Report an Error

NVIDIA GA106

SM Diagram
SM Diagram
NVIDIA's GA106 GPU uses the Ampere architecture and is made using a 8 nm production process at Samsung. With a die size of 300 mm² and a transistor count of 13,250 million it is a large chip. GA106 supports DirectX 12 Ultimate (Feature Level 12_2). For GPU compute applications, OpenCL version 2.0 and CUDA 8.5 can be used. Additionally, the DirectX 12 Ultimate capability guarantees support for hardware-raytracing, variable-rate shading and more, in upcoming video games. It features 4096 shading units, 128 texture mapping units and 64 ROPs. Also included are 128 tensor cores which help improve the speed of machine learning applications. The GPU also contains 32 raytracing acceleration cores.
Further reading: Ampere Architecture Whitepaper

Graphics Processor

GPU Name
GA106
Codename
NV176
Architecture
Ampere
Foundry
Samsung
Process Size
8 nm
Transistors
13,250 million
Density
44.2M / mm²
Die Size
300 mm²
Released
2020

Graphics Features

DirectX
12 Ultimate (12_2)
OpenGL
4.6
OpenCL
2.0
Vulkan
1.2
CUDA
8.5
Shader Model
6.5
PureVideo HD
VP11
VDPAU
Feature Set k

Render Config

Shading Units
4096
TMUs
128
ROPs
64
SM Count
32
FP16 Units
4096
FP64 Units
64
INT32 Units
2048
Tensor Cores
128
RT Cores
32
SFUs
512
TPCs
16
GPCs
4
Tex L1 Cache
64 KB per SM
L1 Cache
128 KB per SM
L2 Cache
3072 KB
Max. TDP
180 W

All Ampere GPUs

NVIDIA GPU Architecture History

Graphics cards using the NVIDIA GA106 GPU

Name Chip Memory Shaders TMUs ROPs Base Clock Boost Clock Memory Clock
NVIDIA GeForce RTX 3060 GA106-300-A1 6 GB 3840 120 64 1500 MHz 1725 MHz 1750 MHz

GA106 GPU Notes

Speculation