107
click to vote
ARITH
2005
IEEE
15 years 8 months ago
2005
IEEE
We are proposing a micro-architecture for highperformance IEEE floating-point addition that is based on a (non-redundant)high-radix representation of the floatingpoint operands....