Report an Error

NVIDIA GF110

GF110

GF110-040-A1

GF110-270-A1

GF110-275-A1

GF110-351-A1

GF110-375-A1

GF110-380-A1

GF110-876-A1

Fritzchens Fritz

Die Shot

Block Diagram

SM Diagram

NVIDIA's GF110 GPU uses the Fermi 2.0 architecture and is made using a 40 nm production process at TSMC. With a die size of 520 mm² and a transistor count of 3,000 million it is a very big chip. GF110 supports DirectX 12 (Feature Level 11_0). For GPU compute applications, OpenCL version 1.1 and CUDA 2.0 can be used. It features 512 shading units, 64 texture mapping units and 48 ROPs.

Further reading: Fermi Compute Architecture

Graphics Processor

GPU Name: GF110

Architecture: Fermi 2.0

Foundry: TSMC

Process Size: 40 nm

Transistors: 3,000 million

Density: 5.8M / mm²

Die Size: 520 mm²

Released: Nov 9th, 2010

Codename: NVC8

Graphics Features

DirectX: 12 (11_0)

OpenGL: 4.6

OpenCL: 1.1

Vulkan: N/A

CUDA: 2.0

Shader Model: 5.1

PureVideo HD: VP4

VDPAU: Feature Set C

Render Config

Shading Units: 512

TMUs: 64

ROPs: 48

SM Count: 16

SFUs: 64

GPCs: 4

L1 Cache: 64 KB per SM

L2 Cache: 768 KB

Max. TDP: 600 W

All Fermi 2.0 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 GF110 GPU

Name	Chip	Memory	Shaders	TMUs	ROPs	GPU Clock	Memory Clock
NVIDIA GeForce GTX 570	GF110-275-A1	1280 MB	480	60	40	732 MHz	950 MHz
NVIDIA GeForce GTX 580	GF110-375-A1	1536 MB	512	64	48	772 MHz	1002 MHz
NVIDIA GeForce GTX 590	GF110-351-A1	1536 MB	512	64	48	608 MHz	854 MHz
NVIDIA GeForce GTX 560 Ti 448	GF110-270-A1	1280 MB	448	56	40	732 MHz	950 MHz
NVIDIA Tesla C2075	GF110-351-A1	6 GB	448	56	48	574 MHz	783 MHz
NVIDIA GeForce GTX 560 Ti OEM		1280 MB	352	44	40	732 MHz	950 MHz
NVIDIA GeForce GTX 560 OEM	GF110-040-A1	1280 MB	384	48	40	552 MHz	802 MHz
NVIDIA Quadro Plex 7000		6 GB	512	64	48	574 MHz	750 MHz
NVIDIA Quadro 5010M	N12E-Q5	4 GB	384	48	32	450 MHz	650 MHz
NVIDIA Tesla M2090		6 GB	512	64	48	651 MHz	924 MHz
NVIDIA Quadro 7000		6 GB	512	64	48	651 MHz	851 MHz
NVIDIA Tesla X2090		6 GB	512	64	48	651 MHz	924 MHz
NVIDIA Tesla X2070		6 GB	512	64	48	651 MHz	924 MHz
NVIDIA Tesla M2075	GF110-876-A1	6 GB	448	56	48	574 MHz	783 MHz
NVIDIA Tesla C2090		6 GB	512	64	48	651 MHz	924 MHz
NVIDIA GeForce GTX 580 Rev. 2	GF110-380-A1	1536 MB	512	64	48	772 MHz	1002 MHz
NVIDIA GeForce GTX 570 Rev. 2	GF110-275-A1	1280 MB	480	60	40	732 MHz	950 MHz

GF110 GPU Notes

GF110 has 4 GPCs each capable of 8 pixels per clock. This limits complete GPU to 32 pixels per clock and because of that it can't feed all 48 ROPs when they all require data at the same time.
Additional ROPs however can be used for MSAA (because it doesn't require additional data from rasterizers, while giving more work to ROPs.)

The complete die contains 64 texture address units and 256 texture filtering units.