GeForce 200 series

GeForce 200 series
	Nvidia GeForce GTX 295 released in January 2009; the flagship unit of the 200 series. This particular model manufactured by NVIDIA board-partner, Inno3D.
Release date	June 16, 2008; 16 years ago
Codename	GT200
Architecture	Tesla
Models	GeForce series GeForce GT series; GeForce GTS series; GeForce GTX series;
Transistors	505M 55nm (G94b) 754M 55nm (G92b); 260M 40nm (GT218); 486M 40nm (GT216); 727M 40nm (GT215); 1,400M 65nm (GT200a); 1,400M 55nm (GT200b);
Cards
Entry-level	GT 205; GT 210; GT 220; GT 230
Mid-range	GT 240; GTS 250
High-end	GTX 260; GTX 275; GTX 280
Enthusiast	GTX 285; GTX 295
API support
DirectX	Direct3D 10.0 or 10.1; Shader Model 4.1
OpenCL	OpenCL 1.1
OpenGL	OpenGL 3.3
History
Predecessor	GeForce 9 series
Variant	GeForce 300 series
Successor	GeForce 400 series
Support status
	Unsupported

The GeForce 200 series is a series of Tesla-based GeForce graphics processing units developed by Nvidia.

Architecture

The GeForce 200 series introduced Nvidia's second generation of Tesla (microarchitecture), Nvidia's unified shader architecture; the first major update to it since introduced with the GeForce 8 series.

The GeForce GTX 280 and GTX 260 are based on the same processor core. During the manufacturing process, GTX chips were binned and separated through defect testing of the core's logic functionality. Those that fail to meet the GTX 280 hardware specification are re-tested and binned as GTX 260 (which is specified with fewer stream processors, fewer ROPs and a narrower memory bus).

In late 2008, Nvidia re-released the GTX 260 with 216 stream processors, up from 192. Effectively, there were two GTX 260 cards in production with non-trivial performance differences.

The GeForce 200 series GPUs (GT200a/b GPU), excluding GeForce GTS 250, GTS 240 GPUs (these are older G92b GPUs), have double precision support for use in GPGPU applications. GT200 GPUs also have improved performance in geometry shading.

As of August 2018^[update], the GT200 is the seventh largest commercial GPU ever constructed, consisting of 1.4 billion transistors covering a 576 mm² die surface area built on a 65 nm process. It is the fifth largest CMOS-logic chip that has been fabricated at the TSMC foundry. The GeForce 400 series have since superseded the GT200 chips in transistor count, but the original GT200 dies still exceed the GF100 die size. It is larger than even the Kepler-based GK210 GPU used in the Tesla K80, which has 7.1 billion transistors on a 561 mm² die manufactured in 28 nm.^[1] The Ampere GA100 is currently the largest commercial GPU ever fabricated at 826 mm² with 54.2 billion transistors.^[2]

Nvidia officially announced and released the retail version of the previously OEM only GeForce 210 (GT218 GPU) and GeForce GT 220 (GT216 GPU) on October 12, 2009. Nvidia officially announced and released the GeForce GT 240 (GT215 GPU) on November 17, 2009. The new 40nm GPUs feature the new PureVideo HD VP4 decoder hardware in them, as the older GeForce 8 and 9 GPUs only have PureVideo HD VP2 or VP3 (G98). They also support Compute Capability 1.2, whereas older GeForce 8 and 9 GPUs only supported Compute Capability 1.1. All GT21x GPUs also contain an audio processor inside and support eight-channel LPCM output through HDMI.

Chipset table