[BioC] number of affy-chips needed to fit rma model

[Dipl.-Ing. Johannes Rainer]

hi to all,

i'm in a lab where we made our first affymetrix chips, and now i want to do some
normalization steps to correct for the errors. i'm not really convinced about
the method that affymetrix provides (mas 5) so i read now many papers about how
to normalize affymetrix chips. the method i like is the rma method from
irizarry, but i do not know if we have enough chips to fit the model parameters
well. by now we have only 2 chips, one with a treatment, the other without. does
anybody know how many chips i need to fit models like rma or vsn well into the
data?
other question, how do i know what's the best method to normalize my data?

thanks, jo