<s>
The	O
distributed-computing	O
project	O
Folding	B-Application
@home	I-Application
uses	O
scientific	O
computer	O
programs	O
,	O
referred	O
to	O
as	O
"	O
cores	O
"	O
or	O
"	O
fahcores	O
"	O
,	O
to	O
perform	O
calculations	O
.	O
</s>
<s>
home	O
's	O
cores	O
are	O
based	O
on	O
modified	O
and	O
optimized	B-Application
versions	O
of	O
molecular	O
simulation	O
programs	O
for	O
calculation	O
,	O
including	O
TINKER	B-Application
,	O
GROMACS	B-Application
,	O
AMBER	B-Device
,	O
CPMD	B-Application
,	O
SHARPEN	B-Algorithm
,	O
ProtoMol	O
and	O
Desmond	B-Application
.	O
</s>
<s>
Available	O
for	O
Windows	O
,	O
Linux	O
,	O
and	O
macOS	O
,	O
use	O
Advanced	B-General_Concept
Vector	I-General_Concept
Extensions	I-General_Concept
if	O
available	O
,	O
for	O
a	O
significant	O
speed	O
improvement	O
.	O
</s>
<s>
Cores	O
for	O
the	O
Graphics	B-Architecture
Processing	I-Architecture
Unit	I-Architecture
use	O
the	O
graphics	O
chip	O
of	O
modern	O
video	O
cards	O
to	O
do	O
molecular	O
dynamics	O
.	O
</s>
<s>
The	O
GPU	B-Architecture
Gromacs	B-Application
core	O
is	O
not	O
a	O
true	O
port	O
of	O
Gromacs	B-Application
,	O
but	O
rather	O
key	O
elements	O
from	O
Gromacs	B-Application
were	O
taken	O
and	O
enhanced	O
for	O
GPU	B-Architecture
capabilities	O
.	O
</s>
<s>
These	O
are	O
the	O
third	O
generation	O
GPU	B-Architecture
cores	O
,	O
and	O
are	O
based	O
on	O
,	O
Pande	O
Group	O
's	O
own	O
open	O
library	O
for	O
molecular	O
simulation	O
.	O
</s>
<s>
v0.0.18	O
Available	O
to	O
Windows	O
and	O
Linux	O
for	O
AMD	O
and	O
NVIDIA	O
GPUs	B-Architecture
using	O
OpenCL	B-Application
and	O
CUDA	B-Architecture
,	O
if	O
available	O
.	O
</s>
<s>
v0.0.20	O
Available	O
to	O
Windows	O
and	O
Linux	O
for	O
AMD	O
and	O
NVIDIA	O
GPUs	B-Architecture
using	O
OpenCL	B-Application
and	O
CUDA	B-Architecture
,	O
if	O
available	O
.	O
</s>
<s>
TINKER	B-Application
is	O
a	O
computer	O
software	O
application	O
for	O
molecular	O
dynamics	O
simulation	O
with	O
a	O
complete	O
and	O
general	O
package	O
for	O
molecular	O
mechanics	O
and	O
molecular	O
dynamics	O
,	O
with	O
some	O
special	O
features	O
for	O
biopolymers	O
.	O
</s>
<s>
An	O
unoptimized	O
uniprocessor	B-Operating_System
core	O
,	O
this	O
was	O
officially	O
retired	O
as	O
the	O
AMBER	B-Device
and	O
Gromacs	B-Application
cores	O
perform	O
the	O
same	O
tasks	O
much	O
faster	O
.	O
</s>
<s>
Available	O
for	O
ATI	O
series	B-Device
1xxx	I-Device
GPUs	B-Architecture
running	O
under	O
Windows	O
.	O
</s>
<s>
Although	O
mostly	O
Gromacs	B-Application
based	O
,	O
parts	O
of	O
the	O
core	O
were	O
rewritten	O
.	O
</s>
<s>
This	O
core	O
was	O
retired	O
as	O
of	O
June	O
6	O
,	O
2008	O
due	O
to	O
a	O
move	O
to	O
the	O
second	O
generation	O
of	O
the	O
GPU	B-Architecture
clients	O
.	O
</s>
<s>
Available	O
for	O
Windows	O
x86	B-Operating_System
,	O
Mac	O
x86	B-Operating_System
,	O
and	O
Linux	O
x86/64	B-Device
clients	O
,	O
this	O
was	O
the	O
first	O
generation	O
of	O
the	O
SMP	B-Operating_System
variant	O
,	O
and	O
used	O
MPI	B-Application
for	O
