/********************************************************************** * * * IBM VISUALIZATION DATA EXPLORER - Frequently Asked Questions * * * * A.0 Overview * * A.1 Modules * * A.2 Client-Server * * A.3 Organization * * A.4 User Interface * * A.5 Windows * * A.6 Executive * * A.7 Data Management * * A.8 Data Flow * * B.0 General Questions * * B.1 What are the Names and Version Numbers of the software? * * B.2 Are there different ways of interacting with DATA EXPLORER? * * B.3 Is Data Explorer distributed or client-server? * * B.4 Is Data Explorer Object Orientated? * * B.5 Is the Power Visualization System Mult-user? * * B.6 Do you supply X-terminal support? * * B.7 What 3rd party modules are offered? * * B.8 What run-time libraries are needed by Data Explorer? * * B.9 Does the API differ across platforms? * * B.10 What Support Structures Exist for Data Explorer? * * C.0 User Interface and Interaction * * C.1 Does the User Interface have a command language interface? * * C.2 Does the GUI have multiple menu selection techniques? * * C.3 Is there an On-line Help Facility? * * C.4 Are there predefined modules available? * * C.5 How is the visualization edited? * * C.6 Does the GUI use MOTIF? * * C.7 Is there a macro capability? * * D.0 Data Handling * * D.1 What types of data are supported? * * D.2 What file formats are supported? * * D.3 How do you add new modules to the system? * * E.0 Realization and Rendering * * E.1 What rendering methods/types does the system support? * * E.2 Does Data Explorer support Z-buffering and double buffering?* * E.3 Does the system support multiple plots and animations? * * E.4 Can data from disparate sources be superimposed? * * E.5 Can the same data be displayed in multiple plots? * * E.6 Does the system support stereoscopic output? * * E.7 Can you pan and zoom? * * E.8 How do you rotate objects in the system? * * E.9 Does the image re-size if the window is re-sized? * * E.10 Does the system handle multiple light sources? * * E.11 Are there multiple cameras? * * E.12 Does the system support perspective and depth cueing? * * E.13 How is anti-aliasing applied? * * E.14 What coordinates systems can one use? * * F.0 Color Space and Color Manipulation * * F.1 Does the system dynamically support RGB color adjustments? * * F.2 Does the system support Hue, Saturation and Brightness? * * F.3 Does the system support transparency/opacity? * * F.4 How does one control the color? * * F.5 Do you have control over line - color, thickness, style? * * F.6 Are marker shape, size, color selectable? * * G.0 Data Manipulation * * G.1 Can one slice 3D space to get a 2D representation? * * G.2 Does the system support front/back clipping? * * G.3 Can one capture subsets of the data via GRAPH? * * G.4 Can one determine the number of points at color value? * * G.5 Is it possible to query the visualization data? * * G.6 Is editing selected points possible? * * G.7 How via "point and click" do you get rendered data points? * * G.8 What data point does the 3D Probe pick? * * H.0 Annotation * * H.1 What annotation mechanisms are provided? * * H.2 Is auto scaling on axes provided? * * H.3 Is data point label provided? * * H.4 Do labels stay attached until removed? * * H.5 What fonts are supported? * * I.0 Animation * * I.1 Does the product supply animation capability? * * I.2 Are there dynamic control capabilities? * * J.0 Print/Plot Functions * * J.1 What print/plot options are there? * * J.2 How are print devices selected? * * K.0 Configurations/Graphic Options/ Recommendations * * K.1 What level of AIX is required for Data Explorer? * * K.2 What graphic adapters are supported by Data Explorer? * * K.3 What is the recommended configuration for a RISC/6000? * * K.4 What are the consequences if one reduces the configuration? * * K.5 What are your bench mark results? * * K.6 What publications are available for Data Explore? * * L.0 Is there a public repository for networks, modules,etc.? * * * **********************************************************************/ --------------------------------------------------------------------- A.0 OVERVIEW The Data Explorer graphical user interface allows end users to perform tasks at various levels of sophistication. For example, a user can use the user interface to apply data and adjust input values to an existing visualization process. A slightly more advanced user can can construct a new visualization process, called a visual program, by connecting a network of Data Explorer's modules. An expert programmer can create new modules, using C or FORTRAN, for use with the system modules. Besides the user interface, Data Explorer also provides a scripting language interface, for users who want to build their own visualization functions in a more traditional programming style. Data Explorer's graphical user interface provides an integrated on-line help facility. This facility provides users with on-line access to the Data Explorer user manual, as well as with context-sensitive help information. In addition to the help information provided with Data Explorer, the on-line help facility allows users to document various aspects of their particular visual programs. Other users of these visual programs then have on-line access to this program-specific documentation. --------------------------------------------------------------------- A.1 Modules Data Explorer provides an extensive set of modules that you can use to visualize your data. For example, the Isosurface, Streamline, and AutoColor modules perform the standard visualization functions of creating constant-value surfaces, tracing particle paths through velocity fields, and coloring objects based