Skip to main content

Über dieses Buch

Powerful new technology has been made available to researchers by an increasingly competitive workstation market. Papers from Canada, Japan, Italy, Germany, and the U.S., to name a few of the countries represented in this volume, discuss how workstations are used in experiments and what impact this new technology will have on experiments. As usual for IFIP workshops, the emphasis in this volume is on the formulation of strategies for future research, the determination of new market areas, and the identification of new areas for workstation research. This is the first volume of a book series reporting the work of IFIP WG 5.10. The mission of this IFIP work- ing group is to promote, develop and encourage advancement of the field of computer graphics as a basic tool, as an enabling technology and as an important part of various application areas.



Invited Presentations


1.1. The Impact of Scientific Visualization on Workstation Development

The importance of developing tools for the scientist to present and understand very complex data has become well-established as a major need that spans nearly every scientific and engineering discipline [20]. The enormous diversity and complexity of the problem demands research into multi-dimensional data presentation techniques, studies of human perceptual capabilities, and human factors. The computational, graphical, and human interface demands of scientific visualization will be a dominant force in the growing expectations for workstations of the future.
R. D. Bergeron, G. G. Grinstein

1.2. Issues in User Interface

A survey of the participants identified a broad list of user interface issues, which were then grouped in the following categories:
  • Information Exchange
  • Realtime
  • Intelligent Interfaces
  • Support for Problem Solving
For each topic the main issues were identified and possible directions were suggested. The discussions are summarized in the following sections.
J. F. Koegel, John Amason, Lyn Bartram, Alex Brown, Naota Inamoto, Sean McElroy, Enrico Nordelli, Brian Thompson, Marian Williams, John Koegel

Tools, Techniques and User Interfaces


2.1. Employing the Intelligent Interface for Scientific Discovery

This paper presents the fundamental notions underlying the concept of an intelligent interface by the analysis of recent research. An evaluation of the desirability of this form of interface within an environment of scientific experimentation is discussed. Given the thesis that an intelligent interface is in fact desirable, a proposal is made on how to motivate its utilization.
S. McElroy

2.2. A Mixed-Mode Man-Machine Interface for Interactive Problem Solving

In the context of problem solving application, the multi-media environments offered by scientific and technical workstations may be used to realize an effective high level of cooperation between the expert and the problem solver. In this paper we build a framework for the design and the development of an intelligent man-machine interface that integrates multimedia workstations’ tools (3D graphics, animation, combination of voice and sound) for the realization of analogical and synthetical representation of the complex states of the resolution process. Such interface can provide visual presentation of complex concepts and as consequence on-line interactive evaluation and manipulation of search strategies of the resolution process. A three layered schema of interface architecture is presented and various issues on the mappings between layers are discussed. An experimental prototype of the proposed interface on APOLLO 10000 workstation is being developed at IASI in the framework of the ongoing EEC Project PONTIFEX. Moreover experiments on synthetical and analogical representations are carried on in the University of Rome and consequently a new perceptive experimentation methodology is turning out.
P. Dell’Olmo, E. Nardelli, M. Talamo, P. Vocca

2.3. A Graph-Based Visual Tool for Workstations

A software tool has been designed and developed for workstation applications. It is a library of functions based on a graph formalism, and consists of graph manipulation functions and visualization functions. It provides for a standard graph editor and performs animation of changing graph data. We describe the characteristics of this tool and illustrate an extended Petri net simulator and a prototype robot system as its applications. In the prototype robot system, the graph editor is used for robot data input and the formalized graph data structure is used for data manipulation, for example in 3D set operations.
Naota Inamoto, Tosiyasu L. Kunii

2.4. Temporal Issues of Animate Response

Increased capacities of personal workstations enable graphical user interfaces to offer natural human computer interaction. This leads to the possibility of animate response, i.e., natural transitions from one state to another. It is recognized that for optimal efficiency such animate response should be “tuned” to the time frame of the user. It is not sufficient to speed up the response as much as possible: the temporal characteristics of the response should be based on human perceptual capabilities. This paper is a survey of notions related to animate response. Throughout the paper we will touch upon subjects which need consideration and/or further investigation.
CR Categories and Subject Descriptors:
[Computer Graphics]: Picture/Image Generation — Display algorithms;
[Computer Graphics]: Computational Geometry and Object Modelling — Curve, surface, solid, and object representations;
[Computer Graphics]: Methodology and Techniques — Interaction techniques
A. A. M. Kuijk

2.4. Applications and Comparison of Different Mathematical Methods to Compute Form Factors for “Radiosity Images”

A fundamental problem in synthetic images is to determine a correct light interaction between surfaces in an environment.
The radiosity method is an algorithm based on theories used in thermal engineering that includes effects of reflection between ideal diffuse surfaces. The computation of geometric factors (Form Factors), which describes the energy exchange between surfaces, represents the most expensive part of computational time necessary to generate a synthetic image.
This paper aims to analyze different approaches to compute form factors. Subsequently, we make a numerical comparison of the results and provide the evaluation of the relative errors and of the computational time.
At the end we present some simple images generated with the different approaches discussed in this paper.
F. Bresciani, P. P. Rinaldi, F. Tapparo

