Product in the spotlight

What's new in CrystalClear™ 1.3.6 SP1 

There are a few new features in CrystalClear 1.3.6 SP1, as well as many bug fixes and enhancements.

New Features

The new features that are available in CrystalClear 1.3.6 SP1 are:

• Upgrade to version 9.2 of the d*TREK processing package.
• Support for capability to search Cambridge Structural Database for currently indexed (or user-specified) cell parameters and crystal system using installed ConQuest software.
• Incorporated new R-AXIS RAPID and D-MAX RAPID shadow masks.
• Support for multi-scan strategy.
• Updated/improved CCD, R-AXIS, R-AXIS VII (HTC), RAPID, and MARCCD instrument servers.
• Incorporated various site default changes for all instruments.
• Support for AFC11 instrument (with Saturn70 and Saturn92 detectors in left/right configurations).
• Support for automatic exposure time evaluation.
• Support for ranking of crystals.
• Upgraded CrystalClear help file.
• Support for alarm e-mail notification capability for JXG-generated alarms.
• Support for multiple cycle attenuation factor shots.
• Support for toggling Director-started CrystalClear server between server and client (stand-alone) modes while still operating Director.
• Updated Adaptec ASPI installation.

Bug fixes/Enhancements

The bug fixes/enhancements that are available in CrystalClear 1.3.6SP1 are:

• Enhancements
• Implemented warning dialogs which display if user attempts to use Processing State Display dialog to significantly change the 2-theta and distance detector goniometer parameters.
• Implemented JXG and RMailer installation, executable path, and executable command line specifications in the new application.configuration file located in the CrystalClear installation directory. These parameters are now user-specifiable through this configuration file allowing flexibility in installing both JXG and RMailer independent of CrystalClear.
• Enabled functionality of the d*TREK Centricity and Spacegroup Check commands in CrystalClear whenever accessed via the Processing menu.
• Changed text for CIF write flowbar button from "Write CIF File" to "Write CIF" in CrystalClear.
• Restored the strategy results summary table in dtstrategy.log.
• Added a new Processing->Run JDTPlot... menu item to allow running of JDTPlot during integration or elsewhere.
• Improved d*TREK Set Resolution dialog behavior including the addition of an Apply to all processing dialogs checkbox. Also added capability to select resolutions by 2-theta in degrees dependent on the resolution units user preference.
• Implemented change to CrystalClear interface to dtstrategy to prompt user whether to accept the best obtainable solution if dtstrategy fails because requested completeness and/or redundancy could not be met rather than just issuing a failure message. Also added an image width field that specifies image width to set for strategy result scans.
• Implemented user preference allowing file open dialogs to always initialize to current sample directory (or sample image directory) or to initialize with last directory path (as was previously done).
• Improved abort confirmation dialog behavior and aesthetics and removed automatic and continual retry of aborts if initial confirmation timout is exceeded.
• Upgraded version of JDTPlot.
• Implemented alarm e-mail notification capability for JXG-generated alarms.
• Implemented new Adminstration Utilites dialog capable of supporting multiple cycle attenuation factor shots.
• Implemented alarm/failure e-mail notification capability for sever-generated failures and aborts for all processing and collection servers. E-mail subject identifies the type of server which had a failure or abort.
• Added logging of vacuum recover attempts/failures/problems. Error log is placed in the installation directory and called vacuum_error.log.
• Modified the R-AXIS server vacuum recovery procedure: The new procedure will be to try to turn the vacuum on 10 times. If it fails, then datum the IP and then rotate the plate all the way around 5 times. It will loop back up to try turning on the vacuum and rotating the plate. It will do the outer loop 5 times before failing.
• Bug fixes: CrystalClear
• Fixed problems in image viewer associated with pixel selections in the "dead region" of wide RAPID images where there are valid display pixels but invalid image pixels. Among the fixes is to not display pixel information if user selects pixels in this invalid region.
• Fixed problem with scan table behavior where scheduling images for collection, then unscheduling them by blanking out the Scheduled field, causes image corresponding to 0 deg rotation angle (omega) to be scheduled for collection when losing focus from either Scheduled Start or Scheduled End fields.
• Fixed problem with overlay image dialog initializing the default value as 1 for the initial image sequence field rather than the first image sequence in the actual scan. Also fixed display of number of images field to be the actual number of images in the scan if number of images is less than 10. If number of images exceeds 10, then the default is limited to 10.
• Fixed problem with Set Resolution dialog setting minimum resolution value numerically to zero on the d*TREK -reso command line option (which requires resolution in angstroms) of several commands if dialog value was previously #< 0.001 or user entered a value #< 0.001 and the current CrystalClear resolution unit user preference is degrees. Now, if such a value is #< 0.001, CrystalClear places an arbitrarily large value of 100 angstroms as the minimum for the -reso option.
• Fixed problem with d*TREK Strategy dialog using refined parameters for 2-theta and distance default values displayed in the respective entry fields.
• Fixed problem with rotation limits being adjusted properly in d*TREK Strategy dialog according to value of 2-theta. Also implemented automatic adjustments of these fields as 2-theta value is changed by the user. Both limits are changed so as to keep the relative angular width between the limits as 2-theta is changed.
• Fixed problem image viewer not initially loading proper 2-theta and distance information from image header and then properly adjusting direct beam position. Behavior is changed now to only display generic (amanually loaded) and Collection image displays with with image header information ONLY (no refined scan parameter values). Processing-loaded images will still be displayed as before using refined scan parameters for 2-theta and distance obtained from CrystalClear processing database.
• Fixed problem with CrystalClear hanging when d*TREK Laue check prompts user if he/she wants to reindex to find a unique cell and the user answers yes.
• Fixed problem with CCD collection dialog applying dialog collection parameters (binning mode and dezinger flag) to all schedule regardless of whether one or more images have already been collected for the scan. Fixed problem so that these parameters can only be modified and applied to purely scheduled scans (i.e. no images yet collected).
• Fixed problem with image viewer not properly reading image header information which affected the proper display of the direct beam crosshairs.
• Fixed problem with d*TREK strategy interface not incorporating into the resultant strategy scan the 2-theta and crystal to detector distance refined parameters from the last scan previously processed. Before, this information was always being obtained from the first screen scan in the first screen schedule. Also added a Distance entry field allowing user to specify a desired distance for the strategy scan. Also fixed defaults for the 2-theta and distance entry fields to be those of the previously processed scan unless user has explicitly modified one of them in the Strategy dialog from a previous session. In such a case, default value(s) is obtained from the dialog parameter database for the Strategy dialog.