on a data value, respectively. In addition to these expected functions, Data Explorer also provides tools to perform more sophisticated manipulation of data. The Map module is a general purpose module that can map a data field onto an arbitrary object--whether it is a streamline, an isosurface, or even another data field's computational mesh. The Compute module can perform pointwise arithmetic or trigonometric computation not only on your data, but also on the grid itself; this makes the task of warping a grid a simple matter of entering an expression. Even standard tools, such as Isosurface, operate on multiple types of input grids. For example, if the input field to Isosurface is two-dimensional, the module automatically creates contour lines. The Data Explorer renderer can handle opaque or translucent surfaces, translucent volumes, and opaque or translucent lines or points--all in the same image. In addition, data on different computational or observational grids can be visualized together, allowing correlative analysis without requiring the re-gridding of any of the grids. The power and interoperability of the modules is possible because of the underlying data model, which is capable of describing a wide variety of types of input data. Because the data itself is self-describing, modules can be flexible in the types of data they accept, and can perform their actions appropriately based on their input. --------------------------------------------------------------------- A.2 Client-Server Data Explorer is designed as a client-server model. The Data Explorer client-server architecture is built using open system components such as TCP/IP, sockets, X Window System, and Motif. In addition, Data Explorer can be extended with user-developed visualization modules. Each scientific and engineering specialty has its own visualization requirements. Data Explorer's open system design allows room for growth and individual needs and supports cross-fertilization of these approaches among disparate disciplines. In this client-server model, the user interface is the client. The executive, modules, and data management components, often referred to collectively as the, make up the server portion. The user interface client can be on a different platform than the server (executive). Data Explorer allows you to easily switch among servers running on different hardware platforms. --------------------------------------------------------------------- A.3 Organization Components of the System and Their Relationships Each of the four layers of the Data Explorer system has a defined interface. The layers are described in the order in which you are likely to encounter them. --------------------------------------------------------------------- A.4 User Interface The graphical user interface is built upon the X Window and Motif standards. These tools manage multiple application windows that allow a user to create and control the visualization process easily and effectively. The graphical user interface provides two levels of service. First, non-programmers or users with fixed requirements can execute previously created visual programs. These visual programs may consist of various menus, dials, sliders, and other interactors that provide fixed functions. Second, programmers can create customized visualizations by using the interface to interconnect modules in flexible ways, and to create new combinations of modules in the form of macros. The Data Explorer graphical user interface lets you create or work with a visual program to easily realize sample, select, and transform data during visualization. You can use the visual program editor to create new scenarios by simply connecting module icons on the screen in any logical sequence. --------------------------------------------------------------------- A.5 Windows Data Explorer provides the following primary windows: Visual Program Editor: Lets you create and alter visual programs. Control Panel: Lets you set and control the variable input parameters of the tools used in a visual program. Image Window: Displays the image created by a visual program and allows direct interaction with the visualized image. Help Window: Provides on-line access to the Data Explorer user manual and context-sensitive help information. Colormap Editor: Data Explorer provides a Colormap Editor window, which lets you map colors to specified data values and display the results in the visual image. Sequencer window: It provides several functions, including controlling how a sequence of images is displayed (with forward and backward direction, repetition, and so forth. --------------------------------------------------------------------- A.6 Executive The executive is the component of the system that manages the execution of the modules specified using the scripting language. This scripting language is generated by the graphical user interface to invoke visualization functions for visual programs. More sophisticated users can also use the scripting language to write their own programs, as described in Programming with the Data Explorer Scripting Language. Data Explorer provides an extensive, powerful set of highly interoperable visualization modules. The modules used for visualization functions are available in three ways: You can use their icons, seen through the user interface as nodes in a visual programming network. You can use them as functions calls, available in the script language interface provided by the executive layer. You can use many of them for integrated applications, as part of the visualization library programming interface. --------------------------------------------------------------------- A.7 