AMD Radeon Pro WX 4100 vs NVIDIA H100 SXM5

GPU Specs Compare

G

NVIDIA H100 SXM5

NVIDIA 2022-03

Graphics Card

Release Date
Nov 2016
Mar 2022
Generation
Radeon Pro
Tesla Hopper
Type
Desktop
AI GPU
Bus Interface
PCIe 3.0 x8
PCIe 5.0 x16

Clock Speeds

Base Clock
1125 MHz
1590 MHz
Boost Clock
1201 MHz
1980 MHz
Memory Clock
1500 MHz
1313 MHz

Memory

Memory Size
4GB
80GB
Memory Type
GDDR5
HBM3
Memory Bus
128bit
5120bit
Bandwidth
96.00GB/s
3350GB/s

Render Config

Compute Units
16
-
Shading Units
1024
16896
TMUs
64
528
ROPs
16
24
L1 Cache
16 KB (per CU)
256 KB (per SM)
L2 Cache
1024 KB
50 MB
SM Count
-
132
Tensor Cores
-
528

Theoretical Performance

Pixel Rate
19.22 GPixel/s
47.52 GPixel/s
Texture Rate
76.86 GTexel/s
1045 GTexel/s
FP16 (half)
2.460 TFLOPS
267.6 TFLOPS
FP32 (float)
2.460 TFLOPS
66.91 TFLOPS
FP64 (double)
153.7 GFLOPS
33.45 TFLOPS

Graphics Processor

GPU Name
Baffin
GH100
GPU Variant
Baffin XT
-
Architecture
GCN 4.0
Hopper
Foundry
GlobalFoundries
TSMC
Process Size
14 nm
4 nm
Transistors
3 billion
80 billion
Die Size
123 mm²
814 mm²

Board Design

TDP
50W
700W
Suggested PSU
250 W
1100 W
Outputs
4x mini-DisplayPort 1.4a
No outputs
Power Connectors
None
8-pin EPS

Graphics Features

DirectX
12 (12_0)
N/A
OpenGL
4.6
N/A
OpenCL
2.1
3.0
Vulkan
1.2
N/A
Shader Model
6.4
N/A
CUDA
-
9.0

📊 Benchmark

3DMark Time Spy
1,520
-
3DMark Time Spy Extreme
774
-