Search
This issue contains the Report from the Graphics Coordinator, a summary of AGOCG Technical Reports, information on progress in the evaluation of image processing software, and details of the latest release of the PHIGS Toolkit from the University of Manchester. Graphics around the Country looks at Computer Graphics in the AFRC. Reports include SIGGRAPH 92 in Chicago and Visualization 92 in Boston.
This issue is being published in December, so we wish all our readers a Happy Christmas and a successful and productive New Year!
I thought it might be useful to let you know some of the politics behind some of the things AGOCG get up to. I have just been putting together the proposal for the graphics budget for the Software and Information Services Committee. SISC is a committee of the Information Systems Committee of the Universities Funding Council. They recommend to the ISC the ways in which the software budget should be spent. My post and that of Steve Larkin (the Visualization Support Officer based at the University of Manchester) are funded under their allocation for graphics as are the various events, training materials etc. There is an allocation of funds for small projects which can be approved relatively quickly through AGOCG but all other proposals have to go through AGOCG, then SISC and then to the ISC. As you might imagine the relative timings of the meetings are critical as SISC and AGOCG meet 4 times a year. It also gets interesting as we move towards the end of financial years and monies have to get approval in time to spend against budgets. The system does seem to work and it does ensure proper consideration of proposals at different levels. SISC are also responsible for budgets such as CHEST, Newcastle mailbase and the Lancaster archive.list chaired by John Slater.
The ISC budget is just one part of the AGOCG funding. The SERC has allocated manpower and funds for graphics, primarily in the area of visualization. The SERC, for example, fund the Graphics and the Engineering Newsletters. The NERC and the AFRC also play major roles in the work of AGOCG. NERC, for example, were involved in the setting up of the Support Workshop. AFRC are involved in the postgraduate course being held in January. The AGOCG work really is a team effort between the funding bodies and the higher education institutions in the UK.
News on the AGOCG front is that Bob Hopgood will be retiring as Chairman at the end of the year due to other pressures on his time. Bob has played a major role in the setting up and running of AGOCG. He is being replaced by David Boyd who is also from the Rutherford Appleton Laboratory. I am sure we will miss Bob's guiding hand. However he has set us all standards which we will attempt to maintain. I am sure that Bob will be pushing us all as a user of the AGOCG services!
The following technical reports are the list of up to date reports. Some of the older ones have been deleted from this list.
AGOCG also have sets of training materials on: graphics standards in general, GKS, PHIGS, advanced PHIGS slideset, CGM, Cricket software, Uniras software. If anyone is interested in these then please contact Joanne Barradell.
I have some Mac software from the San Diego Supercomputer Center which introduces colour concepts. Interactive Colour is a hypermedia guide to colour in computer graphics. It is designed for browsing and allows the user to change colours using various palettes. The program is a stand-alone program that runs on a colour Macintosh. System 6.0.2 or later is needed as are at least 4 megabytes of memory, 4.5 megabytes of free space on the hard disk and an 8 bit video card. To get the correct text appearance you also need Avant garde and Geneva fonts and Adobe Type Manager or TrueType. If you would like a copy of the software please send me two 3.5in high density formatted disks plus a suitably addressed envelope for the return.
Send to Dr A Mumford, Computing Services, Loughborough University, Loughborough,As indicated in the last issue, a meeting was held on September 14/15th in Loughborough at which evaluators reported back their findings and suppliers who had been short-listed were invited to present their products, inform the evaluation team of any recent developments and answer questions which had arisen as a result of the evaluation exercise.
The meeting turned out to be extremely useful. It soon became clear that developments were taking place rapidly in this market especially in the area of user interfaces and tutorial and on-line help systems. Because of these developments it was necessary to plan a limited amount of extra evaluation of new versions of some products. It was also possible to eliminate several products from further consideration since they clearly would not satisfy the community's requirements on the planned time-scale.
A report on the evaluation has been compiled in draft form and this should prove invaluable to those contemplating purchasing software in this area.
On October 20th CHEST and representatives of the evaluation group met with suppliers and final recommendations to be submitted to the funding body (SISC) for a site licence deal were made. The result of these negotiations is, of course, commercially sensitive and cannot be reported at the present time. It should, however, be possible to report the final outcome in the next Newsletter. Anyone interested in the availability of the evaluation report (which should have been published by the time this newsletter reaches you) should contact Anne Mumford.
There is still no concrete news regarding the deal for a PC drawing package. Contractual negotiations are still in the final stages. Hopefully there will be good news for the next newsletter.
Cricket Graph Version 3 for the Macintosh has been distributed to sites. The equivalent version for PC DOS is due in early '93.
The working party is compiling a list of PC packages which handle contouring and surfaces and 3D graphics with a view possibly to carrying out an evaluation. Any suggestions for inclusion in the list are welcome. Also views on whether a site licence deal in this area is desirable would be appreciated.
