OXFORD UNIVERSITY COMPUTING LABORATORY

ANNUAL REPORT 1962-3

  1. Computer Services

    The year under review is the second full year in which the Mercury machine has been "saturated", and for the first time the figures in the table indicate a distribution, rather than a considerable overall increase in the hours of Mercury use. The machine has continued to function satisfactorily, even though at least one-third of its operation has been "unmaintained", with no engineer in attendance.

    1962-31961-21960-1
    Clarendon Laboratory1655 1372430
    Chemical Crystallography1192983 650
    Other University departments1142946 530
    Computing Laboratory6021000710
    Programme testing (not itemised)578583 460
    Other universities49627990
    Firms514492670
    Miscellaneous293360
    Routine engineering822894760
    Faults402297190

    Totals743268794550

    The delay between producing a programme and obtaining results has inevitably increased. This could have been shortened by refusing paid work from industry and other universities, but the income so obtained ensures that the new KDF9 machine will have a reasonably large store, several magnetic tape units and a fast printer, and the new computing facilities, which should become available in the last Summer of 1964, will be a substantial improvement on the old. The work of other universities will decline rather sharply as the various new machines come into use.

    With the arrival of KDF9 the Computing Laboratory will move into part of the old Engineering Laboratory. This is therefore probably the last report from 9, South Parks Road, and provides an opportunity to congratulate and thank the University Surveyor for his work in converting an old house into a convenient and comfortable laboratory.

  2. Staff

    Mrs E Chin, B.Sc., University of West Indies, joined the laboratory in December 1962, and helps Dr C E Phelps in the organisation of the computer service.

    Mr H Wright, B.A. (Oxon), who recently studied at the School of Forestry at Yale University, become the second Ferranti Fellow in November 1962 and pursues research on the application of computing and computers to problems of analysis and management in Forestry.

  3. Research students, visitors, lectures

    Two research students, Mr K Wright (numerical solution of ordinary non-linear differential equations) and Mr L I Hodgson (numerical methods of crystal structure analysis) were granted the degree of D. Phil. Miss T Darwent, supervised in the laboratory by Mr J S Rollett, obtained the degree of M.Sc. at London University. Two new students Miss J Tothill (Imperial College) and Mr M Stern (Cambridge) made successful applications for D.S.I.R. grants for the academic year 1963-4.

    Professor J G Herriot, Stanford University, visited the laboratory and talked about "Algorithmic methods (ALGOL) in the solution of linear partial differential equations by Bergman's method of integral operators".

    The Director gave 28 lectures on Numerical Analysis for Mathematical Moderations and 16 for Mathematics Finals. Mr D F Mayers gave 24 lectures for Mathematics Finals and 8 in the School of Engineering Science. Mr J S Rollett also gave 4 lectures for a special paper in Engineering Finals.

  4. Summer School

    The 1962 Summer School had the title "Computing Methods in Crystallography", and was attended by about 50 representatives of government, industry and the academic world. The Directors of Studies was Mr J S Rollett, and eight other lecturers included Professor Dorothy Hodgkin and Dr C K Prout of the Oxford University Department of Chemical Crystallography, and experts from Pennsylvania State College, US, Naval Research Laboratory, Imperial College, Manchester College of Science and Technology, and the universities of Cambridge and Glasgow. The material is to be published in book form under the editorship of Mr Rollett.

  5. Other activities

    The Director spent a Sabbatical leave as Visiting Research Fellow at the Digital Computer Laboratory of the University of Illinois for the first semester of the academic year 1962-3, and was invited to give ten lectures on Numerical Analysis at the Argonne National Laboratory in April 1963. He became a consultant to the new Atlas Laboratory at Harwell and first chairman of the Users Committee for that laboratory. He was elected to the Faculty Board of the new Faculty of Mathematics, to the Delegacy of Linacre House, and in May 1963 to the Professorial Fellowship at Balliol.

    Mr J S Rollett looked after the affairs of the laboratory in the absence of the Director, and in particular worked out with the University Surveyor the details of the proposed move to the new premises. He visited the University of Goteborg, Sweden, to give lectures and advice to their Department of Medicinsk Biokemi, attended a conference of the X-ray analysis group of the Institute of Physics and the Physical Society, and acted as Chairman of the Scientific and Engineering Committee of the British Computer Society.

    Dr C E Phelps organised and gave all the lectures in the two Autocode courses which were held during the year and which had a total attendance of 200 from the University. He performs the major part of the work of lecturing and demonstrating the machine to a variety of visitors from schools, colleges and learned societies, is preparing ALGOL programming courses for use with the new KDF9 machine, and represents the laboratory at most of the KDF9 Users' meetings organised by the English Electric-Leo Company.

    Dr D C Handscomb completed, with Dr J Hammersley, the manuscript of a book on Monte Carlo Methods, and represented the laboratory at the statistical and data-processing specialised meetings of the KDF9 Users' Committee.

    Mr D F Mayers, Mr J S Rollett and the Director each gave two lectures in Numerical Analysis at a two-week "Refreshing course in mathematics" organised for school teachers by the Mathematics Faculty.

  6. Publications
  7. L. Fox Partial differential equations. Computer J. 6, 69-74 (1962)

    - Teaching and research in numerical analysis in universities. Proceedings, Summer School on Advanced Topics in Computer Sciences. University of North Carolina (1962).

    J S Rollett and L I Hodgson. An acceleration device and standard deviation estimates in least-squares refinements of crystal structures. Acta Cryst. 16, 329 (1963).

    - Automatic methods for heavy atom phase determination. Phys. Verhandl. Dtsch. 13, 117 (1962).

Dr J W Linnett (Chairman) Professor G E Blackman, Professor C A Coulson, Dr D ter Haar, Dr N T J Bailey, Mr E F Jackson, L Fox (Secretary)