power analysis / sample sizes for factorial designs


I'm interested in calculating the power (or sample sizes required for a power) for factorial designs. As an example, if I have a 3-factor, 2-level design with 3 center points (11 runs total), and I know the difference in the response and the standard deviation of the experiments, how do I calculate power?

If you could point me to some articles and/or books that discuss this in particular, I would appreciate it.

Also, does anyone know if this is possible to do in JMP version 6? (main question will be posted in software forum)

Just as a followup, I found this paper that (apparently) discusses exactly what I'm looking for:

Sizing Fixed Effects for Computing Power in Experimental Designs
Gary Oehlert and Pat Whitcomb

However, if anyone knows of anything simpler, or with more examples, please let me know. This paper looks like it will take a while to digest.