Readme.txt I have now started to add bug fixes and known bugs to this file. If you know of any bugs, quirks, or any other annoyance please let me know. Thanks to Lachlan Cranswick (LC) who has helped in the development of Rietica. bah@ansto.gov.au Brett Revisions 1.77 21/11/01 Some minor improvements in the Phase dialog (drop down atom types) Fixed problem with background types being different and changing from 6 to 12 terms. Improved file handling (now doesn't overwrite an input without warnings) Added a Parameters dialog - not fully functional. 1.76 5/9/01 Improved the TOF peakshape Able to read GSAS ALT format (in both the refinement and the File|Open) Fixed d-space plotting option Improved error catching. 1.74 5/4/01 Updated the f' and f'' scattering factors Fixed a problem when the International Settings for the decimal point was a ',' Added Energy Dispersive x-ray data analysis (use only for LeBail) Fixed some inconsistencies in the peak shape Improved performance by rearranging some arrays and using the latest F90 conventions Changed the some of the structure of the program so that you can have non-constant step sizes. Useful for TOF. Plotting can now be in raw x (time, 2theta, energy) or in d-spacing No need for Integrated Intensities to be turned on for LeBail or Fourier. 1.72 11/1/01 Automatically turns on Integrated Intensities when doing a LeBail (so that the hkl file is properly written and it not starting from scratch again) Fixed the Rietbasic examples Added the ability to read in CPI and XY file formats from File|Open File (previously it said not supported yet). Now automatically switches the Data format if you select a different format in the Refine dialog. For example, if you select a xxx.gs file it will automatically select the GSAS Data format as shown in Model|General. The only problem is for LHPM and XY data format when more than one option is availble, so it selects the first, eg for XY it will set it to X, Y, esd format. But ff you have previously set the format to X, Y format it will not change it. Fixed some problems with the RDF. Will still only work on the firt phase. Fixed the peakshape plot - the widths were not plotting even though the shape was. Fixed a problem when selecting the Jorgensen peak shape (left the asymmetry inputs showing). Added the ability to change the hkl marker position and background position. Accessible in the plot window under Chart|Scales. The hkl marker height is not working yet. Change the Cheby polynomials to be orthogonal - much more stable. Also fixed up a format problem with the background values. Background calculation before cycling is out of whack. Added a "Stay on top" option to the Refine Dialog. Accessible in File|Preferences|Run options. 1.71 7/12/00 Fixed a problem with LeBail not reading the hkl file in (only a new bug since 1.70 - thanks Lachlan) Fixed a major bug when refining more than 6 background parameters (didn't read in properly - thanks Florent) Added a Recycle checkbox in the Phases dialog when doing a LeBail (same as the recycle checkbox in General) 1.70 4/12/00 Added an RDF/PDF/G(R)/D(R).... plot feature (only partially operational). When you manually type in an inputfilename into the Refine Dialog it will read that file in (used to do this only when using the Browse button). Added a Lorentzian anisotropic term for the Voigt functions. Added damping for LeBail fitting - uses the Damping/Relaxation value normally used for x,y,z (the first damping value). Changed it so that GSAS, XY, and CPI files are also automatically assigned. Fixed bug when changing the filename in the Refine Input box, it did not update the datafilename automatically. Fixed a bug when you use the F^2 calculation limit and the first peak was rejected. Added a number of new functions to RietBasic and fixed up some errors in it. Added the Startup/Library RietBasic code, so you can run a script on startup, as well as store much used code in the library script. One of the new RietBasic features is the addmenu & clearmenu functions. This allows scripts/macros to be associated with User menus. Very handy. Added a plot routine to RietBasic. Included a script for Ortep-3 integration. Added an Automatic smoothed background (no refineable parameters). It's a no-brainer, but really is not good as it puts the background lower than it should be. Fixed a bug with the PO checkbox not updating when changing phases. Fixed a bug when calculating a pattern and the 2theta range was larger than the number of points allocated Changed some of the Fortran code over to Fortran-90 style, and tried to fix the readability. Fixed it so that when the "Large Font" setting is used in Windows the Rietica windows will change in size (sorry for taking so long on this one). Fixed the missing Damped Amorphous background function (although the function itself is not working well). Fixed bug when LeBailing x-ray data - Fourier options output only Ka1. 1.67a 3/7/00 Fixed a minor bug: Couldn't open a file from the Refine Dialog. Added a StartUp Directory in preferences. 1.67 29/6/00 Added an Offset value for the polynomial backgrounds, so that you can now have yback = b0 + b1*(theta2-offset) + b2*(theta2-offset)^2 +...