• Fixed problem with Write CIF flowbar button being reactivated each time CrystalClear sample is opened if it was last command previously run.
• Fixed problem with input cell table on the d*TREK Reduce Cell dialog in CrystalClear.
• Fixed problem with CCD collection dialogs saving currently selected binning mode and dezinger exposure flag to all scans in scan table regardless of how any images already collected in the scan were actually collected. Fixed behavior of CCD image collection dialogs to only save these parameters to purely scheduled scans. Also incorporated an "autofix" feature for existing samples with already collected scans to automatically and silently save proper image header values for these parameters to the collected scans based on the image header of first collected image if the image header and CrystalClear database saved values differ.
• Fixed problem with ranking command getting wrong scan 2-theta.
• Fixed problem with ranking command getting wrong direct beam values based on getting wrong CCD binning mode.
• Fixed problem with Final Refine Cell (SM version) step choosing wrong reflection file from what user selected for the d*TREK dtrefine command line. Also fixed the default reflection file to be dtprofit.ref if only one scan available and dtprofit1.ref if two or more scans are available.
• Fixed problem with zooming in on certain images in image viewer crashing CrystalClear.
• Fixed problem with dtstrategy issueing an error if it can't satisfy the completeness or redundancy level in one scan. This problem is now remedied with the incorporation of mult-scan strategy.
• Fixed problem with RMailer e-mail notification of hardware errors not attaching the instrument log files from the proper directory. The proper log files are now obtained from the image directory rather than CrystalClear installation directory. Furthermore, all versioned log files are also attached and sent.
• Fixed problem with strategy failing if running from the Strategy main tab without having visited the Advanced tab first due to chi and phi check string defaults not beining initialized and saved to the strategy Hobject database. Fixed default files to contain a non-empty string value to prevent this type failure.
• Fixed problem with CrystalClear leaving copied jdtplot.jar files in a sample directory after JDTPlot was used. CrystalClear will delete any jdtplot.jar found in the current sample directory (as long as it can be deleted -- e.g. if an instance of JDTPlot is not running) upon closing application, closing sample, or opening another existing/new sample.
• Fixed problems associated with direct beam and non-uniformity information not being updated if detector information is being requested or updated in d*TREK. Problems include but are not limited to direct beam crosshairs not displaying properly on the second or subsequent images if two or more images are opened in image viewer from images of different instruments or images of same instrument having different direct beam positions. Fix also included improvements in the behavior of the entry dialog of the image viewer.
• Fixed problem with image header viewer dialog unnecessarily requiring a sample be open. Enhanced image header viewer dialog to display name of image being viewed.
• Fixed problem with not being able to rank crystals when run from a user directory having spaces in path such as from the default Administrator account.
• Fixed CrystalClear installation to properly check and upgrade the ASPI driver to version 4.71.2 .
• Made detector distance and 2-theta parameters issue a GetCollectionTime instrument server command whenever changed in the scan table. This allows the exercise of the collision detection check built into this command and the possible early warning of a problem. For 2-theta, this is necessary in the AFC11.
• Fixed collection scan table defaults for all CCD instruments except the AFC7 Mercury to prevent scan start and end angles from exceeding allowable ranges. This became necessary after the upgraded CCD instrument server in SP1e now detects and issues server warnings when this happens.
• Fixed problem with issuance of ForceUpdateInstrumentDisplay server command in upgraded server causing the Abort mechanisms to disappear before instrument initialization and other commands finished, preventing user from being able to abort such commands.
• Fixed problem with garbage due to floating point overflow from being displayed in the d*TREK Set Resolution dialog minimum/maximum entry fields.
• Fixed problem with instrument collection server not being reset to the installation directory upon restarting if server was killed after being initially started.
• Fixed problem with resolution unit labels not changing at the appropriate times in image viewer. Also fixed problem with generically opened images (not associated with sample) using the sample's wavelength to determine resolution values, a problem if these wavelengths differ. Also fixed problem with pixel information not blanking out whenever exiting Select mode in image viewer.
• Added second display field in image viewer to allow simultaneous display of pixel resolution in units of angstroms and degrees. Also enhanced numeric labels of resolution and ice ring arcs by 1) showing resolution values in both angstroms and degrees, and 2) displaying the unit symbols with the values. The default unit value (as specified in User Preferences) is shown above the alternate init value enclosed in parenthesis.
• Fixed problem with Processing State Display not updating direct beam position and pixel size values. This is especially important to keep track of these values after such operations as indexing with the beam check utility on.
• Fixed problem with Set Resolution dialog displaying incorrect maximum resolution values for the image edge and corner whenever sample images are wide RAPID or D-MAX RAPID images due to the cylindrical geometry of the image plates.
• Fixed problem with Processing State Display showing incorrect pixel size.
• Fixed problem with scans not having a default value for rotation axis name in the Collect Initial Images dialog for the R-AXIS HTC instrument.
• Fixed problem with image unable to load a zipped image file.
• Fixed problem with Twinsolve spacegroup check results dialog not displaying all results.
• Fixed problem with Rebuild Scans function introducing a "_screen" into the scan templates of strategy results scans whenever the sample has no collection images.
• Fixed problem with new version of JDTPlot not properly reading dtcaleaverage log files by rolling back to previous working version.