Data Management The data management layer is the portion of the programming interface that provides modules with access to the data model, which is discussed in Understanding the Data Model. This layer includes general system services as well as routines for creating and managing the set of data objects. The data management layer also provides an application programming interface (API) for adding new modules to Data Explorer and for accessing the power and flexibility of the data model. The execution model of Data Explorer is based on two principals. First, parallelism, in general, is essential to achieving the highest possible performance at any given time in the progress of computer technology. Second, the data-flow processing model implied by the visual programming language places some constraints on the behavior of the modules. --------------------------------------------------------------------- A.8 Data Flow The visual programs and programs generated using the script language represent data flow networks. The data flow networks place some constraints on the behavior of the modules. In particular, modules generally must be pure functions of their inputs. That is, a module's outputs must be completely predictable on the basis of its inputs, rather than on the basis of some internal (or external) state, derived from a previous execution. Pure-function semantics also require that modules treat their input as read-only. If a module is set to modify its input, it makes a copy of that input and changes the copy. Because of object sharing allowed by the data model, this copying process generally involves little overhead. This requirement also arises from parallel execution of modules. As a side benefit, the pure-function semantics allow the executive to cache intermediate results for re-use when appropriate. --------------------------------------------------------------------- B.1 WHAT IS THE NAME AND VERSION NUMBER OF THE VISUALIZATION SOFTWARE? The name of the workstation application is the IBM AIX Visualization Data Explorer/6000 Version 1.2 The program number is 5765-057. The name of the server environment is the IBM POWER Visualization Server/Data Explorer Version 1.2 IBM Visualization Data Explorer Version 1.1 for Sun, Hewlett Packard and Silicon Graphics. For Sun, the software operates under SunOS 4.1 for SPARCstation 2, and OpenWindows 3.0. For HP, the software is intended to operate under HP-UX 8.07 for HP 9000/700 series workstations. For SGI, the software is intended to operate under IRIX 4.0.1 for Iris Indigo and Crimson workstations. --------------------------------------------------------------------- B.2 ARE THERE DIFFERENT MEANS OF INTERACTING WITH DATA EXPLORER? The flexibility of the system accommodates various levels of end-user participation by permitting "casual" to "expert" control of the visualization process. The novice user may use the menu-driven front end to apply previously created visual programs to a specific data set. Facilities exist for such users to easily control module parameters. At the next level of sophistication, users may interact with a visual program editor to modify the visualization program, adding or deleting modules, or changing the order of module execution. Finally, the expert user may couple system modules with user-created modules for customized application solutions. Both a windowing user interface and a scripting (command) language interface is available to all levels of users. --------------------------------------------------------------------- B.3 IS DATA EXPLORER DISTRIBUTED OR CLIENT/SERVER? The Power Visualization System's Data Explorer package was developed to take advantage of the flexibility pro- vided by code running in a client/server relationship. In this case, the Data Explorer User Interface (UI) runs as a client on a workstation platform which supports industry standard X-Windows(TM) and Motif(TM) protocols. The server function may be provided either by the work- station itself or by the IBM Power Visualization Server. In the case where the server function is provided by the workstation, local graphics hardware may be used to increase the performance of the rendering stage. --------------------------------------------------------------------- B.4 IS DATA EXPLORER OBJECT-ORIENTED? Data Explorer is written in object orientated style. Data Explorer contains "objects", or regions of global memory that contain an identification of its type, and additional type-dependent information. Access to objects is through a set of information-hiding access routines. Data sharing between objects is possible, even encour- aged. Data Explorer was not written in an Object- Oriented programming language. --------------------------------------------------------------------- B.5 IS THE POWER VISUALIZATION SYSTEM A MULTI-USER SYSTEM? The Power Visualization Server is a multi-user compute and visualization server. The Data Explorer environment was developed around a client/server model, multiple users could access the server simultaneously. Again, size of the data and computational response will determine whether the server is shared or used in a queued request mode. --------------------------------------------------------------------- B.6 DO YOU PROVIDE X TERMINAL SUPPORT? Data Explorer has the capability of providing remote X-window display. We do not guarantee full function on all hardware platforms. Dithering (converting from 24-bit color to 8-bit color) is done if the display has only 8-bit color. Performance is dependent on network throughput. --------------------------------------------------------------------- B.7 