Document Actions
latline
This is a program to do fitting of curves to either Intensity or F,Phi data for individual layer lines.
C*LATLINEK.FOR ****************************************************************** C * C This is a program to do fitting of curves to either * C Intensity or F,Phi data for individual layer lines. * C Individual weights are used for F's and phases (or I's) * C Normally a trial set of A,B for each lattice line are * C input (optionally this can be calculated internally). * C This trial set is then refined to minimize the weighted * C least-squares error between F's and Phases. * C * C Version 1.01 LATLINE 18.01.82 DAA VAX * C Version 1.03 17.02.82 DAA VAX * C Version 1.10 25.02.82 DAA VAX * C Version 1.11 09.03.82 DAA VAX * C Version 1.20 23.10.82 DAA VAX * C Version 1.21 07.11.82 DAA VAX * C Version 1.22 10.12.82 DAA VAX * C Version 1.23 LATLINEA 24.12.82 DAA VAX * C Version 1.24; 07.12.88; JMB; Option IWP=-2: sigphi input in degrees * C Guessed phase is weighted * C AK adjusted by factor nfobs/nphases * C Symbols on phase plot show input sigphi * C Version 1.25 LATLINEB 15.3.91 RH removed determinant test * C Version 1.26 20.9.93 RH allow NPARM=1, for NOBS>0 * C Version 1.27 LATLINEC 21.9.93 RH min phase sigma 1.0 degree * C Version 1.28 LATLINED 5.8.94 RH overall statistics added * C Version 1.29 30.12.94 RH debug CLNFIT - WPHI * C Version 1.30 31.12.94 RH debug phase residual (NP) * C Version 1.31 2.1.95 RH debug above CLNFIT debug * C Version 1.32 22.1.95 RH to UNIX, NCALC=2 in CLNFIT * C Version 1.33 19.1.98 RH prevent VMETRIC looping * C Version 1.34 17.9.99 RH parameter statements for NOBSMX * C NPLTMX,NPROMX,NCALMX,NPRMX,NOUTMX * C Version 2.00 30.8.00 RH plot2000 plots postscript directly * C Version 21.6.00 TSH fix wrong NOBSMX value in CLNFIT * C Version 2.01 03.7.01 TSH set XPLTSIZ and YPLTSIZ * C * C Remember to change version number in WRITE statement * C * C******************************************************************************** C * C Input parameters (unit 5): * C * C TITLE 80 Character title for this data set * C * C IPG Plane group number (1-17) * C * C IPAT 0 for F & Phase 1 for Intensity data * C * C AK,IWF,IWP AK = relative weights for phases with * C respect to the F's * C IWF,IWP are flags to use individual weights * C for F's and Phases in input file: * C -1 : use individual sigmas (1/sigma**2) * C 0 : set all weights = 1.0 * C 1 : use individual weights * C IWP= -2 : use individual sigmas; (wt=1/sigma**2); * C : phase and sigma both input in degrees * C ALAT,ZMIN,ZMAX,DELPLT * C ALAT = lattice size in angstroms * C ZMIN = minimum z* value expected for data set * C ZMAX = maximum z* value expected for data set * C DELPLT = z* interval for plotting * C if DELPLT = 0.0, then no plotting * C * C DELPRO,RMIN,RMAX,RCUT,PFACT * C DELPRO = real-space or patterson-space * C sampling interval for profile function * C (in angstroms) * C RMIN,RMAX lower,upper boundaries for * C profile function (angstroms) * C in general RMIN = -RMAX * C note, overall boundary width must be 2x * C as wide for Intensities compared to F,Phase * C RCUT = distance from either boundary to * C point where tapering of profile is to start * C (in angstroms) see also PFACT * C PFACT controls mode of tapering, starting * C RCUT