Known Bugs in Diamond Version 3.1f

The latest version of Diamond is version 3.1f, released on February 26, 2008. This new version is an intermediate update version and resolves some bugs as well as offers some minor changes.

Version 3.1f is available as a software patch for users who are still using version 3.0 through 3.1e (see Update page).

Please note: Some more bugs are still left and known and thus listed below. (Bugs found or reported after February 26, 2008 are emphasized and use a date as suffix. Comments from the developer are given in blue.)

Severe bugs

• Diamond crash right after start: Some users report that Diamond crashes when it is started and before a file is opened or new data are entered. The problem persists even after new installation. Reason: A (not yet known) bug in Diamond writes erroneous settings into the Registry.
  Workaround:
  1) Open the Registry by running the command "regedit" from the Windows Start menu ("Start" -> "Run...").
  2) In the tree view in the Registry editor, go to "HKEY_CURRENT_USER\Software\Crystal Impact\Diamond 3".
  3) (This is not necessary for the workaround but may help us to find the error: Run the "File/Export..." command and save the "Diamond 3" node as a REG file and send us that file in an email attachment. Maybe we can reproduce the crash on one of our systems.)
  4) Delete the branch "Diamond 3", or rename it (if you want to keep a copy) to e.g. "Diamond 3 old".
  When you now start Diamond 3, it should run with "factory default settings", like after the first installation. [Jun 8, 2005]
• At least one user reported that after an update from e.g. version 3.0e to 3.1e no more space group information is available.
  Workaround:
  (Re-)install the latest (demo) version from our web server:
  1) Make a backup copy of your license file, e.g. copy "yourlicense.lic" from your Diamond program directory to e.g. "C:\Temp".
  2) Uninstall Diamond. (Start -> All Programs -> Diamond 3 -> Uninstall Diamond 3)
  3) Download and install current demo version from:
  http://www.crystalimpact.de/diamond/download.htm
  4) Write back your "yourlicense.lic" file to "C:\Program Files\Diamond3" in order to replace the demo license file.
  (Note: Your "latest settings" (recent files, atom colors, etc.) remain, because these are stored in the Windows registry and not removed during uninstallation.)
• At least one user reported that Diamond crashes when in (distances, angles) measuring mode right after you click the last (2nd, 3rd., rsp.) atom each. (The result is written in the status bar but the info tip window does not appear.) [Jul 6, 2006] Another user reported that Diamond does not crash but hangs up in that situation. [Jul 11, 2006]
  The following workaround suppresses the info tip window:
  1) Close Diamond, if it is still running.
  2) Open the Windows Start menu and start the Registry by entering "regedit" after the "Run..." command.
  3) In the registry tree, go to "HKEY_CURRENT_USER\Software\Crystal Impact\Diamond 3\Picture View".
  4) In the list on the right half, add a new "DWORD value" with the "New" command of the context menu. This new value must be named "EnableMeasureTip" and have the value 0. (1 enables the measure info tip window.) [Jul 12, 2006]
• The pasting of structural parts does often work incorrectly (cf. "Edit/Copy" and "Edit/Paste"). In some cases Diamond crashes. If the copied part contains polyhedra, these are not pasted into the target structure picture. There are sometimes no atom group pairs in the target, making it impossible to edit the connectivity.
• If you import the file "faujasite.cif" in the "Tutorial" folder and let Diamond create a picture automatically, the program will crash. The primary reason is the false non-centrosymmetric setting of the space group.
• Crash reported on some computers when drawing in rendering mode. (It does not appear when drawing the same picture in flat mode.)
  We can reproduce the crash on one of our computers (runs with MS Windows 98). It does not appear, if you switch off "double buffering" ("Tools -> Options -> Desktop"), but remains when printing or saving into a graphics file. [Oct 15, 2004]
  On at least one computer the problem remains even when switching off "double buffering". [May 31, 2006]

