Report an Error

NVIDIA TU106B

Default
TU106B
Block Diagram
Block Diagram
SM Diagram
SM Diagram
NVIDIA's TU106B GPU uses the Turing architecture and is made using a 12 nm production process at TSMC. With a die size of 445 mm² and a transistor count of 10,800 million it is a very big chip. TU106B supports DirectX 12.0 (Feature Level 12_1). It features 2304 shading units, 144 texture mapping units and 64 ROPs. Also included are 288 tensor cores which help improve the speed of machine learning applications. The GPU also contains 36 raytracing acceleration cores.
Further reading: Turing Architecture Whitepaper

Graphics Processor

GPU Name
TU106B
Architecture
Turing
Foundry
TSMC
Process Size
12 nm
Transistors
10,800 million
Density
24.3M / mm²
Die Size
445 mm²
Codename
NV166
Released
Jan 29th, 2019

Graphics Features

DirectX
12.0 (12_1)
OpenGL
4.6
OpenCL
1.2
Vulkan
1.1.103
CUDA
7.5
Shader Model
6.4
PureVideo HD
VP10
VDPAU
Feature Set J

Render Config

Shading Units
2304
TMUs
144
ROPs
64
SM Count
36
Tensor Cores
288
RT Cores
36
GPCs
3
TPCs
18
SFUs
576
Tex L1 Cache
32 KB per SM
L1 Cache
64 KB per SM
L2 Cache
4096 KB
Max. TDP
115 W

All Turing GPUs

Graphics cards using the NVIDIA TU106B GPU

Name Chip Memory Shaders TMUs ROPs GPU Clock Boost Clock Memory Clock
NVIDIA GeForce RTX 2070 Max-Q N18E-G2-A1 8 GB 2304 144 64 885 MHz 1185 MHz 1500 MHz
NVIDIA GeForce RTX 2060 Mobile N18E-G1-KD-A1 6 GB 1920 120 48 960 MHz 1200 MHz 1750 MHz
NVIDIA GeForce RTX 2070 Mobile N18E-G2-A1 8 GB 2304 144 64 1215 MHz 1440 MHz 1750 MHz