<s>
The	O
GeForce	B-Device
256	I-Device
is	O
the	O
original	O
release	O
in	O
Nvidia	O
's	O
"	O
GeForce	B-Application
"	O
product-line	O
.	O
</s>
<s>
Announced	O
on	O
August	O
31	O
,	O
1999	O
and	O
released	O
on	O
October	O
11	O
,	O
1999	O
,	O
the	O
GeForce	B-Device
256	I-Device
improves	O
on	O
its	O
predecessor	O
(	O
RIVA	B-Device
TNT2	I-Device
)	O
by	O
increasing	O
the	O
number	O
of	O
fixed	O
pixel	O
pipelines	O
,	O
offloading	O
host	O
geometry	O
calculations	O
to	O
a	O
hardware	B-Algorithm
transform	I-Algorithm
and	I-Algorithm
lighting	I-Algorithm
(	O
T&L	B-Algorithm
)	O
engine	O
,	O
and	O
adding	O
hardware	O
motion	B-Algorithm
compensation	I-Algorithm
for	O
MPEG-2	B-Algorithm
video	O
.	O
</s>
<s>
It	O
offered	O
a	O
notably	O
large	O
leap	O
in	O
3D	O
PC	O
gaming	O
performance	O
and	O
was	O
the	O
first	O
fully	O
Direct3D	O
7-compliant	O
3D	B-Architecture
accelerator	I-Architecture
.	O
</s>
<s>
The	O
chip	O
was	O
manufactured	O
by	O
TSMC	O
using	O
its	O
220	O
nm	O
CMOS	B-Device
process	B-Architecture
.	O
</s>
<s>
There	O
are	O
two	O
versions	O
of	O
the	O
GeForce	B-Device
256	I-Device
the	O
SDR	O
version	O
released	O
in	O
October	O
1999	O
and	O
the	O
DDR	O
version	O
released	O
in	O
mid-December	O
1999	O
each	O
with	O
a	O
different	O
type	O
of	O
SDRAM	O
memory	O
.	O
</s>
<s>
GeForce	B-Device
256	I-Device
was	O
marketed	O
as	O
"	O
the	O
world	O
's	O
first	O
'	O
GPU	B-Architecture
 '	O
,	O
or	O
Graphics	B-Architecture
Processing	I-Architecture
Unit	I-Architecture
"	O
,	O
a	O
term	O
Nvidia	O
defined	O
at	O
the	O
time	O
as	O
"	O
a	O
single-chip	O
processor	O
with	O
integrated	O
transform	O
,	O
lighting	O
,	O
triangle	O
setup/clipping	O
,	O
and	O
rendering	O
engines	O
that	O
is	O
capable	O
of	O
processing	O
a	O
minimum	O
of	O
10	O
million	O
polygons	O
per	O
second	O
"	O
.	O
</s>
<s>
The	O
"	O
256	O
"	O
in	O
its	O
name	O
stems	O
from	O
the	O
"	O
256-bit	O
QuadPipe	O
Rendering	O
Engine	O
"	O
,	O
a	O
term	O
describing	O
the	O
four	O
64-bit	O
pixel	O
pipelines	O
of	O
the	O
NV10	B-Device
chip	O
.	O
</s>
<s>
In	O
single-textured	O
games	O
NV10	B-Device
could	O
put	O
out	O
4pixels	O
per	O
cycle	O
,	O
while	O
a	O
two-textured	O
scenario	O
would	O
limit	O
this	O
to	O
2multitextured	O
pixels	O
per	O
cycle	O
,	O
as	O
the	O
chip	O
still	O
had	O
only	O
one	O
TMU	O
per	O
pipeline	O
,	O
just	O
as	O
TNT2	B-Device
.	O
</s>
<s>
In	O
terms	O
of	O
rendering	O
features	O
,	O
GeForce256	B-Device
also	O
added	O
support	O
for	O
cube	O
environment	O
mapping	O
and	O
dot-product	O
(	O
Dot3	O
)	O
bump	O
mapping	O
.	O
</s>
<s>
The	O
integration	O
of	O
the	O
transform	B-Algorithm
and	I-Algorithm
lighting	I-Algorithm
hardware	O
into	O
