Page 136 - Neutrons for Sciences and Society

Neutrons for Science
scattering, time of flight, mass spectrometer). The GfK, Karlsruhe, provided the expertise for the realization. At the ILL, Michel Grevaz (1970) took charge assisted by Helga Schwab, Jean- Jacques Tschofen and Günter Anderlohr. Two of the nuclear physics experiments each benefited from using a PDP11 mini- computer under the responsability of Philippe Ledebt assisted by Philippe Blanchard.
This support of national industry soon proved to be a handicap. Using time-sharing on slow systems like CARINE and NICOLE led to evident absurdities. For example, on CARINE each command sent was followed by a wait often exceeding 5 seconds. While not too debilitating for a slow instrument like a triple-
axis this was catastrophic for others performing rapid scans, and also greatly limited graphical applications. For example for the four-circle diffractometer D8 this resulted in a loss of efficiency of more than 50% (50% measuring time, 50% awaiting the computers response); it amounted to losing more than half the neutron flux available. This untenable situation had to await the arrival of Mössbauer to abandon these national preferences and install one computer per instrument (PDP11 or SEMS Solar85) which resulted in a substantial gain in efficiency for most experiments.
It was, of course, also necessary to have access to more powerful means to treat the data. At the start, until 1973, this was possible using a remote job-entry terminal (RJE, terminal lourd) connected
85 These French computers were the last bought to satisfy the national preferences. While fast they lacked soft- ware. They had to be replaced fairly quickly (though not fast enough for the users) by PDP11 systems.
127