Intel Gaudi 3 vs AMD Radeon E9550

Intel Gaudi 3
AMD Radeon E9550
16384 Shaders
128GB HBM2e
438MHz
2304 Shaders
8GB GDDR5
1243MHz
Peak AI Performance
1.84 PFLOPS
FP8 Tensor (FP32 Accumulate)
Peak AI Performance
5.73 TFLOPS
FP16
FP32
14.35 TFLOPS
FP32
5.73 TFLOPS
FP16
28.7 TFLOPS
FP16
5.73 TFLOPS
Form Factor
OAM Module
-
Form Factor
MXM-B
-
TDP
900W
TDP
95W
-
-
-
-
-
-
-
-
-
-
GB6 OpenCL N/A
0%
GB6 Metal N/A
0%
GB6 Metal N/A
0%
GB6 Vulkan N/A
0%
GB5 OpenCL N/A
0%
GB5 OpenCL N/A
0%
GB5 CUDA N/A
0%
GB5 CUDA N/A
0%
GB5 Metal N/A
0%
GB5 Metal N/A
0%
GB5 Vulkan N/A
0%
GB5 Vulkan N/A
0%
OCT 2020.1 N/A
0%
OCT 2020.1 N/A
0%
OCT Metal N/A
0%
OCT Metal N/A
0%
Peak AI Performance
1.84 PFLOPS
FP8 Tensor (FP32 Accumulate)
Peak AI Performance
5.73 TFLOPS
FP16
-
-
-
-
-
-
FP8
-
-
-
1.84 PFLOPS Tensor (FP32 Accumulate)
-
-
-
-
-
-
-
FP16
28.7 TFLOPS
-
-
458.75 TFLOPS Tensor (FP32 Accumulate)
-
FP16
5.73 TFLOPS
-
-
-
-
FP32
14.35 TFLOPS
229.38 TFLOPS Tensor
-
FP32
5.73 TFLOPS
-
-
-
-
-
FP64
360 GFLOPS
-
BF16
28.7 TFLOPS
1.84 PFLOPS Tensor
-
-
-
-
-
TF32
458.75 TFLOPS Tensor
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
Pixel Fillrate
-
Pixel Fillrate
39.776 GPixel/s
-
-
-
-
Texture Fillrate
-
Texture Fillrate
178.992 GTexel/s
Manufacturer
Intel
Manufacturer
AMD
Chip Designer
Intel
Chip Designer
AMD
Architecture
Gaudi 3
Architecture
GCN 4
Family
Gaudi
Family
Embedded
Codename
Gaudi 3
Gaudi 3
-
-
Codename
Ellesmere
Polaris 10
Variant
Polaris 10 XT
Market Segment
Server
Market Segment
Server
Release Date
4/9/2024
Release Date
9/1/2016
Foundry
TSMC
-
Foundry
GlobalFoundries
-
Fabrication Node
N5
-
Fabrication Node
14LPP
-
-
-
-
Die Size
232 mm²
-
-
-
Transistor Count
5.7 Billion
-
-
-
-
Transistor Density
24.57M/mm²
-
Form
OAM Module
Form
MXM-B
Shading Units
16384 Shaders
-
Shading Units
2304 Shaders
-
Texture Mapping Units
-
Texture Mapping Units
144 TMUs
Render Output Units
-
Render Output Units
32 ROPs
Tensor Cores
64 T-Cores
-
-
-
-
-
-
-
-
-
-
Compute Units
8 CUs
Compute Units
36 CUs
-
-
-
-
-
-
-
-
-
-
-
438MHz
-
-
-
1243MHz
-
-
-
-
L1
-
-
-
Unknown
L1
-
-
16KB/CU
-
L2
Unknown
L2
2MB Shared
-
-
-
-
-
-
128GB
HBM2e
ECC
8GB
GDDR5
-
Bus Width
8192Bit
Bus Width
256Bit
Clock
1808MHz
Transfer Rate
3.6GT/s
Bandwidth
3702.8GB/s
Clock
1250MHz
Transfer Rate
5GT/s
Bandwidth
160GB/s
-
-
-
-
eSRAM
96MB
12800GB/s
-
-
-
-
-
-
-
-
-
-
-
TDP
900W
TDP
95W
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
No Ports
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
No Ports
Max Resolution
Unknown
Max Resolution
5120x2880
Max Resolution Refresh Rate
-
Max Resolution Refresh Rate
60Hz
Variable Refresh Rate
-
-
-
Variable Refresh Rate
-
FreeSync
-
Display Stream Compression (DSC)
Not Supported
Display Stream Compression (DSC)
Not Supported
Multi Monitor Support
Unknown
Multi Monitor Support
3
-
-
Content Protection
HDCP 2.2
No Encoders
-
Model
VCE 3.0
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
Codec
-
-
-
-
-
-
-
-
AVC (H.264)
HEVC (H.265)
-
-
-
-
No Decoders
Model
UVD 6.3
-
-
-
-
-
-
-
-
-
-
-
-
-
-
Codec
MPEG-1
MPEG-2
MPEG-4
JPEG
VC-1
-
-
-
AVC (H.264)
HEVC (H.265)
-
-
-
-
-
-
-
-
Direct X
12
Direct 3D
12_0
-
-
OpenCL
3.0
-
-
OpenGL
4.6
OpenCL
2.1
Vulkan
1.3
-
-
-
-
-
-
-
-
-
-
Shader Model
6.7
-
-
GFX
8
-
-
-
-
Not a Card
-
-
-
Not a Card
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
PCIe Version
5.0
PCIe Lanes
16
-
-
PCIe Version
3.0
PCIe Lanes
16
Multi GPU Support
Supported
Type
RoCE
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
Change Comparison