Haven't gotten it yet (haven't done much with PLD at all lately), but yes, a controlled test where you are only taking damage from 1000 needles would be ideal for this since you know how much damage you'd take.
Was thinking you might want to control either tp feed or damage taken, but since you're trying to take the full 1000 damage, you don't need to worry about making sure you have an untouched stoneskin, making this really easy. Grab a cactaur that can't kill you outside of this (probably one in Kuftal tunnel) and melee the sucker to death using a weapon that will do damage, but not much of it. Parse the results if you could, and it would be possible to generate an enmity vs. damage taken vs. time graph to allow us to see the results.
As a side note, I totally misread the "mitigates damage based on enmity" as a "mitigates enmity loss from damage" note.... Not sure how I did that, but I'm much more intrigued by the piece now then I was.
Edit: You'll need more than 1 1000 needles, though, and preferably from the same mob where you're NOT trying to spike your enmity from the get go. I.e., we'll need to see several at a low enmity value (primarily from TP spam on him while not doing much else) and some at high. Ideally, if you could sleep it reliably, you could let all your VE disintegrate after a brief period, allowing you to test the results of CE vs. VE, as well. It's not completely straight forward, but we need to decide if its TE, VE, or CE that affects it and to what degree do they affect it.
Edited, Feb 2nd 2011 2:36pm by Dracoth
Leviathan - PLD90, BLU90, MNK 90, DRG 78