The latest release of the PHIGS Toolkit from the University of Manchester is now available. The new version (3.1) supports SunPHIGS 2.0, HP PHIGS 2.2 and MIT's PEX-SI.
Help for the Application Programmer Tools in the PHIGS Toolkit are divided into two categories: programming tools, and high-level tools. Programming tools are generally quite simple single-purpose procedures, and are designed to help applications programmers to construct PHIGS programs more quickly and reliably. The high-level tools are more powerful, and provide programmers with means for visualizing and debugging structure networks. A complete description of the PHIGS Toolkit is given in the Graphics Newsletter, Number 23, June 1992.
A training course for users of the PHIGS Toolkit was held at the University of Manchester on September 23rd 1992. Course materials including four programming exercises and three step-through tutorials are provided in this release of the Toolkit.
We would like to encourage all people interested in the PHIGS Toolkit to register as PHIGS Toolkit users. This will ensure that all PHIGS Toolkit users will receive notice of new versions, course dates, bug reports and any other useful information. Please send the following information to phigstoolkit@uk.ac.man.cs.
Name: Organisation: email: Telephone/FAX: PHIGS implementations used:
Even if the PHIGS Toolkit is not currently available for your particular PHIGS implementation, please register. We are currently working on ports to several other PHIGS implementations and it would be very useful to know what the demand is for different versions.
The PHIGS Toolkit is available from two sites in the UK:
The PHIGS Toolkit is available from HENSA (Higher Education National Software Archive) at the University of Kent. The HENSA Service at the University of Kent can be accessed in a number of ways:
address: uk.ac.hensa.unix login: guest path: <ARCHIVE>/filenameFor example:
% fcp -b "<ARCHIVE>/uunet/ ls -IR.Z"@uk.ac.hensa.unix ls -IR.Z User name on uk.ac.hensa.unix? guest Password on uk.ac.hensa.unix? jn@ukc.ac.uk
Any general queries regarding the HENSA service at The University of Kent should be directed to hensa@unix.hensa.ac.uk, queries specific to the Netlib service should be sent to netlibadmin@unix.hensa.ac.uk and stuff concerning the source archive to archive-admin@unix.hensa.ac.uk.
The relevant files for the PHIGS Toolkit on HENSA are:
PTK 3.1 /unix/phigstk/ PhigsToolkit3.l.tar.Z for SunPHIGS 2.0 on SunOS, HP PHIGS 2.2 on HP-UX, PEX-SI on SunOS. PTK 2.0 /unix/phigstk/PhigsToolkit.tar.Z for SunPHIGS l.x on SunOS. /vms/phigstk/ptk.hex for DEC PHIGS 2.3A on VAX/VMS.
By anonymous ftp from uk.ac.mcc.hpb. (130.88.200.7) Usemame anonymous, and your network address as password. The files are:
PTK 3.1 pub/cgu/ptk/ptk3.1.tar.Z for SunPHIGS 2.0 on SunOS, HP PHIGS 2.2 on HP-UX, PEX-SI on SunOS. PTK 2.0 pub/cgu/ptk/ptk.tar.Z for SunPHIGS l.x on SunOS. pub/cgu/ptk/ptk.shar* for DEC PHIGS 2.3A on VMS.
For VMS the Toolkit is stored as a collection of SHAR files. There are 288 files in total, each 15K in size. They are called ptk.shar_X where X is 1 .. 288. To rebuild the Toolkit directory structure the files must be concatenated together and run as a command file.
$ copy ptk.shar_%, ptk.shar_ %%, ptk.shar_ %%% ptk.shar $ @ptk.shar
Send a 1/4 inch cartridge for SunOS, and a 1/2 inch open reel magnetic tape for VMS to Tim Hopkins, Computing Laboratory, University of Kent or Toby Howard, Department of Computer Science, University of Manchester.
The first phase of the PHIGS Toolkit does not include support for PHIGS PLUS. Work is now under way to expand the Toolkit to include extensive support for PHIGS PLUS functions, and the expanded PHIGS Toolkit will be released in April 1993. We are currently looking for people to act as official beta-testers for the PHIGS PLUS work. The beta-test version will be ready on 1st January 1993 and the test will run until the end of February . A report outlining the test results will be required. Please contact us if you would like to help.
A toolkit providing support for NURBS curves and surfaces has also been developed at Manchester, and is designed to be complementary to the PHIGS Toolkit. It is available from the same sites as PTK.
The netlib archive run from the University of Kent has been extended and now forms part of the Higher Education National Software Archive (HENSA) service.
