NVIDIA H100 CNX vs AMD Radeon PRO W9070
GPU Specs Compare
Graphics Card
Release Date
Mar 2023
Unknown
Generation
Tesla Hopper
Mining GPUs
Type
AI GPU
Desktop
Bus Interface
PCIe 5.0 x16
PCIe 3.0 x16
Clock Speeds
Base Clock
690 MHz
1120 MHz
Boost Clock
1845 MHz
1206 MHz
Memory Clock
1593 MHz
2100 MHz
Memory
Memory Size
80GB
4GB
Memory Type
HBM2e
GDDR5
Memory Bus
5120bit
256bit
Bandwidth
2039GB/s
268.8GB/s
Render Config
SM Count
114
-
Shading Units
14592
2048
TMUs
456
128
ROPs
24
32
Tensor Cores
456
128
L1 Cache
256 KB (per SM)
16 KB (per CU)
L2 Cache
50 MB
2 MB
Compute Units
-
32
RT Cores
-
64
L3 Cache
-
64 MB
Theoretical Performance
Pixel Rate
44.28 GPixel/s
38.59 GPixel/s
Texture Rate
841.3 GTexel/s
154.4 GTexel/s
FP16 (half)
215.4 TFLOPS
4.940 TFLOPS
FP32 (float)
53.84 TFLOPS
4.940 TFLOPS
FP64 (double)
26.92 TFLOPS
308.7 GFLOPS
Graphics Processor
GPU Name
GH100
Polaris 20
Architecture
Hopper
GCN 4.0
Foundry
TSMC
GlobalFoundries
Process Size
4 nm
14 nm
Transistors
80 billion
5.7 billion
Die Size
814 mm²
232 mm²
GPU Variant
-
Navi 48 XTW
Board Design
TDP
350W
180W
Suggested PSU
750 W
450 W
Outputs
No outputs
1x HDMI 2.0b
Power Connectors
8-pin EPS
2x 8-pin
Graphics Features
DirectX
N/A
12 (12_0)
OpenGL
N/A
4.6
OpenCL
3.0
2.1
Vulkan
N/A
1.3
CUDA
9.0
-
Shader Model
N/A
6.7