Less severe bugs

• The following functions do not work (correctly) and need to be exchanged in one of the next software patches:
  • The copying of both graphics in Windows metafile format and Diamond 3 document object now work but the "in-place" editing of a Diamond 3 document object in a container application, such as Microsoft Word or PowerPoint, does not work yet. Instead, if you want to change a Diamond object after embedding in a e.g. Word or PowerPoint document, run the "Open Diamond 3 document object..." rather than the "Edit..." command ("Edit" or context menu).
  • The sub-menu "Structure Pictures" of the "View" menu lists a maximum of eight pictures to select from. Workaround: If the current dataset uses more pictures, select either from the Navigation bar or tree or go to "View/Structure Overview..." and select from the thumbnails.
• File import:
  • If you open/import a crystal structure from a CIF, the atoms of the parameter list may get wrong assignments to atom type (e.g. atom "O 4" is interpreted as iron), when "_atom_site_type_symbol" appears *before* "_atom_site_label" in the atomic parameter loop. [Jun 12, 2008]
    Workaround: Before you open the CIF file in Diamond, swap the columns "_atom_site_type_symbol" and "_atom_site_label" and swap the corresponding lines under the "loop_" command.
  • CIF files from the Cambridge Crystallographic Data Centre's CSD often contain symmetry-equivalent positions in their "_atom_site_xxx" loops but Diamond expects an asymmetric unit. Thus there may be too many atoms in the structure picture when using the normal building-up functions like fill cell range, generate molecules, etc. [Jul 17, 2006]
    Workaround: If possible, delete the symmetry-equivalent positions from the atomic parameter list ("Structure/Atomic parameters..." command). These are usually indicated by a letter A, B, etc. and/or an asterisk in the atom label.
• Atomic parameter editing:
  • For space group "P42mc" (Int. Tables no. 105), the Wyckoff position "c" (2mm.: 0,1/2,z) is given with the wrong multiplicity 4, but it should be correctly 2. This leads to false calculated chemical formulas or densities. [Nov 3, 2008]
  • Diamond may crash after a while when you frequently make changes to the atomic parameter list and this parameter list has mixed sites.
• Molecule creation:
  • The number of atoms created during the molecule creation process (e.g. with "Build/Get molecules...") reported in the status bar does often not match the real number of created atoms.
    Workaround: Check the table of created atoms ("View/Table/created atoms") for the real number of (created) atoms.
  • The command "Get molecules directly" (toolbar button or Shift+Ctrl+M) sometimes does not create molecules or complete fragments to molecules.
• Moving:
  • When you use central projection (perspective), you may enter small values, e.g. "20" (cm) in the "Picture/Representation..." dialog but it is actually rejected to at least 25 cm, if you press '+' or '-' on the numeric keypad.
  • When you try to shift a structure picture with '8', '6', '4', or '2' on the numeric keypad (with NumLock on), it rotates rather than shifts when the SHIFT key (used to accelerate for factor 5) is involved. Without SHIFT and with CTRL key it works correctly.
• Atom and bond labels:
  • Bond labels do sometimes deviate significantly from their correct positions in rendering mode but not in flat mode. (The selection box as well as info tip appear at the correct positions, each.)
  • Besides this, atom and bond labels do often deviate slightly from their correct positions when using both rendering mode and central projection. This makes it difficult to select them with a mouse click.
• Printing:
  • Some users report a problem when printing a structure picture. The picture has a miniature size of approximately 1 x 1 cm only. It obviously appears only in rendering mode. Obviously the 2nd, 3rd etc. printing after the erroneous printout is o.k. It is also o.k, if you run a print preview before the actual printing. 
• POV-Ray:
  • When you do a POV-Ray preview (Tools/POV-Ray/Preview or from Tools/POV-Ray/Global Settings -> Preview) or render into a bitmap (Tools/POV-Ray/Render Into Bitmap), there might be a situation where no graphic is displayed or created.
    The critical point is that "POV-Ray for Windows" has a developer user interface ("pvengine.exe") and there is no separate "kernel" executable for the rendering. So Diamond has to execute the "pvengine.exe" and tries to hide as much of the developer user interface as possible. (The splash screen cannot be hidden, reminding that POV-Ray is freeware.) There are critical situations where "pvengine.exe" waits for user input (and Diamond waits and waits...): a) When POV-Ray is started for the first time after installation, it shows a "Tip of the day" dialog window. This means, you should run POV-Ray at least once after installation before you let Diamond launch it and disable the "Show Tip of the day next time" check box in that dialog. b) If the automatic search for updates is active and you are connected with the Internet and there is a newer version available, POV-Ray also shows a dialog. We recommend to disable the option in POV-Ray: "Options/Online Operations/Check For New Versions".
• Powder pattern:
  • When you copy the diffraction diagram graphics to the Windows clipboard using the "Copy Diagram" command from the diagram's context menu, the graphics is pasted incompletely or not at all into other applications, e.g. MS Word. When you use "Edit/Copy" from the main menu, the clipboard receives a severe error. You cannot paste graphics nor selected row(s) from reflection table and further Copy actions in Diamond will fail due to "error opening clipboard". [Apr 19, 2007]
    Workaround: Use "Edit/Paste special..." rather than "Edit/Paste" (Ctrl+V) in the target application and import "device-independent bitmap" format. Do not use "Edit/Copy" from the main menu in the powder pattern view but use "Copy Table" from reflection table, if you want to copy the selected rows OR "Copy Diagram" from diagram's context menu to copy the graphics to the Windows clipboard. 
  • When you decrease 2theta-max in the "Powder Pattern Settings" dialog, Diamond still draws the x-axis up to the previous higher maximum 2theta value but calculates up to the new lower 2theta maximum only. [Apr 19, 2007]
    Workaround: Use the two input fields left beneath reflection table and diagram, enter the new value(s) and confirm with the Enter key.
  • A user from sdpd@yahoogroups.com reported: I just tried to compare four different software: FindIt, Diamond, PowderCell and FullProf to calculate structure for alpha-alumina using ICSD #51687 file (Materials Science Forum(2001), 378, 288-293,Unit Cell   4.7597(1) 4.7597(1) 12.9935(3)). I used the same geometry, wavelength, profile parameters, no background, same space group settings. But all four patterns have different intensity, and structure factors also (for software that gives a possibility to get one). I cannot understand why the structural factors differ. The visualized structures are exactly the same. [Nov 20, 2006]
    