HENSA is an ISC funded service to provide the academic and research communities with access to a number of software archives. The new HENSA UNIX Archive run from Kent provides mostly UNIX software and is accessible via an interactive browsing facility, called fur as well as e-mail, DARPA ftp and NI-FTP (Blue Book) services. This archive currently makes available a large part of the uunet software archive including usenet's comp.sources and the gnu sources as well as all the existing netlib and statlib software and an archive of miscellaneous UNIX items. The archive is updated daily and is rapidly increasing its software base.
Although the new netlib service will continue to be accessible from netlib@ukc.ac.uk for the near future, current users are encouraged to change to netlib@unix.hensa.ac.uk. In addition all netlib software may now be obtained via both Internet ftp and Blue-book ftp (ni-ftp) and through the interactive browsing facility.
A number of graphics packages are available from the archive; these include
The NURBS Library from David Yip and Terry Hewitt at the University of Manchester. This library provides a number of functions which includes the creation and manipulation of NonUniform Rational B-splines (NURBS). The library is available for both Unix and Vax/VMS systems. For Unix the package is available as the compressed tar file misc/unix/ nurbs/nurbs.tar.Z. The VMS version is available as misc/vms/nurbs/nurbs.hex.
The PHIGS Toolkit, also from the University of Manchester, is available for Unix machines at version 3.1 and for VaxNMS at version 2. A compressed tar file containing the Unix version may be found in misc/unix/phigstk/PhigsToolkit3.1.tar.Z.
The VMS version is available as misc/vms/phigstk/ptk.hex.
The RAL-CGM toolse for manipulating Computer Graphics Metafiles is available for Unix systems only as a compressed tar file in misc/unix/ralcgm/ralcgm.tar.Z. The Vax/VMS files are in a compressed hexified format. The tools for unpacking these files may be found in the directories misc/vms/hexifyandmise/vms/lzw. The following may also be of interest
In addition there is a wealth of X based software in the directory uunet/pub/window-sys/x/ eontrib.
The University at Kent is also acting as the distribution agent for the RALGKS and RALGKS-3D packages. These are available under licence to UK academic institutions at a cost of £40 per package. Unix and VMS versions of RALGKS-3D are available; RALGKS is only available for Unix. To obtain copies of this software contact Tim Hopkins at the address given at the end of this article.
<ARCHIVE>path-of-file from uk.ac.hensa.unix Username: guest Password: <your e-mail address>will retrieve the requested file.
send path-of-file to archive@unix.hensa.ac.uk.For more details on how to find out where things are, see the section on fbr below. For example, more details of the HENSA service may be obtained using the request
send info/hensa.unix.general
Welcome to HENS A - the Higher Education National Software Archive at the University of Kent at Canterbury Use the login name 'archive' to access the Unix archive. login:archive Welcome to the UK Source Archive at the University of Kent at Canterbury. Please enter your email address:<your e-mail address>You will then be placed inside fbr's restricted shell. Use the help command for up to date details of what commands are available.
Any queries/problems/suggestions concerning the HENSA Archive should be E-mailed to the Archive server administrator at archive-admin@unix.hensa.ae.uk
We continue this month on the grand tour of the Research Councils with an article from John Owen from AFRC. John's comments on migration to the latest Software of the Year and the leapfrogging between PC packages must strike a chord with all graphics supporters.
Again it is worth remembering that this section is intended to give space to those working in, or having a relation to, education and research to write on topical subjects. This is a good opportunity, which will perhaps help you attract support and finance. Contributions should be sent to me.
Lastly, only for this issue, there is an unrepeatable and completely irresistible offer. I have been editing this section for a little time, and now feel it is time to hand over to some fresh blood. I have been aware that there are fields out there that have not been touched, like image analysis and research into computer graphics.
Don't be shy now... this is a unique opportunity to widen your own view of computer graphics and to expand the horizons for the rest of the community. If you feel that you would like to edit this section, please contact me for further information.
AFRC's scientists use computer graphics at a relatively low level compared with those at SERC and NERC. The vast majority of scientific computing is done on VAX/VMS machines, though UNIX-based machines are starting to be introduced at our Sites.
The major expansion in the AFRC's work over the last few years has been in molecular biology, which does not (yet) make great use of anything other than simple graphics.
We have interfaced all the Molecular Biology software which we support to UNIGKS. A user can select the output device (and filename, if appropriate) from a menu, and then all the molecular biology programs which are run thereafter will write their graphics to that device. Output devices include the usual terminals, as well as PostScript and UNIPICT files.
The pictures from the molecular biology software are fine as working drawings, but are usually considered to be of insufficient quality for publication. For smartening up a picture, a variety of PC-based drawing packages are used. Pictures are transferred via CGM flIes, and the text is replaced with a suitable high quality font.
