<s>
The	O
F16C	B-Device
(	O
previously/informally	O
known	O
as	O
CVT16	O
)	O
instruction	B-General_Concept
set	I-General_Concept
is	O
an	O
x86	B-Operating_System
instruction	B-General_Concept
set	I-General_Concept
architecture	I-General_Concept
extension	O
which	O
provides	O
support	O
for	O
converting	O
between	O
half-precision	O
and	O
standard	O
IEEE	O
single-precision	O
floating-point	O
formats	O
.	O
</s>
<s>
The	O
CVT16	B-Device
instruction	I-Device
set	I-Device
,	O
announced	O
by	O
AMD	O
on	O
May	O
1	O
,	O
2009	O
,	O
is	O
an	O
extension	O
to	O
the	O
128-bit	O
SSE	B-General_Concept
core	O
instructions	O
in	O
the	O
x86	B-Operating_System
and	O
AMD64	B-Device
instruction	B-General_Concept
set	I-General_Concept
.	O
</s>
<s>
CVT16	O
is	O
a	O
revision	O
of	O
part	O
of	O
the	O
SSE5	B-General_Concept
instruction	B-General_Concept
set	I-General_Concept
proposal	O
announced	O
on	O
August	O
30	O
,	O
2007	O
,	O
which	O
is	O
supplemented	O
by	O
the	O
XOP	B-General_Concept
and	O
FMA4	B-General_Concept
instruction	I-General_Concept
sets	I-General_Concept
.	O
</s>
<s>
This	O
revision	O
makes	O
the	O
binary	O
coding	O
of	O
the	O
proposed	O
new	O
instructions	O
more	O
compatible	O
with	O
Intel	O
's	O
AVX	B-General_Concept
instruction	O
extensions	O
,	O
while	O
the	O
functionality	O
of	O
the	O
instructions	O
is	O
unchanged	O
.	O
</s>
<s>
In	O
recent	O
documents	O
,	O
the	O
name	O
F16C	B-Device
is	O
formally	O
used	O
in	O
both	O
Intel	O
and	O
AMD	O
x86-64	B-Device
architecture	O
specifications	O
.	O
</s>
<s>
There	O
are	O
variants	O
that	O
convert	O
four	O
floating-point	O
values	O
in	O
an	O
XMM	O
register	O
or	O
8	O
floating-point	O
values	O
in	O
a	O
YMM	B-General_Concept
register	I-General_Concept
.	O
</s>
<s>
VCVTPH2PS	O
ymmreg	O
,	O
xmmrm128	O
convert	O
eight	O
half-precision	O
floating	O
point	O
values	O
in	O
memory	O
or	O
an	O
XMM	O
register	O
(	O
the	O
bottom	O
half	O
of	O
a	O
YMM	B-General_Concept
register	I-General_Concept
)	O
to	O
eight	O
single-precision	O
floating-point	O
values	O
in	O
a	O
YMM	B-General_Concept
register	I-General_Concept
.	O
</s>
<s>
VCVTPS2PH	O
xmmrm128	O
,	O
ymmreg	O
,	O
imm8	O
convert	O
eight	O
single-precision	O
floating	O
point	O
values	O
in	O
a	O
YMM	B-General_Concept
register	I-General_Concept
to	O
half-precision	O
floating-point	O
values	O
in	O
memory	O
or	O
an	O
XMM	O
register	O
.	O
</s>
<s>
The	O
8-bit	O
immediate	O
argument	O
to	O
VCVTPS2PH	O
selects	O
the	O
rounding	B-Algorithm
mode	O
.	O
</s>