• Other:
  • In flat mode, the styles "Crescent", "Shadow", and "Globe" are not displayed in the atoms in the structure picture (but in the legend, if you define these styles on atom group level). [May 31, 2007]
  • If two or more atoms are selected having different transparency, and you call the "Edit Atom Designs" dialog, "-999" is given in the "Transparency" input field. (Using the up-down arrow beneath adjusts this value between 0 and 1.) [Mar 26, 2007]
  • If you use the option "Transparent" on the "Background" page of the "Picture/Layout" dialog, transparent objects may still appear too dark or too bright in the target graphics file when exporting the structure picture graphics into e.g. a TIFF file. [Feb 7, 2007]
    Workaround: If possible, switch off the option "Transparent" when using transparent objects. 
  • "Picture/Polyhedron Designs...": If more than one polyhedron is selected, the preview sometimes reads "Not unique", although all selected polyhedra use the same design settings.
  • On some computers, the toolbars reset when Diamond is started, i.e. they do not appear at the positions they had the last time you closed Diamond.
  • Dialog "Picture/Representation...": The controls on the three pages need to be adjusted
  • The counts for atoms, bonds, polyhedra at the right end of the status bar are still missing
  • On some Windows XP systems, the tooltips are sometimes not displayed correctly
  • On some Windows XP systems, the atom design preview is not displayed (but black) in rendering mode (dialog from Picture/Atom Designs) 
  • A long structure's or picture's title is not abbreviated when displayed in navigation bar

Bugs resolved in version 3.1f

Version 3.1f was released on February 26, 2008.

• File import:
  • There was still a limitation of 1600 atoms in the parameter list when you read data from a ShelX formatted file.
  • If a CIF file contained only a "data_global" block AND blank lines preceded the "data_global" line, Diamond reported an error message "No structure data found...".
• Symmetry:
  • The unconventional settings of space-group no. 2, e.g. "C-1" were not recognized when you used the "-1" rather than "1-" syntax.
  • For space group Pn-3n (Int. Tables no. 222), wrong Wyckoff positions were assigned in the atomic parameter table. For origin choice 1, the positions for origin choice 2 were assigned and vice versa. This did not affect structure picture or interatomic distances etc.
  • When importing from a ShelX RES file, space groups P3₁ (no. 144), P3₂ (145), P3₁21 (152), and P3₂21 (154) were not recognized from the "SYMM" cards.
• Atomic parameter editing:
  • When using a space group setting where no informations about the Wyckoff sites (letters, multiplicity) is available (cf. "Structure/Space-group..."), Wyckoff letters were erased even if you entered them manually in the dialog "Structure/Atomic parameters...".
• Structure picture:
  • The option "Save view as default orientation" did not take any effect, i.e. the next time you created a new picture for a new or imported data set, the view was along c-axis.
  • In a structure picture that was exported by Pearson's Crystal Data the atom group designs (and thus the legend) deviated from the designs of the corresponding atoms in the picture.
  • In a structure picture that was exported by Pearson's Crystal Data some atoms appeared at wrong positions and had wrong colors and designs, especially if the atomic parameter list contains mixed sites. (Requires Pearson's Crystal Data version 1.0b; available in March 2008.) 
• Printing:
  • When printing the selected row(s) of a table, e.g. the table of created atoms (cf. "View/Table/Created Atoms"), no or only a part of the selected rows appeared in the printout. Additionally, a rather simple "Footer..." was placed at the page bottom.
  • When printing both distances table and distances histogram in distances view (cf. "View/Distances and Angles"), the graphics was not printed (the last page), if the table was too long to fit on one single printout page.
• Distances and angles:
  • When you calculated distances in pm or nm, then closed and re-launched Diamond, it could hang up the next time you went to distances (and angles) view (View/Distances and angles), especially if pm was preset, because Diamond tried to calculate up to e.g. 300 Angstroems rather than picometers.
• Others:
  • Pearson code and Wyckoff sequence were not erased after transformation of a crystal into a molecular structure. Cell edges and corners are now changed to pseudo-atoms with real bonds between during that transformation.
  • "Pearson's Crystal Data" was not stated as origin in the data sheet or data brief.

Bugs resolved in version 3.1e

Version 3.1e was released on May 18, 2007 and contains some bug fixes and minor changes in context with the release of the Pearson's Crystal Data demo version, which can launch Diamond to post-process a crystal structure picture.