Output is then usually to laser printers. Many PC-based packages are used, on both IBM PC clones and Apple MACs. Although some packages are more popular than others, there is no AFRC-wide standard set of programs. We do take part in CHEST deals where appropriate, but it is difficult to move people from their favourites, especially if a new program is not easy to learn. The use of interfaces with similar look-and-feel, helps migration to the latest Software of the Year (or month, ...week, ...). The leapfrogging between the most popular packages has caused us problems, and will probably continue to do so. It also causes packages to get more and more complex, so a package which starts out simple (and therefore easy-to-use), rapidly gets fiendishly complex, as it competes with similar products.
The majority of graphics production on VMS and UNIX is from UNIGRAPH with some UNIMAP hotspots. There are also a few groups who write their own programs, using UNIRAS libraries for the graphics. We currently make no use of PHIGS, and programs using GKS are written only at the Computing Centre.
We have written a program (called PLOT) to ease the production of hardcopy and allow flexibility in location and replacement of printers. PLOT can process CGM and UNIPICT files in batch, and submit the final file to the selected queue. Intermediate and final plotfiles are located on a scratch disk, so do not come out of the users normal disk quota. PLOT can also be used to display pictures on a terminal or X-server.
The use of PLOT means that the user does not need to know the name of the UNlRAS device driver, the name of the device ('Colour printer in Room A29, Grenville Building' is sufficient), or the name of the VMS queue.
Most users do not know (or care) what the underlying graphics package is. They are offered a menu of devices, and they know that they should use PLOT to process UNIPICT files, and that seems to be sufficient for the majority.
PLOT is the only program produced in-house which we support. All the others are commercially produced, and we at the Computing Centre act as contact to the vendors for the whole of AFRC.
Our agricultural engineering institute (Silsoe Research Institute) does make use of more powerful graphical methods. CFD is used to study wind-flow over agricultural buildings, and full CAD and mechanical linkage interaction systems as well as PEA systems are used in the design of agricultural machinery. Multidimensional data are now beginning to be collected. For example, logging of temperature, humidity and carbon dioxide levels at many positions in a greenhouse, over time. Analysis of these data will need a scientific visualization system.
At other Institutes, medical-type image analysis is also used, both at the micro level (investigating cell structure) and at larger scales (food structure). Again, a scientific visualization system could be very useful in this.
Contouring programs are used to show the spread of aphids outwards from the south-east during the summer weeks. The growth in population of insect species in areas varying in size from a single field to the whole of mainland Great Britain are also displayed using contour maps.
Currently, we use VT340 terminals and PCs running EMU- TEK for display of graphics from VMS and UNIX applications. Hardcopy is to laser printers for monochrome, and A4 inkjets for colour. We are starting to see the use of X-clients displaying on workstations and PCs running HCL eXceed. We believe this will be very valuable to us in hiding the operating system (whether it is VMS or UNIX) from the user. The scientist sees the operating system as something that gets in the way of what he wants to do. The sooner we can remove this obstacle, the better for all. This is our long-term plan.
Over 34,000 computer graphics enthusiasts packed the McCormick Centre in Chicago, July 26-31, to see and experience the latest computer graphics technology at SIGGRAPH '92. Official SIGGRAPH activities opened Sunday afternoon with a Fundamentals Seminar chaired by Wayne Carlson. This buzzwords seminar explained the terms and concepts attendees would encounter during the week. A welcoming reception took place in the art show area following the Fundamentals Seminar. This juried art show presented computer graphics in 2D works, 3D sculptures, animations, and interactive works, and the open location for the show this year provided an inviting, comfortable setting for viewers.
On Monday and Tuesday there were 27 in depth full day and half day courses on topics in modelling, rendering, animation, video technology, and applications such as scientific visualization and interactive multimedia publishing.Tuesday also marked the opening of the 225 vendor exhibition floor where exhibitors demonstrated the latest in networked computer graphics hardware and software and products for engineering, entertainment and scientific visualization. Popular virtual reality exhibits and HDTV displays dotted the floor. This year's special events included Showcase, G-Tech, and SIG Kids. Showcase included 36 demonstration projects, illustrating interactive and collaborative research and applications that rely on high performance computing and communications. These projects included the Distributed Virtual Wind tunnell by NASA Ames Research Centre, the Distributed Laboratory, an interactive visualization environment for electron microscopy and 3D imaging, by San Diego Supercomputer Centre, and Distributed Visualization of Large Atmospheric Data Sets by University of Wisconsin. G-Tech presented unusual interactive projects such as Apple Computer's interactive, electronic museum and Hypermedia Productions' exploration of memory and forgetfulness in humans and computers. SIGKids was a learning lab and showcase for students in grades 6-12.
The technical papers, state of the art in computer graphics, and panels, the forum for discussion and debate on graphics topics, took place Wednesday through Friday. SIGGRAPH '92 presented cross- disciplinary, video, multimedia, and pedagogical papers, in addition to the research and systems papers. Panels covered a broad range of topics such as multimedia, virtual reality, standards, colour and hardware.
