Intel Arc A350 vs NVIDIA H100 CNX

GPU Specs Compare

G

Intel Arc A350

Intel 2022-01
G

NVIDIA H100 CNX

NVIDIA 2023-03

Graphics Card

Release Date
Jan 2022
Mar 2023
Generation
Alchemist(Arc 3)
Tesla Hopper
Type
Desktop
AI GPU
Bus Interface
PCIe 4.0 x8
PCIe 5.0 x16

Clock Speeds

Base Clock
2000 MHz
690 MHz
Boost Clock
2000 MHz
1845 MHz
Memory Clock
1937 MHz
1593 MHz

Memory

Memory Size
4GB
80GB
Memory Type
GDDR6
HBM2e
Memory Bus
64bit
5120bit
Bandwidth
124.0GB/s
2039GB/s

Render Config

Shading Units
768
14592
TMUs
48
456
ROPs
24
24
Tensor Cores
96
456
RT Cores
6
-
L2 Cache
4 MB
50 MB
SM Count
-
114
L1 Cache
-
256 KB (per SM)

Theoretical Performance

Pixel Rate
48.00 GPixel/s
44.28 GPixel/s
Texture Rate
96.00 GTexel/s
841.3 GTexel/s
FP16 (half)
6.144 TFLOPS
215.4 TFLOPS
FP32 (float)
3.072 TFLOPS
53.84 TFLOPS
FP64 (double)
768.0 GFLOPS
26.92 TFLOPS

Graphics Processor

GPU Name
DG2-128
GH100
GPU Variant
ACM-G11
-
Architecture
Generation 12.7
Hopper
Foundry
TSMC
TSMC
Process Size
6 nm
4 nm
Transistors
7.2 billion
80 billion
Die Size
157 mm²
814 mm²

Board Design

TDP
25W
350W
Suggested PSU
200 W
750 W
Outputs
No outputs
No outputs
Power Connectors
-
8-pin EPS

Graphics Features

DirectX
12 Ultimate (12_2)
N/A
OpenGL
4.6
N/A
OpenCL
3.0
3.0
Vulkan
1.3
N/A
Shader Model
6.6
N/A
CUDA
-
9.0