SLICER DICER Release Notes

Copyright 2003-2012 Pixotec, LLC

2 October 2012


CURRENT VERSION: 5.2 (Slicer Dicer) and 1.0.4 (3VO)

V5.2  Slicer Dicer

Axis scaling behavior is improved. In the case of arbitrary scales, negative values are permitted, having the effect of reversing the axis. Previously, there were situations in which the 3VO view would have incorrect "handedness", i.e., it would be the mirror image of the portrayal in the Slicer Dicer window. This is now fixed.

V5.1.1  Slicer Dicer

Minor memory deallocation problems (memory leaks) have been fixed in this version.

V1.0.4  3VO

In certain circumstances, 3V0 would fail to open properly. A loop would be entered in the initialization processing and the application window would appear with a black area and display the busy cursor indefinitely.

Some isosurfaces were not being rendered correctly in 3V0. When this happened, there would be a large gap between the edges of these isosurfaces and data rendered on planar (non-isosurface) objects. Also, some isosurfaces were "off-by-one" relative to their rendering in the Slicer Dicer window. That is, the 3V0 isosurface would be computed for an opacity threshold one less than that in Slicer Dicer.

In the 3V0 window the colors assigned to any data corresponding to either the minimum or maximum of the assigned range were incorrect.

All of these problems have been fixed in this version.

V5.1  Slicer Dicer

Switched the online help system from winhelp to htmlhelp. This achieves two benefits: 1) the more modern functionality of htmlhelp, and 2) compatibility with Windows Vista, since Vista does not support winhelp. Seeker's online help has also been converted to htmlhelp.

Under Vista, previous versions would occasionally crash due to a peculiarity in the way Vista frees memory objects. This problem is now fixed.

V5.0.1  Slicer Dicer

Some Japanese users reported a problem with the operation of the the "Transparency" popup control in the Transparency dialog. (The popup would not expand correctly when clicked.) Although most installations appear to be unaffected by this problem, new code supporting the operation of this particular control has been inserted, which appears to fix it for the problem cases.

V1.0.1  3VO

This release fixes a bug in 3VO. Previous versions would, in a certain circumstance, use an incorrect color scale. Specifically, this would happen when the data range (minimum - maximum) specified by the user in the File Parameters dialog in Slicer Dicer would extend beyond the actual data range. The color scale in Slicer Dicer would be correctly assigned to the selected range, but in 3VO the colors would be assigned to the range defined by the intersection of the two intervals -- actual and selected. For example, suppose the actual data range were 0 - 1000. If the user selected a range of 100 - 2000 (minimum - maximum), the color scale in 3VO would be assigned (incorrectly) to 100 - 1000.

This problem is now fixed in v1.0.1.


V5.0  (Slicer Dicer) and 1.0 (3VO)

3VO - OpenGL Viewing Window

3VO is an accessory application installed with Slicer Dicer, starting with this version. It provides an enhanced 3D viewer for scenes created in Slicer Dicer. The standard Slicer Dicer document window provides a viewing environment that is well-suited for interactive data exploration, but this view is static, in that you can't interactively rotate or zoom the scene. With 3VO installed, you can open a second document window with enhanced interactive 3D viewing. The scene in this window is drawn with polygon rendering algorithms supported by the OpenGL subsystem on your computer. You will still use the standard Slicer Dicer window to add/delete/move display objects ("slicing and dicing"),  and you will use the 3VO window to inspect the results. The portrayal here will be more realistic, and you will have much more control over the scale and rotational orientation of data displayed in this window.

In addition to its interactive viewing capabilities, 3VO can also export 3D model data to a file. You can use this feature to export your scene to a  file in any of several standard formats that are used in the 3D modeling industry. This makes it possible to extract 3D objects from your data set and merge them into other modeling environments.

Large File Support

NetCDF "Large File Support" (LFS) is now included in Slicer Dicer. Previous versions were linked to v3.4 of the netCDF libraries from Unidata (UCAR). This version is linked with netCDF v3.6, which adds LFS. The following is quoted from the FAQ entitled "What is Large File Support?" on the Unidata web site (

"Large File Support (LFS) refers to operating system and C library facilities to support files larger than 2 GiB. On many 32-bit platforms the default size of a file offset is still a 4-byte signed integer, which limits the maximum size of a file to 2 GiB. Using LFS interfaces and the 64-bit file offset type, the maximum size of a file may be as large as 2**63 bytes, or 8 EiB. For many current platforms, large file macros or appropriate compiler flags have to be set to build a library with support for large files. This is handled automatically in netCDF 3.6."