• The color of hatching (for thick bonds in flat mode only) was not stored in the Registry, so the default bond design was not restored completely the next time you started Diamond.
• The selection of atom groups to be displayed in the legend ("Objects/Legend..." -> "Select..." -> "Select Atom Groups for Legend" dialog) was not stored in the Diamond document and was not considered during Undo/Redo.
• In data sheet and data brief, super- and subscripts as well as Greek characters are now considered.
• In data sheet and data brief, the publication year was not mentioned under "Citation".
• In the dialog "Model and Radii" a "Reset" button is now available to assign the model that is currently selected in the combobox to all (if no atom is selected) or to the selected atoms (and associated bonds). It also assigns the radii of the atom groups to the created atoms.
• If the mouse cursor hoovers over an invisible atom or bond, the info tip now mentions "invisible" in the info tip text.
• In the diffraction diagram, intensitites are now scaled to Imax=1000 (and not approximately 200) when using profile pattern and the corresponding option is active.
• The commands to rotate or shift selected atoms ("Structure/Change Position/...") was available for crystal structures but should not (nothing happened, if you selected the command yet). Now it is enabled for molecular structures only.
• The color values for color differentiation for symmetry operations were not correctly initialized.

Bugs resolved in version 3.1d

Version 3.1d was released on July 5, 2006 and resolved a severe bug in printing that obviously slipped in version 3.1b:

Printing: Diamond renders a graphics in multiple horizontal bands for printer output (the more bands the bigger the drawing area and the higher the printer resolution). Versions 3.1b and 3.1c sent the bottom-most band to the printer multiple times as the sample below shows (5 bands).

Besides this, the background color was ignored (and white used) in flat mode, when the option to print on white background (default: on) was switched off (cf. "Picture/Layout...").

Version 3.1c

Version 3.1c was released on June 12, 2006 and introduced a new automatic online update checking feature but no bug fixes.

Bugs resolved in version 3.1b

Version 3.1b was released on April 10, 2006.

• Installation:
  • The installation of Diamond 3 did not work under MS Windows 2003 Server.
• File import/export:
  • When you exported a structure into CIF format and at least one atom symbol was blank (cf. command "Structure/Atomic parameters..."), the CIF caused trouble when reading into Diamond again.
  • CIF import: If the origin choice (e.g. origin at centre = centrosymmetric) is not available from the space-group symbol, the symmetry matrices are searched for an "-x,-y,-z" entry to make a decision for the right setting.
• Space groups:
  • Diamond often derived the wrong settings, e.g. "_symmetry_space_group_name_H-M 'P 21/n'" in combination with "_symmetry_Int_Tables_number 14" led to "P 1 21/c 1".
  • The new Hermann-Mauguin symbols "Aem2" (no. 39), "Aea2" (no. 41), "Cmce" (no. 64), "Cmme" (no. 67), and "Ccce" (no. 68) were not considered (cf. Int. Tables for Crystallography, vol. A, 5th ed., ch. 1.3). For instance, "Cmce" is now redirected to the "old" "Cmca".
• Line objects:
  • The components of the vector of a l.s. linethroughthreeormorepointswereoftencalculatedinawrongway as m1=m2=m3=0.
  • The terminal points of the visual line object in the structure picture were sometimes calculated in a wrong way so that it did not appear or at an arbitrary place in the picture.
  • Line objects with at least one of the components m1,m2,m3 zero (parallel to at least one orthogonal axis) did not appear in the structure picture.
• Other:
  • Big transparent atoms that swallowed other objects were not rendered correctly. The restriction of max. 5 A for an atom radius has been removed.
  • "Model and Radii" dialog: The option "Superimpose space-filling" did not work.
  • "Picture/Layout": On the "Background" page, the option "Transparent" did not work (was ignored). Now it deviates, if necessary, the RGB values of the background slightly to a color that no other foreground pixel has in the bitmap.
  • Command "Select atom(s)" from the context menu of the table of atomic parameters (cf. command "View/Table/Atomic parameters") took no affect, if the selected atom is part of a mixed site and not the "representative" atomic parameter of the site.
  • Diamond could crash when you deleted the last atom from the table of atomic parameters ("Delete" command from the context menu) or showed wrong information in the properties view when you deleted another atom than the last one.
  • The commands "Space group..." and "Cell parameters..." are not more available, if the structure has no translational symmetry. (If you entered a space group and/or cell parameters, this had no effect. To introduce translational symmetry, use the command "Add Translational Symmetry..." from the "Structure" menu.)
  • "Edit/Select Molecule" did not work or was not available, if you clicked on a bond rather than on an atom of the molecule to be selected. The properties view was not updated in response of the select molecule command, i.e. did not reflect the new atom selections.
  • The unit cell axes a,b,c were sometimes drawn even for a structure having no translational symmetry. The command "Show or Hide axes a,b,c" is not more available from the drop-down menu in the "Picture" toolbar for structures without translational symmetry.
  • In the navigation bar (document > structure > picture 1 > data sheet, etc.) the link for the structure node was often underlined in a wrong way when the mouse was moved over.
  • The histogram graphics were not copied to the Windows clipboard in response to the "Copy" command of the properties view's context menu.

Bugs resolved in version 3.1a

Version 3.1a was released on December 16, 2005.

