Coprocessor communications
In the past decade, various techniques and strategies have been introduced by a number of leading designers for controlling the operation of one or more Arithmetic Processing Units, coprocessors, in a system which share common address lines, data lines, and main memory with a central processor. Specifically, techniques and strategies as applied to some commercially available and some non-available Floating Point Processors which utilize either an 8-bit, 16-bit, or 32-bit architecture are reviewed. A comparative analysis of the performance of these devices is then presented to determine the more desirable features of each leading to a recommendation pertaining to the future design of communications systems for Arithmetic Processing Units.
Thesis85C336.pdf
7.1 MB
Unknown
a63119466f5f274c385a72e5ff00a65c