NetCDF datasets with LFS can be created on 32-bit Windows platforms (using netCDF v3.6). An error occurred when these files were opened with previous versions of Slicer Dicer. As a result, the file was misinterpreted as a generic file format. Now, both LFS and, older, non-LFS netCDF files can be successfully opened.

Better Tick Marks

When you have tick marks showing and you zoom to a larger scale, the tick marks will be redistributed according to the default ("pretty ticks") scheme. This prevents the tick mark density from becoming extreme when the scale change is significant. Previously, when you opened a very small data set and then zoomed up so that the datacube had a convenient size, the tick marks were not reallocated and would become very dense.


1. There was an error in the algorithm used in the Format Analysis command in the Generic File dialog. As a result, the number of columns reported for columnar data files was off-by-one (one less than the actual number).

Status: Now fixed.



The Generic File dialog has been extensively revised. The modifications are designed to both simplify the process of opening a generic file and to support additional functionality. This dialog has two versions, one for text files and one for binary files. In both versions, the various controls are presented on tabbed pages, instead of all together in one window. On each page, the user is presented with a relatively small, coherent set of issues, and there is room for explanatory text and hints to help the user deal with these issues. The help topics displayed when the Help button is pressed are now specific to the current tab and are more detailed than in previous versions.

The interface for entering dimensional information in these dialogs is streamlined. Previously, one clicked on a row in a table and edited the fields of the selected row in a separate set of edit windows. Now, you can click on any field and edit it in place.

It is now much easier to open x-y-z-value data files and other similar columnar formats. In previous versions, it was necessary to include a special dummy dimension of size four for the purpose of enumerating the x, y, z, value components. This is no longer necessary. Now one treats the individual columns as variables. This has the advantage of eliminating the need for the dummy dimension and makes it possible to conveniently handle multi-variate data sets where the array elements are n-tuples of variables, e.g., (x, y, z, value) or (u, v, w). The n-tuple components can be any mixture of dependent and independent variables, and may even include non-numeric columns.

The Slicer Dicer generic metadata file (.gm) format has been revised in order to support the new multi-variate options. These extensions apply to both text and binary data files. Most (but not necessarily all) old version .gm files can still be read. A few old version .gm files cannot be read and will need to be converted to the new format.

The generic text file dialog includes a new option to generate a generic metadata (.gm) file. This is a convenient way to create metadata files, although it is still easy to prepare these files with any text editor.

For text data, an optional format analysis command is available that parses a file and attempts to determine how fields are delimited and the "shape" of the array (number of dimensions and dimension sizes).

There is now limited support for missing data in text arrays. Missing data or markers representing missing data will be detected by Slicer Dicer. Survey operations will skip these data. Upon visualization, such data will be clipped automatically as "out of range."

Data file survey results presented in the File Parameters dialog now list the five lowest and five highest values found, rather than just the two extremes (maximum and minimum). These are listed in popup controls, so you can easily select which is to be used for scaling purposes. This will be useful for data sets that include wild values or special absurd values intended to represent special data, such as missing values. These should show up in the survey results as outliers together with the true range limits.

The following new data formats are supported: Analyze 7.5 (Mayo Clinic). Surfer Grid (Golden Software).     

The operation of the AVI (animation) playback window has been improved. The image aspect ratio is preserved when the window is resized or maximized. The updating of the slider position and frame count annotation is more responsive.

There is a new "Maintain Aspect Ratio" option in the Scales and Rotations dialog. When this option is checked, the dimension scale factors are linked together so that when you change one, the others are adjusted automatically to preserve their ratios. This is convenient when you want to zoom the size of the view volume in all dimensions simultaneously. 

Some formats (e.g., DICOM, Analyze, Surfer Grid) may include voxel shape information. Slicer Dicer now detects this information and makes it conveniently available for scaling the dimensions.


1. In many Slicer Dicer dialogs that included numeric entry windows, when you pressed Cancel and a validation error existed in one of those windows, you would be forced to fix the error unnecessarily.

Status: Now fixed. The dialog will be dismissed silently, even if there are validation errors. Since the operation effected by the dialog is being cancelled, the error has no significance.



This version of Slicer Dicer is designed to work with Seeker, Pixotec's new DICOM file browser. Slicer Dicer automatically detects when Seeker is installed and adds two commands to the File menu: "DICOM Browser" and "DICOM Information." The first command activates Seeker, which can be used to locate a DICOM dataset. Seeker's Slicer Dicer button can then be used to return to Slicer Dicer and open a new document for the selected dataset. The second command is enabled when a DICOM dataset is already open in Slicer Dicer. When selected, it activates Seeker where the DICOM header information for that dataset is displayed.