• Insert structure(s) from file(s):
  • The command "Structure/Insert From File...", which was enhanced in version 3.1 to enable import of multiple structure data sets in one step, did not work for diamdoc files.
• Jerky rotation:
  • The rotation of a structure picture in "tracking mode" (cf. "Move/Rotate along...axis") became very slow ("jerky"), if the structure table for the current Diamond document ha been viewed (cf. "View/Structure table").
• Lasso selection:
  • When you were in "bitmap" or "printout page" layout (cf. "Picture/Layout..."), the lasso selection did not correctly match the objects inside the free-hand figure (the "lasso").
  • There was no online help available for lasso selection.
• Plane and line objects:
  • Plane and line objects (cf. "Objects" menu) are not displayed in flat mode, only in rendering mode. This restriction was not mentioned in the online help. There is now also an "i" hint in the lower left corner of the structure picture.
• Measuring distances or angles:
  • When you were measuring distances or angles (commands "Tools/Measure[...]"), the yellow info tip window showing the result often disappeared after less than a second in bitmap or printout page layout.
• Properties view:
  • Under some circumstances, Diamond crashed or (closed without error message) when you used one of the options "Distances histogram" or "Planarity/Linearity of selected atoms" in the properties view pane and had a lot of atoms selected.
• Powder pattern:
  • Intensities of reflections could be wrong, caused by typing errors in the scattering factor table. Affected atom types were: H (neutral), Sc3+, Ru4+, and Bi5+. Besides this, wrong intensitites could be caused by wrong determination of multiplicites.
• Other:
  • In the Distances and Angles view, the central atom(s) were not listed in the title bar of the histogram.
  • In some cases, the definition of a (least-squares) plane through four or more atoms led to incorrect plane with m1 = m2 = m3=d=0.
  • A picture thumbnail was not updated in response to the "Picture/Hide" command. "Undo" and "Redo" did not work for "Picture/Hide", too.

Bugs resolved in version 3.1

Version 3.1 was released on September 21, 2005. A list of changes and enhancements can be found here.

• When Diamond read from a CIF file ("File/Open" or "Structure/Insert from file"), it cancelled reading under circumstances, if the file had UNIX format (i.e. simple line feeds rather than return/line feed sequences) and there were comments between the CIF data.
• The export of structure data in CIF format contained errors:
  "_geom_hbond_xxx" informations was not written when exporting structure data in CIF format.
  When exporting "_geom_bond_xxx", "_geom_contact_xxx", "_geom_angle_xxx", and "_geom_torsion_xxx", only two atom symbols but no symmetry info or distance or angle values were written.
• Deleting of structures from structure table (command "View/Structure Table") did not work correctly, if multiple rows were selected.
• A picture thumbnail was not updated when you shifted (or duplicated) labels or text in the structure picture.

Bugs resolved in version 3.0e

Version 3.0e was released on July 20, 2005.

• "Undo" and "Redo" did not work for functions:
  • "Build/Create H-Bonds..." as well as function "Destroy all" in the "H-bonds" dialog.
  • "Build/Polyhedra/Construct Polyhedron...".
• POV-Ray:
  • Dialog "Global POV-Ray Settings", page "Lights" (command "Tools/POV-Ray/Global Settings..."):
    a. When you added a light source, some of the settings of the currently selected light source could be overwritten.
    b. An undefined "point_at" was sometimes given as "<-999,-999,-999>".
• Polyhedra:
  • Diamond could crash, if a polyhedron that was selected in the table of polyhedra and that was built using "Build/Polyhedra/Construct Polyhedron..." contained less atoms than had been selected to construct. (This happens when some of the atoms are eliminated due to the fact that they do not contribute to a convex polyhedron.)
• Other:
  • After you had deleted one or more or all bonds, e.g. with the "Destroy Bonds" command of the context menu, and then tried to fill a coordination sphere of one of the previously bonded atoms, the function failed.
  • When tracking with hidden mouse cursor (see "Tools/Options..." -> "Options" dialog, "Tracking" page, checkbox "Hide mouse cursor while tracking"), the picture was not updated unless you released the mouse button.
  • Structure data sets often did not appear in the original order in the structure table (cf. command "View/Structure Table"), if "No sort" was set in the "Structure Table Settings" dialog.
  • Picture thumbnails in the "Thumbnails" docking window were not updated when you ran the "Undo" or "Redo" command.
  • "Tools/Options" -> "Assistants" page: The options "Show welcome page" did not work correctly.

Bugs resolved in version 3.0d (and minor changes)

Version 3.0d was released on May 4, 2005.

