Report an Error

NVIDIA GA102

NVIDIA's GA102 GPU uses the Ampere architecture and is made using a 7 nm production process at TSMC. With a die size of 700 mm² and a transistor count of 45,000 million it is a very big chip. GA102 supports DirectX 12 Ultimate (Feature Level 12_2). It features 7552 shading units, 472 texture mapping units and 96 ROPs. Also included are 672 tensor cores which help improve the speed of machine learning applications. The GPU also contains 168 raytracing acceleration cores.

Graphics Processor

GPU Name
GA102
Codename
NV172
Architecture
Ampere
Foundry
TSMC
Process Size
7 nm
Transistors
45,000 million
Density
64.3M / mm²
Die Size
700 mm²
Released
Unknown

Graphics Features

DirectX
12 Ultimate (12_2)
OpenGL
4.6
OpenCL
2.0
Vulkan
1.2.140
CUDA
8.5
Shader Model
6.5
PureVideo HD
VP11
VDPAU
Feature Set k

Render Config

Shading Units
7552
TMUs
472
ROPs
96
SM Count
118
Tensor Cores
672
RT Cores
168
L1 Cache
192 KB per SM
L2 Cache
6144 KB
Max. TDP
320 W

All Ampere GPUs

NVIDIA GPU Architecture History

Graphics cards using the NVIDIA GA102 GPU

Name Chip Memory Shaders TMUs ROPs Base Clock Boost Clock Memory Clock
NVIDIA Quadro RTX 8200 48 GB 7552 472 96 1110 MHz 1500 MHz 1750 MHz
NVIDIA GeForce RTX 3080 Ti GA102-400-A1 12 GB 5376 336 96 1770 MHz 1925 MHz 1750 MHz
NVIDIA GeForce RTX 3080 GA102-200-K1-A1 10 GB 4352 272 88 1535 MHz 1905 MHz 1750 MHz
NVIDIA GeForce RTX 3090 24 GB 5376 336 96 1410 MHz 1935 MHz 1750 MHz

GA102 GPU Notes

Speculation