the	O
GPU	B-Architecture
itself	O
set	O
the	O
GeForce256	B-Device
apart	O
from	O
older	O
3D	B-Architecture
accelerators	I-Architecture
that	O
relied	O
on	O
the	O
CPU	O
to	O
perform	O
these	O
calculations	O
(	O
also	O
known	O
as	O
software	O
transform	B-Algorithm
and	I-Algorithm
lighting	I-Algorithm
)	O
.	O
</s>
<s>
This	O
reduction	O
of	O
3D	O
graphics	O
solution	O
complexity	O
brought	O
the	O
cost	O
of	O
such	O
hardware	O
to	O
a	O
new	O
low	O
and	O
made	O
it	O
accessible	O
to	O
cheap	O
consumer	O
graphics	B-Device
cards	I-Device
instead	O
of	O
being	O
limited	O
to	O
the	O
previous	O
expensive	O
professionally	O
oriented	O
niche	O
designed	O
for	O
computer-aided	B-Application
design	I-Application
(	O
CAD	B-Application
)	O
.	O
</s>
<s>
NV10	B-Device
's	O
T&L	B-Algorithm
engine	O
also	O
allowed	O
Nvidia	O
to	O
enter	O
the	O
CAD	B-Application
market	O
with	O
dedicated	O
cards	O
for	O
the	O
first	O
time	O
,	O
with	O
a	O
product	O
called	O
Quadro	B-Application
.	O
</s>
<s>
The	O
Quadro	B-Application
line	O
uses	O
the	O
same	O
silicon	O
chips	O
as	O
the	O
GeForce	B-Application
cards	O
,	O
but	O
has	O
different	O
driver	O
support	O
and	O
certifications	O
tailored	O
to	O
the	O
unique	O
requirements	O
of	O
CAD	B-Application
applications	O
.	O
</s>
<s>
Compared	O
to	O
previous	O
high-end	O
3D	O
game	O
accelerators	O
,	O
such	O
as	O
3dfx	B-Device
Voodoo3	I-Device
3500	I-Device
and	O
Nvidia	B-Device
RIVA	I-Device
TNT2	I-Device
Ultra	I-Device
,	O
GeForce	B-Application
provided	O
up	O
to	O
a	O
50%	O
or	O
greater	O
improvement	O
in	O
frame	O
rate	O
in	O
some	O
games	O
(	O
ones	O
specifically	O
written	O
to	O
take	O
advantage	O
of	O
the	O
hardware	B-Algorithm
T&L	I-Algorithm
)	O
when	O
coupled	O
with	O
a	O
very-low-budget	O
CPU	O
.	O
</s>
<s>
The	O
later	O
release	O
and	O
widespread	O
adoption	O
of	O
GeForce2	O
MX/4	O
MX	O
cards	O
with	O
the	O
same	O
feature	O
set	O
meant	O
unusually	O
long	O
support	O
for	O
the	O
GeForce256	B-Device
,	O
until	O
approximately	O
2006	O
,	O
in	O
games	O
such	O
as	O
Star	B-Application
Wars	I-Application
:	I-Application
Empire	I-Application
at	I-Application
War	I-Application
or	O
Half-Life	O
2	O
,	O
the	O
latter	O
of	O
which	O
featured	O
a	O
Direct3D7	O
path	O
,	O
targeting	O
the	O
fixed-function	O
pipeline	O
of	O
these	O
GPUs	B-Architecture
.	O
</s>
<s>
Without	O
broad	O
application	O
support	O
at	O
the	O
time	O
,	O
critics	O
pointed	O
out	O
that	O
the	O
T&L	B-Algorithm
technology	O
had	O
little	O
real-world	O
value	O
.	O
</s>
<s>
Initially	O
,	O
it	O
was	O
only	O
somewhat	O
beneficial	O
in	O
certain	O
situations	O
in	O
a	O
few	O
OpenGL-based	O
3D	O
first-person	O
shooters	O
,	O
most	O