• Copy and Paste of structure pictures:
  • Use of "Copy" and "Paste" ("Edit" menu) to create multiple pictures for one structure data set may lead to Diamond crash. In the thumbnail overview ("View/Structure Overview") there may be more pictures than are really present. After you have saved the document in a "diamdoc" file, you might be unable to open that file again.
• POV-Ray:
  • In some cases, polyhedron faces with four or more edges were missing in the POV-Ray output. Besides this, the polyhedron design "Front faces open" was ignored, i.e. open faces were represented as closed yet. 
• Program start:
  • If the option "Create new (blank) document" was active on the "Desktop" page of the "Options" dialog ("Tools" menu), and you started Diamond, just the "New Document" dialog was shown (no Diamond main window, no splash screen). When you continued with the option "Insert data set(s) from a file", the "Insert Structures From File" dialog opened, followed by the "Insert From File" dialog. The "Insert From File" dialog was closed automatically after some seconds. The same happened when you used the "Create a document and type in structure parameters" option in the "New Document" dialog. (Then the "New Structure" assistant was shut down after some seconds.)
• Insert structure:
  • Inserting of one ore more structures from a file into a Diamond document (command "Structure -> Insert From File") failed, if that file was a *.diamdoc file. In most cases you got the error message "An attempt was made to access an unnamed file past its end", in some cases the program hung up, since it tried to allocate hugh amount of memory. 
• Sample file "unknown.diamdoc":
  • When you opened the tutorial sample file "unknown.diamdoc", then switched to flat mode ("Picture -> Representation", "Representation" page: switch checkbox "Rendering" off), all polyhedra turned to black. If you then clicked on a central atom, e.g. the P atom in the center, and ran the "Polyhedron design" command from the "Edit" sub-menu of the context menu, the fill color was correctly given as "Central atom". If you chose another color, say green, then pressed "Apply Now", it worked, but if you returned to "Central atom", it turned back into black (although the color button now has the correct purple face of P...). The black color was not updated even if you changed the central atom's color ("Edit -> Atom design" in context menu).
• Other:
  • The command "Insert Bonds Directly" (through button in "Build" toolbar) let Diamond crash, if more than two atoms were selected.
  • Some errors in the Stoe PKS format when exporting reflection parameters (powder pattern view, context menu): I(rel) referred to max= 1000 instead of 100, the intensities of the I(abs) column might follow the I(rel) values immediately without blank between, and the I(int) column was missing. The version was also missing.

Bugs resolved in version 3.0c (and minor changes)

Version 3.0c was released on March 30, 2005.

