The GRFPU is an IEEE-754 compliant floating-point unit, supporting both single and double precision operands. The pipelined design combines high throughput with low latency, providing up to ... The ...

The GRFPU is an IEEE-754 compliant floating-point unit, supporting both single and double precision operands. The pipelined design combines high throughput with low latency, providing up to 200 MFLOPS ...

Also called a "floating point unit" (FPU), the math coprocessor may be a stand-alone chip or circuits built into the CPU. Floating point capability is very important to computation-intensive work ...