[ccp4bb]: mtz2various SHELX warning

[Dirk Kostrewa <dirk.kostrewa@psi.ch>]

***  For details on how to be removed from this list visit the  ***
***          CCP4 home page http://www.ccp4.ac.uk         ***

Dear CCP4 users,

I've encountered a rather strange behaviour of mtz2various in CCP4 4.2.1 for 
SHELX output: contrary to previous versions, mtz2various now ALWAYS squares the 
input values, thus tacitly assuming that you want to use the SHELX HKLF4 format! 
I think, this is not a good idea: the user should still decide whether to use 
HKLF3 (F, SIG(F)) or HKLF4 (F^2, SIG(F^2)), as was possible in previous versions 
of mtz2various just by giving or not the FSQUARE keyword which is, puzzlingly 
enough, still present. SHELX does NOT always require or assume HKLF4 format - 
for instance, the SHELXD manual for ab initio (sub-)structure solution describes 
input files with HKLF3 format. If a user follows these examples and just 
believes that mtz2various outputs F, SIG(F) if the FSQUARE keyword is not given, 
you might waste computer time and end up with no solution at all (this is what 
happened to me).
My request to the author of mtz2various (is it still Eleanor?) is to remove the 
default-squaring of input values for SHELX output format and let the user decide 
again via giving the FSQUARE keyword. Not all users read the log-files or 
compare input with output values (what I finally did).

Best regards,

Dirk.

-- 

***************************************************************
Dirk Kostrewa
Paul Scherrer Institut             E-mail: dirk.kostrewa@psi.ch
Life Sciences, OSRA/007             Phone: +41-56-310-4722
CH-5232 Villigen PSI                  Fax: +41-56-310-4556
Switzerland                      Internet: http://www.sb.psi.ch
***************************************************************