• When copying the structure picture with "Edit/Copy" or Ctrl+C, a bitmap was created but no Windows Metafile, just a blue square instead. Now three formats are available when you do "Edit/Paste Special..." in your client application, e.g. Microsoft Word or PowerPoint:
  a) Diamond 3 Document Object, b) Graphics, c) Bitmap. (Some applications list more formats, such as "Device-independent bitmap"). "Edit/Paste" (Ctrl+V) will paste the Diamond 3 document object.
• The checkbox "Create broken-off bonds" was missing in "Auto-Build Scheme Settings" dialog.
• The commands "Copy Style" and "Paste Style" from the "Edit" menu were not available.
• The command "Copy To New Picture" from the "Edit" menu was not yet available.
• The command "Table Settings..." from the context menu of the table of plane or line objects did not work correctly. The list did not show any items to select from, so the table could be configured.
• Structure picture display on computers with aspect ratio different from 4 : 3:
  • On computers with an aspect ratio different from 4 : 3, e.g. a wide-screen notebook display, objects were mostly distorted, that means you saw ellipses rather than circles for atoms etc. This happened in flat mode and rendering mode, except rendering mode with central projection.
  • As a consequence of the above mentioned distortion, in rendering mode and central projection, the selection of objects could fail, since the object rectangles might not match the objects' representation on the screen.
• Assistants:
  • If the automatic file format recognition in the "File Import Assistant" did not suceed and you chose a format from the combobox manually, the "Next" button was still disabled and you could not proceed the file conversion.
  • A structure picture created automatically or by an assistant, did not show coordinate system and/or legend of atom groups, if these objects had been disabled before in the "Coordinate System" and "Legend" dialog, rsp., ("Objects" menu).
• "Big" structures (with some thousand atomic parameters):
  • The function that checks for mixed sites often assigned multiple atoms of the parameter list to one site, especially when the atomic coordinates were given in cartesian coordinates, e.g. in an XYZ file. The legend of atom groups then showed something like "Fe; FeFe; Fe;Fe;Fe; ...".
  • The "Atom Groups" dialog ("Build" menu) behaved slowly or it took a long time before it opened.
• Atom vectors:
• • Default settings for length and radius (1 and 0.2, rsp., Angstroem) led to hardly visible cones.
  • Under circumstances, an atom vector (cone) could not be selected with a single mouse-click but it was caught with rectangle selection, but if you clicked into the selection rectangle, the selection got lost, so you were unable to edit the atom vector properties.
• Plane objects:
  • Diamond failed or even crashed, if you defined a plane by hkl through 0 (n= 0) (command "Objects/Planes/Create Lattice Plane...").
  • A plane through three or more atoms that passes the origin (i.e. d= 0) was not drawn or drawn incorrectly in the structure picture.
  • A lattice plane with hkl = (0kl), e.g. (0,12,5) and n = 1, was drawn perpendicular to the c-axis and the distance from origin was given as 0 Angstroems rather than 0.5321 Angstroems in the table of planes.
  • If you tried to add another plane with (0kl), e.g. hkl = (0,5,12), Diamond reported wrong error message "Plane already exists".
  • The view coordinates of a new lattice plane were not updated unless you changed/updated the orientation, e.g. by rotating the structure picture "a little bit".
  • When you ran "Edit Plane Style" dialog, the button "Apply Now" worked once but not more for subsequent changes.
• Line objects:
  • On the "Style" page of the "Add Line" or "Edit Line Style" dialog, the words "plane" were replaced by "line".
  • A line through two atoms sometimes ended in one of the atoms or in both atoms.
  • When you ran "Edit Line Style" dialog, the button "Apply Now" worked once but not more for subsequent changes.
• Function "Tools/Measure Planes etc":
  • Wrong mouse cursor symbol (the symbol for torsion angle measuring mode was shown instead).
  • If you measured the angle between two objects, e.g. two planes or between two lines or between a plane and a line, you could not select an atom for the second object that already had been assigned to the first object.
• Properties view (lower right pane):
  • For atom vectors, no object info was given.
  • For legend and coordinate system, the wrong relative position was shown, e.g. "center of picture" rather than "relative to upper left corner".
• Navigation tree ("View/Navigation"):
  • When you changed a structure picture's title ("Picture/Title..."), the title was not updated in the navigation tree. Besides this, the title of the 2nd, 3rd, etc. structure data set was not updated when changed.
  • When you clicked on the structure level node in the navigation tree (the main view then showed thumbnails of the pictures of that structure dataset), then you clicked into one of the picture thumbnails in the main view (the full structure picture view came), then again on the structure level node ==> the thumbnail overview did not return.
• Walk/in out mode ("Move/Walk In/Out"):
  • When you walk out, the structure picture might disappear at "high" camera distance values, but it re-appeared, if you walk back in.
• VRML export:
  • The VRML export (command "File/Save As/Save Graphics As...") often created WRL files that only contained cell edges but no atoms, bonds, polyhedra, etc. An error message was displayed in the status bar of Diamond after the command had been executed.
  • Cortona VRML player displayed an error message "Wrong SFBitMask Value; File...; Line...; Column..." in its console window, if the WRL file contained text. (This text may come from atom or bond labels or from user-defined text.)
• "Start" page: The table of the most recently used files was still missing.
• Dialog "Picture/Representation...": The latest used page is now shown at first when you re-open the dialog.
• When you did tracking in bitmap or printout page layout mode (cf. "Picture/Layout...), the dark gray frame around the bitmap or page, rsp., turned into black. It was restored when you finished tracking mode and updated the graphics ("View/Update" or F5 key).
• Docking window "Thumbnails" (cf. "View/Thumbnails"): The picture thumbnails were often not updated or at least not all of the pictures (if you were using more than one structure picture in your Diamond document) were shown in the "Thumbnails" window.
• The commands "Save Structure As..." and "Save Graphics As..." in the "File/Save As" sub-menu failed, if the structure's and/or picture's title contained characters that are not allowed in MS Windows file names ('"', '*', '/', ':', '<', '>', '?', '\', '|').
• In the dialog "Cell Edges Design" ("Picture" menu), the spin button for "Weight" did not work.
• Document title was not updated after "File/Save As/Save Document As..." in navigation bar and navigation tree.
• If there were more than eight structure pictures for one data set, some of the commands in the "Structure Picture" sub-menu of the "View" menu were visible but not available (disabled).
• "Adjust" did not take into account the camera position, so it did not work if the user had walked through the structure behind the center of rotation.

Bugs resolved in version 3.0b (and minor changes)

Version 3.0b was released on February 4, 2005.