Inter-process	B-Operating_System
communication	I-Operating_System
.	O
</s>
<s>
Available	O
only	O
to	O
x86	B-Operating_System
Macs	O
and	O
x86/64	B-Device
Linux	O
,	O
this	O
core	O
is	O
very	O
similar	O
to	O
Core	O
a1	O
,	O
as	O
it	O
uses	O
much	O
of	O
the	O
same	O
core	O
base	O
,	O
including	O
use	O
of	O
MPI	B-Application
.	O
</s>
<s>
However	O
,	O
this	O
core	O
utilizes	O
more	O
recent	O
Gromacs	B-Application
code	O
,	O
and	O
supports	O
more	O
features	O
such	O
as	O
extra-large	O
work	O
units	O
.	O
</s>
<s>
Also	O
known	O
as	O
the	O
SCEARD	O
core	O
,	O
this	O
variant	O
was	O
for	O
the	O
PlayStation	B-Operating_System
3	I-Operating_System
game	O
system	O
,	O
which	O
supported	O
a	O
Folding	B-Application
@Home	I-Application
client	O
until	O
it	O
was	O
retired	O
in	O
November	O
2012	O
.	O
</s>
<s>
This	O
core	O
performed	O
implicit	O
solvation	O
calculations	O
like	O
the	O
GPU	B-Architecture
cores	O
,	O
but	O
was	O
also	O
capable	O
of	O
running	O
explicit	O
solvent	O
calculations	O
like	O
the	O
CPU	O
cores	O
,	O
and	O
took	O
the	O
middle	O
ground	O
between	O
the	O
inflexible	O
high-speed	O
GPU	B-Architecture
cores	O
and	O
flexible	O
low-speed	O
CPU	O
cores	O
.	O
</s>
<s>
This	O
is	O
the	O
original	O
Gromacs	B-Application
core	O
,	O
and	O
is	O
currently	O
available	O
for	O
uniprocessor	B-Operating_System
clients	O
only	O
,	O
supporting	O
Windows	O
,	O
Linux	O
,	O
and	O
macOS	O
.	O
</s>
<s>
Available	O
to	O
Windows	O
,	O
Linux	O
,	O
and	O
macOS	O
uniprocessor	B-Operating_System
clients	O
only	O
,	O
this	O
core	O
uses	O
the	O
Gromacs	B-Application
3.3	O
codebase	B-General_Concept
,	O
which	O
allowing	O
a	O
broader	O
range	O
of	O
simulations	O
to	O
be	O
run	O
.	O
</s>
<s>
This	O
core	O
uses	O
the	O
Serial	O
Replica	B-Algorithm
Exchange	I-Algorithm
Method	O
,	O
which	O
is	O
also	O
known	O
as	O
REMD	B-Algorithm
(	O
Replica	B-Algorithm
Exchange	I-Algorithm
Molecular	O
Dynamics	O
)	O
or	O
GroST	O
(	O
Gromacs	B-Application
Serial	O
replica	B-Algorithm
exchange	I-Algorithm
with	O
Temperatures	O
)	O
in	O
its	O
simulations	O
,	O
and	O
is	O
available	O
for	O
Windows	O
and	O
Linux	O
uniprocessor	B-Operating_System
clients	O
only	O
.	O
</s>
<s>
This	O
may	O
allow	O
Folding	B-Application
@home	I-Application
to	O
more	O
efficiently	O
sample	O
the	O
transitions	O
between	O
folded	O
and	O
unfolded	O
conformations	O
of	O
proteins	O
.	O
</s>
<s>
Available	O
for	O
Windows	O
and	O
Linux	O
uniprocessor	B-Operating_System
clients	O
only	O
.	O
</s>
<s>
Available	O
for	O
uniprocessor	B-Operating_System
clients	O
,	O
this	O
core	O
uses	O
SSE2	B-General_Concept
processor	O
optimization	O
where	O
supported	O
and	O
is	O
capable	O
of	O
running	O
on	O
Windows	O
,	O
Linux	O
,	O
and	O
macOS	O
.	O
</s>
<s>
Very	O
similar	O
to	O
Core	O
79	O
,	O
and	O
initially	O
released	O
for	O
Linux	O
and	O
Windows	O
in	O
April	O
2008	O
for	O
Windows	O
,	O
Linux	O
,	O
and	O
macOS	O
uniprocessor	B-Operating_System
clients	O
.	O
</s>
<s>
Available	O
solely	O
for	O
all	O
uniprocessor	B-Operating_System