+ b_1/(theta-offset) Fixed CopyFile function in RietBasic Fixed GetValue function in RietBasic (asymmetry now reading ok) Fixed Plotting of TOF FWHM. Fixed bug when you push the finish button before actually finishing. Now have versions with/without internet ability. Fixed the shadow border when copying plots as bmp's from the clipboard Changed the (x,y,weight) format to (x,y,esd). Improved spelling, some layout, and some errors/bugs in the open dialogs. Completed the 'Use Instrument' feature (right mouse click). Now can easily 'use' instruments in Model|Histogram and Model|Sample. Fixed the TOF Jorgenson Peakshape. Derivatives are now correct and the lorentzian is properly working. Fixed the 'Reload' button when Viewing files (only ever reloaded the input file, now reloads output file, etc) Improved FWHM plot in Model|Sample (but not asymmetry effects or TOF peak yet). Fixed background grid in Histograms. Now has checkboxes when going from 6-->12. Also changed the last value to read B-1 for the simple polynomial. 1.66 10/5/00 Improved GSAS importing Fixed a memory leak in the DLL. Fixed the monoclinic space group settings (now can have alpha and gamma. eg can have "P 1 1 2/a", or the standard "P 1 2/C 1" or "P 2/C" Added a new "XYZ Transformation" dialog. To access it you right mouse click the atom grid. It allows any changes in xyz (rotation and displacement). Added parameter limiting in the fortran - not yet in Rietica. 1.65 26/4/00 Added a rudimentary CIF file import (right mouse click in Phases). You can only import a single CIF entry, so if there are more than one it will pick up the last one. Better to manually delete all but one at this stage. Added a input file import so you can read a phase in from another file. You access this via righ mouse click in Phases. Added simple EXPO xyz file importing. You right mouse click in the atoms grid. Fixed the generic xyz atom file importing (used to have to be fixed format), now it reads in free format (Atom Label, Atom Type, x, y, z). 1.64 10/4/00 Cleaned some memory allocation (so that it shouldn't take up so much memory now) Fixed the 'Low memory' problem when using NT. Fixed the LeBail, so that you can have non 1.0 ratio. This allows you to have a mix of LeBail and Normal in an x-ray pattern when ratio is not 1.0 (you could always do this previously in neutron patterns since ratio=1.0) Now can read GSAS files in File|Open|Data Files. Must now have GSAS files as a *.gs extension instead of *.dat (sorry) Fixed a problem with 'New Input File' and LeBail (switched back to Normal) Improved Fullprof importing. 1.64 30/3/00 Fixed a few bugs, including the annoying bug of 'Unstable refinement' when there is no lorentzian and a large Base Width. Fixed Sample Displacement (the refine value is actually s/R*180/pi) 1.63 7/3/00 Fixed a new bug (since v1.62 I think) when you have a combination of lebail and normal phases in the one refinement. You had to manually set the Lebail/Normal every time you went into Model|Phases. Added a beta version of Sample Displacement correction for x-rays Changed the Automatic Marquardt to set itself to Newton-Raphson if the value becomes less than 0.00001 Fixed a problem in Fullprof importing a large Base Width value (now sets it to a reasonable 7 if larger) Added TOF Extinction, but still has a rounding error when the value is very small (not a major problem). Fixed a problem when reading GSAS STD format with no detector number (gave large chisq's) 1.62 10/2/00 Hopefully fixed the web updater to work properly Added hotkeys (to the Refine dialog, eg Start = F2, Step = F3, Finish = F4) 1.61 7/2/00 Fixed up correlation matrix when using Marquardt (sometimes caused a divide by zero error) Fixed a problem with the TOF peak shape when numbers were larger than 100.0 or less than -9.99 1.60 2/2/00 Included a web updater. You should be able to download new versions of the program and/or dll's and/or help files, without downloading the distribution. (i.e. it will download only the changed parts of the program.) Now has the Marquardt least-squares algorithm. The easiest is to set it to automatic in General. Remember to turn it off once you have a stable refinement (although the automatic will eventually reduce it to zero after a number of cycles). Fixed bug when going from Normal to LeBail. Can now drag and drop constraints in the atom grid. If you have two varying parameters (the checkboxes must be checked) then drap and drop from one parameter to the other will bring up a dialog for the constraining value. Note: at the moment it will not overright existing constraints. 1.59 20/1/00 Fixed bug when x=0 in plot window Fixed bug when intensity = 0 when using 10F8 data format Now when you change from Lebail to Normal and back again it will remember the atoms. But if you refine while in LeBail you will loose the atoms if updating is on. Fixed minor bug when going from Lebail to Normal and the site multiplicity was still the '?' in the atom list. I think the installation 'error' (although nothing bad happened) has been fixed. 1.58 24/12/99 Fixed up some more bugs with the last two versions. 1.57 17/12/99 Sorry about the last release - it had many new 'features' due to a change in the compiler I was using. The 'features' that have been removed from (just) the last release are: the atom grid now does not let the focus roam to the cursor position, the background grid now lets you change the refinement flags, and there was a missing DLL in the distribution. If you did not get 1.56 then the above is irrelevant. Hope thats all. Added a (simple) SIRPOW export, as well as an EXTRA/SIRPOW .fou file option. The combination has been used to good effect when using the LeBail for structure solution via SIRPOW97 (that is it works). Fixed more flow control, such as when you exit, etc. Added the DOS version of LHPM (LHPMCON) to the distribution. 