Search
The SIMA project is funded by the HEFC's Joint Information Systems Committee (JISC) as part of their New Technologies Initiative. As reported in the last newsletter, this project has been funded and is being advised by Advisory Group on Computer Graphics (AGOCG) with myself as project director. The project will consist of a Support Officer, a series of projects including pilot studies and evaluations, and workshops.
A number of projects have been agreed at the time of writing. These are:
The results of these projects will be available as AGOCG Technical Reports and will be advertised in the Newsletter when they are ready.
This looks like an exciting project and I look forward to the results!
A team of support staff from various institutions have been working on new versions of workbooks for Unigraph, Unimap and Uniedit as well as new booklets on Unigraph and Unimap Solutions. These are being made available electronically and details of their content and availability are given inside this Newsletter. This continues to be an excellent collaborative exercise in the community. It is now 4 years since 17 of us got together in Edinburgh to produce the first set. This is now the second update set.
A meeting of graphics art and design staff was held in mid-April. The aim was to discuss the needs of this community and the ways in which AGOCG and other JISC funded initiatives might help this community. A report is given in this Newsletter.
AGOCG are keen to get projects under way in the following 2 areas.
The first project is an evaluation of colour management software. We would like someone to look at the various commercial and public domain packages and to write a report which can be distributed by AGOCG.
The second concerns the amount of on-line information available. I guess I am not unusual in often not looking at anyon-line information because there is too much and I am too busy. Filtering infonnation that is available on the network is at best, semi-manual. AGOCG would be interested to receive a proposal from anyone who would monitor the information available relevant to AGOCG interests to filter it and make key information available in a concise way.
Please reply to me by 15 July with an outline proposal for either of the above topics.
Other project proposals are welcome and wi11 be considered by AGOCG.
This is the report of an open meeting held at Loughborough University on 16 April 1994 to discuss the assistance available through initiatives to support the use of IT in graphic art and design in research and teaching in UK higher education. The meeting was organised by Dr Anne Mumford under the auspices of the Advisory Group on Computer Graphics (AGOCG). AGOCG is funded by the Joint Information Systems Committee (JISC) of the HEFCs. The meeting was attended by 18 participants.
Graphic art and design is an area of research and teaching which is, in general, a new addition to the subject areas which have been the major concern of the initiatives in the area of information services in UK Higher Education (HE). The majority of the work taking place is in the "new" universities and in specialist colleges in both HE and Further Education (FE).
There was a wide-ranging discussion of peoples' current activities and experience. This report summarises the major points made. The recommendations wi11 be taken to the AGOCG meeting in June. The points are, in no particular order:-
It was noted that the community involved both HE and FE institutions. JISC (and thus AGOCG) is primarily concerned with HE. The community is very variable with some art and design departments being small departments in large institutions and others dominating in small institutions. In small institutions it may be hard to acquire the skills necessary to introduce and support IT aspects and use of network tools. This may be easier in large institutions, but in these it is often the science/engineering based subjects that dominate resource provision and not the art and design departments. Thus, graphic art and design can suffer in both.
The requirements of art and design departments are different to those for science and engineering. 111is needs to be recognised in IT strategies in institutions. The differences include the following. There is a major bias towards Macs. Art and design departments would rather have high specification equipment capable of running industry standard software than lots of lower specification equipment. Art and design is not moving towards a lap-top for every student but towards shared highend equipment. There is also a difference in terms of the measured output of work - this may be a work of art and not an academic paper. Institutions - and assessors - may not consider this to be quite as serious as papers and this needs a culture change if resources are to be allocated to graphic art and design.
It was agreed that the CHEST style agreement and contacts with suppliers sounded as though they would overcome some problems experienced by sites. The inclusion of documentation copying and the development of associated training materials was felt to be beneficial. Action is to be taken to discover what software is needed by the graphic arts community.
Most sites did not know what other sites were doing. People tended to meet according to their particular craft interest.
This was seen as a major problem for many sites. Most sites are in the new sector. It is believed that it is intended to ensure connections to all HE sites by the end of the year.
