Hi all, in http://www.slac.stanford.edu/~daniele/vub/toterrmax.eps I show the total error as a function of the Mx cut. This is the sum in quadrature of - statistical error (green curve) - theoretical error, mb (red curve) - theoretical error, a (blue curve) - B-D decay systematics (yellow curve) The statistics is ~equivalent to the data one and extracted from MC. Theor. errors come from plots shown by Urs in his last posting. I used the ones with mb = 4.800 +/- 150 MeV and a = 1.29 +/- 1.0. The B-D systematics was extracted directly on data (see one of my last posting and the plot http://www.slac.stanford.edu/~daniele/vub/mxstudy.ps). The "best" cut is for 1.55 GeV. Then, the error we should expect on data is 23%. The remaining systematics have to be added. The total error probably will not exceed 27% (=>13/14% on Vub). Daniele