Both Seeker and Slicer Dicer are stand-alone applications and will function independently, but when they are both installed they work together cooperatively. Seeker is distributed free with Slicer Dicer. It can also be downloaded from


1. Since support for the DICOM file format was added in v3.0, a number of shortcomings of the handling of this format have emerged: 1) Not all DICOM files were recognized. 2) The ensemble of files constituting a multiple-image dataset was not always detected correctly. 3) Similarly, in a multiple-image (multi-file) dataset, the frames were not always placed in the correct order. The latter two problems were a consequence of applying Slicer Dicer's standard filename conventions for associating and ordering the files in a multiple-file dataset. This approach is not appropriate for DICOM data, where header information is available for this purpose.

Status: Slicer Dicer's support for the DICOM format is significantly improved in this version. Low-level support for the DICOM format has been completely revamped, with the result that reading these files is much more robust. DICOM files are recognized and opened if at all possible, even in some instances where the format is non-standard. File names and name extensions are ignored in the handling of multi-file datasets. When a DICOM file is selected in the Open command, all other DICOM files in the same directory are processed (assuming the Open Multiple-File Dataset option has been selected) to find the set of files sharing the same Series Instance UID. If available in the headers, slice locations or positions, or image numbers, are used to order the frames. Only if none of these is available is the order based on the lexicographical filename order.

2. In previous versions a crash would occur when a blank line in the Variables popup in the File Parameters dialog was selected.

Status: Now fixed. No extraneous blank lines are displayed in the popup controls in Slicer Dicer dialogs.

3. Recent Slicer Dicer versions have been built using HDF 4.1.r2 a code library distributed by NCSA. There were bugs in that version affecting the reading of HDF files. These bugs would generally be innocuous, except for large files where they could very significantly slow down the input process.

Status: Slicer Dicer is now built with HDF 4.1.r4, which appears to have fixed these bugs. It is now possible to read certain very large HDF files that were prohibitively slow with previous versions.




1. In certain circumstances, opening a multiple-file data set was not working properly. This problem would arise when uppercase characters were used in the filenames and the option "Allow all uppercase names" was not enabled in Windows Explorer settings. The workaround was to make sure all characters after the first in the filenames in a multiple-file dataset are lowercase, or to select the "Allow all uppercase names" option. The latter option is not available for all users. It apparently is available on NT, for example, when the Active Desktop is installed.

Status: This problem is fixed in v3.0.4. You can use either upper or lower case in all filenames. Slicer Dicer converts everything in these names to uppercase when it parses filenames when opening a multiple-file dataset.

2. Some users were unable to open sample document files (.dcr) distributed with Slicer Dicer. This problem would occur when disk volume E: is a CD drive in an unmounted state. In this situation, when the user attempted to open one of the sample documents, the open would fail with the message "Can't access e:\home-d\filename" where "filename" was the name of the data file associated with the document.

Status: This problem is fixed in v3.0.4. For a normal installation, the sample dcr file and associated data files are placed in a directory named Samples, which is located in the Slicer Dicer installation directory. When the document file is opened, Slicer Dicer will find the data file automatically. If for some reason (the file has been moved, or renamed, etc.) the data file is not found, Slicer Dicer will present a dialog asking the user to either find that file or choose an alternate.




1. In previous versions the File Parameters Use Survey command did not function correctly when a data survey found a zero range, that is when the minimum and maximum values were equal.

Status: Now fixed in v3.0.3.

2. Multi-file data sets were not being read correctly for generic binary files. (Image files, non-generic binary formats, and text file imports were ok.) Typically, the first file in the set would be read correctly, but all subsequent files were not being read correctly. In fact all data representing these subsequent files were in fact copies of the first file.

Status: Now fixed in v3.0.3.




1, Version v3.0.1 had a bug that would cause a crash when the user attempted to open a .dcr file pointing to multi-file data set that had been moved to a new location after the .dcr file was created.

Status: Now fixed in v3.0.2.



This release fixes a few problems that showed up in v3.0.


1. Version v3.0r1 had a serious bug affecting multiple-file data set reads. In certain cases, this bug would cause the program to adopt a very inefficient mode for importing the data. When this happened, it could take many minutes, or even hours, to read a data set consisting of a few dozen image files. Typically, this would happen when the image pixels were 16-bit grayscale integers or in other cases with other than 8-bit pixels.