• Navigation bar: There are now two arrows on the right-hand side which makes it possible to scroll through all pictures' titles, if the navigation bar is too short to display the titles of all pictures at one time.
• Navigation bar and picture thumbnails: There was a wrong mouse cursor symbol ("grabbing hand" like e.g. in Adobe Acrobat Reader), if the mouse pointer was placed over a "link" in the navigation bar. The same hand occured when mouse was over a picture thumbnail.
• "New Structure Picture" assistant: Execution of a selected Auto-Build scheme did not work. 
• "Create Structure Picture" assistant ("Picture/Guidance/Picture Creation Assistant..."): Diamond "forgot" the background color that you set on the 4th page ("Picture Design") of this assistant. Besides this, several settings under "Layout" led to other results than expected.
• "Atom Groups" dialog (command "Build/Atom Groups..."):
  • In Diamond 3.0(a) - unlike in Diamond 2 - the occupancies of the atoms of the parameter list were considered when atom groups are defined, leading e.g. to two different groups "AlCo" and "CoAl".
  • If you created a new atom group, the "Info" field at the was not updated, even if you assigned the new group to other atoms of the atom site list below.
  • If you clicked "Reset all", empty lines appeared in the "atom groups" list at the top.
  • The assignments of atom groups to atom sites were not stored in DIAMDOC document.
• "Atom [Group] Designs" dialog (command "Picture/Atom Designs..."):
  • "Atom Group Designs" dialog: Wrong color in preview (default white), which was not corrected unless you clicked into the atom group list or changed to another style.
  • If atom style is "Octant" at the beginning of the dialog, the "More Settings..." button was not visible unless you clicked into the atom group list.
  • If you changed the color of the octant walls (in flat mode: hatching lines), the new color was not displayed in the atom design preview.
  • For style "Mixed/defect site sectors" or "Mixed site color mix", the atom preview was not displayed correctly (no e.g. half spheres or color mix).
  • If you had added a new atom group and assigned that atom group to at least one atom (site) of the parameter list, the assignment of atom group designs to the created atoms went wrong. (Assignment of created atoms to atom groups was not updated.)
• "Add Atoms" dialog and command "Build/Add All Atoms":
  • The function that creates atoms of the parameter list may access the wrong atoms or even crash, if mixed sites are present in the structure.
  • Both function "Add all atoms" and the atoms selected in the dialog may create multiple atoms on one site for mixed sites.
  • A code number range, e.g. "1555011-4555011", specified at the bottom input line of the dialog was not evaluated correctly.
• Auto-Build function:
  • "Auto-Build" dialog: The summary text for the selected Auto-Build scheme did not match the selected scheme's settings.
  • The Undo/Redo function did not work for changes made by an Auto-Build scheme.
• Left mouse button double-click on an object (atom, bond etc.) did not open the corresponding dialog to edit the selected object(s) (as Diamond 2 does).
• "Del" key did not delete selected labels.
• The command "Molecule Cage...", listed in the context menu that appears when you click on a filter or cage symbol in the lower left corner of the structure picture view, led to a message box "This command has not yet been implemented or completed in this pre-release".
• In the "Connectivity" dialog ("Build/Connectivity..." command), the push button "Connect now" ignored the current selection of atom group pairs and their bond ranges.
• Dialog "Viewing Direction": Function "Define axis" did not work.
• There was a wrong sign for y when you shifted the structure picture vertically either by stepping (cursor up [8] or down [2] on numeric keypad) or by dialog (Move/Shift Incrementally). There was also wrong information in status bar.
• "Reset" in "Color differentiation" -> "Color settings" dialog did not work correctly. Obviously all but the currently selected option were reset when "Reset" was clicked.
• "Picture/Hide.": "Hide all atoms" did not work in rendering mode (but did already in flat mode).
• Wrong default setting for z (0) for atom labels, which made labels usually partially invisible (label was swallowed by depth buffer).
• A label hides the tool tip for the corresponding atom.
• Labels were usually drawn too dark (normal vector not set in OpenGL drawing).
• Shift of selected labels and text with the cursor keys (one pixel per keystroke) did not work yet.
• "Edit Text" dialog: "Apply Now" button was renamed to "Close" rather than "Cancel".
• Background color was not saved in Registry after "Layout" dialog, so every new picture started with the default black background.
• Wrong mouse cursor symbol in "Walk in/out mode". (The symbol for "Shift mode" was shown instead)
• "Objects/Coordinate system.": Font attributes italics and/or bold were ignored.
• "Picture/Layout." dialog: If you switched to "bitmap layout" and used a rather small size, e.g. 200x150 pixels and the option "Scale to fit" was active (cf. "View/Zoom."), Diamond could hang up.

Bugs resolved in version 3.0a (and minor changes)

Version 3.0a was released on November 15, 2004.

• POV-Ray export:
  • Wrong scaling of fragmentated bonds. (In most cases, a fragmentated bond was longer than it should be.)
• Atom labels:
  • Individual label text (Edit/Add Atom Labels dialog: Content = "(individual text)") was ignored.
• Layout dialog ("Layout" command from "Picture" menu):
  • Whenever you called this dialog for a 2nd, 3rd etc. time, the actual settings for width, height, and resolution were lost but the default values given, e.g. 800 x 600 pixel, 96 dpi for bitmap layout.
  • The "Apply Now" button did not work when switching between "Window" and other layout, but if it worked, then twice.
  • Actually not a bug, but if you now change the (bitmap) resolution and the width and height of the bitmap are given in pixel units, the width and height are converted, that means the size of the bitmap (in cm or inches) remains the same. (For this reason, we swapped the controls "Resolution..." and "Width...Height..." in the dialog vertically.)
• The page size was calculated in a wrong way, when you defined bitmap dimensions in centimeters or inches rather than pixels (see "Layout" dialog), or when you defined printout page dimensions in inches rather than in centimeters.
• Function "File/Save As/Save Graphics As...":
  • When you saved a structure picture in bitmap layout into a graphics file (bitmap, GIF, etc.) the picture was scaled in a wrong way, usually the frame around the picture was wider than on the screen.
  • The resolution (given in "Layout" dialog) was not stored in the BMP file.
• Functions "Undo" and "Redo":
  • The structure picture graphics on the screen was not correctly restored, when you ran "Undo" or "Redo" after you changed between window and bitmap or printout page layout ("Picture/Layout...").
• Print preview:
  • After you did a print preview for a structure picture in bitmap or printout page layout, the computer did not recalculate the correct size of the drawing area for the screen and thus attempted to render into a relativ huge drawing are instead, which could lead to Diamond hang-up.