clients	O
on	O
Windows	O
,	O
Linux	O
,	O
and	O
macOS	O
.	O
</s>
<s>
Available	O
for	O
Windows	O
,	O
Linux	O
,	O
and	O
macOS	O
,	O
this	O
core	O
was	O
originally	O
released	O
in	O
early	O
October	O
2010	O
,	O
and	O
as	O
of	O
February	O
2010	O
uses	O
the	O
latest	O
version	O
of	O
Gromacs	B-Application
,	O
v4.5.3	O
.	O
</s>
<s>
The	O
next	O
generation	O
of	O
the	O
SMP	B-Operating_System
cores	O
,	O
this	O
core	O
uses	O
threads	O
instead	O
of	O
MPI	B-Application
for	O
inter-process	B-Operating_System
communication	I-Operating_System
,	O
and	O
is	O
available	O
for	O
Windows	O
,	O
Linux	O
,	O
and	O
macOS	O
.	O
</s>
<s>
Short	O
for	O
Car	B-Application
–	I-Application
Parrinello	I-Application
Molecular	I-Application
Dynamics	I-Application
,	O
this	O
core	O
performs	O
ab-initio	O
quantum	O
mechanical	O
molecular	O
dynamics	O
.	O
</s>
<s>
Unlike	O
classical	O
molecular	O
dynamics	O
calculations	O
which	O
use	O
a	O
force	O
field	O
approach	O
,	O
CPMD	B-Application
includes	O
the	O
motion	O
of	O
electrons	O
in	O
the	O
calculations	O
of	O
energy	O
,	O
forces	O
and	O
motion	O
.	O
</s>
<s>
This	O
is	O
a	O
double-precision	O
variant	O
for	O
Windows	O
and	O
Linux	O
uniprocessor	B-Operating_System
clients	O
.	O
</s>
<s>
SSE2	B-General_Concept
optimization	O
on	O
Intel	O
CPUs	O
is	O
supported	O
.	O
</s>
<s>
Due	O
to	O
licensing	O
issues	O
involving	O
Intel	O
libraries	O
and	O
SSE2	B-General_Concept
,	O
QMD	O
Work	O
Units	O
were	O
not	O
assigned	O
to	O
AMD	O
CPUs	O
.	O
</s>
<s>
In	O
early	O
2010	O
Vijay	O
Pande	O
said	O
"	O
We	O
've	O
put	O
SHARPEN	B-Algorithm
on	O
hold	O
for	O
now	O
.	O
</s>
<s>
This	O
core	O
uses	O
different	O
format	O
to	O
standard	O
F	B-Application
@H	I-Application
cores	I-Application
,	O
in	O
that	O
there	O
is	O
more	O
than	O
one	O
"	O
Work	O
Unit	O
"	O
(	O
using	O
the	O
normal	O
definition	O
)	O
in	O
each	O
work	O
packet	O
sent	O
to	O
clients	O
.	O
</s>
<s>
The	O
software	O
for	O
this	O
core	O
was	O
developed	O
at	O
D	B-Application
.	I-Application
E	I-Application
.	I-Application
Shaw	I-Application
Research	I-Application
.	O
</s>
<s>
Desmond	B-Application
performs	O
high-speed	O
molecular	O
dynamics	O
simulations	O
of	O
biological	O
systems	O
on	O
conventional	O
computer	O
clusters	O
.	O
</s>
<s>
Desmond	B-Application
and	O
its	O
source	O
code	O
are	O
available	O
without	O
cost	O
for	O
non-commercial	O
use	O
by	O
universities	O
and	O
other	O
not-for-profit	O
research	O
institutions	O
.	O
</s>
<s>
Possible	O
available	O
for	O
Windows	O
x86	B-Operating_System
and	O
Linux	O
x86/64	B-Device
,	O
this	O
core	O
is	O
currently	O
in	O
development	O
.	O
</s>
<s>
Short	O
for	O
Assisted	O
Model	O
Building	O
with	O
Energy	O
Refinement	O
,	O
AMBER	B-Device
is	O
a	O
family	O
of	O
force	O
fields	O
for	O
molecular	O
dynamics	O
,	O
as	O
well	O
as	O
the	O
name	O
for	O
the	O
software	O
package	O
that	O
simulates	O
these	O
force	O
fields	O
.	O
</s>
<s>
AMBER	B-Device
was	O
originally	O
developed	O
by	O
Peter	O
Kollman	O
at	O
the	O
University	O
of	O
