Report an Error

NVIDIA GP104B

Default
GP104B
Die Shot
Fritzchens Fritz
Die Shot
Block Diagram
Block Diagram
SM Diagram
SM Diagram
NVIDIA's GP104B GPU uses the Pascal architecture and is made using a 16 nm production process at TSMC. With a die size of 314 mm² and a transistor count of 7,200 million it is a large chip. GP104B supports DirectX 12 (Feature Level 12_1). For GPU compute applications, OpenCL version 3.0 and CUDA 6.1 can be used. It features 2560 shading units, 160 texture mapping units and 64 ROPs.

Graphics Processor

Released
Aug 15th, 2016
GPU Name
GP104B
Codename
NV134
Architecture
Pascal
Foundry
TSMC
Process Size
16 nm
Transistors
7,200 million
Density
22.9M / mm²
Die Size
314 mm²
Package
BGA-2150

Graphics Features

DirectX
12 (12_1)
OpenGL
4.6
OpenCL
3.0
Vulkan
1.3
CUDA
6.1
Shader Model
6.8
WDDM
3.2
NVENC
6th Gen
NVDEC
3rd Gen
PureVideo HD
VP8
VDPAU
Feature Set H

Render Config

Shading Units
2560
TMUs
160
ROPs
64
SM Count
20
SFUs
640
TPCs
20
GPCs
4
L1 Cache
48 KB per SM
L2 Cache
2048 KB
Max. TDP
150 W

All Pascal GPUs

NVIDIA GPU Architecture History

Graphics cards using the NVIDIA GP104B GPU

Name Chip Memory Shaders TMUs ROPs Base Clock Boost Clock Memory Clock
NVIDIA GeForce GTX 1070 Mobile
N17E-G2-A1 8 GB 2048 128 64 1443 MHz 1645 MHz 2002 MHz
NVIDIA GeForce GTX 1080 Mobile
N17E-G3-A1 8 GB 2560 160 64 1557 MHz 1734 MHz 1251 MHz
NVIDIA GeForce GTX 1070 Max-Q
N17E-G2-A1 8 GB 2048 128 64 1215 MHz 1379 MHz 2002 MHz
NVIDIA GeForce GTX 1080 Max-Q
N17E-G3-A1 8 GB 2560 160 64 1277 MHz 1366 MHz 1251 MHz

GP104B GPU Notes

NVENC: 6th Gen
NVDEC: 3rd Gen
PureVideo HD: VP8
VDPAU: Feature Set H

Latest Drivers:
Windows 7 / 8 / 8.1 (x32 / x64):
GeForce Release 391.35 / 474.89

Windows 10 / 11 (x32 / x64):
GeForce Release 391.35 / Latest
Nov 13th, 2024 02:25 EST change timezone

Latest GPU Drivers

New Forum Posts

Popular Reviews

Controversial News Posts