A number of sites reported favourable experience with this means of accessing information across the network. It was felt to be an appropriate interface. It was proposed that a WWW server for this community could be set up through AGOCG.
It was noted that there was a general bias towards the use of Macs in many departments and in the industry. This was with the exception of the design tools where CAD packages tended to be used on PCs and unix workstations.
It was agreed that a discussion list would be useful. There is a need for people to be able to find others working in their subject domain. Mailbase might help this as might other network tools.
There is a need to show people how they can use the tools that are available. This could be done at a future AGOCG event for the community.
People should be encouraged to include training on network tools and services as part of their courses. Students are coming into HE with more IT skills than before. There was a feeling that creating and updating training materials was unrealistic given funding and changes in software. It was felt that what was required were study skills for both staff and students in how to use manuals supplied with software - give them the courage to open manuals!
It was agreed that we should make contacts with Design Librarians.
This is needed for shared output; stereo lithography was mentioned.
It was suggested that a series of case studies of what people are doing would be useful. This might cover tools being used and site policies.
There was interest in this area and in the area of video conferencing. There are clearly a number of people working in this area and it would be useful to share experience. It was noted that the Support Initiative for Multimedia Applications being funded by the JISC New Technologies Initiative is going to look at this area and to produce reports through AGOCG.
There is a need to continue to have available online catalogues of information. The information is visual as well as textual and image databases wi th supporting text are needed. There is a need for curatorial and co-ordinating effort in this area. This may link with JISC data sets activity.
There was interest in this and an agreed need for evaluations of the market. The Mac/PC crossover is important to the art and design community and tools needed to go across platforms.
AGOCG will be discussing this report at its next meeting in June and will be progressing the recommendations.
Stock building continues and a detailed list of material available has been added to the files attached to the AGOCG-resources list on mailbase. Comments and suggestions are welcome, particularly as regards the configuration of the information. We are always grateful for any information on new sources.
However, the amount of relevant material published is extremely small so we are now turning our attention to other areas such as courseware and promotional material. An appeal for information (not necessarily actual materials} went out to relevant lists on mailbase and contributions from G&V readers are actively sought. A template for reporting is available from AGOCG-resources on mailbase.
As network tools become increasingly pervasive we felt that some of the information we have accumulated could be made more conveniently available with gopher technology and accordingly we will be launching an experimental gopher service early in the summer term. Topics covered will be heavily influenced by any feedback we receive - especially in advance!
Finally, we have taken a deep breath and decided that there is no alternative but to conduct a survey in order to collect, in a structured way, the information we need to inform our forward planning. We realise that everyone hates surveys and this is the wrong time of year but we would be extremely grateful if you could spare a few minutes to let us have your thoughts: we will try to make it as painless as possible.
With the ever increasing performance and availability of price competitive PC platforms many traditional workstation applications are now migrating to the PC.
Microsoft's 32 bit windows based operating systems, Windows NT and Windows 4, provide the ideal development and delivery environments for workstation-class Graphics and Visualization applications.
To aid the migration of workstation applications to the PC platform, Template Graphics Software Inc (TGS) have produced a ful1y functional version of their renowned Phigs Plus implementation FIGARO+ for the Windows 32 environment.
A demonstration of Figaro+ for Windows NT is available via anonymous ftp from:- The European Microsoft Windows NT Academic Centre (EMWAC)
emwac.ed.ac.uk:pub/figarofigaront.zip
With the high volume and easy upgrade paths available on the PC platform many independent hardware vendors will soon be offering graphics accelerators capable of providing performance previously only available on expensive dedicated graphics workstations. By adhering to graphics standards such as Phigs Plus, developers and users will be able to take speedy advantage of these accelerators and bring powerful visualization tools to the everyday user's desktop.
The Advanced Interfaces Group at the University of Manchester has been commissioned by the Advisory Group on Computer Graphics (AGOCG) to conduct a survey of current activity in the field of Virtual Reality in the United Kingdom.
AGOCG is an initiative of the Joint Information Systems Committee of the Higher Education Funding Councils and the Research Councils.