WHAT 3RD PARTY MODULES ARE AVAILABLE? There are no third party modules currently. There are some special purpose modules developed internally. The Cornell anonymous FTP (ftp.tc.cornell.edu) is a repository for public domain networks, macros, modules etc.. --------------------------------------------------------------------- B.8 WHAT RUNTIME LIBRARIES ARE NEEDED FOR DATA EXPLORER? Data Explorer requires Xlib, Xt, Xm, libC . --------------------------------------------------------------------- B.9 HOW DOES THE API DIFFER ACROSS PLATFORMS? Data Explorer allows modules written in C or FORTRAN to be compiled and linked into the system. Any computer running the Data Explorer software would have the same API. Modules specifically developed for the POWER Visualiza- tion System will use the Data Explorer tasking library calls to implement parallelism. Modules developed spe- cifically for a workstation will most likely not make use of the tasking library calls. However, the API, including tasking calls, is identical. Code developed on one machine would be able to run on the other. --------------------------------------------------------------------- B.10 WHAT SUPPORT STRUCTURE EXISTS FOR DATA EXPLORER? The Cornell Engineering and Theory Center will function as a repository for newly developed modules, as well as general information regarding Data Explorer. The Visualization Systems group will maintain an archive- server: data-exp@watson.ibm.com for public forum and dissemination of time topics and information. --------------------------------------------------------------------- C.0 DESCRIBE THE USER INTERFACE (UI). The Data Explorer graphical user interface provides a visual programming paradigm. It works on a client- server model where the user interface is independent of the execution of the visual program. The user interface runs on top of the X Window system and OSF/Motif(TM). It has five major window types: o The Visual Programming Editor provides the capability of defining and editing visual programs (including defining macros that may be used in other visual pro- grams or macros). o Control Panels which enable a user to control parame- ters within their visualization via Motif dials, sliders, steppers or widgets. o The Sequencer provides the ability to easily create animations by providing an interface to control frame increments, step, loop and halt. o The Image window where the visualization is dis- played, provides functions to directly manipulate the image. The User Interface has been designed so that it is not necessary for a user to see a previously created visual program. Interaction with the visual program may be accomplished, if desired, solely through the use of Image window, Control Panels and/or Sequencer window. The user interface provides on-line help which is similar to the User's Guide. It has contextual associations, "hot-links", to connect related topics. --------------------------------------------------------------------- C.1 DOES THE USER INTERFACE PROVIDE COMMAND LANGUAGE FACILITIES? Data Explorer provides a scripting language to specify a visualization in a more traditional programming style. The scripting language is basically a procedural language with macro capabilities. See Chapter 4 of the IBM POWER Visualization User's Guide or IBM AIX Visual- ization Data Explorer/6000 User's Guide for more informa- tion. --------------------------------------------------------------------- C.2 DOES THE USER INTERFACE PROVIDE MULTIPLE MENU SELECTION TECHNIQUES? The user interface provides mouse driven pull-down menus as well as menu accelerators. --------------------------------------------------------------------- C.3 IS THERE AN ON-LINE-HELP FEATURE? On-Line-Help is available using two different methods: Use the Help pull-down from the menus. This opens a window with a list of the major subtopics of the User Manual. Select the desired subtopic by clicking with mouse. Or use the center mouse button to click on various elements within the Data Explorer window. The help window will open and display the portion of the User Manual related to the selected item. --------------------------------------------------------------------- C.4 ARE PREDEFINED MODULES AVAILABLE? Data Explorer provides a host of visualization modules that may be applied in a "building-block approach" to transform, realize, and render data. Multiple visualiza- tion techniques (for example, volume rendering, isosurfaces, and streamlines) may be combined to create a single image or sequence of images. --------------------------------------------------------------------- C.5 HOW IS THE VISUALIZATION EDITED? The visualization may be changed by directly editing the visual program, or more commonly, by merely changed pre- defined control panel values. --------------------------------------------------------------------- C.6 HOW DOES THE USER INTERFACE MAKE USE OF MOTIF? All User Interface features are implemented using Motif widgets and follow the Motif style guide. The interactors are defined as Motif dials, sliders, steppers or widgets. Users of Motif will recognize file selection dialog box, menu pull-downs, iconification, and other windowing func- tions. --------------------------------------------------------------------- C.7 IS THERE A MACRO FACILITY? Data Explorer allows macros to be defined in both the Visual Programming Environment and the scripting language which look to the user like modules. Within the Visual Programming Environment, they are constructed in the same manner as a visual program. They are assigned by default to the Macros category, but may be assigned to any category when they are