inside of RMIN/RMAX and going towards * C RMIN/RMAX. * C PFACT <= 0 then use linear drop off to zero * C at RMIN/RMAX * C PFACT > 0 then use gaussian roll-off where * C PFACT = value of gaussian at RMIN/RMAX * C (a value of .1 is reasonable) * C * C IGUESS,BINSIZ * C IGUESS = 0 to refine input set of lattice data * C = 1 to generate a new set * C BINSIZ = delta z* for binning observations * C This is only used to generate the initial * C guess, which is then refined against the * C actual obsered data points. * C (a value of .005 to .002 is reasonable) * C * C NCYCLS,MPRINT Number of refinement cycles to perform. * C If <0 then output initial guess. * C If =0 then minimizer goes till end * C else (25-50 is quite reasonable) * C 19.1.98 impose a limit of 2000 * C If MPRINT >0, print var/covar matrix * C =0 do not * C * C * C * C Data formats (all are free-format): * C * C Observed data [logical unit = OBS] : * C H, K, Zstar, Fobs, Phiobs, Weight on Fobs, Weight on Phiobs * C * C If no Fobs for this Zstar you must set Fobs=-999. * C If no Phiobs for this Zstar you must set Phiobs=-999. * C * C For Intensity data: * C Fobs actually = Iobs * C Phiobs should be 0 * C * C It is assumed that the H,K are in the correct asymmetric * C unit for the symmetry operators to be valid. * C The data can be un-sorted on Zstar. * C * C * C Initial guess for the lattice data [logical unit = GUESS] : * C H, K, Zstar, AMP, PHASE * C * C H,K must correspond to the H,K that is on the observed * C data file. Data must be sorted in order of increasing Zstar * C and MUST be in equal intervals of Zstar. The actual sampling * C interval is not critical, as long as it is the same or finer * C than the critical sampling interval of 1./(RMAX - RMIN) * C If the plane group has an inversion-center along Zstar * C then only Zstar >= 0 are allowed; the first value must * C correspond to Zstar=0. * C * C * C Output data [logical unit = OBS] : * C * C H, K, Zstar, Fcalc, Phicalc, SigmaF, SigmaPhi, Figmerit * C the figure of merit = cos(Sigmaphi) * C * C Plots are generated on [logical unit = PLOT] * C * C************************************************************************ C * C SPACE GROUP INFO * C * C Number Spacegroup Asymmetric Unit Real Imaginary * C * C 1 P1 H>= 0 * C * C 2 P21 H,Z>=0 Z=0 * C * C 3 P12 H,K>=0 K=0 * C * C 4 P121 H,K>=0 K=0 * C * C 5 C12 H,K>=0 K=0 * C * C 6 P222 H,K,Z>=0 H=0;K=0;Z=0 * C * C 7 P2221 H,K,Z>=0 (0,2N,Z) (0,2N+1,Z) * C (H,K,0) * C (H,0,Z) * C * C 8 P22121 H,K,Z>=0 (H,K,0) * C (2N,0,Z) (2N+1,0,Z) * C (0,2N,Z) (0,2N+1,Z) * C * C 9 C222 H,K,Z>=0 (H,K,0) * C (H,0,Z) * C (0,K,Z) * C * C 10 P4 H,K,Z>=0 (H,K,0) * C * C 11 P422 H,K,Z>=0 (H,K,0) * C K>=H (H,0,Z) * C (0,K,Z) * C (H,H,Z) * C * C 12 P4212 H,K,Z>=0 (H,K,0) * C K>=H (H,H,Z) * C (2N,0,Z) (2N+1,0,Z) * C (0,2N,Z) (0,2N+1,Z) * C * C 13 P3 H,K>=0 * C * C 14 P312 H,K>=0 (H,H,Z) * C K>=H * C * C 15 P321 H,K>=0 (H,0,Z) * C K>H (0,K,Z) * C * C 16 P6 H,K,Z>=0 (H,K,0) * C * C 17 P622 H,K,Z>=0 (H,K,0) * C K>=H (H,H,Z) * C * C************************************************************************ C * C NON-STANDARD ROUTINES USED: * C * C 1) Plotting routines in subroutine GRAPH * C * C 2) MOVE(B,A,NBYTES) transfers NBYTES from A to B * C * C************************************************************************ C
