Chapter 7 Floating-Point Support
In 1987, the MIPS FPU set a new benchmark for microprocessor math performance in affordable workstations. Unlike the CPU, which was mostly a rather straightforward implementation relying on its basic architecture for its performance, the FPU was a heroic silicon design bristling with innovation and ingenuity.
Later on, the MIPS FPU was pulled onward by Silicon Graphics’s need for math performance that would once have been the preserve of supercomputers. The use of floating-point computations in embedded systems has grown fairly slowly; but the trend toward FP is a classic trade-off of clever hardware that makes software simpler and more maintainable, and that trade-off (over time) only goes one way. Some day ...
Become an O’Reilly member and get unlimited access to this title plus top books and audiobooks from O’Reilly and nearly 200 top publishers, thousands of courses curated by job role, 150+ live events each month,
and much more.
Read now
Unlock full access