notably	O
Quake	B-Application
III	I-Application
Arena	I-Application
.	O
</s>
<s>
Benchmarks	O
using	O
low-budget	O
CPUs	O
like	O
the	O
Celeron300A	O
would	O
give	O
favourable	O
results	O
for	O
the	O
GeForce256	B-Device
,	O
but	O
benchmarks	O
done	O
with	O
some	O
CPUs	O
such	O
as	O
the	O
PentiumII	O
300	O
would	O
give	O
better	O
results	O
with	O
some	O
older	O
graphics	B-Device
cards	I-Device
like	O
the	O
3dfx	O
Voodoo	O
2	O
.	O
</s>
<s>
3dfx	O
and	O
other	O
competing	O
graphics-card	O
companies	O
pointed	O
out	O
that	O
a	O
fast	O
CPU	O
could	O
more	O
than	O
make	O
up	O
for	O
the	O
lack	O
of	O
a	O
T&L	B-Algorithm
unit	O
.	O
</s>
<s>
Software	O
support	O
for	O
hardware	B-Algorithm
T&L	I-Algorithm
was	O
not	O
commonplace	O
until	O
several	O
years	O
after	O
the	O
release	O
of	O
the	O
first	O
GeForce	B-Application
.	O
</s>
<s>
Early	O
drivers	O
were	O
buggy	O
and	O
slow	O
,	O
while	O
3dfx	O
cards	O
enjoyed	O
efficient	O
,	O
high-speed	O
,	O
mature	O
Glide	B-Operating_System
API	I-Operating_System
and/or	O
MiniGL	B-Library
support	O
for	O
the	O
majority	O
of	O
games	O
.	O
</s>
<s>
Only	O
after	O
the	O
GeForce256	B-Device
was	O
replaced	O
by	O
the	O
GeForce	B-Application
2	O
,	O
and	O
ATI	O
's	O
T&	O
L-equipped	O
Radeon	B-Device
was	O
also	O
on	O
the	O
market	O
,	O
did	O
hardware	B-Algorithm
T&L	I-Algorithm
become	O
a	O
widely	O
utilized	O
feature	O
in	O
games	O
.	O
</s>
<s>
The	O
GeForce	B-Device
256	I-Device
was	O
also	O
quite	O
expensive	O
for	O
the	O
time	O
and	O
did	O
n't	O
offer	O
tangible	O
advantages	O
over	O
competitors	O
 '	O
products	O
outside	O
of	O
3D	B-Architecture
acceleration	I-Architecture
.	O
</s>
<s>
Additionally	O
,	O
some	O
GeForce	B-Application
cards	O
were	O
plagued	O
with	O
poor	O
analog	O
signal	O
circuitry	O
,	O
which	O
caused	O
display	O
output	O
to	O
be	O
blurry	O
.	O
</s>
<s>
As	O
CPUs	O
became	O
faster	O
,	O
the	O
GeForce256	B-Device
demonstrated	O
that	O
the	O
disadvantage	O
of	O
hardware	B-Algorithm
T&L	I-Algorithm
is	O
that	O
,	O
if	O
a	O
CPU	O
is	O
fast	O
enough	O
,	O
it	O
can	O
perform	O
T&L	B-Algorithm
functions	O
faster	O
than	O
the	O
GPU	B-Architecture
,	O
thus	O
making	O
the	O
GPU	B-Architecture
a	O
hindrance	O
to	O
rendering	O
performance	O
.	O
</s>
<s>
The	O
GeForce	B-Device
256	I-Device
introduced	O
motion	B-Algorithm
compensation	I-Algorithm
as	O
a	O
functional	O
unit	O
of	O
the	O
NV10	B-Device
chip	O
,	O
this	O
first-generation	O
unit	O
would	O
be	O
succeeded	O
by	O
Nvidia	O
's	O
HDVP	O
(	O
High-Definition	O
Video	O
Processor	O
)	O
in	O
GeForce2	O
GTS	O
.	O
</s>
<s>
NVIDIA	O
has	O
ceased	O
driver	O
support	O
for	O
the	O
GeForce	B-Device
256	I-Device
series	O
.	O
</s>
