Hi Tassos,
It's a bit early in the year to have our annual I/sigI discussion on the CCP4BB, isn't it? Usually, we wait until at least April 1. :-) >. >d*TREK could confuse people further as to which one should be used for reporting and >decision making. > >. >The Big Question Again: >Which the infamous "I/sigma" that "should be above 2.0" and should be in "Table 1" ...? Since d*TREK gives you a Table 1 to place in the supplementary material of your paper where no one can read it, you should use that one. >If we agree we want <I/sigma(I)> (after sigmas are corrected I presume) here is a solution: >Since Phil is generous enough to offer to rename the Mn(I/sd) column to I/SIGMA >and maybe rename I/sigma to I/rms-scatter or so, if Jim could also rename >one of his columns, and Wladek can add one, we can have a standard. >Or even better if everybody would use <I/sigma(I)> then we would even start >getting the tables right in the papers? I got no problem changing labels. In d*TREK dtscaleaverage one column is the mean I/sigma(I) of the input observations, one column is the mean I/sigma(I) of the output unique reflections. This is clearly stated in the output and one does not have to read the manual. I believe these are identical to SCALA's columns and do convey useful information. Jim
