precision of optics

[mfittere]

After rematching the squeezed optics, not the same precision in beta, alpha and dispersion is reached as in previous layout. E.g. study_squeeze/rematchir6/opt_round_new.madx:
betxIP2b1= 9.999999; betyIP2b1= 10.000000;
alfxIP2b1 = -0.000000; alfyIP2b1 = -0.000000;
dxIP2b1 = 0.000002; dpxIP2b1 = -0.000000;
betxIP2b2= 10.000000; betyIP2b2= 10.000001;
and IR4 is not properly matched any more:
betxIP4b1= 172.8079 50; betyIP4b1= 230.297 895;alfxIP4b1 = 0.580679; alfyIP4b1 = -0.325498;dxIP4b1 = -0.000009; dpxIP4b1 = 0.000000;
betxIP4b2= 228.562 607; betyIP4b2= 302.5659 70;alfxIP4b2 = -0.537702; alfyIP4b2 = 0.588783;dxIP4b2 = 0.000008; dpxIP4b2 = -0.000000;