The technical sessions opened with a welcome by Maxine Brown, SIGGRAPH '92 Chair, and the State of SIGGRAPH by Jim Thomas, SIGGRAPH Chair. Henry Fuchs received the 1992 SIGGRAPH Computer Graphics Achievement Award for his contribution to high-performance, parallel display architecture. Fuchs is Professor of Computer Science and Adjunct Professor of Radiation Oncology at the University of North Carolina. Robert W. Lucky, executive director of the Communications Sciences Research Division of AT&T Bell Laboratories, delivered the keynote address, stressing creativity in understanding abstract images. Guest Speaker Jim Clark, chairman and founder of Silicon Graphics, Inc., talked about the telecomputer of the future.
The Electronic Theatre attendees packed the auditorium for three evenings to see the latest juried selection of computer graphics animations, in HDTV, as well as traditional film and video. My favourite selections included S.C.A.M. Starving Computer Artist's Market, a very humorous spoof from New York Institute of Technology, and Caustic Sky: A portrait of regional acid deposition, a superb visualization from North Carolina Supercomputer Centre. The Computer Graphics Screening Room showed additional animations in a more intimate setting throughout the week. SIGGRAPH Video Reviews, containing most of these selections and other educational and technical animations, are available for purchase.
The annual SIGGRAPH conference illustrates the constantly evolving nature of the field of computer graphics. Scholarly paper presentations provide information about advances in computer graphics hardware, software and theory, as well as topics for future panels. Topics that are in panel discussions will evolve into future courses, and many of the techniques covered in the day long courses will appear in future exhibits in new hardware or application packages. Next year's conference will celebrate the 20th anniversary of SIGGRAPH conferences and promises to be very exciting. SIGGRAPH '93 will be in Anaheim, CA, August 1-6. Mark your calendars now.
October 19-23, 1992 Boston, USA
The IEEE Visualization 92 Conference took place in Boston, 1923 October. There was a full program of tutorials, papers, panels, and case studies, and many items of interest for visualization users, developers, and researchers. This Report summarises the principal results presented.
Around 400 people attended Visualization '92 in Boston, October 19-23. The first 2 days consisted of Tutorials and Workshops on a range of topics including volume visualization, modelling, fluid flow, human perception, virtual reality, and education. There was a Khoros Workshop and also the ACM SIGGRAPH 2-day Volume Visualization Workshop.
On the Tuesday evening there was a Special Evening Session on How to Lie and Confuse with Visualization led by Nahum Gershon, Mitre Corporation, who made some interesting and serious points about how easily we are misled by the way we intuitively deduce things from information presented in pictures, the way we contrast large and small areas sometimes wrongly, and how we construe perspective, particularly in works of art. Carol Hunter, Lawrence Livermore National Laboratory, described how the choice of colours for information on video can dramatically affect the information presented. Al Globus, NASA Ames Research Centre, outlined 13 ways in which it was possible to lie with scientific visualization!
The opening session of the Conference began with an IEEE Meritorious Award to Dr Larry Rosenblum for his service in the areas of developing the theme of data visualization within the IEEE Technical Committee on Computer Graphics, the Visualization Conference, the special theme issues of IEEE Computer Graphics and its Applications and the Computer Magazine.
The Keynote Address was delivered by Dr Eugene Wong, Associate Director for Industrial Technology, Office of Science and Technology Policy, Office of the President. Dr Wong was a Professor of Computer Science at Berkeley and founder of Ingres Corporation. Dr Wong began by outlining the Presidential initiatives in 1991 - Global Change, 1992 Mathematics and Science Education, and High Performance Computing and Communications, 1993 - Advanced Materials and Biotechnology, 1994 Manufacturing. He defined science as knowledge of general truth and technology as the means to sustain and improve human life. The Science and Technology Policy was the government policy in the areas of supporting science, producing technology, and maximizing their benefits.
Participants in the High Performance Computing and Communications initiative (FY 92) were DOD (DARPA), DOE, NASA, NSF, ED, EPA, NIH, NIST, and NOAA. High performance measures included speed, reliability, usability, and capacity. There were four major components - Systems - hardware and software, Applications - the 'grand challenges', Networking - involving the National Research and Education Network (NREN), and Basic Research and Human Resources. Visualization was the grandest of the challenges and had a symbiotic relationship with the High Performance Computing and Communications (HPCC) initiative, for obvious reasons. These were as follows -
The budgets for the HPCC program were as follows - FY 92 638M, FY 93 789M, FY94 956M, FY 95 1087M, and FY 96 1202M. The vision and the strategy was to accelerate the networking technology forward since this was the underlying infrastructure upon which other developments depended. The two cultures of science versus commerce, and applications verses technology could be brought together in a unified way via scalable and modular systems. The two key areas were Gigabit networks ('one for all') and parallel architectures ('all for one). Examples of speed-ups accomplished in current application areas included the following: structural mechanics (1019 times), wave mechanics (1020 times), fluid mechanics (1009 times), electrostatics, and ODE's.