The aim of the survey is to obtain an accurate profile of who is doing what in the field in the UK, including:
The results of the survey will contribute to AGOCG's advisory activity in this area, and will be widely disseminated, published as an AGOCG Technical Report in September 1994 and will be available to all UK academic and industrial VR workers. This is an excellent opportunity for workers in the field to find out exactly who's doing what and where.
To obtain a copy of the questionnaire, please contact Toby Howard at the address below, or you can ftp the PostScript directly from anonymous ftp site m1.cs.man.ac.uk, as /pub/toby/questionnaire.ps. Log in as user anonymous, giving your full email address as password.
We appreciate your cooperation, and we look forward to sharing the results of the survey with you.
During the Eurographics UK Conference in Oxford in March there was an AGOCG-sponsored demonstration of document exchange using SGML. This was tied in to the conference theme of graphics in publishing. SGML, the Standard Generalized Markup Language, is an International, IS 8879, and British Standard. It allows documents to be exchanged between different applications on different platforms.
The first demonstration showed SGML tags being inserted into text documents. A selection of papers from the Conference Proceedings were used as the example documents. Two different products, Intellitag and SGML Tagger, were used for the tagging. Those SGMLencoded documents were then displayed with the lADS viewer. The second demonstration of BASIS SGML server showed how SGMLencoded documents could be stored and manipulated in a database. Guided tours were organised throughout the conference and many delegates took the tour.
As the coordinator for the demonstration I learned quite a lot about SGML and the problems associated with its use. The standard seems to work satisfactorily in a controlled environment, such as a large corporation, where there are systems integrators on hand to set up the systems and specify Document Type Definitions (DTDs). It does not seem to be the kind of standard that an end user can pick up off-the-shelf and use successfully on their own. For successful document exchange there has to be quite a lot of agreement between the creator of the SGML document and the reader.
SGML does not specify what graphics formats are supported. This has to be done within the DTD. For the demonstration both CGM and BMP formats were used. The BMP graphics were in-line in the SGML file. The CGM graphics were in separate files pointed to via hypertext links.
The demonstration showed that SGML could be used to exchange documents. It took a lot of effort to set it up and the involvement of a SGML systems integrator to make it work, but it can be done successfully.
The original Uniras Training Materials for the interactive packages have been invaluable but new releases of Uniras have come and gone, methods of teaching have changed and the time had come for an update. The result five totally new books of which four are available now.
For each of Unigraph and Unimap, there is a new workbook, short enough for even the slowest typist to work through in less than an hour, and which gives just enough information to get the user started and drawing plots or maps with their own data. Companion books have been written called "Unigraph Solutions" and "Unimap Solutions" which describe briefly how to do those extra bits for which the manuals are not always crystal clear. Each of these two books has a very comprehensive index.
A new workbook is in preparation for Uniedit. No new books are planned for the Picture Manager or for the subroutine libraries.
All new books are stored as tar archives containing compressed PostScript files. The workbooks are stored with each page as a separate PostScript file so that, as new versions of Uniras come out, single pages can easily be updated.
The books are available by anonymous ftp from src.doc.ic.ac.uk and are in subdirectories of packages/uniras.
Colour PostScript as well as monochrome is available. See the README file for details.
The production of these new books has been funded by AGOCG.
As noted in issue number 34 of Graphics & Visualization, the European Microsoft Windows NT Academic Centre (EMW AC), operated by Edinburgh University Computing Services (EUCS), has a number of special projects associated with it. One of these is concerned with porting a Fortran application to Windows NT and providing a GUI front-end to it.
This project was completed in April this year. It was carried out by Francis van Millingen and Alex Nolan of the EUCS Graphics Team.
The report will be of interest to anyone who has a need to develop a GUI front-end to a Windows NT application (Fortran based or otherwise). It provides a step-by-step description of the main stages involved in building the graphical user interface. For users with Fortran applications, it also deals with how to transfer these.
The methodology employed utilises a Client-Server technique based on the use of Remote Procedure Call. This allows the GUI front-end and the application to be separated, with the GUI situated on the Client and the application on the Server. This approach enables the user to place the application and GUI on systems of appropriate power.
