NVIDIA H100 CNX vs AMD Radeon Pro W6800
GPU · Specs Compare
Graphics Card
Release Date
Mar 2023
Jun 2021
Generation
Tesla Hopper
Radeon Pro
Type
AI GPU
Desktop
Bus Interface
PCIe 5.0 x16
PCIe 4.0 x16
Clock Speeds
Base Clock
690 MHz
2075 MHz
Boost Clock
1845 MHz
2320 MHz
Memory Clock
1593 MHz
2000 MHz
Memory
Memory Size
80GB
32GB
Memory Type
HBM2e
GDDR6
Memory Bus
5120bit
256bit
Bandwidth
2039GB/s
512.0GB/s
Render Config
SM Count
114
—
Shading Units
14592
3840
TMUs
456
240
ROPs
24
96
Tensor Cores
456
—
L1 Cache
256 KB (per SM)
128 KB per Array
L2 Cache
50 MB
4 MB
Compute Units
—
60
RT Cores
—
60
L3 Cache
—
128 MB
Theoretical Performance
Pixel Rate
44.28 GPixel/s
222.7 GPixel/s
Texture Rate
841.3 GTexel/s
556.8 GTexel/s
FP16 (half)
215.4 TFLOPS
35.64 TFLOPS
FP32 (float)
53.84 TFLOPS
17.82 TFLOPS
FP64 (double)
26.92 TFLOPS
1114 GFLOPS
Graphics Processor
GPU Name
GH100
Navi 21
Architecture
Hopper
RDNA 2.0
Foundry
TSMC
TSMC
Process Size
4 nm
7 nm
Transistors
80 billion
26.8 billion
Die Size
814 mm²
520 mm²
Board Design
TDP
350W
250W
Suggested PSU
750 W
600 W
Outputs
No outputs
6x mini-DisplayPort 1.4a
Power Connectors
8-pin EPS
1x 6-pin + 1x 8-pin
Graphics Features
DirectX
N/A
12 Ultimate (12_2)
OpenGL
N/A
4.6
OpenCL
3.0
2.1
Vulkan
N/A
1.3
CUDA
9.0
—
Shader Model
N/A
6.5
Benchmark
3DMark Time Spy
—
15,963
▲ Specs Compare
Data sourced from official specifications and benchmark tests.