1.56 13/12/99 Fixed the hexagonal setting constraint Fixed the fourier file (now matches what it says) Cleaned up some 'features', such as when you click on the checkboxes in the atoms grid Can now drag and move an atom row in the atom grid Fixed up the "Print U's instead of B's" A 'copy' in the atoms grid not only copies the selected atom, but also puts all the atoms in the Windows clipboard (for input into Word etc) Has a rudimentary 'sanity' check of the input (as a really bizzare input file can cause a program crash where the program vanishes instantly) Added a check for zero scale factors in the fortran backend 1.55 26/10/99 Added a 'Semi-Automatic' background interpolation. It does a n-point average of the counts around the clicked 2theta position. You can change the 'n' in the option menu. [n=0, no averaging, n=1, 3-point average, n=2, 5-point..] Fixed a problem in background interpolation with multi-histograms. Added the Sietronics CPI format. Can now set the Step size to 0, and you can now Finish directly after a Start without doing any cycles. This allows a better (but not perfect) way to add background points without doing any cycles. Added a 'Save on exit' to the Macro dialog. Can now cut,paste,copy from the Macro ouput window. Fixed the Plot File droplist in General. Added an 'Export to Excel/Sigma/Axum/Origin/Kaledia' to the Plot window under file. This allows an easy way to produce an import file for those programs and hence a publishing quality plot. 1.54 8/10/99 Added a Full Illumination model (sin(theta)) for x-rays. 1.52 14/9/99 Fixed a bug with the database (not saving xyzB values properly). This was a new introduced bug, so older versions worked properly. Fixed a bug with the fortran code that was causing a problem with multiple histrograms and constraining the lorenztian components. There is still a bug with the installation. This is due to the migration to a new version of the database software. It does not cause a problem with Rietica functioning. Hopefully it will be fixed soon. 1.51a 2/8/99 Fixed a minor problem with Trigonal space group setting Deleted the strange number that was in the Phase form 1.51 8/7/99 Fixed a bug when reading Flat plate absorption (mixed codewords up because it never read the FP codewords if you had more than one histogram) 1.50 8/7/99 Fixed problem with Find/Replace in the edit forms Added automatic space group constraints for the lattice parameters. Now prints out what Laue group you are in (as well as an 'Unknown' if you have entered an invalid space group) Corrected the list of space groups (some errors in the cubic ones) Can now enter the space group number in the edit box. It will automatically select it from the list. As the list is only standard settings you will need to manually adjust if you want a non-standard setting. Added cosmetic changes to the atoms list and a site multiplicity column. Added a new popmenu item 'Set to full occupancy', for those of us who don't want to look up the international tables. Fixed bug to allow LeBail to work when there is a different inputfilename to the datafilename Fixed bug when LeBail F^2's where small (such as what happens in x-ray data sometimes) 1.43 13/5/99 Fixed problem with excluded regions (when whole numbers) Fixed bug with the plot when 'Finished' had been pushed When doing LeBail's it now sets Scale=1.0 and turns off the refinement parameter (if a user didn't know that he/she had to do this to stop unstable refinements) Starting to add a feature in Histograms that allows quick linking and a quick 'Use Instrument #x' feature. If you use the manual edit feature and then 'Save' it, it will read that changes and reflect them in the normal dialogs. This allow a quick and seamless flip to and from the dialogs and the manual edit (as sometimes it is quicker to do a manual edit for some operations) 1.42 18/3/99 Fixed some problems with Project (kept reading the old input file in and would try to plot [causing an error] even if plotting was not selected) Now asks if you want to save on exiting/closing. 1.41 16/2/99 Fixed the missing space group (#189) Fixed the wavelength readout in the Watch window 1.40 1/2/99 Added a new file .hkl for LeBail recycling. If LeBail is selected it automatically creates this file and the update file has been set for automatic recycling. Curerntly you can have multiple histograms with one Lebail phase or multiple LeBail phases with one histogram, but not multiple histograms with multiple LeBail phases. Fixed a problem with win95 filenames (if you had a space in them) Added a Cu Ka as a default instrument 1. Will not update current settings if the title of Instrument 1 <> 'Instrument 1 not set' Changed to Install Shield program for installation - now includes the database BDE automatically. Cleaned out current database. Fixed a problem with the 'Add to database' - it wasn't adding to the right directory. 