Status: Now fixed in v3.0.1. It now takes only a few seconds or less to read these data sets.

2. When a multiple-file data set was opened with version v3.0r1, and when each file was a multiple-image file, only the first image in each file would be extracted. This wasn't really a bug. It was designed that way.

Status: This limitation had been removed in v3.0.1. The case of a multiple-file data set with multiple-image files becomes a 4D data set.



In addition to bug fixes and minor changes, this version includes several important new features. The online documentation has been amended to reflect these changes.

Multiple-File Data Sets

In many applications 3D data sets exist in the form of multiple image files. It is no longer necessary to concatenate these images into a single file prior to importing them into Slicer Dicer.

Slicer Dicer's open dialog now offers an "Open multiple-file data set" option. When a file is selected for opening with this option enabled, Slicer Dicer searches that file's directory for all files satisfying a certain name rule. Specifically, it checks to see if the selected file name has an imbedded numeric substring, and then it searches for other files with names that are identical except for the value of this numeric field. Each file must be of similar type (e.g., netCDF, TIFF, etc.) and contain an array with the same rank and size. The files, individually representing n-dimensional sets, are sorted on the numeric field and opened collectively to form an (n+1)-dimensional set.

In the case of image data, a set of 2D image files forms a 3D data set, but there is no limitation on the rank of the data sets. For example, a set of 3D (x, y, z) arrays representing a time series can be opened as a 4D (t, x, y, z) set.

Support for Standard Image Formats

Slicer Dicer now recognizes over 45 standard image formats. The list includes AVI, BMP, CLP, DICOM, DIB, EPS, HDF, JPG, PICT, TIFF, WMF, and many others. This has been achieved by incorporating powerful image processing technology from AccuSoft Corp.

It should now be possible to open virtually any image file, regardless of format, bit depth, compression, etc. This capability will be especially useful in conjunction with multiple-file data sets. Support for the DICOM format will facilitate many applications in medical arenas.

Multiple images (pages) in one file can be extracted to form a 3D data set. (Some image formats, e.g., TIFF and PNG, support multi-page files.)

Options for saving images have also been expanded to include most commonly used formats. This will facilitate the use of Slicer Dicer results in other applications, publications, etc.

Improved Isosurface Rendering

The computation and rendering of isosurfaces have been improved. This results in greater speed and realism. The lighting model is more rigorous and permits independent control of shading (relative portions of ambient and directional light) and light direction. A new light-direction control makes it easy to adjust the lighting for best effect.


Auto/Generic Open Option

This option makes it possible to force a file to be opened generically, regardless of its type. This will be useful in situations where, for some reason, the normal format-specific processing fails to read certain data. This might happen, for example, if the file has been corrupted, or if the format is not fully supported by Slicer Dicer. If structural details are known for that file its data can still be extracted via the generic file interface.

Survey Selected Data Button in File Parameters Dialog

This button is convenient when you want to perform a range survey (maximum, minimum) on a selected subvolume without resetting the view volume.

Improved HDF Support

Slicer Dicer's HDF support has been brought up-to-date. Most HDF image formats are now recognized. The reading of SDS (Scientific Data Set) data is now more reliable. Multiple data sets in a single file and all appropriate metadata (dimension names, labels, etc.) are now extracted.



This release fixes a few problems that showed up in v2.6.


1. v2.6 had a bug that prevented the opening of netCDF files with non-byte (e.g., short, float, double, etc.) data.

Status: Fixed in v2.6.1. All netCDF files can be opened.

2. There was a problem in the File Parameters dialog affecting the setting of the range (minimum, maximum) for variables of numeric type "double". If either limit was less than zero (or some very small positive quantity), the range would be clipped to zero.

Status: Fixed in v2.6.1.

3. Opening a data file (or a document file) with a template would occasionally crash when the template file had fewer variables than the file being opened.

Status: Fixed in v2.6.1.

4. The Transparent/Opaque popup menu in the Transparency Animation and Transparency dialogs was not sized properly, making it difficult to operate.

Status: Fixed in v2.6.1. These popups should now expand to show both items without a scroll bar.

5. Printing to certain Canon color printers was not working correctly. Colors were not being reproduced accurately.

Status: Fixed in v2.6.1.



In addition to bug fixes and minor changes, this version includes several important new features. The online documentation has been amended to reflect these changes.