• Fixed problem with Processing State Display not showing the refined beam positions and pixel size for collect-type tasks.
• Fixed problem with collection scans erroneously getting scan templates set with a "_screen" causing problems during refine and prediction.
• Fixed problem with abort confirmation dialogs not displaying properly in Windows with small font setting.
• Fixed problem with d*TREK strategy results being set with -1 wavelength.
• Fixed problem with d*TREK strategy failing when using the "Use Previous Reflections" option.
• Fixed problem with D-MAX RAPID orientation vector definition in d*TREK interface.
• Fixed crash problem in d*TREK Strategy when saving after modifying phi and chi checks edit boxes.
• Fixed problem opening d*TREK Strategy dialog from menu option when in a Process-only task.
• Fixed CrystalClear installation to check for TwinsolveLicense.lic license file in a 1.3.5 installation and, if available, copy to CrystalClear 1.3.6 installation during an upgrade.
• Fixed problem with Find Spots command not providing dtfind with the correct image-specific binning dimensions via the input header for CCD images. This is manifested whenever screen images are collected at 2x2 binning causing spots to be found in only a quarter of collect images when such images are collected at 1x1 binning.
• Fixed CrystalClear crash in Index dialog right-button "View.." menu item when no reflection files are available.
• Fixed problem with *.raw reflection files being required when running TwinSolve Space Group check on orientation reflections.
• Fixed problem with image viewer viewport not configured to properly display a new image file's binning dimensions if the viewport was initially configured to display an image of another different set of binning dimensions.
• Fixed problem with R-AXIS VII Chi (HTC) site default file have empty rotation axis values in scan tables.
• Fixed problem with not being able to save certain image collection dialog parameters (e.g. Read Head) to any level except Sample level for the R-AXIS V and R-AXIS VII Chi instruments.
• Incorporated capability to search Cambridge Structural Database for currently indexed (or user-specified) cell parameters and crystal system using installed ConQuest software.
• Moved "Apply mask" parameter on the d*TREK Predict Spots Advanced tab dialog to the Main tab dialog . Since this parameter was the only parameter remaining on the Advanced tab, this tab was removed from display.
• Fixed problem with command-generated modeless result dialogs not being closed when opening the Open Sample dialog. The fix includes notifying user that the command is still running (and thus the result dialog is still displayed) and prompting him whether to continue with his intent to open the open Sample dialog. If he wishes to continue, the currently running command is aborted, closing the persistent results dialog.
• Fixed problem with potential confusion when user changes task before changing sample in the Open Sample dialog. This fix involved improving visual cues that the task and image directory parameters "belong" to the currently-selected sample. Also changed New Project and New Sample radio buttons to GUI-standard checkboxes. Also, the dialog title is changed to reflect current purpose of the Open Sample dialog depending on which of these checkboxes are checked or which File menu item is selected to open this dialog.
• Fixed problem with 1.3.6 installation not copying configuration files (instrument and others) from a 1.3.5 installation to the 1.3.6 installation when upgrading.
• Fixed problem with 1.3.6 installation not checking for and carrying over the aspci.dll from a 1.3.5 installation. The 1.3.6 installation now checks the availability of this dll and copies it to the Windows system directory where it really belongs.
• Fixed problem with TwinSolve Numerical Absorption Correction dialog incorrectly asking for a non-zero radius when Gaussian integration method is selected.
• Fixed problem with running strategy through Director and in Process-only samples
• Updated Screen and Collect schedule defaults for AFC10
• Fixed scan tables to reflect fixed chi on AFC10
• Fixed problem with non-standard/anode-element wavelength with synchrotron source type not getting transferred to instrument server properly
• Fixed problem with instrument servers not being started up in correct directory when switching tasks.
• Fixed problem with processing commands getting minimum resolution of 0.0 when preferences set to degrees.
• Bug fixes: Director-related
• Fixed problem in CrystalClear/Director interface where strategy would fail because no reflections could be found due to 2-theta and distance each being assigned to the value -99999.0.
• Fixed problem in CrystalClear/Director interface when running a recipe with both Screen and Collect where strategy is also run and the ranking fails because it the input.head header file it is input with contains the collection scan template and not the correct screen collection scan template upon which the ranking should be based.
• Fixed problem in CrystalClear when running in Director server mode where user could open a sample and collect an image without going into client (stand-alone) mode which interfered with the current recipe and caused a crash. Now, if user is in server mode and attempts to open a sample and then collect, a warning message will appear instructing user to get out of server mode before collecting. In this case, the collection dialog is prevented from opening. Also added the string "(Director server mode)" to mainframe title indicating to user the current Director server mode status.
• Fixed problem with images not being processed for the ranking command when Crystalclear is run as a server from Director and user only collects screen scans and runs strategy.
• Fixed problem with CrystalClear/Director interface where Collect-only recipes caused the CrystalClear server to crash.
• Bug fixes: Instrument servers
• Reenabled old-style syntax in ExposeAndRead (for CCDs) command so that it will work with CrystalClear. Changed the command so that overload correction is not done, even if turned on in configuration file.
• Fixed bug in goniometer movement commands that set the goniometer in a failed state if a pre-condition to the move (e.g., attempting to move beyond the allowable range of an axis), thus requiring a reinitialization. Reinitialization is no longer needed if a precondition fails.
• Added configuration file option to allow specification of the order of datuming the axes when initializing (for CCDs). The configuration file (MSCServDetCCD.configuration) option is