1.39 6/1/99 Fixed a problem with the constraint list (when the parameter was a constraint) Did not affect refinements but it made some parts of the list unreadable. Added some extra Rietveld Basic functions, and changed the name of some. eg. You can now read a Rietveld file, change some values and then save the file without doing a refinement. 1.38 15/12/98 Fixed a problem with the phase/atom database - not correctly reading the directory path. Added an automatic hkl in the plot form - the hkl label updates as you move and displays the closest peak. No more need to click the markers. Fixed the Find/Replace in the Editing form(s) The Undo/Redo now works in the Editing form(s) Fixed some problems with the summary/distance/bond valence file. Fixed a few minor 'features'. 1.37 26/10/98 Fixed a bug with pattern calculations (didn't read in properly in GUI, didn't write properly in the fortran code when Update was checked) Changed plot so that difference doesn't plot on pattern calculations. 1.36 23/10/98 Added a Unzoom option on right mouse click in the plot window Added Patterson maps in Fourier plotting. Fixed the Fourier obs, calc, dif - it was previously doing |F|^2's (i.e. a patterson with phases), not |F|'s. More fixes to the background output. 10I8 format now reading correctly. Improved FULLPROF import; now reads in free format files (the import routines are not perfect and some values are not read eg. Lorentzian values) Changed Edit Form Save button from a Save As to a Save (added a save menu item as well) The error that sometimes occured when zooming the refinement plot after FINISH being pushed has been fixed. It was found to occur when the Update checkbox was checked. Hopefully it is now fixed. Changed the Fourier Plot axes to be between 0.0 and 1.0. Changed the Plot file and Fourier File selection in General. 1.35 13/10/98 Added a simple Fourier plotting routine. Fixed an error in dll that caused a crash on reading some datafiles. Changed the plotting so that the difference plot does not show when no cycles have been done. Added a 'Quick start' button, when you want to just run and not fuss. (It is in fact just a button saving device on existing procedures) Fixed bug in General meu to do with 'Other' in Fourier options 1.32 8/10/98 Fixed display of the 'd value' in the status bar of the plot (for TOF only) Fixed another problem on background numbers not being read properly (large numbers this time) Added a 'Number of Recent files' option in the Options (needs to restart the program to take effect) Fixed a problem in interpolated background display Improved Export Shelxs Fixed a bug in GSAS importing (still not perfect) 1.31 6/10/98 Added a way of editing the background points in 'User Point Background' mode in the Histogram dialog. The point and click in the refinement plot is still there. Added a sort routine to the interpolated background, excluded regions and interpolated scattering factors Fixed a problem with interpolated scattering factors Fixed reflection markers in TOF refinement plots Fixed writing of the background numbers in the input file and the writing of the Incident spectrum (TOF) (if the numbers were small they were be written as 0.0) Fixed Instruments section in Options. Fixed bug when the 'Update' checkbox was checked and an unstable refinement occured (an old refinement file was read in) Added a 'Make the rest of the Histograms the same' option in the New Input dialog 1.30 28/9/98 Added a dialog for the New Input selection. Users can generate a quick input file by selecting various items in a small dialog. The new Instruments section in the Options now stores commonly used parameters. Added a checkbox in Options so users can do a refinement cycle on pushing START Previous revisions not recorded. Major Feature List (a compilation of features most demanded by users and likely completion dates) * Energy Dispersive (2000) * Spherical Harmonic Preferred Orientation (2000) * Magnetic Structure (2000) * Strain Tensor for peak shape (2000) - a anisotropic term is now in for the Lorentzian * Fourier Transform Peak Shape (2000) * Bond Constraints (already in, but not fully working, soon) * CIF export/import (added rudimentary import) * inline viewing of structures and manipulation (out forever - now have Ortep-3. Has saved a load of work) Features to be added * CIF import/export * Fullprof/DWBS export * Possibility of running Fullprof/DWBS from command line? That is, allow different backends. Known Bugs: * the 'Refinement is unstable...' error message is uninformative * export/import functions not perfect * cut/copy/paste does not work everywhere * does not clean up (ie delete) temporary files in the temp directory * Fourier output occurs only when 'Integrated Intensities' is checked. * LeBail .hkl file only occurs when 'Integrated Intensities' is checked. Definitions Simple = May or may not work on complex input, in fact may not work on simple input Improved = Fixed some bugs that should never have been there Fixed = There was a significant problem and hopefully it has gone away (at least until the next crash) Changed = Initially didn't really think about it, but now someone has pointed it out it has been fixed up, but probably will not work at tleast for a year Added = It really should have been there in the first place, and now after a dozen emails we have been forced to put it in. Of course, it goes in as Simple. Soon = This lifetime. A Date = We start work on this date and hope it gets done by 12 months