Support for Text Data Files

Previous versions would open only binary data files, specifically netCDF, HDF, and so-called generic format files. The latter consisted of arbitrary arrays stored in binary form. Now, the generic category has been extended to include text files (AKA ASCII or formatted data). In these files, values are represented by character strings, e.g., "8.2342" and "2.332e-002". These files are convenient in many applications because they are human-readable and because they are portable. (The ASCII character representation is recognized on all platforms, whereas binary data generated on one platform will often be unreadable on another.) Previously, users had to translate their text data sets into binary form before viewing them with Slicer Dicer. As of version 2.6, this is no longer necessary.

When you select a text file with the File Open command, Slicer Dicer will present the Generic File dialog. This dialog is the same as before, except that "text" has been added to the list of data types, along with "byte", "short", etc. That's all you need to deal with: if it's a text file, select "text" for the data type; if it's a binary file, select the appropriate numeric type. The other generic file parameters -- variable name, number of dimensions, and the names, sizes, and header and trailer sizes for the dimensions -- are essentially the same as before, except that headers and trailers are measured in lines rather than bytes for a text file. See "Text Data File" in the online help for a full discussion of these issues.

Generic Metadata File Format

A generic metadata file is a text file containing the metadata associated with a particular generic format file. The metadata format is defined as a convenience for users who use generic format files extensively and wish to avoid dealing with the Generic File dialog every time a file is opened. When one of these files is opened, Slicer Dicer will read the metadata and open the referenced data file without presenting the Generic File dialog.

Any text editor can be used to create a metadata file, which typically is given a  ".gm" file name extension. The format includes multiple lines containing a mixture of string and numeric values. In this context, a string value is a double-quote-enclosed sequence of characters. Here's an example:

     "This is a Slicer Dicer Generic metadata file"

     "Pete 63x101x101"



     0 255


     "X" 63 0 0

     "Y" 101 0 0

     "Z" 101 0 0

The first line must include the string "generic metadata". This is a signature used by Slicer Dicer to identify a generic metadata file. (The filename extension is ignored.)

The second line is the pathname of the data file, which must be a generic format file. This path can be either absolute or relative to the location of the metadata file itself.

The third line is the name of the dependent variable, "density" in this case. This name can be anything you wish and will be used in Slicer Dicer in various displays that refer to the variable.

The fourth line contains the numeric type. This string value must be one of the numeric types recognized by Slicer Dicer, e.g., "byte", "signed byte", "short", "long", "float", or "double".

The fifth line contains two values specifying the range that you want to use for visualizing the dependent variable. These numbers will be read in as double precision floating-point values and converted to the variable's type.

The sixth line contains a single value, the number of dimensions.

Each of the remaining lines, as many as the number of dimensions, define the name, size, header, and footer associated with a dimension. See "Generic File Format" in the online help for a description of these parameters.

Suppose the file named "" contains the above lines. If you use Slicer Dicer's File Open command to open this file, the metadata in this file will be read and used to open the data file "Pete 63x101x101".  In this case, since the data file reference is a relative pathname, that file must be in the same directory as the metadata file.

Template Files

Normally, when a data file is opened, a new document is created in a certain default state, and when a document file is opened, the state information is read from that file. Alternatively, in v2.6, you can elect to open a data or document file using another document as a template. In this case the state information for the newly opened document will be set to match that in the template document.

This will be very useful for users who prefer to use a particular configuration for viewing a variety of files. This "configuration" might include a color table, transparency range, variable names and units selections, data scaling and rotation, and many other document parameters. It might also include a particular set of display objects (slices, blocks, etc.).

There is no special "template" file format. Any existing Slicer Dicer document file can serve as a template file. The current template file pathname, if one has been selected, is shown in the Open dialog. Controls are also available there for selecting and using this file. There is also a new command-line switch for controlling template file options. See below.


1. The positions of ticks and labels are recomputed more frequently. Previously, after changing the scale factor or units for a dimension, tick positions would be totally inappropriate and tick marks would be either squashed together too densely or spaced too widely. Now, changes such as these automatically force "pretty" (default) tick settings.

2. The "char" numeric type is now referred as "signed byte", which more clearly states the nature of this type.

3. Parsing of command line arguments has been extended to support template files. This is intended as a convenience for users who wish to launch Slicer Dicer from other processes. Previous versions interpreted the first command line argument as a file pathname and attempted to open this file. Other command line arguments were ignored. Version 2.6 checks the other arguments for an occurrence of either "/t", "/T", "-t", or "-T". This will be interpreted as a template flag, and the next argument, if one exists, is interpreted as the pathname of a template file.