The project report is available via anonymous FTP from host emwac.ed.ac.uk in the directory emwac/workshops.
This report concludes the Computer Graphics Unit summer student project Geological applications of AVS, and is a summary of possible uses of the application builder Application Visualization System (AVS). The project aims to demonstrate the use of AVS in the geosciences, and its progress throughout the 10 weeks is documented elsewhere. This report summarises and exemplifies the findings.
Potential fields vary continuously In space and time, making them unusual amongst geological datasets, which are very often discretized (Pajon & Bui Tran, 1992). That is, they are usually spatially physically heterogenous. However, AVS, like most visualization packages, is best suited to data which are point samples of continuous functions, like potential fields. This is one problem with AVS which may need to be solved before it can make genuine contributions to geology.
The data are from Dr Bill Sowerbutts, in the Department of Geology, University of Manchester. He has provided gravity anomalies, magnetic anomalies and topography for the SK 100km grid square (Solway Firth area). Geometries are easily rendered from the data, and any dataset may be mapped on to any other.
Other techniques include:
These techniques are very useful to many branches of geology, particularly geophysical and mining research. Other potential fields to which the techniques may be simply applied include geoid anomalies and VLF perturbations.
Multibeam side-scan sonar is now capable of high resolution (less than 100m) deep (up to about 8km) sea-floor imagery, and results in large datasets. Sample data have been obtained from NOAA CD-ROMs, of a canyon in the Gulf of Mexico and a sea-mount in the Pacific, west of Hawaii.
As these data are visualized as topographic mesh geometries, techniques are similar to those which may be applied to potential field data. In particular, enhancing the topography by increasing the z-scale in real-time may be most useful.
It is possible to model displacement vectors in sedimentary basins, such as the soft, young, hard-rock-bounded basins seaward of the Western Cordillera of North America. This is of interest to earthquake scientists, as ground motion frequency and amplitude may coincide with the resonance character of buildings, in which case they fall down. The direction of ground motion polarization and the location of nodal (zerodisplacement) surfaces at depth are also significant.
There are sample data sets in ITTI/Vis2/data/basin, containing displacement vectors for one quarter of a half-ellipsoidal basin for four frequencies. The curved surface is the basal bedrock contact, the flat top is a free surface (the ground). The data set is in AVS ucd (unstructured cell data) format, and has been successfully visualized. Techniques include:
There are many areas in which AVS could be useful within the geosciences, but these are proving difficult to explore owing to the unavailability of viable datasets. Some applications that might benefit most are:
Seismic Profiling in 3D and 4D (3D variance over time). As oil becomes harder to find, industry invests more money in exploration, and 3D sub-surface imaging becomes a possibility. Processing these data is timeconsuming and technical1y troublesome, but AVS could aid interpretation, rendering surfaces and sections easily.
Basin Analysis and Sequence Stratigraphy are important branches of geology at the moment, because of their significance for the oil industry. Sedimentary basins are complex, large, 3D structures that vary over time (like al1 geological data, although the timescales are very long). AVS is wel1-suited to visualizing data like this, and it would be as wel1 to implement its use early as it remains a relatively young field. The real significance of 3D surfaces is becoming apparent, but the science lacks the visualization tools.
Phase Equilibria have always been complicated, and are becoming more so with the study of 9-component systems. Certainly, visualization of 4-component systems in 3-space would be useful, especial1y for teaching purposes. The system (e.g. Ca-Mg-Al-Si) is represented by a tetrahedron, which may be coloured, rotated and sliced in the geometry viewer. However, the elucidation of phase systems is an old science, and much of the data is unavailable in digitised form.
Deep Earth Tomography is a relatively new remote sensing techniques, used very effectively to model the mantle and lower mantle (up to about 3000km depth), and in particular to model convective flow ( e.g. King 1993) and mantle plumes. This is a matter of profound importance, as it may directly affect sea-level change, basin evolution and even mass extinctions. Convective flow model1ing may also have applications in the study of nearsurface processes such as magma chamber behaviour.
