Object-oriented languages such as C++ allow class definitions which eliminate the need for writing special-purpose simulators in many cases. Simulating digital logic components and computer architectures is one case where the simulation model can be effectively and conveniently expressed in the programming language itself.

No abstract available.

Students are introduced to the object-oriented paradigm at the beginning of this implementation of CS 1.A number of window objects are available and are incorporated in the first laboratory assignments.These window objects communicate through the standmd input and output mechanisms of the operating system.These operating system mechanisms allow piping among objects and the simultaneous execution of multiple objects.Some assignments are threaded throughout the term to become sizable culminations, Although the emphasis is primarily on the objects involved in the solution designs, there is some amount of algorithm development.The window objects used provide some exposure to graphical user interfaces and scientific visualization, and give visuat feedback on the performance of the algorithms.

A workstation laboratory allows undergraduate students to build complete, functioning computers--in simulation. The simulation extends down to the gate and signal levels, with effective modeling of delays and transitions, so reasonable assurance of the validity of the designs can be obtained. The computers constructed in this laboratory are complete with peripheral equipment including tapes and disks, and the students furnish a rudimentary operating system. The student operating-systems receive executable binary objects on the tapes, transfer them to the disk, and initiate their execution. The executable objects and the operating system are produced by a standard C compiler. Students must limit their C-language source code and data types used so the compile binaries will be executable on the 16-bit wide, Motorola 680X0-like computer they construct. The computers are constructed in an hierarchical manner from the gates and medium-scale components available in the digital simulator used in this laboratory.

This paper describes the tools and techniques used to allow students to produce a compiler in a ten-week term.Individualized subsets of ANSI-C source code are compiled to Motorola 68020 assembly language.The source-language specifications consist of eleven language features common to alI students, plus ten individuallyprescribed features.The individuahzations am derived from student numbers, and ensure a reasonable degree of independent work.Students can aid each other, and are asked to give acknowledgements in their code to others who have aided them.Completed compilers are judged using a set of programs that invoke common language features only, plus an appropriate set containing individualized-language features.

Correlation matrix memories are formed iteratively given a single cue-response pair at a time to be stored in an associative anner. If the cues presented are not orthogonal, the memory provides imperfect recollections after a single training exposure; however, if the cues are linearly independent, repetitions of the exposures can lead to ultimate perfect memorization and recollection ability.