27 Jan
1999
27 Jan
'99
6:53 a.m.
Dick Erlacher wrote:
The 9511 was the floating point unit, and the 9512 was
for 64-bit integer
arithmetic, I believe.
Close. They were both floating point. The AMD 9511/Intel 8231 used a
non-IEEE 32-bit format, and included transcendentals.
The AMD 9512/Intel 8232 used IEEE 64-bit format, and only had the four basic
arithmetic functions. Maybe it might have had square root, but definitely
no transcendentals. Like the 8087, it predated the adoption of the final
IEEE standard, so there may be minor differences, primarily in how infinities,
denormalized numbers, NANs, and rounding are handled.