Highlights of current progress included the delivery of scalable systems (e.g. Thinking Machines Corp), the doubling of NREN traffic, and the grand challenge application areas. Internet (the precursor to NREN) had measured a 30% increase in network traffic per MONTH. A massive culture change was therefore anticipated, brought about by the rapid growth in electronic communications.
The twelve largest computer companies had (via their CEO's) come to an agreement with the semiconductor industry and Japan. The National Coordination Office for HPCC was established on September 1, 1992 and the Director was Dr D .A.B. Lindberg, currently Director of the National Library of Medicine and a pioneer in the use of computers in the area of pathology. He would also chair the Federal Coordinating Council for Science and Technology (FCCSET). Its functions were to coordinate the program, provide a single point of contact, formulate policy, and provide institutional continuity.
The Keynote Panel on Interactions between Science and Visualization investigated the relationship between visualization and science. How could good visualization techniques help us to do science better? How could the needs of science influence visualization research more effectively? Dr Larry Rosenblum (Office of Naval Research, European Office), Dr Stephen Benton (MIT Media Lab), Dr John Conway (Princeton University), Dr Jeff Dozier (UC Santa Barbara), and Dr Robert Langridge (UC San Francisco) outlined the interactions in their own fields of interest. Dr Stephen Benton covered stereopsis, computational holography, and MRI Imaging, and outlined image cues spatial organisation, and spatial display technologies. He cited a reference from Jacob Bronowski from a 1978 Yale lecture on The Origins of Knowledge and Imagination where imaging and visualization were brought together.
Dr Jeff Dozier outlined Sequoia 2000, a next generation information system for assessing global change. The severity of the overall global situation was indicated by the increasing greenhouse gases, the decreasing stratospheric ozone, the decreasing biodiversity, deforestation, and the increasing human population. Just to preserve the status quo in the environment would need an overall reduction of 70% in car exhaust gases, and global HCFs would continue to go up for 15 years even if they were completely stopped now. Thus in this situation of severe crisis it was essential to have improvements in observations, models, and information systems to enable an accurate picture of the environment to be obtained. The study, prediction, and understanding of changes in the Earth's system all had a visualization component which was essential to their success. A $12 million DEC flagship project had been set up in this area, which involved database management systems, access and data, and visualization. Remote visualization was to be used for browsing, managing, interactive analysis and computational steering. Challenges included dealing with large objects and interactive visualization.
Dr Robert Langridge outlined the development of molecular graphics from 1964 to 1992, and the ways in which molecular biologists used visual tools to explore existing structures, design new molecules, and investigate relationships between molecules, e.g. by fitting one inside another to form a new compound. He presented a graph of log (typical cost of a computer for molecular modelling) against time, beginning with 2 million in 1964 for Project Mac to 40K for an IRIS 3030 in 1986. If this trend continues, a comparable system was expected to cost 300 dollars in the year 2000 (Dr Jim Clark has similar figures in his predictions - see ACM SIGGRAPH 92 Proceedings).
The remainder of this report provides a snapshot of various items in the Conference Program, since it was possible to attend only a proportion of the total due to the parallel streams of Papers, Panels, and Case Studies.
One of the key problems in the future is linking large databases to visualization tools, and a Panel on Managing Large Scientific Databases set out to address these issues. Robert Cromp of the NASA Goddard Space Flight Center outlined the Intelligent Data Management Project and methods for extracting metadata, and how an intelligent information fusion system could be designed to address the areas of metadata management and user interface. Ray Wall of the Jet Propulsion Laboratory discussed animating space imagery. Animation was very useful to observe spatial relationships, structure, and process dynamics. Complex models and process simulations could be overlaid on observational data which was essential for identifying correlations or deviations. The image data could be queried or browsed. Animation could also be related directly to virtual environments. Rendering of images could be done on parallel architectures. These methods were to be used on the Magellan data from the Venus probe; the rendered was also being interfaced to the JPL animation tool Surveyor.
Two Panels of further interest were Grand Challenge Problems in Visualization Software chaired by Dr Lloyd Treinish (IBM), and Real Virtual Environments Applications Now chaired by Paul Breen, Mitre Corp. Lloyd Treinish began by outlining the problem and the tools required. The issues in visualization software included matching the software more closely to the scientist's needs and requirements, integration of visualization and computation, usability of visualization software, managing large data sets, increasing the dimensions of interaction with the data, and real-time interaction with data. Matching the software to the scientist's needs required greater knowledge of the discipline characteristics of each area. Systems were required that could provide different levels of access, accommodate multiple sources of data, support domain-driven problems, required a minimum of detailed system knowledge for their use, and which could create intelligent customised interfaces. A tall order!