Remote Sensing in the more conventional sense (from aircraft and spacecraft) produces images for processing and interpretation. Al1 of the techniques described in the section on potential fields, and all of those for visualizing 20 images, are relevant here, as much potential field data is col1ected from above the ground.
Palaeontology might also benefit, as CAT scans of delicate or large fossils become more accessible. AVS makes it very easy to visualize 3D data such as these, as it can colour, rotate and slice volume data, and can even simulate X-ray images.
Looking further afield, there are avenues for research in management and exploitation of subsurface fluids, statistical geology and geography, and cartography (e.g. Brodlie et al. Eds. 1992). There is much potential for fully integrating AVS into the well-developed field of GIS (Geographic Information Systems, important examples of which are Idrisi and ARC/INFO), and the younger field of Geological Information Systems. Most notably, AVS offers a superior user interface compared to most other visualization or analytic applications, such as ARC/INFO.
Advanced Visual Systems Inc. (1992). AVS Technical Overview, AVS Module Reference, AVS Developer's Guide, AVS User's Guide. Waltham, Mass. Advanced Visual Systems Inc.
Earnshaw, R.A. & Wiseman, N. (1992). An Introductory Guide to Scientific Visualization. Berlin: Springer-Verlag.
Brodlie K. W. et al. (Editors, 1992). Scientific Visualization Techniques and Applications. Berlin: Springer-Verlag.
Jenkins, K.F. (1992). The response of coral reefs to glacio-eustatic sea-level cyclicity: an application of computer modelling. Unpub. PhD thesis, University of Manchester.
King, S.D. (1993). Seeing the mantle in the round. Nature 361, pp688-689.
Lamport, L. (1986). LaTeX: A document preparation system. Reading, Mass. Addison-Wesley.
Marschal1inger, R. (1991). Interface programs to enable full 3D geological modelling with a combination of AutoCAD and Surfer. Computers & Geoscience 17(8), pp1383-1394.
Pajon, J.L. & Bui Tran, V. (1992). Discrete data visualisation in geology. In: Advances in scientific visualisation (Ed. Post, F.H. & Hin, A.J .s.), Berlin: Springer-Verlag. 208pp, 22-35.
Reynolds, J.M. (1991). The need for recognised standards of applied geophysical software, Computers & Geoscience, 17(8), pp1099-1104.
Tackley, P.J., Stevenson, D.J., Glatzmaier, G.A. & Schubert, G. (1993). Effects of an endothermic phase transition at 670km depth in a spherical model of convection in the earth's mantle. Nature 361.
NCSA xmosaic was used to access information in the US and Manchester; sample images, some raw data and AVS modules were copied from directories at the fol1owing sites:
ftp://ftp.mcc.ac.uk/pub/cgu/avs/avs\_modules the Manchester ftp site, where there is a large module library http://oemg.er.usgs.gov/data - United States Geological Survey database gopher://rs5.1oc.gov:70/g10bal/sci/geo the Geogopher, containing information from al1 over the US; mostly climatology and oceanography.
Also, useful data were obtained from Dr Bill Sowerbutts in the Department of Geology at Manchester. Some of his data (the bathymetry datasets) were from the National Oceanic and Atmospheric Administration (NOAA) and the National Ocean Service (NOS).
We wish to record our thanks to the fol1owing who supplied us with data and valuable advice; Dr Bill Sowerbutts at Manchester for the potential field data, Dr George Helffrich at Bristol for the earthquake data, and Dr S. A. T. Redfern at Manchester for the crystallographic data.
Report on Special Session at the Institute of Physics Congress, 11-14 April 1994, Brighton Metropole, UK
Co-Chairs: Dr Rae A. Earnshaw and Prof John A. Vince
Scientific visualization tools and facilities are now making a major impact on many areas of the physical sciences. They offer facilities for analysing large and complex data sets, and pinpointing areas of significance in timevarying multivariate data. They enable supercomputers and desktop workstations and PCs to be coupled with applications and data sources in highly effective and useful ways. These tools are becoming increasingly user-friendly (always long overdue!) and many advanced facilities can now be utilised by the user without any programming expertise. Point and click menus enable the user's requirements to be selected with ease. More advanced requirements can be catered for by the user building an application using a visual language on the workstation screen. It is anticipated that more effective interaction devices (e.g. Spaceball, and virtual reality input) will extend the capability for the user to interact natural1y with multidimensional data sets for the purposes of exploration, analysis, presentation, and publication. The latter category includes not just traditional paper publication but new media such as electronic journals, slides, video, and multimedia databases and archives.
