On Fri, Jul 1, 2011 at 8:47 AM, israel vainsencher <israel@mat.ufmg.br>
 wrote:

 waw!
 best news of the week, perhaps of the month!
 the question now is how to ensure, a priori, if the accuracy sufficed!
 cheers
 i

 On Thu, Jun 30, 2011 at 6:07 PM, Jonathan Hauenstein
 <jhauenst@math.tamu.edu> wrote:
  Israel,
 
  You are correct.  Since the system is overdetermined, Bertini has to
  determine which solutions of the randomized system it actually solved
  are solutions of the original system.  By increasing the accuracy of the
  solutions to the randomized system, we have the following for your
  system:
 
   Multiplicity  |  Number of real solns |  Number of non-real solns
 
  ------------------------------------------------------------------------------------------
         1       |       0               |       5440
 
  Thanks,
  Jon