created. Macros within the scripting language resemble subroutines, in that they have a header, which lists inputs and outputs, and a body, which defines the operations to be performed. --------------------------------------------------------------------- D.1 WHAT TYPES OF DATA ARE SUPPORTED? The Data Explorer data model supports a variety of scientific simulation and observational data, as well as any derived data within the system (including graphics). Field objects are the fundamental objects in the Data Explorer Data Model. A field is defined over some space and is represented by a set of sample points and information on how to interpolate between them. The domain of the mapping is specified by a set of positions and (generally) a set of connections that allow interpolation between the positions. The value of the mapping at all points is represented implicitly by data samples dependent either on the sample points or on the connections between the sample points. The information in a field is represented by some number of named components. Each component has a value, which is an object. In general, components are array objects. Fields may share components. Components have attributes associated with them. For example, the "dep" attribute of a component records the dependency of that component on another component; thus the "data" component will have a "dep attribute of either "positions" or "connections", depending on whether that data is associated with the sample points or with the connections between them. A component may also have a "ref" attribute that indicates that it refers to another component, or a "der" attribute specifying that it is derived from another component. A "connections" component must have an "element type" attribute naming the type of connections, such as "triangles", "quads", tetrahedra", and so on. Fields may be combined into groups, and you may also have groups of groups. Data structures that may be represented include data defined on a regular orthogonal grid, on a deformed or curvilinear regular grid, and on irregular grids such as triangular, quadrilateral or tetrahedral meshes: or ungridded data with no connections between the data samples. The data samples may be defined over spaces of any dimensionality, and, independently, may also be connected by primitives of various dimensional- ities. The data values may be associated either with the sample points or with the connection points or with the con- nections between the sample points. Available data types include real data; scalar, vector, tensor data; and byte, integer, and floating point data. It is possible to represent data types in the Data Explorer data model that will not be processed by the currently defined modules within Data Explorer. For example, in the current release of Data Explorer, only single precision floating point data are universally sup- ported. However, tools are provided which would allow the user to convert to floating point as part of the visual program. Currently, most pre-defined modules support one-, two-, or three- dimensional positions. Series of data, such as time or temperature steps, are handled easily by the system. Please refer to Chapters 2 and 5 in the IBM POWER Visualization User's Guide or IBM AIX Visualization Data Explorer/6000 User's Guide for more information. --------------------------------------------------------------------- D.2 WHAT FILE FORMATS ARE SUPPORTED FOR INPUT DATA? Three data formats are directly accessible: Data Explorer File Format, the Unidata Network Common Data Form (netCDF) and the General Array Format. --------------------------------------------------------------------- D.3 HOW ARE NEW MODULES ADDED TO THE SYSTEM? There is an Application Programming Interface (API) pro- vided for adding modules. New modules may be added to the system by programming in C, or in FORTRAN. The new programs are added to the system through the Data Explorer programming interface. This provides access to the data model and extensive system services thereby facilitating ease of programming. This procedure is documented in the IBM POWER Visualization Programmer's Reference or IBM Visualization Data Explorer Programmer's Reference. --------------------------------------------------------------------- E.1 DOES THE SYSTEM SUPPORT THE FOLLOWING RENDERING METHODS? +---------------------------+---------------------------+ | Points | yes | +---------------------------+---------------------------+ | Lines | yes | +---------------------------+---------------------------+ | Specular lighting | yes | +---------------------------+---------------------------+ | Flat shading | yes | +---------------------------+---------------------------+ | Gouraud shading | yes | +---------------------------+---------------------------+ | Phong shading | no | +---------------------------+---------------------------+ | Backface culling | yes | +---------------------------+---------------------------+ | Volume rendering | yes | +---------------------------+---------------------------+ | Translucent surfaces | yes | +---------------------------+---------------------------+ | Z-buffering | yes | +---------------------------+---------------------------+ --------------------------------------------------------------------- E.2 DOES THE DATA EXPLORER SUPPORT Z-BUFFERING AND DOUBLE BUFFERING? The software supports z-buffering and double buffering. Data Explorer uses this features in the software as well as the hardware when available. --------------------------------------------------------------------- E.3 CAN THE USER DISPLAY MULTIPLE GRAPHS OR ANIMATIONS? The user can simultaneously show multiple