David Butler, Sandia National Laboratory, analysed the work being done in the area of the integration of visualization and computation where the user had direct access to computation as it proceeded rather than outputting results of previous post-processing stages. Spontaneous computing was an area where all objects were 'visible' as opposed to interactive computing where only a fixed number of user dialogues were possible. The computation agent managed the numerical objects, the visualization agent managed the graphical objects, the storage agent managed the persistent objects, and control managed other agents. The enabling technologies were the object-oriented paradigm, and the visualization reference model. Current research challenges included domain-specific data models, algorithm visualization, new visual programming paradigms, and 3D interaction techniques.
Hikmet Senay, George Washington University, assessed visualization tools from a usability perspective, and proposed that it was important to design visualization systems which provided guidance and support for users for domain specific visualizations, and task-specific visualizations. Automating the design of visualizations was an area of current research. Some current systems for 2D included APT, SAGE, and BOZ, and for 3D - CCAD, NSP, VISTA, and Autovisual. The second challenge was to integrate visualization with other technologies and tools - computation simulation and analysis; data - storage and models; interaction - multisensory and multimedia. The overall objective was to be able to design intelligent information systems capable of understanding users' goals specified in an appropriate high-level language, e.g. 'Show me the effects of acid rain on the wildlife in Queensland'. Such systems should know under which circumstances a particular visual presentation was more effective than another. The two aims therefore were to provide support and guidance to the user, and integrate visualization with other technologies and tools.
Georges Grinstein, University of Massachusetts at Lowell, outlined three challenges: large data sets, human perception capabilities, and increasing the dimensions of interaction via sound, color, motion, surfaces, and volumes. It was going to be important to be able to explore massively large data sets, and integrate the database with the visualization.
Steve Bryson, NASA Ames, argued for real-time interactive scientific visualization, or reification, defined as the ability to regard or treat an abstraction or data as if it had concrete material existence. This involved the use of Virtual Reality type of environments. It was easier to explore and understand a 'thing' than a picture of a thing, and active exploration was more useful and dynamic than passive observation. Real-time visualization was easy to do for small tasks but very difficult for real-time applications in science and engineering. Computation, rendering, and data management had all to be done in less than 0.1 sec. For unsteady CFD a typical data set was 5-10 GB, and a time-step was 100MB. For some visualizations (e.g. particle paths) all data must be available. Some visualizations such as isosurfaces and streamers are computationally intensive. Hardware improvements will alleviate some of the problems, but currently there are design compromises that have to be made, viz, speed versus accuracy, speed versus limited data access, and speed versus quality of the rendering. Interactivity must make sense for the particular scientific problem, so the designer must understand the science. The interface must be as invisible as possible. The user must be given control over the compromises, and high performance must be built in from the beginning.
In the Panel on Real Virtual Environment Applications - Now Dr David Zeltzer (MIT Media Laboratory) began by outlining the need for a better term than Virtual Reality to cover all the areas encompassed by the matrix of immersive interfaces, autonomous agents, and the complex work domain. His work was concerned with the visualization of human behaviour and human motor behaviour, and the modelling and simulation of virtual actors as humans.
David Mizell, Boeing Computer Systems, outlined the use of CAD and VR systems in the design of the 777 plane. Full scale mock-ups of the plane were no longer used; everything was done by CAD. VR was now being used to provide 'augmented reality' where graphics was superimposed on the real world, to provide 3D immersive views of engineering work as it proceeds, and also do accessibility and maintainability checks. On the 747 there were 40000 form boards for wiring circuits; these were now being done by using a graphics display to ascertain where to put the wiring.
Richard Satava MD, Silas B. Hays Army Hospital, described the increasing use of VR tools and techniques in surgery. Remote manipulation techniques were now being used with appropriate feel and feedback. The primary benefit of the new techniques was the minimal impact upon the patient since much smaller incisions were performed compared to manual procedures. Hospital recovery time was less and patient recovery time was much less. These techniques have great potential for surgical training and for remote surgery via telepresence. If there is no reason for the surgeon to believe the patient is not in front of him on the TV screen, then he will operate as if he were. But the actual patient could be on the other side of the country. Using these methods other modalities can be brought in and overlaid on the patient, e.g. CT scans. SRI are currently working on this technology to make it available to surgeons. Clearly the issues of accuracy, reliability, stability, and maintainability have to be fully addressed.
Michael Stephens, Computer Sciences Corporation, outlined the uses of VR in a number of application areas including the Waterways Experimental Station (WES) - six major laboratories for hydraulics, structures, geotechnical, coastal, environment, and information technology. VR applications included architectural walk-through, fluid mechanics, surface exploration, and volume visualization.