2.6. Comparative Operations in Solid Modeling

Functions for manipulating objects are essential for scientific visualization systems because it is necessary to handle all related data efficiently and to generate objects satisfying given conditions to objects. If operations to compare objects both topologically and geometrically when visualizing the real world, we can model objects more credibly and validly. We introduce comparative operations for this purpose. They are based on a boundary representation incorporating the concept of connected component as a separate geometric entity. Quantitative and qualitative characteristics for all geometric entities which compose objects are analyzed during comparison. The necessary algorithms for performing comparative operations have been developed, and implemented on a commercial solid modeler.
Myeong Won Lee, Toshiaki Satoh, Tosiyasu L. Kunii

Highperformance and Multimedia Workstations


3.1. 3DGRP — A High Performance Graphics System

Modern high performance workstations are based on hardware support for geometry and rendering processing in conjunction with a flexible window management system. These areas are addressed by existing and forthcoming graphic systems and standards like GKS-3D, PHIGS, PHIGS+ and the X- Window System. The 3DGRP is designed to fulfil the different requirements of hardware support in these areas. The architecture presented is a homogeneous solution based on a multi-processor approach with automatic load balancing. The processors are connected tightly to a distributed frame buffer which provides high memory bandwidth. An efficient support of BitB1Ts and window addressing mechanisms is given.
H. Selzer, T. Haaker, H. Joseph

3.2. An Auditory Display for Exploratory Visualization of Multidimensional Data

This paper describes an auditory display being developed for a workstation designed to facilitate the visualization of multidimensional data. To establish the context of this work, the paper first discusses the psychophysical basis for using sound to represent multidimensional data, surveys the related literature, and describes the novel graphical technique employed by the workstation to display multidimensional data visually. Next, it gives an overview of the auditory display and its relationship to the visual display, emphasizing the roles played by the apparent location of sound sources and the repetition of sounds in time. Finally, the paper discusses the current prototype sound system and some of the difficulties to be faced in implementing the proposed facility. This work is being done as part of the Exploratory Visualization (Exvis) project at the University of Lowell.
S. Smith

3.3. A Model for the Use of Different Sound Description Layers Within a Multimedia Environment

The number of applications on graphics workstations using sound in order to enhance the human-computer interaction capabilities is increasing. Using and handling sound in addition to text and graphics seems to be the next forthcoming step towards a multimedia environment.
This paper outlines a first approach for an audio content architecture consisting of different description layers which are able to handle different levels of abstraction. Therefore a comparison between image rendering and sound synthesis is made. Then a short overview on digital sound synthesis techniques and sound color models is given in order to show hardware requirements as well as psychophysical difficulties. Finally, the audio layer model is presented. It contains three layers for parametric and symbolic audio description, one layer for digital audio, and one presentation layer.
C. Blum

Experimental Workstations and Environments


4.1. Some Thoughts on a Computer Workstation for the Biological Sciences

The biological sciences are in need of a mechanism to consolidate vast amounts of morphologic information. Currently, both academic and professional communities are struggling with the organization and accessibility of current technologies. The computer workstation provides a powerful solution to this management problem because of its ability to adaptively store, access, and communicate information. The computer can adapt to the progression of knowledge, helping the user to assimilate and easily access current technology. The effective use of a computer for this purpose will require complex biological models derived from all the applicable sciences, including anatomy, physiology, biochemistry, kinesiology, and biomechanics.
T. O. McCracken, T. L. Spurgeon, L. G. Lodise, S. K. Koch, R. Miranda, C. Fedde

4.2. Issues in the Design of Workstations for Psychology Experimentation

Three issues must be considered in the design of a computer-based system for psychology experimentation: flexibility and generality, data logging and temporal precision, and stimulus presentation. This paper discusses these issues and presents solutions to them based on the design of a graphics workstation used for running experiments in perception, human factors and cognitive psychology in the Computer Graphics Lab of the University of Waterloo.
L. R. Bartram, K. S. Booth, W. B. Cowan

4.3. Protocol Toolkits for Distributed Real-Time Experimental Workstations

An experimental workstation is often an end system in a network of systems generating, processing, and presenting data. Graphic software standards for workstations often extend the virtual display device model but leave little flexibility for application programs using task and process concurrency in all systems, including the workstation. Such applications are true distributed systems; they benefit from use of standard services in communications facilities, and from formal methods of design and implementation for custom protocols, where standard services are inadequate or non-existent.
As an example, a protocol for synchronization of clocks by exchange of timebase messages (based on an existing network timestamp protocol) is generalized to a method for tracking and synchronization of other real-time databases, which can be applied to distributed interactive world-coordinate clipping of a real-time data stream for graphic display. This method of distributed clipping is illustrated in a system for interactive display of real-time weather data.
A. S. Brown

4.4. Data Analysis in a User Interface for Distributed Process Control

IBIT is a user interface toolkit for distributed process control and manufacturing applications. It has been used to build a direct manipulation interface for a family of ion beam implanters. In the course of controlling production, a large volume of information can be collected for both online and offline analysis. In this paper we discuss extensions to the toolkit related to data collection and analysis. These extensions, which are currently under development, include a spreadsheet for event definition, icons for controlling charting, and visualization of the ion beam density. We relate these results to data analysis in manufacturing applications in general, and to facilities needed in workstations for experimentation.
J. Koegel, S. McElroy, B. Chanasyk

4.5. Scientific Visualization in Supercomputing Environments

This paper describes the CINECA supercomputing environment in which many graphical applications are being realized using local network facilities. The general guidelines of a CNR project on scientific visualization promoted with the cooperation among several Italian research groups are mentioned here. Moreover, the applications considered in this project are provided with graphical workstations remotely connected to a supercomputer via a high speed network.
G. Fabiani, M. Lanzarini, L. Moltedo


Weitere Informationen