   
        InitializeOrder = {axis1 axis2 ... } 
     

• Added a recover initialization option (for CCDs), via the command line (MSCServDetCCD -recover). This is intended for situations where omega/2theta are in a collision condition, and needs to be reinitialized. The user will need to specify the current approximate positions of the axes, and then the axes are "datumed in place". Then phi is datumed, then distance is datumed, followed by chi. Then omega and 2theta are moved roughly together back to 0. The server then exits.
• Improved the estimate of collection time remaining for a scan.
• Updated CCD instrument server for AFC8 so that the 2-theta and omega movements use the AFC11 movement mechanism (the PF command is no longer used for such movements for the AFC8).
• RAPID, HTC, RAXIS: for 1/4 chi systems, the maximum omega speed was slowed down.
• On occasion with the R-AXIS IV++, the response for the vacuum state status command read from the serial port does not contain the last couple of characters, causing a vacuum error state. This is now treated as a recoverable error and the command is resent.
• Changed serial port reads to read all the characters up to the delimiter in a single read, rather than reading the characters one at a time.
• Moved the one second sleep to earlier in the vacuum cycle, so that it is always done before interpreting the vacuum state response. This addresses the problem found in some instrument where the vacuum is report as off by the controller but isn't completely off.
• Added a check for intermediate states when turning vacuum on/off, found in some instances on IVs with mechanical problems. If an intermediate state is found, then the vacuum status is requeried.
• Fixed a problem which caused some IP movements to treat an abort request as an IP movement error.
• Turned off checking for socket reset errors after one such error has occurred (as would happen if the client crashed).
• Changed the manner in which the dark current with the MAR CCD so that it is taken so that it conforms to the latest MAR procedure.
• Fixed problem with CCD instrument server so that it sends the Setup command to send the binning mode of the detector before it gets the header used for updating the scan's header. This fixed problem where crystal-to- detector distance comes up at half correct value because pixel sizes in header of images are wrong.
• RAPID instrument server contains necessary changes to properly handle D-MAX RAPID collection.
• Fixed CCD instrument server problem where no oblique incidence factors not being written in image header for Saturn92 copper.
• Removal of backlash correction when movement near minimum axis position for CCDs.
• Fixed bug in taking dark current images for overload correction images when the overload correction time is longer than 60 seconds.
• Removed collision detection check when generating collection time estimate for AFC7.
• Bug fixes: d*TREK
• Fixed problem with stand-alone d*TREK strategy.
• Incorporated new R-AXIS RAPID shadow masks.
• Restored summary table listing at the end of the d*TREK strategy log file.
• Fixed problem with d*TREK crashing when mask image and image to be masked have different dimensions.
• Fixed problem with Strategy calculating an invalid omega range due to incorrect default rotation limits for the R-AXIS RAPID-S, R-AXIS RAPID-F, and D-MAX RAPID instruments.
• Fixed d*TREK Scale and Average dialog to validate that a non-empty batch ID template was entered by user if the Batch ID flag is checked.
• Revised d*TREK to place SOURCE_WAVELENGTH keyword in the output dtscale.ref file.
• Fixed d*TREK multi-scan strategy.