Dr Earnshaw began the session by outlining current trends in visualization facilities and the availability of software tools at a variety of levels. His recent article in Physics World on "Scientific Visualization: The State of the Art" (Sept 1993, pp 48-51) outlined current facilities ranging from packages and turnkey systems up to higher level application builders and volume visualization systems. The article also lists sources of further information and ftp addresses from which free publicdomain software can be obtained. A further article on What to look for in PC Graphics to appear in Physics World in May 1994, presents information on the rapid developments that have taken place in the PC area in graphics and multimedia facilities.
Judy Brown, Manager of Advanced Research Computing Services at the University of Iowa, and past Chair of SIGGRAPH, gave a presentation on Visualization and Scientific Applications. She outlined the developments in visualization tools and facilities in the USA. There is now better hardware and software, at lower cost, which has led to more widespread use of visualiza tion systems. Visualization is also now widely accepted as a research area in its own right.
Current data challenges include the fol1owing
Networks are bringing datafication to cities and also rural areas. This allows the possibility of doctors from remote sites to interactively access information across the network.
Information design challenges include:
Ethical challenges include:
It is become current practice in law suits in the USA for visualizations to be used for anything, and to prove any point that lawyers want to prove! By appropriate selection of display methods almost any aspects of any data set can be highlighted and the remainder suppressed. How meaningful are such visualizations? Clearly ethical boundaries in the use and display of data are being crossed. Visualization and visualization tools need more careful specification, as do the data sets on which they operate, otherwise visualizations will be used to prove anything.
Visualizations are also being used to support and promote learning methods:
Interactivity challenges in research and education include :
Prof John Vince, Hughes Rediffusion Simulation Ltd, outlined the use of visualization and virtual environments in the context of flight simulators. Very realistic 3D images can now be generated in simulators giving the pilot the full experience of immersion. Colour, texture and transparency are all used. Different lighting scenarios can be employed from morning to night, along with atmospheric effects such as fog, clouds, rain, and snow. Objects such as the terminal building, other aircraft, roads, traffic, trees can be modelled in the system, along with their shadows. All these features contribute to realistic scenes and enable a full immersive environment to be created that can be varied at will by the flight instructor who wishes to test out the pilot's capabilities in different circumstances and lighting conditions.
Prof Brian Wyvill, University of Calgary, gave a presentation on "Visualization of Natural Phenomena with Animation and Video" and used three examples-
In fractures of the heel 25% of current patients don't heal properly after the operation. Closer studies of the curvature of the surfaces is being done to ensure a good match to the surfaces after the operation. 3D MRI scan data is converted into implicit form and the two surfaces located and checked for congruity. Currently this cannot be done directly from the data. An algorithm has been devised to convert a set of contour surfaces from the data to identical surfaces generated from an implicit definition.
3D data values on a grid are the raw data from which isosurfaces are obtained. Implicit surfaces are being used to gain insight into the fields. A technique known as 'shrink-wrap' is being used to shrink the surface so that it adapts to the one the user is looking for. Octree data structures are used in which leaf nodes represent the pores and the solid parts of the rock. Graphical representations of the simulated rocks are produced as well as slices which faithfully match electron micrographs of real rocks. A neighbour finding algorithm is at the heart of the fluid flow visualization.
A model of the lightning strike is being used to illustrate how lightning operates. Visualization is being used to run this model for output as a video sequence. Particular attention is paid to simulating the appearance of lightning including the glow around the channel as well as the effect of the lightning as it strikes an object. This is achieved by using a scalar field from which both the glow and the model can dervied. A detailed paper presenting this work will be given at SIGGRAPH 94.
