McClure simulation language(ID:469/)TI digtal languageUnnamed (FORTRAN-like) simulation language for describing computer logic design devised by Robert McClure of Texas Instruments in 1964 Related languages
References: The author's language appears well thought-out and his paper is well written, but the present reviewer is left with a strong feeling of objection. This stems from the fact that the techniques described are well known, and solutions to the various problems obvious to all who seek them. Compilers, with their main properties apparently very similar to the properties of the compiler which the author plans to describe n a future paper, must be in almost daily use the world (or at least the USA) over, and the present paper presents little, if anything new. Basically, criticism is here being directed at the ACM referees, or at the editorial policy, that permits publication in the Journal of material that represents a description of just another language. The Communications would have been a more appropriate medium for publication, and there a short note might have sufficed to draw attention to the fact that the language exists, is presumably in use, and is -- hopefully -- available to other users of the CDC 1604. Since the paper has been published, some comments are in order. A compiler -- apparently in 1604 machine language, though this is not clearly stated -- has been written which permits the simulation of digital systems where these have been described in a special language formally defined by the present paper. The language also includes procedural statements that help define the characteristics of the required simulation run. The author mentions some characteristics of a first simulation, in which an object-machine-speed of one hundred clock-steps per second was achieved. This represents a simulation to real-time ratio of 5000, but was achieved only by suppressing all output, i.e., by doing little useful work. The ratio is about average for this type of activity, but surely only if some rerults can be obtained; and it would have been better to present a more meaningful figure. Some detailed criticism is also in order. It is not true to say that earlier "simulators were devised for a particular problem at hand", nor can the author claim, in contrast, to have provided a perfectly "general purpose tool". The present simulator is restricted to synchronous devices, though a simulation for asynchronous device is now planned. Moreover, all devices are manned ideal, and this is usually an assumption which may be justified only during early tests on the accuracy of logic. Exception may also be taken to the author's claim as to the "naturalness" of his language, a claim that is meaningless in a scientific context, and which implies a rather low opinion of his engineers' abilities. To sum up, one might say that this paper is typical of so much that appears in print nowadays, and which is rapidly turning the journals of "learned societies" into media for the general dissemination of progress reports, manuals and raw material for resume references. DOI in ACM Computing Reviews 6(03) May-June 1965 view details in [ACM] JACM 12(1) January 1965 view details in [ACM] JACM 12(1) January 1965 view details |