visualizations. The visualizations are then displayed within a single image window. --------------------------------------------------------------------- E.4 DOES THE SOFTWARE ALLOW DATA FROM DISPARATE SOURCES TO BE SUPERIMPOSED? Data Explorer provides correlative analysis capabilities, enabling data from various sources to be visualized together. Mapping between arbitrary grids (using the compute module), and warping of grids are supported. --------------------------------------------------------------------- E.5 DOES THE SYSTEM ALLOW FOR THE SAME DATA TO BE DISPLAYED IN MULTIPLE PLOTS? The system allows for simultaneous presentation of the same data in multiple visualizations. --------------------------------------------------------------------- E.6 DO YOU PROVIDE STEREOSCOPIC OUTPUT? Stereoscopic output is available using the IBM POWER Visualization System with the optional video controller but is not directly supported by Data Explorer. It may be accomplished, under Data Explorer, using two cameras at different offsets and saving the resulting images for later display. Stereoscopic output is not directly supported with Data Explorer on a workstation. --------------------------------------------------------------------- E.7 DOES THE PRODUCT INCLUDE PAN AND ZOOM FEATURES? There are built in pan and zoom features. Pan and zoom are controlled with the mouse, or by explic- itly calling out values using the keyboard. The visual- ization is automatically adjusted to accommodate the new view within the window. --------------------------------------------------------------------- E.8 DOES THE USER HAVE CONTROL OVER THE ROTATION OF OBJECTS? The rotation of visualized objects is controlled via "grab and drag" mouse operation. --------------------------------------------------------------------- E.9 IS THE IMAGE DYNAMICALLY RE-SIZED IF WINDOW IS RE-SIZED? The image is automatically re-sized when the image window is re-sized. --------------------------------------------------------------------- E.10 DOES THE PRODUCT SUPPORT MULTIPLE LIGHT SOURCES? Multiple light sources are available and are independ- ently controllable. --------------------------------------------------------------------- E.11 DOES THE PACKAGE SUPPORT MULTIPLE CAMERAS? Multiple cameras are supported. --------------------------------------------------------------------- E.12 IS DEPTH-CUEING OR PERSPECTIVE SUPPORTED? The current release of Data Explorer supports perspective, but not depth-cueing. --------------------------------------------------------------------- E.13 HOW IS ANTI-ALIASING APPLIED? Anti-aliasing is currently not directly supported by the Data Explorer renderer; however this effect may be achieved by generating a high resolution image and sub- sampling it (e.g. create an image 4 times larger than the final image will be and using a "reduce" module to filter the image 2 times in each direction, then using the "display" module to show the final result). This may be combined into a macro if there is common need for this functionality. --------------------------------------------------------------------- E.14 WHAT COORDINATE SYSTEMS ARE SUPPORTED? Modules use cartesian coordinates internally. By using a "compute" module, users are able to convert from various coordinate systems, (including curvilinear, polar and those based on longitude and latitude), to cartesian without requiring preprocessing. Axes, in 2-D or 3-D, may be automatically generated and manually labeled using the Data Explorer user interface. Other classes of anno- tation may be produced from provided tools (e.g. 1-D axis, color bar, clock). --------------------------------------------------------------------- F.1 DOES THE SYSTEM DYNAMICALLY SUPPORT RGB COLOR ADJUSTMENTS? RGB color adjustments may be made to displayed objects. --------------------------------------------------------------------- F.2 DOES THE SYSTEM SUPPORT HUE, SATURATION AND BRIGHTNESS ADJUSTMENTS? A full color map editor is available which controls hue, saturation, value and opacity and may be changed while viewing the image. --------------------------------------------------------------------- F.3 DOES THE SYSTEM DYNAMICALLY SUPPORT TRANSPARENCY AND OPACITY? True transparency and opacity of displayed objects is supported and this may be dynamically altered via the color map editor or a control panel. --------------------------------------------------------------------- F.4 HOW DOES THE USER CONTROL THE COLOR? The user controls the color explicitly or via a Colormap editor. A single color value may be selected by the Color module. An AutoColor module is provided which applies a default range of color to the data values. The Colormap Editor provides access to value, hue, satu- ration, and opacity for the full color range. Modules are also available to convert from RGB to HSV and back. --------------------------------------------------------------------- F.5 DOES THE USER HAVE CONTROL OVER LINES (COLOR, STYLE, THICKNESS)? The user has direct control over the color of lines. A line may be transformed into a tube which may be varied in thickness or color. Only solid and translucent lines are supported currently. --------------------------------------------------------------------- F.6 ARE MARKER SHAPES, SIZES AND COLORS SELECTABLE? The color and size of markers may be manipulated by the user. There are a variety of types available: text, spheres, diamonds, rockets, needles. --------------------------------------------------------------------- G.1 IS IT POSSIBLE TO SLICE 3D SPACE TO OBTAIN A 2D VIEW? The user may take slices from 3D space and obtain a 2D view. All interactive features are