The size of the group has temporarily increased. The usual members:
have been augmented by the arrival of David Ralphs, a Brunel University Sandwich Student who will be with us for most of the summer. In addition. Professor Judah Schwartz from the Education Research Centre at MIT arrived in mid-June and will be working with the group for 8 weeks this summer. He is hoping to produce one or more films on Quantum Mechanics during his stay.
The random vector generator for the display has still not arrived from DEC. This piece of hardware allows full screen vectors to be drawn on the display in any direction. Without it we are forced to draw lines in one of the eight basic directions. This means that in general lines have to be approximated by a set of short lines defined in these basic directions. Apart from the rather poor result on the display, it does mean that each line requires a large number of core locations for its display file. Consequently we are continually embarrassed by lack of storage.
The hardware faults on the display itself are still with us and it has been agreed that DEC will replace the hardware in August. We intend to move the PDP 15 to the new building near the end of July and once it is installed and working. we hope to have the display hardware replaced and the additional 16k of core storage and disc added. Due to the long delay in receiving the display hardware, DEC have agreed to give us an input tablet which it is hoped will be fitted at the same time.
After working reasonably well for the first six months, the PDP15 came up with a bad timing fault near the beginning of June. This took the engineers several weeks to locate. The fault was, unfortunately, temperature dependent and, due to the variability of the local air conditioner, it did make it difficult for them to get consistent results from tests. This has now been cured and work has got going again. It has however severely disrupted the work on the PDP15 during June.
The advance repeat fault on the SC4020 was finally solved in May after 10 months of incorrect working. It will be interesting to see whether this is a complete solution or just a temporary one. Since then little work has been done on the SC4020 due to a number of minor faults which have kept the machine out. The worst of these was a decision by Kodak to change its specifications of the hardcopy paper. The new paper requires a much higher light intensity to get a photographic image. After a week's work, the Datagraphix engineer has managed to get the hardcopy to a reasonable quality. To some extent this has been a blessing in disguise as the light intensity on the film output is much higher now than it has been for a long time.
The manuals for the D-MAC package have now arrived. Copies have been sent to all Universities. Internal staff interested in using the pencil follower can obtain manuals from the library.
We have been waiting for a device handler for the display keyboard from DEC ever since the display arrived. We have now implemented two handlers for the keyboard ourselves. The first, LUA., accepts input from the display keyboard and is very similar to the handler for the D-MAC. The second, LTA, is a multi-device handler which is derived from a handler we obtained from Delft. It will support both the display keyboard and also the KSR 33 teletype. Details are given in SPROGS Paper No 7.
Both the DEC BSI and ICL BSI for interfacing the two computers are built and being tested in the factory. Work has started on the specification of how the link will be driven and what telecommunications testing programs we shall need. Details of this will be given in SPROGS Paper No 6 which is still only in draft form at the moment.
After a lot of effort, we finally obtained a copy of the POLYGRAPHICS package from Michigan University. This is a FORTRAN package similar to GROATS and is one of the most frequently used packages in the USA. It is hoped to get this running on the 360/75 and have it available for selected users on the 360/195 next year.
FOCAL is an interpretive interactive language like JEAN and JOSS. It is available on the PDP 15 as part of the standard DEC software. Commands can be obeyed as soon as they are typed or they can be held in core and obeyed later. Facilities available include assignment statements involving any complexity of expression, looping, one-dimensional arrays and a crude subroutine facility. It is also possible to write machine-code algorithms which appear to FOCAL to be one-valued functions, and this facility is being used to hook on a simple display package.
We hope that being able to use the VT15 within a desk-calculator language will make it possible to rapidly produce demonstration programs and to experiment with picture producing algorithms.
Fortran is the only high level language that we use on the PDP15 and each member of the section has found it necessary to write some subroutines in MACRO, the assembler language, to remove inadequacies of the FORTRAN language, the PDP15 dialect or in some cases for efficiency.
The most useful of these are being gathered together into a documented library. FORTRAN routines and programs of general interest will also be added. So far we have routines for bit and character manipulation, input/output and storage allocation. The higher-level routines include some display debugging aids and a program for reformatting SC4020 tapes.
