NVIDIA Tesla K20 Server vs AMD FirePro S9010
NVIDIA Tesla K20 Server
$3,199
AMD FirePro S9010
$2,499
|
2496 Shaders
5GB GDDR5
758MHz
|
1792 Shaders
3GB GDDR5
900MHz
|
|
Peak AI Performance
3.78 TOPS
INT32
|
Peak AI Performance
3.23 TFLOPS
FP32
|
|
FP32
3.78 TFLOPS
|
FP32
3.23 TFLOPS
|
|
Form Factor
PCIe Card
2.0-Slots
|
Form Factor
PCIe Card
2.0-Slots
|
|
TDP
225W
|
TDP
225W
|
|
Power Connectors
1x 6-Pin
1x 8-Pin
|
Power Connectors
-
1x 8-Pin
|
Peak AI Performance
|
|
Tesla K20 Server - 3.78 TOPS INT32
x1.17
FirePro S9010 - 3.23 TFLOPS FP32
x1
FP32
|
|
Tesla K20 Server - 3.78 TFLOPS FP32
x1.17
FirePro S9010 - 3.23 TFLOPS FP32
x1
|
|
Tesla K20 Server - 208GB/s
x1
FirePro S9010 - 264GB/s
x1.27
|
|
Tesla K20 Server - 225W
x1
FirePro S9010 - 225W
x1
|
0%
|
0%
|
|
0%
|
0%
|
|
0%
|
0%
|
|
0%
|
0%
|
|
0%
|
0%
|
|
0%
|
0%
|
|
0%
|
0%
|
|
0%
|
0%
|
|
0%
|
0%
|
|
Manufacturer
NVIDIA
|
Manufacturer
AMD
|
|
Chip Designer
NVIDIA
|
Chip Designer
AMD
|
|
Architecture
Kepler
|
Architecture
GCN 1
|
|
Family
Tesla K
|
Family
FirePro S
|
|
Codename
NVF0
GK110
Variant
GK110-885-KA-A1
|
Codename
Tahiti
-
Variant
Tahiti Pro GL
|
|
Market Segment
Server
|
Market Segment
Server
|
|
Release Date
1/5/2013
|
Release Date
8/27/2012
|
|
Foundry
TSMC
|
Foundry
TSMC
|
|
Fabrication Node
28nm
|
Fabrication Node
28nm
|
|
Die Size
561 mm²
|
Die Size
352 mm²
|
|
Transistor Count
7.1 Billion
|
Transistor Count
4.3 Billion
|
|
Transistor Density
12.62M/mm²
|
Transistor Density
12.25M/mm²
|
|
Form
PCIe Card
|
Form
PCIe Card
|
|
Shading Units
2496 Shaders
|
Shading Units
1792 Shaders
|
|
Texture Mapping Units
208 TMUs
|
Texture Mapping Units
112 TMUs
|
|
Render Output Units
40 ROPs
|
Render Output Units
32 ROPs
|
|
Streaming Multiprocessors
13 SMs
|
-
-
|
|
-
-
|
Compute Units
28 CUs
|
|
706MHz Base
758MHz
|
-
900MHz
|
|
Peak AI Performance
3.78 TOPS
INT32
|
Peak AI Performance
3.23 TFLOPS
FP32
|
|
FP32
3.78 TFLOPS
|
FP32
3.23 TFLOPS
|
|
FP64
1.26 TFLOPS
|
FP64
810 GFLOPS
|
|
INT32
3.78 TOPS
|
-
-
|
|
Pixel Fillrate
30.32 GPixel/s
|
Pixel Fillrate
28.8 GPixel/s
|
|
Texture Fillrate
157.664 GTexel/s
|
Texture Fillrate
100.8 GTexel/s
|
|
L1
48KB/SM Tex
16KB/SM
-
|
L1
-
-
16KB/CU
|
|
L2
1.5MB Shared
|
L2
768KB Shared
|
|
5GB
GDDR5
|
3GB
GDDR5
|
|
Bus Width
320Bit
|
Bus Width
384Bit
|
|
Clock
1300MHz
Transfer Rate
5.2GT/s
Bandwidth
208GB/s
|
Clock
1375MHz
Transfer Rate
5.5GT/s
Bandwidth
264GB/s
|
|
TDP
225W
|
TDP
225W
|
|
-
|
1x DisplayPort 1.2
|
|
Max Resolution
Unknown
|
Max Resolution
4096x2160
|
|
Max Resolution Refresh Rate
-
|
Max Resolution Refresh Rate
60Hz
|
|
Variable Refresh Rate
G-Sync
FreeSync
|
Variable Refresh Rate
-
-
|
|
Display Stream Compression (DSC)
Not Supported
|
Display Stream Compression (DSC)
Not Supported
|
|
Multi Monitor Support
Unknown
|
Multi Monitor Support
4
|
|
Model
NVENC 1
|
Model
VCE 1.0
|
|
Codec
AVC (H.264)
|
Codec
AVC (H.264)
|
|
Model
NVDEC 1
|
Model
UVD 3.2
|
|
Codec
MPEG-1
MPEG-2
MPEG-4
VC-1
AVC (H.264)
|
Codec
MPEG-1
MPEG-2
MPEG-4
VC-1
AVC (H.264)
|
|
Direct X
12
Direct 3D
11_0
|
Direct X
12
Direct 3D
11_1
|
|
OpenGL
4.6
OpenCL
3.0
Vulkan
1.2
|
OpenGL
4.6
OpenCL
2.1
Vulkan
1.2
|
|
Shader Model
6.5
CUDA
3.5
-
-
PureVideo HD
VP5
VDPAU
Feature Set D
|
Shader Model
6.5
-
-
GFX
6
-
-
-
-
|
|
Power Connectors
1x 6-Pin
1x 8-Pin
|
Power Connectors
-
1x 8-Pin
|
|
Slots Required
2.0
PCIe Version
3.0
PCIe Lanes
16
|
Slots Required
2.0
PCIe Version
3.0
PCIe Lanes
16
|
|
Multi GPU Support
Supported
-
-
|
Multi GPU Support
Supported
Type
CrossFire XDMA
|
|
Height
111 mm (4.37 in)
Width
267 mm (10.51 in)
Depth
39 mm (1.54 in)
|
Height
111 mm (4.37 in)
Width
267 mm (10.51 in)
Depth
37 mm (1.46 in)
|
Copy Link