NVIDIA H100 CNX vs NVIDIA RTX 2000 Ada Generation

GPU Specs Compare

G

NVIDIA H100 CNX

NVIDIA 2023-03

Graphics Card

Release Date
Mar 2023
Feb 2024
Generation
Tesla Hopper
Quadro Ada
Type
AI GPU
Desktop
Bus Interface
PCIe 5.0 x16
PCIe 4.0 x8

Clock Speeds

Base Clock
690 MHz
1620 MHz
Boost Clock
1845 MHz
2130 MHz
Memory Clock
1593 MHz
2000 MHz

Memory

Memory Size
80GB
16GB
Memory Type
HBM2e
GDDR6
Memory Bus
5120bit
128bit
Bandwidth
2039GB/s
256.0GB/s

Render Config

SM Count
114
22
Shading Units
14592
2816
TMUs
456
88
ROPs
24
48
Tensor Cores
456
88
L1 Cache
256 KB (per SM)
128 KB (per SM)
L2 Cache
50 MB
12 MB
RT Cores
-
22

Theoretical Performance

Pixel Rate
44.28 GPixel/s
102.2 GPixel/s
Texture Rate
841.3 GTexel/s
187.4 GTexel/s
FP16 (half)
215.4 TFLOPS
12.00 TFLOPS
FP32 (float)
53.84 TFLOPS
12.00 TFLOPS
FP64 (double)
26.92 TFLOPS
187.4 GFLOPS

Graphics Processor

GPU Name
GH100
AD107
Architecture
Hopper
Ada Lovelace
Foundry
TSMC
TSMC
Process Size
4 nm
5 nm
Transistors
80 billion
18.9 billion
Die Size
814 mm²
159 mm²

Board Design

TDP
350W
70W
Suggested PSU
750 W
250 W
Outputs
No outputs
4x mini-DisplayPort 1.4a
Power Connectors
8-pin EPS
None

Graphics Features

DirectX
N/A
12 Ultimate (12_2)
OpenGL
N/A
4.6
OpenCL
3.0
3.0
Vulkan
N/A
1.3
CUDA
9.0
8.9
Shader Model
N/A
6.7

📊 Benchmark

3DMark Time Spy
-
8,147
3DMark Time Spy Extreme
-
3,601
OctaneBench
-
295
3DMark Speed Way
-
1,970
Blender
-
2,649