I'll noodle it a little and we'll see.

Thanks, this was a really interesting stat problem for me. :)

[aristarchus1000]

I love the math and explanation. That said, in basic terms:

When you first can equip disruptors, they are pretty nice, since lots of mobs will fail the save DC of 14.
At mid-high levels they become much less useful, and using a cursespewer can help mitigated the loss of effectiveness, but you may want to consider switching to a Greater Undead Bane anyway.

At high levels, disruptors are only useful as a "on vorpal" effect (basically when the mob rolls a 1 on a save, 5% of the time). Very few undead have enough hitpoints that make this worthwhile (while not being immune boss or red-named). Off hand, I can only think of the Cinderspawns in Inferno quest and Sins of Attrition.

[Templarion]

*bump*

Very good thread.