Search
AGOCG had its fourth meeting on 27 June, we have now been in operation nearly a year. The meeting this time was in Manchester so that we could see the state of progress on the GKS-3D implementation. Progress is quite good although a Beta test version will not be available until September. Terry Hewitt showed us quite an amusing bug they have to find in the Viewing Pipeline, which looked more like modern art than computer graphics! Negotiations with BTG on the release of the GKS-UK source to HEIs is proceeding and we are hopeful that some arrangement can be concluded in the near future.
We have begun to put together some criteria for evaluating PHIGS implementations with the intention of producing a Report of those on the market. Once we have established the criteria we are likely to be looking for volunteers to help with that. If anybody has been using a particular implementation and would like to be involved we would be interested in hearing from !hem (either to me or Anne Mumford). We are beginning to get some training material together that we plan to make available. Toby Howard at Manchester has produced a PHIGS Introductory Guide and we are putting together two Viewgraph sets (one introductory and one more advanced). Both should be available some time this year in at least Beta test form. We will eventually have this available in an SGML/CGM format to allow people to incorporate their own material but probably not in the initial release. Progress on defining the Document Type Definitions has been delayed while we awaited the DAPHNE software. This is now here so progress should speed up on the SGML front.
At the meeting we finished off one or two pieces of work that have been worked on during the year. The Graphics Operational Requirement information that we wish to see in relevant purchases in the academic community was agreed. We have a good definition of the responsibilities of AGOCG relative to IGWP and CHEST so that we do not tread on each other's toes! The colour printer assessment is complete and should be available at about the time of this Newsletter. The training aids for UNIRAS Version 6 are ready for distribution and you should read Anne Mumford's article for more details of that. It was gratifying to see some items completed and coming off the Agenda so that we can move on to new areas.
We had our first look at what is needed in the Visualisation area and are putting together a two year plan to go to the Computer Board and SERC for approval of other activities.
It has been a busy year with some good deliverables. Anne Mumford has been invaluable as Coordinator. Having a full time person available to AGOCG has meant that good progress has been made in between meetings.
I am writing this article at the end of June, confident that the training materials will be with people by the end of August. The workshop at the start of this month gathered together 17 people who worked on the draft materials prepared primarily by Norman Wiseman of NERC and Bob McGonnigle and his colleagues at Edinburgh. The documents were discussed and tested and further material written. The results of this will be a range of materials that are useful to support staff and users of the Uniras software.
This article describes the nature of the materials which are being prepared and how you might use them at your site. The set will include:
Other documents on conversion from Ghost and Gino are also included together with a program for the Gino conversion. The program is to be included on a disk together with any data sets and command files used in the training materials.
The materials will be available towards the end of August and order forms will be sent out. A master set can be ordered. This set includes all the materials described above as single sheet masters as well as examples of made up workbooks, student notes and factscards. There are a limited number of master sets being made up. Further sets of made of documents will be available from me. The only online material at this stage will be a set of UNIPICT files for the OHP set. It is intended to consider moving the materials to an SGML markup as an experiment in the future. At this stage the documents are on good old fashioned media!
I am sure that these materials will be useful at sites with the software. If you have not received an order form and you would like one then please mail me with your postal address.
Precision Visuals' Workstation Analysis and Visualisation Environment (PV - WAVE) is currently under investigation by the IGWP and this is a preliminary report prior to a full on-site evaluation.
It is a data analysis and display system which currently runs on Dec VMS and SUN Unix workstations and central VAX machines linked to Tektronix 40xx, 41xx, 42xx and emulating terminals.
Data analysis functions include:
Graphics Display functions include:
There appear to be no facilities for 3D modelling although the manipulation of 3D surfaces and Hidden Line Hidden Surface removal functions are able to take advantage of any special purpose graphics hardware in display devices.
A major drawback is the absence of any hardcopy output other than by screen dump or Postscript file. This limits the size of output to either A4 or A3 at present, on fairly expensive plotting devices.
Further investigation of the package appears to be merited, especially to provide a comparison with the Uniras package. At the present time there are no plans to offer PV-Wave under any kind of CHEST deal.
The process of revising GKS is now well under way. There is general agreement that a revision is necessary, but there is continuing conflict over the extent of the changes. The UK, together with Netherlands and Italy, favour a fairly radical approach so that GKS can be significantly improved and be an attractive product for years to come. Other countries, notably the US, take a very conservative line wanting as few changes to the document as possible. All agree that it is important to preserve compatibility for existing GKS programs.
The UK put forward a complete proposal, known as GKS-N, to ISO last year. This was rejected, not on technical grounds, but on the perceived extent of the changes. GKS-N is built on the following model: a program creates a picture in NDC space; this picture acts as a database of primitives from which a workstation can select those primitives it wishes to view. Pictures are created either by adding primitives directly, or by adding previously prepared picture parts. GKS-N also includes more precisely defined geometric primitives called shapes.
A major advantage of the GKS-N system, and its concept of an NDC picture, is that interworking with the Computer Graphics Metafile (CGM) standard is well-defined: one simply copies the current NDC picture to a CGM.
The document style also received attention. Use of abstract datatypes allows related GKS functions to be grouped into a single GKS-N function. The overall effect is a much simpler, more concise document, less than half the size of the GKS document.
Of necessity new concepts are introduced. But this is done in such a way that the old GKS concepts, such as segments, can be expressed in terms of the new. Thus it is possible to describe the old GKS in terms of GKS-N, and to show that compatibility for existing GKS application programs can be preserved. For example, a walk through of the GKS validation suite suggested that a GKS-N implementation could satisfy virtually all the tests.
However, such a solution was too big a leap for the majority of nations. Hence work at the ISO level has begun on a more modest tack. At a recent meeting in Paris, a number of changes were agreed:
These changes are being made in an incremental manner so as to minimise changes to the document - rather than starting again as was done with GKS-N. However there is likely to be great difficulty in keeping consistency.
The UK still favours the approach in GKS-N as a proper solution to the GKS Revision issue, but is committed to working within ISO. It was never the UK intention to present GKS-N as a package which had to be accepted in its entirety, but rather as a clean starting point for evolution towards the next GKS.
A new draft document for the revised GKS is expected very shortly, and will be discussed at an ISO meeting in the UK this July. The UK will participate in the review of this document, seeking some middle ground between the ISO direction and GKS-N, which can act as a basis for future progress towards a new GKS standard.
Anyone wishing to learn more about the GKS revision process is welcome to contact me - and any comments on failings in the existing GKS would also be appreciated.
