[ccp4bb]: Refmac SCAL keywords

[Frank von Delft <frank@cryst.bioc.cam.ac.uk>]

***  For details on how to be removed from this list visit the  ***
***    CCP4 home page http://www.dl.ac.uk/CCP/CCP4/main.html    ***

Dear ccp4ers

I think I've missed something in Refmac scaling.  The following
combinations of SCAL keywords all give different results:

(1)
SCALe MLSC NCYC 15
(2)
SCALe LSSC NCYC 15 ANISO
(2)
SCALe MLSC NCYC 15
SCALe LSSC NCYC 15 ANISO
(3)
SCALe LSSC NCYC 15 ANISO
SCALe MLSC NCYC 15
(4)
SCALe LSSC NCYC 15 ANISO MLSC NCYC 15

Because I'm doing ML refinement, I assumed that MLSC scaling was applied
by default, and I was expecting nearly identically behaviour in terms of
sigmaA estimation, R-factors etc.  (apart from the first option, where
ANISO is not specified).  This is not at all the case.  Extremes occur
in (4) where sigmaA is output as 0.00 for all resolutions, and in (2),
where sigmaA vs. reso varies between flat and curved from one cycle to
the next.

This is obviously mainly a matter of syntax, so how exactly are the
above interpreted?

Thank you for any comments
Frank