<s>
Block	B-Algorithm
floating	I-Algorithm
point	I-Algorithm
(	O
BFP	B-Algorithm
)	O
is	O
a	O
method	O
used	O
to	O
provide	O
an	O
arithmetic	O
approaching	O
floating	B-Algorithm
point	I-Algorithm
while	O
using	O
a	O
fixed-point	O
processor	O
.	O
</s>
<s>
BFP	B-Algorithm
assigns	O
a	O
group	O
of	O
significands	B-Algorithm
(	O
the	O
non-exponent	O
part	O
of	O
the	O
floating-point	B-Algorithm
number	I-Algorithm
)	O
to	O
a	O
single	O
exponent	O
,	O
rather	O
than	O
single	O
significand	B-Algorithm
being	O
assigned	O
its	O
own	O
exponent	O
.	O
</s>
<s>
BFP	B-Algorithm
can	O
be	O
advantageous	O
to	O
limit	O
space	O
use	O
in	O
hardware	O
to	O
perform	O
the	O
same	O
functions	O
as	O
floating-point	B-Algorithm
algorithms	O
,	O
by	O
reusing	O
the	O
exponent	O
;	O
some	O
operations	O
over	O
multiple	O
values	O
between	O
blocks	O
can	O
also	O
be	O
done	O
with	O
a	O
reduced	O
amount	O
of	O
computation	O
.	O
</s>
<s>
Block	B-Algorithm
floating-point	I-Algorithm
algorithms	I-Algorithm
were	O
extensively	O
studied	O
by	O
James	O
Hardy	O
Wilkinson	O
.	O
</s>
<s>
BFP	B-Algorithm
can	O
be	O
recreated	O
in	O
software	O
for	O
smaller	O
performance	O
gains	O
.	O
</s>
