Aaagh! Vast hordes of em

[Crystalion]

Oop, yeah I did mean solo. But [snip]

You raise so many good points I'm going to slice and dice my observations in reply...

> but it requires a resource of uncontrollable quantity to do so: corpses.

Ironic, isn't it, that you are proposing that the Bowazon is *precisely* such a generator. Ergo, team up the awesome zon and potentially awesome CE Necro and surprise, surprise, together they kill more than twice as fast as they would individually. By time you factor in play speed due to confidence in the Necro's superior crowd control and the fact that the pre-CE Amp damage will be cast well before, in order to also help the the zon [lower resist instead in special cases] the net efficiency could be quite high.

> You must factor in that time, since it is a part of the whole "killing things with CE" process.

Solo, assuredly. This would be why the "smart" xp harvesting necro would look for opportunities like the cow level to do only a few starter corpses once, for the whole level.

> CE is twice as likely to get at least half its damage through, since it's half fire and half physical, but it's also resisted twice as often to do half damage by way of immunities, and also subject to the 1.09 global 50% phys resists.

I'm more interested in the v1.10 state of affairs, in which global resists are gone but mob HPs are up [so your old damage output didn't scale, but CE got better and does] and in which Necro's can actually remove phys or elemental immunity from some mobs.

> The only confounding factor to damage / sec being proportional to the kill rate is regeneration rate.

This is an error in logic and of omission. You forget that a necro's kill rate is not proportional to a monster's total HPs [any more than Static Field was before it was capped]. In the same way, if a Necro's number and duration of revives were not capped there would also not be a proportion to the "damage/sec" of revive [a non-sensical idea in any event] as the potential growth in the Necro's army would eventually come to dominate the entire D2 universe. If you think I'm being silly, then you really ought to try a Necro in one of the HOMM games and see what happens when you emerge from every battle with half your defeated enemy raised as skeletons for you.

> So, you take all the monsters' regeneration rates, calculate ...

Yes, when one doesn't have an apples and oranges comparision problem, this works. As another poster pointed out, this calc is even simplier if you have prevent monster heal, or, vs. non-immunes, poison. Nonetheless, often in the game the mobs will not be homogenous in terms of their HPs or resists. In the innocuous case you may "quickly" kill the easy mobs and be left with the "hard" ones, which, simply by virture of there being *less* of them, will greatly reduce your damage/sec. rate [not to mention the RoK ;) ]. In a less pleasent case, a nasty boss or overlarge champions will "ignore" your probablistic crowd control method and breathe down your neck. Naturally this will not be a common occurance unless v1.10 does a much better job at mixing up monsters, density and AI than I expect. But certainly it means you should refer to "peak theoretical" or "under normal conditions" when you cite your damage/sec. calculations.

> Clearly rate of kill is the ultimate proof, but that's not quantifiable in the same way that damage per second is quantifiable.

On the contrary, damage/sec is much much harder to make a "real" number. Rate of Kill can be tracked quite easily--for example by turning on a voice recorder [or VCR on your session for later call out], noting the time, calling out kills, noting the time, doing the math. This would be an actual number. Bona Fide. D1 was nice enough to tell you how many of a particular mob you'd killed. Damage/sec. is complicated by far more factors (such as overkill--you weren't planning on counting that any more than you'd count misses, were you?).

> accounted for the largest discrepency between the two.

I would guess, up through v1.09, players=1, that the largest discrepancy is likely to be monster acquision time. As v1.10 seems to have made "progress" in making the game harder largely by increased mob HPs you may be right [certainly calculation RoK for, say, Izual, pretty much correlates with your damage output into his massive HPs].

Finally I suppose one should always check the price of charges on items that grant CE, as any class capable of killing a few mobs could potentially obliterate a mass of enemies efficiently near the same spot with a few casts. This begs the question, if charges are expensive, whether the new Rune repair recipe repairs them.