This program has been developed further and has now reached the stage whereby the paths objects are to follow can now be stored. The lower right-hand box remains to be implemented successfully.
|
SAVE |
ADVANCE FILM |
STORE |
|---|---|---|
|
THICKEN |
PICTURE SPACE |
NEWLINE |
|
SPATH |
FINISH (RESTART) |
TAKE * |
* Renamed since last report
Using the keyboard it is hoped to be able to enter keywords which will produce standard shapes such as squares, circles, etc.
When the Rand tablet arrives, input may come from this and the display will show a representation of the above diagram.
A modified sub-set of the above package is being used to produce SD4020 output from CAMP on the PDP15.
The section, with the help of Dianne Prince, is organising a symposium to be held in the Lecture Theatre at the Rutherford Laboratory on Friday, July 30 1971. We are hoping to get a large number of attendees from both the University's teaching and research staff and also the Film Industry. We have been lucky in getting John Halas, a world famous animator (Animal Farm for example), to open the symposium. He has been interested in computer animation from the start and is soon to publish a book on the subject. In addition we have managed to get most of the people in the UK who have made significant contributions to give papers. We were also careful to organise it while Professor Schwartz was with us so that he could talk of his own work.
The full programme is as follows:-
All the groundwork has been done and to date approximately 40 applications have been received. Several films have been promised from the USA and manufacturers of microfilm recorders will be represented among the attendees.
We are currently investigating a method whereby a black and white picture can be used to produce colour film by a video process. The aim is to make a black and white video recording from the VT15 display. The input will have a variety of intensity levels. It is possible then to process the video tape so that differing intensity levels give rise to different colours. A film print can then be obtained. from the colour enhanced video tape.
It is hoped to have this ready for the Symposium. However hardware faults on the PDP15 have delayed this experiment.
SD4020 faults have delayed progress in this area. A technique for producing colour films by chemical methods from black and white partially reversed film is currently under investigation.
A survey of potential successors to the SD4020 has commenced. A detailed questionnaire has been circulated to the manufacturers, some of whom have already replied. A series of test tapes will be generated and run on the machines selected. It is anticipated that a conclusion will be arrived at in the Autumn of this year. This will then afford the opportunity of seeing a Calcomp 1670 installed and working at the Meteorological Office before a final decision is reached.
The main piece of work here has been to transfer the Atlas version to the PDP15. Due to the rather restricted dialect of FORTRAN available on the PDP15, this has meant a large number of changes to the package. Also it has been necessary to break the package down into a number of chains so that it will fit into the PDP15 core. The system is nearly debugged now and it is hoped that it will be working within the next month. The space available for display storage is minimal at the moment. It is hoped that a useable system will be available once the additional core store is added to the PDP15. A manual for the CAMP system is also under preparation.
A draft outline of the SPROGS system has been defined (SPROGS Paper No 8). The aim has been to provide a system which can be used in an interactive mode through the PDP15 and also in a batch mode from the 1906A. The system borrows ideas from most of the existing graphics packages and cures the faults that we have noticed in our own systems. It is hoped to solidify the ideas put forward in this paper in the next few months.
Progress this quarter bas been very slow due mainly to hardware faults on the PDP15 and SC4020. A number of objectives which should have been realised this quarter will we hope be achieved in the next. The Animation Symposium will take up quite a large amount of our time in July and holidays could also seriously disrupt work in this quarter. Our main aim will be to finish existing programs being debugged on the PDP15 and get nearer to defining the SPROGS system.
A number of minor projects have been started. Work on what should be the final form of the Hash Table films has been started. The Nuffield Foundation are interested in us collaborating with them on a number of films. Jon Ogborn is interested in a film showing the form of the magnetic field as charge builds up in a capacitor. Neil Ryder has a number of two-colour films that he wishes to produce on Enzymes. He is interested in seeing whether we can produce these at a cost comparable to using conventional animation. We intend to cost out the first of these films in order to see if this is a sensible way to generate them. This will give us a realistic test for colour film generation. If successful, Ryder hopes to program the remaining films himself.
