NVIDIA GeForce RTX 4070 SUPER vs NVIDIA H100 CNX

GPU Specs Compare

G

NVIDIA H100 CNX

NVIDIA 2023-03

Graphics Card

Release Date
Jan 2024
Mar 2023
Generation
GeForce 40
Tesla Hopper
Type
Desktop
AI GPU
Bus Interface
PCIe 4.0 x16
PCIe 5.0 x16

Clock Speeds

Base Clock
1980 MHz
690 MHz
Boost Clock
2475 MHz
1845 MHz
Memory Clock
1313 MHz
1593 MHz

Memory

Memory Size
12GB
80GB
Memory Type
GDDR6X
HBM2e
Memory Bus
192bit
5120bit
Bandwidth
504.2GB/s
2039GB/s

Render Config

SM Count
56
114
Shading Units
7168
14592
TMUs
224
456
ROPs
80
24
Tensor Cores
224
456
RT Cores
56
-
L1 Cache
128 KB (per SM)
256 KB (per SM)
L2 Cache
48 MB
50 MB

Theoretical Performance

Pixel Rate
198.0 GPixel/s
44.28 GPixel/s
Texture Rate
554.4 GTexel/s
841.3 GTexel/s
FP16 (half)
35.48 TFLOPS
215.4 TFLOPS
FP32 (float)
35.48 TFLOPS
53.84 TFLOPS
FP64 (double)
554.4 GFLOPS
26.92 TFLOPS

Graphics Processor

GPU Name
AD104
GH100
GPU Variant
AD104-350-A1
-
Architecture
Ada Lovelace
Hopper
Foundry
TSMC
TSMC
Process Size
5 nm
4 nm
Transistors
35.8 billion
80 billion
Die Size
294 mm²
814 mm²

Board Design

TDP
220W
350W
Suggested PSU
550 W
750 W
Outputs
1x HDMI 2.13x DisplayPort 1.4a
No outputs
Power Connectors
1x 16-pin
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
CUDA
8.9
9.0
Shader Model
6.8
N/A