Dr Jeremy Walton, NAG Ltd, presented a paper on Using a Dataflow Toolkit to Display Scientific Data. He explained how toolkits like IRIS Explorer can be used to build up visualization application interactively from pre-existing modules which are selected and connected together by the user. The user then plugs in their data and produces the visualization., Several examples of IRIS Explorer's use for the display of results from molecular mechanics, computational fluid dynamics, geophysics, numerical simulation and other application areas were presented.
IRIS Explorer comes with about 200 modules (more are available) which perform tasks such as reading in data, filtering it, transforming it; creating graphical objects like line graphs, histograms, contours, surfaces, isosurfaces, volumes, vector plots, etc; and displaying them together in a window with full 3D interaction. A number of modules are built using Silicon Graphics' ImageVision library, and provide a large amount of image processing functionality. Many modules embody state-of-the-art algorithms published in recent international conferences such as SIGGRAPH and IEEE Visualization. Thus users can have confidence they are using the best tools to analyse their data sets.
Where the user has a requirement that is not covered by the set of existing modules, then a new module can be created using a tool called the Module Builder. This takes existing routines (in the form of C, C++ or FORTRAN source, or even as pure executablcs) and transforms them into modules for use from within IRIS Explorer. Another point-and-click tool called the DataScribe can be used to build modules to read or translate data. Over 200 userwritten modules which perform a variety of functions already exist; these are available from anonymous ftp sites. Finally, IRIS Explorer provides the application developer with the ability to customise the look and feel, of the appllcation before handing it over to an end-user.
Working in a modular environment such as a dataflow toolkit leads naturally to the requirement that it should be possible to distribute various parts of the application across different hosts in a computer network, This is a particular feature of IRIS Explorer and, in this context, it was mentioned that - following NAG's involvement in its development - the package was being made available on a wider variety of machine types such as Sun, IBM RS/6000, HP and DEC, together with the already-existing versions for Silicon Graphics and Cray.
Those wishing to find out more about this work arc invited to contact IRIS Explorer Center (Europe).
Walton, J.P.R.B., Get the picture new directions in data visualization, in Animation and Scientific Visualization (Watson, D. and Earnshaw, R.A, Eds), Academic Press, London, 1993, p. 29.
Walton, J.P.R.B., Now you see it interactive visualization of large datasets, in Applications of Supercomputers in Engineering III (Brebbia, CA. and Power, H., eds), Computational Mechanics Publications/Elsevier Applied Science, 1993, p. 139.
Walton, J.P.R.B., "Visualization of sphere packs using a dataflow toolkit", J. Molec. Graphics, 1994, (in press).
Dr Mikael Jern, Vice-President Technology, AVS/UNIRAS, presented Interactive Visualization for Large, Complex Data Sets. He outlined the next generation of visualization technology based on open systems and direct interaction paradigms. Current problems with dataflow models are as follows:
Current requirements of users are as follows:
As part of this move to more direct interaction a more effective interactive device is needed. Tests with the Spaceball (TM) have demonstrated the benefits in giving the user more degrees of freedom in which to explore their data sets.
Data sets from 50Mb to 100Mb have been interactively visualized using this approach. Videos can be produced by recording the session in real time.
The following are the principal benefits of this new approach:-
This prototype system is being further developed by a Consortium.
For further information contact Dr M. Jern.
The final presentation From Ruins to Reality by Dr Brian Collins, IBM UK Laboratories Ltd, Hursley, outlined an area where a project costing £200K had realised £4 million - by the use of visualization techniques. The project was concerned with the visualization of the restoration of a famous 17th century church in Dresden, the Frauenkirche, which was destroyed during the last war in February 1945.
Reconstruction of the church, in the style of the original, is planned and agreed, but will take around 10 years to complete and will cost an estimated £200 million. The purpose of the visualization was to present the church as it would look like re-built in order to assist with publicity on German TV and fund-raising activities.
500MB of data has been collected comprising solids, structures, and user-scanned images from surviving photographs and architectural drawings of the original church. The model was then constructed and rendered using 31 light sources. The sequence was produced in digital video from which HDTV, 01, and Umatic versions were made. The animation sequence shows a flight around the outside and inside of the church.
