[time-nuts] Rubidium (Rb) or Caesium (Cs)

[WarrenS]

Rubidium (Rb) or Caesium (Cs) standard reference oscillator?

What will give the more accurate absolute Frequency source over day to day averages?
A primary Cs (the types available to time nuts) or a optimally disciplined GPS Rb Osc?

By definition Cs is the primary time standard, 
but there are several things that effect a time-nut's "Primary Cs Standard's" absolute frequency including how it is built and maintained, if it has the high stability option and Einstein. 
What I'd like to find out is how accurate a GPS Disciplined_Rb_Osc can be made compared to the typical Cs out there.

I'm experimenting to find out how accurate a freq standard can be made using a LPRO Rb disciplined to a Tbolt.
Using a  temperature compensated and tweaked LPRO Rubidium (Rb) oscillator, 
I'm getting low e-13 per deg F and day to day freq variations (compared to GPS) even before being disciplined.
When the LPRO Rb is disciplined to GPS using a well setup Tbolt with an extended time constant of a few hours, 
their phase difference stays with-in a couple of ns RMS, and of course the difference between them long term is zero.
What I would like to determine is how accurate that really is.

ws

******