Table of Contents
Advertisement
B.1.47
Evaluation of EXP2(x)
Floating-point evaluation of exp2(x) can be performed by representing x as i+f where f is the fractional part
and i is the greatest integer in x that does not exceed x. Then, exp2(i+f) = exp2(f) * (2 ** i). After extracting
the fractional part f, exp2(f) can be evaluated with a polynomial. By scaling by the integer part, exp2(x)
results. Various execution speeds and accuracies may be determined by using different order polynomi-
als.
page
org
polyc
dc
dc
dc
dc
dc
dc
dc
dc
dc
org
;
;
calculate d2=exp2(d0)
;
floor
fsub.x
fclr
int
do
fmpy.x
fadd.x
_log2sig
fscale.s d7,d2
B-116
132,60,1,1
x:0
-0.5770606e-03
0.2093549e-02
-.02777411e-02
0.3357901e-02
0.8940958e-02
0.5558203e-01
0.2402348e+00
0.6931450e+00
0.1000000e+01
p:$100
d0,d7
#polyc,r0
d7,d0
d2
x:(r0)+,d1.s
d7
#9,_log2sig
d2,d0,d2
d7.l,d7.h
d1,d2
x:(r0)+,d1.s
DSP96002 USER'S MANUAL
; ** 8
; ** 7
; ** 6
; ** 5
; ** 4
; ** 3
; ** 2
; ** 1
; ** 0
Totals:
Program
ICycles
Words
2
2
1
1
1
1
1
1
2
3
1
1
1
1
1
1
---
---
10
27
MOTOROLA
Advertisement
Table of Contents