For example, the following command line

     Dicer.exe "d:\Slicer Dicer\Samples\ClarkJet.hdf"

       /t "d:\Slicer Dicer\Samples\ClarkJet_template.dcr"

would cause Slicer Dicer to open the file ClarkJet.hdf using the template file ClarkJet_template.dcr. Subsequent Open commands would have the Use Template option ON and the Current Template File set to ClarkJet_template.dcr. (Note that command line arguments containing spaces should be enclosed in quotes.)

If the command line has no arguments following a "/t" flag, the "Use Template" option in the File Open command will be set to OFF.

The following command line is also valid

     Dicer.exe /t template_filename data_filename

where the template flag and filename precedes the data filename.

Other command line arguments will be ignored. If multiple data filenames are listed in the command line, only the first will be opened.


1. In certain circumstances, when reading a large data file (> 5 MB), the last slice could overwrite the first slice.

Status: Fixed in v2.6.

2. With the system Palette Settings in direct color mode (>256 colors), the current color button in the Infobar wasn't being redrawn when a document window was activated. Consequently, this button could show the wrong color (the current paint color for the deactivated document).

Status: Fixed in v2.6. The current color button should always display the correct color for the active document window.



This release fixes problems that showed up in v2.5.2.

1. Another bug causing color discrepancies in the 16-bit color mode for certain video drivers was discovered.

Status: Fixed in v2.5.3.

2. On many printers, the rear edges of the data volume outline were not being printed, or were being printed very faintly. Many printer drivers do not draw the dashed lines that were being used to render these line. (Note: the data volume outline is displayed only when it is enabled in the Layout options. See Options/Layout dialog.)

Status: Fixed in v2.5.3. Now the rear edges of the data volume are drawn in to the rendered bitmap before being sent to the printer. As in previous versions, all other lines (front edges of the data volume, coordinate axes and tick marks, etc.) and text annotations are not placed in the bitmap but are displayed at full printer resolution.



This release fixes problems that showed up in v2.5.1.

1. In previous versions it was possible to corrupt the colors in a document window by switching the Control Panel Display palette settings either while a document was open, or between saving and reopening a document.

Status: Fixed in v2.5.2. You should be able to switch modes at will between the various modes with 256 or more colors.

2. Certain of the lines drawn in the Print Preview window were slightly misplaced. For example, coordinate axes and tick marks or the data volume outline would occasionally be displaced by one or two pixels. This was a problem only in the preview window, not in the main document window or in the printed result.

Status: Fixed in v2.5.2.

3. In the 256-color mode, but not in palette modes with more than 256 colors, some of the colors in the Print Preview window were incorrect. Usually this would show up in the background wall colors inside the data volume, resulting in discrepancies between the colors inside versus outside the data volume.

Status: Fixed in v2.5.2.

4. There were slight color artifacts in the Print Preview window (all palette modes) when the zoom setting was less than 100%. The various decimated versions would include scattered mis-colored pixels. There was no problem with the printed result or in the preview window when zoomed to full size.

Status: Fixed in v2.5.2.



This release fixes problems that showed up in v2.5.

1. In v2.5, slight color discrepancies may appear in a Slicer Dicer document window under the following circumstances:

a) Windows NT 4.0 with Service Pack 3 installed, and

b) Palette in Display Control Panel (Settings tab) set to either 15-bit (32,768 colors) or 16-bit (65,536 colors) pixels.

With these particular combinations, parts of the background walls may be drawn with the wrong color. Often, the outermost parts of these walls will have a "greenish" tint relative to the interior portions. Similarly, colors for text and the wire outlines may also have incorrect colors. So far as we know, these problems do not occur under Windows 95 or Windows NT 4.0 with Service Pack 1 or Windows NT 5.0 (beta). Also, they do not affect operations under the 8-bit (256 colors) or 24- or 32-bit (16 million colors) Palette settings.

Microsoft has confirmed that the problem is caused by a bug introduced in Service Pack 3 (which has apparently been fixed in NT 5.0 beta).

Status: Fixed in v2.5.1. Alternative graphics functions are used that avoid the MS system bugs.

2. Slicer Dicer v2.5 does not run in the True Color (32-bits/pixel, 16 million colors) Palette mode. (The 24-bits/pixel mode, also 16 million colors, works correctly.)

Status: Fixed in v2.5.1. Slicer Dicer is designed to run in all modes with 256 or greater colors.

3. In v2.5, the Inverse Color Map check box in the Color Mapping dialog occasionally gets out of sync with the actual setting.

Status: Fixed in v2.5.1.