Modcomp FLOATING POINT numbers are represented
by one bit for the sign (S), 9 bits for the exponent, and 22 bits for the
fraction. Floating point numbers are in 2's complement. There
is no normalization.
Modcomp 32-Bit Floating Point Format | |||||||||||||||||||||||||||||||
S | Exponent | Fraction | |||||||||||||||||||||||||||||
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
DOUBLE PRECISION floating point numbers add another 32-bit word which is all fraction bits for added precision.
Negative floating point numbers are in two's complement.
INTEGER numbers are 16 bits long (not 32!), in two's
complement.