NVIDIA A10 vs AMD Instinct MI100
NVIDIA A10
AMD Instinct MI100
9216 Shaders
24GB GDDR6
1695MHz
|
7680 Shaders
32GB HBM2
1502MHz
|
Peak AI Performance
999.75 TOPS
INT4 Tensor Sparse
|
Peak AI Performance
184.57 TFLOPS
FP16 Tensor (FP16 Accumulate)
|
FP32
31.24 TFLOPS
|
FP32
23.07 TFLOPS
|
FP16
31.24 TFLOPS
|
FP16
46.14 TFLOPS
|
Form Factor
PCIe Card
1.0-Slots
|
Form Factor
PCIe Card
2.0-Slots
|
TDP
150W
|
TDP
300W
|
Power Connectors
1x 8-Pin
|
Power Connectors
2x 8-Pin
|
Peak AI Performance
|
|
A10 - 999.75 TOPS INT4 Tensor Sparse
x5.42
Instinct MI100 - 184.57 TFLOPS FP16 Tensor (FP16 Accumulate)
x1
FP32
|
|
A10 - 31.24 TFLOPS FP32
x1.35
Instinct MI100 - 23.07 TFLOPS FP32
x1
FP16
|
|
A10 - 31.24 TFLOPS FP16
x1
Instinct MI100 - 46.14 TFLOPS FP16
x1.48
|
|
A10 - 300.1GB/s
x1
Instinct MI100 - 1228.8GB/s
x4.09
|
|
A10 - 150W
x1
Instinct MI100 - 300W
x2
|
|
A10 - GB6 OpenCL 47,060
x1
Instinct MI100 - GB6 OpenCL 123,425
x2.62
Manufacturer
NVIDIA
|
Manufacturer
AMD
|
Chip Designer
NVIDIA
|
Chip Designer
AMD
|
Architecture
Ampere
|
Architecture
CDNA 1
|
Family
Server
|
Family
Instinct
|
Codename
NV172
GA102
Variant
GA102-890-A1
|
Codename
Arcturus
-
Variant
Arcturus XL
|
Market Segment
Server
|
Market Segment
Server
|
Release Date
4/12/2021
|
Release Date
11/16/2020
|
Foundry
Samsung
|
Foundry
TSMC
|
Fabrication Node
8N
|
Fabrication Node
N7
|
Die Size
628 mm²
|
Die Size
750 mm²
|
Transistor Count
28.3 Billion
|
Transistor Count
25.6 Billion
|
Transistor Density
45.04M/mm²
|
Transistor Density
34.13M/mm²
|
Form
PCIe Card
|
Form
PCIe Card
|
Shading Units
9216 Shaders
|
Shading Units
7680 Shaders
|
Texture Mapping Units
288 TMUs
|
Texture Mapping Units
480 TMUs
|
Render Output Units
96 ROPs
|
Render Output Units
-
|
Tensor Cores
288 T-Cores
|
Tensor Cores
480 T-Cores
|
Ray-Tracing Cores
72 RT-Cores
|
-
-
|
Streaming Multiprocessors
72 SMs
|
-
-
|
-
-
|
Compute Units
120 CUs
|
885MHz Base
1695MHz
|
1000MHz Base
1502MHz
|
Peak AI Performance
999.75 TOPS
INT4 Tensor Sparse
|
Peak AI Performance
184.57 TFLOPS
FP16 Tensor (FP16 Accumulate)
|
FP16
31.24 TFLOPS
124.97 TFLOPS Tensor (FP16 Accumulate)
249.94 TFLOPS Tensor (FP16 Accumulate) Sparse
124.97 TFLOPS Tensor (FP32 Accumulate)
249.94 TFLOPS Tensor (FP32 Accumulate) Sparse
|
FP16
46.14 TFLOPS
184.57 TFLOPS Tensor (FP16 Accumulate)
-
184.57 TFLOPS Tensor (FP32 Accumulate)
-
|
FP32
31.24 TFLOPS
-
|
FP32
23.07 TFLOPS
46.14 TFLOPS Tensor
|
FP64
490 GFLOPS
|
FP64
11.54 TFLOPS
|
BF16
31.24 TFLOPS
124.97 TFLOPS Tensor
249.94 TFLOPS Tensor Sparse
|
BF16
-
92.28 TFLOPS Tensor
-
|
TF32
62.48 TFLOPS Tensor
124.97 TFLOPS Tensor Sparse
|
-
-
-
|
INT4
499.88 TOPS Tensor
999.75 TOPS Tensor Sparse
|
INT4
92.28 TOPS Tensor
-
|
INT8
249.94 TOPS Tensor
499.88 TOPS Tensor Sparse
|
INT8
92.28 TOPS Tensor
-
|
INT32
15.62 TOPS
|
-
-
|
Ray Tracing
61 TOPS
|
-
-
|
Pixel Fillrate
162.72 GPixel/s
|
Pixel Fillrate
-
|
Texture Fillrate
488.16 GTexel/s
|
Texture Fillrate
720.96 GTexel/s
|
L1
64KB/SM Tex
128KB/SM
-
|
L1
-
-
16KB/CU
|
L2
6MB Shared
|
L2
8MB Shared
|
24GB
GDDR6
-
|
32GB
HBM2
ECC
|
Bus Width
384Bit
|
Bus Width
4096Bit
|
Clock
782MHz
Transfer Rate
6.3GT/s
Bandwidth
300.1GB/s
|
Clock
1200MHz
Transfer Rate
2.4GT/s
Bandwidth
1228.8GB/s
|
TDP
150W
|
TDP
300W
|
Max Resolution
Unknown
|
Max Resolution
Unknown
|
Max Resolution Refresh Rate
|
Max Resolution Refresh Rate
|
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
Unknown
|
Model
NVENC 7
|
Model
VCN 2.5
|
Codec
AVC (H.264)
HEVC (H.265)
|
Codec
AVC (H.264)
HEVC (H.265)
|
Model
NVDEC 5
|
Model
VCN 2.5
|
Codec
MPEG-1
MPEG-2
MPEG-4
-
VC-1
VP8
VP9
AVC (H.264)
HEVC (H.265)
AV1
|
Codec
MPEG-1
MPEG-2
MPEG-4
JPEG
VC-1
-
VP9
AVC (H.264)
HEVC (H.265)
-
|
Direct X
12
Direct 3D
12_2
|
-
-
-
-
|
OpenGL
4.6
OpenCL
3.0
Vulkan
1.2
|
-
-
OpenCL
2.1
-
-
|
Shader Model
6.6
CUDA
8.6
-
-
PureVideo HD
VP11
VDPAU
Feature Set K
|
-
-
-
-
GFX
9.4
-
-
-
-
|
Power Connectors
1x 8-Pin
|
Power Connectors
2x 8-Pin
|
Slots Required
1.0
PCIe Version
4.0
PCIe Lanes
16
|
Slots Required
2.0
PCIe Version
4.0
PCIe Lanes
16
|
Multi GPU Support
Supported
Type
NVLink
|
Multi GPU Support
Supported
Type
Infinity Fabric
|
Height
112 mm (4.41 in)
Width
267 mm (10.51 in)
Depth
20 mm (0.79 in)
|
Height
111 mm (4.37 in)
Width
267 mm (10.51 in)
Depth
37 mm (1.46 in)
|






















Copy Link