Scheme of the expected distribution of (π_pho − π_hpc) vs. π_hpc. Here we assume a constant absolute astrometric standard error in the trigonometric Hipparcos-based parallax σ_πhpc. As can be seen in Fig. 17.7 of Hipparcos Catalog this assumption is correct only for bright stars. We checked that the 85% of the stars considered in this analysis, with the requirement to have full Strömgren photometry have , therefore the assumption is acceptable. For photometric parallaxes, a constant relative error is assumed. Again, in this case, as the stars are apparently bright, which means that they are nearby, we assume that reddening effects and a low signal-to-noise ratio do not affect the derived distances.