California	O
,	O
San	O
Francisco	O
,	O
and	O
is	O
currently	O
maintained	O
by	O
professors	O
at	O
various	O
universities	O
.	O
</s>
<s>
The	O
double-precision	O
AMBER	B-Device
core	O
is	O
not	O
currently	O
optimized	B-Application
with	O
SSE	O
nor	O
SSE2	B-General_Concept
,	O
</s>
<s>
but	O
AMBER	B-Device
is	O
significantly	O
faster	O
than	O
Tinker	B-Application
cores	O
and	O
adds	O
some	O
functionality	O
which	O
cannot	O
be	O
performed	O
using	O
Gromacs	B-Application
cores	O
.	O
</s>
<s>
Available	O
for	O
Windows	O
and	O
Linux	O
uniprocessor	B-Operating_System
clients	O
only	O
.	O
</s>
<s>
Available	O
to	O
Linux	O
x86/64	B-Device
and	O
x86	B-Operating_System
Windows	O
.	O
</s>
<s>
These	O
are	O
the	O
second	O
generation	O
GPU	B-Architecture
cores	O
.	O
</s>
<s>
Unlike	O
the	O
retired	O
GPU1	O
cores	O
,	O
these	O
variants	O
are	O
for	O
ATI	O
CAL-enabled	O
2xxx/3xxx	O
or	O
later	O
series	O
and	O
NVIDIA	O
CUDA-enabled	O
NVIDIA	O
8xxx	O
or	O
later	O
series	O
GPUs	B-Architecture
.	O
</s>
<s>
Available	O
for	O
x86	B-Operating_System
Windows	O
clients	O
only	O
.	O
</s>
<s>
Supported	O
until	O
approximately	O
September	O
1	O
,	O
2011	O
due	O
to	O
AMD/ATI	O
dropping	O
support	O
for	O
the	O
utilized	O
Brook	B-Library
programming	O
language	O
and	O
moving	O
to	O
OpenCL	B-Application
.	O
</s>
<s>
This	O
forced	O
F	B-Application
@h	I-Application
to	O
rewrite	O
its	O
ATI	O
GPU	B-Architecture
core	O
code	O
in	O
OpenCL	B-Application
,	O
the	O
result	O
of	O
which	O
is	O
Core	O
16	O
.	O
</s>
<s>
Available	O
for	O
x86	B-Operating_System
Windows	O
clients	O
only	O
.	O
</s>
<s>
Available	O
for	O
x86	B-Operating_System
Windows	O
clients	O
only	O
.	O
</s>
<s>
Available	O
for	O
x86	B-Operating_System
Windows	O
clients	O
only	O
,	O
this	O
core	O
was	O
officially	O
released	O
Mar	O
02	O
,	O
2009	O
.	O
</s>
<s>
These	O
are	O
the	O
third	O
generation	O
GPU	B-Architecture
cores	O
,	O
and	O
are	O
based	O
on	O
,	O
Pande	O
Group	O
's	O
own	O
open	O
library	O
for	O
molecular	O
simulation	O
.	O
</s>
<s>
Available	O
to	O
x86	B-Operating_System
Windows	O
only	O
.	O
</s>
<s>
Available	O
to	O
x86	B-Operating_System
Windows	O
only	O
.	O
</s>
<s>
Released	O
alongside	O
the	O
new	O
v7	O
client	O
,	O
this	O
is	O
a	O
rewrite	O
of	O
Core	O
11	O
in	O
OpenCL	B-Application
.	O
</s>
<s>
Available	O
to	O
Windows	O
and	O
Linux	O
for	O
AMD	O
and	O
NVIDIA	O
GPUs	B-Architecture
using	O
OpenCL	B-Application
.	O
</s>
<s>
Available	O
to	O
Windows	O
for	O
AMD	O
and	O
NVIDIA	O
GPUs	B-Architecture
using	O
OpenCL	B-Application
.	O
</s>
<s>
There	O
are	O
currently	O
issues	O
regarding	O
the	O
stability	O
and	O
performance	O
of	O
this	O
core	O
on	O
some	O
AMD	O
and	O
NVIDIA	O
Maxwell	O
GPUs	B-Architecture
.	O
</s>
<s>
This	O
is	O
why	O
assignment	O
of	O
work	O
units	O
running	O
on	O
this	O
core	O
has	O
been	O
temporarily	O
stopped	O
for	O
some	O
GPUs	B-Architecture
.	O
</s>
<s>
Available	O
to	O
Windows	O
and	O
Linux	O
for	O
AMD	O
and	O
NVIDIA	O
GPUs	B-Architecture
using	O
OpenCL	B-Application
.	O
</s>