Georges Grinstein, University of Massachusetts at Lowell and Mitre Corp, summarised the use of Virtual Environment Architectures for interaction with magnification, satellite control, electronic sand box scenarios, virtual command post, and air traffic control. The Electric Power Research Institute were developing a trainer system by software which could replace a trainer system currently needing 600K dollars in annual maintenance.
Five papers of particular interest were Visualization of Second Order Tensor Fields and Matrix Data by T. Delmarcelle and L. Hesselink, Volume Rendering on a Distributed Memory Parallel Computer by T. Elvins, Visualizing Wind Velocities by Advecting Cloud Textures by N. Max, R. Crawfis and D. Williams, Virtual Space-time: An Environment for the Visualization of Curved Space-times via Geodesic Rows by Steve Bryson, and Network Video Device Control by D. Nadeau and M. J. Bailey.
John Rasure summarised current thinking with regard to Khoros and visual languages. Current issues included interoperability, collaboration, objects in dataflow, distributed computing, fine grain programming, and applications building.
A Panel on Object-Oriented, Dataflow Visualization Systems - A Paradigm Shift? began by looking at current systems and noting the trends. Some points about current systems include the following: current uses, prospects, drawbacks, improvements, alternative approaches, and interoperability. Bob Brown, Manager of the IRIS Explorer Project at Silicon Graphics Inc indicated how large grain data flow systems were likely to develop. There were currently many users of these systems. To date around 25000 copies of Explorer had been shipped! Meaningful results were being obtained by users and visualization researchers alike, and tools like Explorer had brought visualization to many people who had not used it before. However, current systems had limitations. Large data sets and large programs were cumbersome. Large module suites were unmanageable, and systems were not optimised. Extending systems needed experts. Use of distributed computing needed care, and the visual programming style was not natural for all users. He envisaged that in the short term there would be developments for alternate execution modules, better data model, larger data sets, and better coverage of application areas. There were two pressures in the marketplace - one for simpler systems and one for more powerful systems! One approach to this is to make the system more user-friendly and also add advanced features for the experts. He expected more highly interactive interfaces in the future and more desk-top features such as audio and video.
Terry Myerson, International AVS Center, outlined the work of the AVS Consortium. Lloyd Treinish, IBM T J.Watson Research Centre, examined the effectiveness of dataflow systems, particularly in the context of large and complex data sets. There was high level user interface, powerful prototyping, and extensibility, but it required knowledge about the structure of the data, and it was difficult for the practising scientist (and non-computer expert) to extend the system. Interfaces could be too unfamiliar or too simplistic, and dataflow systems did not scale too well to problems of interesting and complex science.
A further evening session on Research Problems Visualization chaired by Georges Grinstein, added to the list of last year's problems. The new list is available from Georges Grinstein. Problems include handling large data sets, perception issues, and higher dimension problems. Issues in VR include creating a real sense of presence, better resolution, and better body tracking. A further challenge was in the area of modelling and visualizing unsteady phenomena where the data was changing with time.
The Capstone Speaker, Dr Margaret Livingstone, Harvard Medical School, outlined the complexities involved in seeing objects, and how the human visual system made sense of an infinite variety of images. This ability is made possible by the brain's capacity to process large amounts of information simultaneously. Recent studies suggest that form, colour, and spatial information are processed along three independent pathways in the brain. An understanding of the human visual system can vastly improve the perceptual quality of the display of images.
This was an excellent conference with many points of interests in the Papers, Panels, and Case Studies. The Proceedings are published by the IEEE Computer Society and there is also a video (PAL or NTSC format) containing extracts of the videos submitted with the papers.
The recently held European X User Group conference and exhibition was a resounding success. This is the fourth year the event has been held, and X Windows continues to gain new ground and attract more players. The exhibition was well-attended, with exhibitors offering a wide range of X terminals, software development tools, training programmes and connectivity products. Exhibitors included NCD, Tektronics, VI, Vision Ware and IST.
Speakers addressed issues such as X for conferencing and group working; current technological trends; X over slow lines; display PostScript and graphics within X Windows.
Continuing the EXUG policy of inviting big names from the world of X, this year three US-based experts headed the programme - Doug Young, Oliver Jones and Bob Scheifler. The programme included presentations by over 20 speakers, from eight countries.
Pre-conference tutorials proved a major attraction. Subjects on offer this year included Using C++ with Motif, Debugging X Programs, and a course designed for newcomers to X - From Awareness to Understanding. The strength of interest shown in this year's conference has given a boost to EXUG plans for 1993, including a rolling programme of tutorials, seminars and workshops, to be held in centres in the UK and other EC countries.
