I'm trying to narrow down the necessary criteria for a PCR collaborative and this seems to be "one" first step?

I think it is a good idea to start with lowest complexity possible, although people using different machines or chemistry are left outside for a short moment. Standardisation of a quantitative realtime PCR includes many aspects of assay format (how many replicates, calibration points, dilutions as well as modes of calculating GM-percentages, control of inhibition effects etc.) and actual performance e.g. in terms of variation coefficients may be dependent on chemistry used and possibly also the machine type. In a second step, not far ahead, a collaborative could be broader in terms of machinery, chemistry ect. and could then be referred to this initial, restricted but more homogenous data set.

I would assume we'd have the largest number of participants if we chose that technology.

Presently, Taqman does seem to be the quantitative technique most widely used for testing. I think that an initial study using this technique on a series of grain samples would be a good first step in evaluating this technique.