available on the "sliced" view. A slice may be viewed as 2D or as part of a 3D scene. Additional functions can be applied with MaptoPlane, Slice and Slab. --------------------------------------------------------------------- G.2 DOES THE SYSTEM SUPPORT FRONT AND BACK CLIPPING PLANES? A single clipping plane is supported in Data Explorer. Multiple clipping planes are supported through the ClipBox module which clips front and back. --------------------------------------------------------------------- G.3 IS IT POSSIBLE TO CAPTURE SUBSETS OF DATA VIA THE GRAPH? The user may capture subsets of data via the visualiza- tion displayed and store that data. This is available on both 2d and 3d views. --------------------------------------------------------------------- G.4 IS IT POSSIBLE TO DETERMINE THE NUMBER OF POINTS AT A GIVEN COLOR LEVEL? To do this you would mark the colors, include those points with the specific color and use a histogram to count them. --------------------------------------------------------------------- G.5 IS IT POSSIBLE FOR THE USER TO QUERY THE VISUALIZATION DATA? Data Explorer provides a 3 dimensional probe into the data which enables the user to query the visualization --------------------------------------------------------------------- G.6 IS EDITING OF SELECTED POINTS SUPPORTED? Editing of data values on the fly is not supported in this release of Data Explorer. --------------------------------------------------------------------- G.7 HOW DO YOU USE "POINT AND CLICK" TO IDENTIFY RENDERED DATA POINTS? Select the "Enable 3D Cursor" menu option. A wire-frame box appears around the object. To add a point, double click inside the wire-frame box. A small square box appears, marking the point. Select the point by pressing and holding the mouse pointer on it. When the mouse button is depressed on the point, the three projections (one for each axis) appear inside the wire-frame box as dots, and the values for the x, y, and z coordinates are displayed on the right side of the Image window menu bar. You may move the point by dragging the selected point inside the box. --------------------------------------------------------------------- G.8 DOES THE 3D PROBE PICK A DATA POINT NEAREST, OR ANY LOCATION IN 3D SPACE? The probe chooses an arbitrary location in 3D space. It is possible to use Map to interpolate a data value to that point. --------------------------------------------------------------------- H.1 WHAT ANNOTATION MECHANISMS ARE PROVIDED? Stationary captions are available. Also text may be linked to the visualization and rotate and re-size as the image is manipulated. Annotated axes may be displayed. An annotated ColorBar module is also available. --------------------------------------------------------------------- H.2 IS AUTO SCALING ON AXIS INTERVALS PROVIDED? OPTIONAL? The axes are auto-scaled. The user has the ability to specify certain parameters and override the default. --------------------------------------------------------------------- H.3 IS DATA POINT LABELING SUPPORTED? Data points may be labeled. The point to be labeled may be selected using a 3-D probe (mouse controlled) into the image window, or labeled automatically when the data is brought into the visualization using text glyphs. --------------------------------------------------------------------- H.4 DO LABELS STAY ATTACHED UNTIL REMOVED? Labels stay attached until removed by the user. The user may selectively remove labels. --------------------------------------------------------------------- H.5 WHAT TEXT FONTS ARE SUPPORTED? There are two text fonts (variable and fixed) supplied with Data Explorer at the current release. --------------------------------------------------------------------- I.1 DOES THE PRODUCT PROVIDE ANIMATION CAPABILITIES? The product supports animation in a very intuitive manner. An optional window is provided which is the Sequencer. The Sequencer may be connected into the Visual Program and set up to provide an incrementing, (or decrementing) count of user defined range, increment, type (single frame or sequence). --------------------------------------------------------------------- I.2 ARE THERE DYNAMIC CONTROL FEATURES? Speed is controlled via a throttle interactor, which specifies frames/second. Animated views may themselves be manipulated while the sequencing is paused. --------------------------------------------------------------------- J.1 WHAT PRINT OR PLOT OPTIONS EXIST? The is a Plot module that will generate a data point plot. The image window may be saved as rgb, PostScript(TM), encapsulated PostScript or tiff files. --------------------------------------------------------------------- J.2 HOW ARE PRINT DEVICES SELECTED? Selection of the printer is made outside Data Explorer. --------------------------------------------------------------------- K.1 WHAT LEVEL OF AIX IS REQUIRED FOR DATA EXPLORER? Users of IBM AIX Visualization Data Explorer for RISC/6000 need AIX 3.2 --------------------------------------------------------------------- K.2 WHAT GRAPHICS ADAPTERS ARE SUPPORTED BY DATA EXPLORER? When running Data Explorer on a workstation the user may specify within the render module that hardware rendering should be done. For software rendering alone, the following graphics adapter is required: +----------------------------------------------+--------+ | PRODUCT NAME | COMPANY| +----------------------------------------------+--------+ | 8 bit color adapter supported under X/MOTIF | | +----------------------------------------------+--------+ When hardware rendering is required, the following cards may be used: +----------------------------------------------+--------+ | High Performance 8-bit 3D Color Graphics Proc|s IBM | +----------------------------------------------+--------+ | 24-bit Z-buffer Solid Rendering Option | IBM | +----------------------------------------------+--------+ | High Performance 24-bit 3D Color Graphics Pro| IBM | +----------------------------------------------+--------+ | 24-bit Z-buffer Solid Rendering Option | IBM | +----------------------------------------------+--------+ | IBM RISC System/6000 7235 Powergraphics GTO | IBM | | Graphics upgrade to RISC/6000 | | +----------------------------------------------+--------+ | POWER GT3X | IBM | +----------------------------------------------+--------+ | POWER GT4 8-bit Feature | IBM | +----------------------------------------------+--------+ | POWER GT4 24-bit Feature | IBM | +----------------------------------------------+--------+ | POWER GT4X 8-bit Feature | IBM | +----------------------------------------------+--------+ | CRX48Z | HP | +----------------------------------------------+--------+ | CRX24Z | HP | +----------------------------------------------+--------+ | ELAN | SGI | +----------------------------------------------+--------+ | XS24 | SGI | +----------------------------------------------+--------+ |and others for IBM, SGI, HP | +----------------------------------------------+--------+ --------------------------------------------------------------------- K.3 WHAT IS A RECOMMENDED CONFIGURATION ON A RISC SYSTEM/6000? Requirements for memory size, disk size and processing power will vary according to the size of the data set, and complexity of the networks. The following is a guideline only. +----------+--------------------------------------------+ | RECOMMEND| ALTERNATE | +----------+--------------------------------------------+ | MODEL 320| any model higher ... | +----------+--------------------------------------------+ | MEMORY 64|MMinimum 16MB | +----------+--------------------------------------------+ | DISK 800 |B350MB | +----------+--------------------------------------------+ +-----------------------------------------------+-------+ | GRAPHICS ADAPTER | FEATUR| +-----------------------------------------------+-------+ | Color Graphics Display Adapter | 2770 | +-----------------------------------------------+-------+ | NOTE: Minimum Configuration, for software rendering | | only | +-----------------------------------------------+-------+ | -OR- | | +-----------------------------------------------+-------+ | High Performance 8-bit 3D Color Graphics | 2780 | | Processor | | +-----------------------------------------------+-------+ | 24-bit Z-buffer Solid Rendering Option | 2782 | +-----------------------------------------------+-------+ | -OR- | | +-----------------------------------------------+-------+ | High Performance 8-bit 3D Color Graphics | 2780 | | Processor | | +-----------------------------------------------+-------+ | 24-bit Z-buffer Solid Rendering Option | 2782 | +-----------------------------------------------+-------+ | -OR- | | +-----------------------------------------------+-------+ | IBM RISC System/6000 7235 POWERgraphics GTO | 7235 | +-----------------------------------------------+-------+ --------------------------------------------------------------------- K.4 WHAT ARE THE CONSEQUENCES OF REDUCING THE CONFIGURATION? PROCESSOR POWER Software response is directly propor- tional to the processing power MEMORY Reductions affect the size of data which can be contained in working memory Increase Paging, decrease response time Software Caching Penalty, images would be recalculated and slow response time DISK SPACE Used for paging space and local data, could be reduced to 400 MB, but you may need even more for particularly large data sets, you may NFS mount across the systems GRAPHICS ADAPTER Only the minimum is required, others enable hardware rendering option --------------------------------------------------------------------- K.5 WHAT ARE YOUR BENCHMARK RESULTS? There are currently no specific benchmarks which cover Data Explorer. --------------------------------------------------------------------- K.6 WHAT PUBLICATIONS ARE AVAILABLE FOR DATA EXPLORER? IBM POWER Visualization System Overview GA22-7137 and Functional Characteristics IBM POWER Visualization System Installation Manual - GC22-7105 Physical Planning Manual IBM POWER Visualization 7245 Server SY22-9474 Installation and Service IBM POWER Visualization System Problem Determination Guide GC38-0500 IBM POWER Visualization System Console User's Guide SC38-0081 IBM POWER Visualization Quick Reference Guide SC38-0082 IBM POWER Visualization User's Guide SC38-0499 IBM POWER Visualization Programmer's Reference SC38-0065 IBM POWER Visualization 7246 Video Controller SY66-0212 Installation and Service IBM POWER Visualization 7246 Video Controller GA23-2066 User's Guide IBM POWER Visualization 7246 Video Controller Supplement GA23-2068 IBM POWER Visualization 7245 Server Safety Notices SY22-9868 IBM POWER Visualization 7245 Server Safety Inspection SY22-9873 IBM AIX Visualization Data Explorer/6000 User's Guide SC38-0496 IBM AIX Visualization Data Explorer/6000 SC38-0498 Quick Reference Guide IBM AIX Visualization Data Explorer/6000 SC38-0497 Programmer's Reference IBM AIX Visualization Data Explorer/6000 GC38-0495 Licensed Program Specifications L.0 Public repository for macros, networks, modeules The Theory Center at Cornell University has agreed to maintain a public repository for IBM Visualization Data Explorer networks, modules, data, macros, etc.. It can be reached via anonymous ftp at ftp.tc.cornell.edu. They also have a Gopher interface at info.tc.cornell.edu (userid: info).