NVIDIA GP108B
GP108B
Block Diagram
SM Diagram
NVIDIA's GP108B GPU uses the Pascal architecture and is made using a 14 nm production process at Samsung. With a die size of 74 mm² and a transistor count of 1,800 million it is a very small chip. GP108B supports DirectX 12 (Feature Level 12_1). For GPU compute applications, OpenCL version 3.0 and CUDA 6.1 can be used. It features 384 shading units, 24 texture mapping units and 16 ROPs.
Graphics Processor
- GPU Name
- GP108B
- Codename
- NV138
- Architecture
- Pascal
- Foundry
- Samsung
- Process Size
- 14 nm
- Transistors
- 1,800 million
- Density
- 24.3M / mm²
- Die Size
- 74 mm²
- Released
- Unknown
Graphics Features
- DirectX
- 12 (12_1)
- OpenGL
- 4.6
- OpenCL
- 3.0
- Vulkan
- 1.2
- CUDA
- 6.1
- Shader Model
- 6.4
- PureVideo HD
- VP8
- VDPAU
- Feature Set H
Render Config
- Shading Units
- 384
- TMUs
- 24
- ROPs
- 16
- SM Count
- 3
- SFUs
- 96
- TPCs
- 3
- GPCs
- 1
- L1 Cache
- 48 KB per SM
- L2 Cache
- 512 KB
- Max. TDP
- 10 W
All Pascal GPUs
NVIDIA GPU Architecture History
- 1998-2000 Fahrenheit
- 1999-2005 Celsius
- 2001-2003 Kelvin
- 2003-2005 Rankine
- 2003-2013 Curie
- 2006-2010 Tesla
- 2007-2013 Tesla 2.0
- 2010-2016 Fermi
- 2010-2013 VLIW Vec4
- 2010-2016 Fermi 2.0
- 2012-2018 Kepler
- 2013-2015 Kepler 2.0
- 2014-2017 Maxwell
- 2014-2019 Maxwell 2.0
- 2016-2021 Pascal
- 2017-2020 Volta
- 2018-2021 Turing
- 2020-2021 Ampere
Graphics cards using the NVIDIA GP108B GPU
| Name | Chip | Memory | Shaders | TMUs | ROPs | Base Clock | Boost Clock | Memory Clock |
|---|---|---|---|---|---|---|---|---|
| NVIDIA GeForce MX250 | 2 GB | 384 | 24 | 16 | 937 MHz | 1038 MHz | 1502 MHz |