GeForce 200 series
Second-generation Tesla architecture with unified shaders and double precision.
The GeForce 200 series is a line of graphics processing units from Nvidia built on the Tesla architecture. These chips represent the second generation of Nvidia's unified shader design, marking the first significant overhaul since the GeForce 8 series.
The GTX 280 and GTX 260 share the same processor core. During manufacturing, chips were tested and sorted by defect testing of core logic. Those that failed to meet the GTX 280's specifications were retested and binned as GTX 260 models, which have fewer stream processors, fewer ROPs, and a narrower memory bus. In late 2008, Nvidia released a revised GTX 260 with 216 stream processors instead of the original 192, creating two distinct GTX 260 cards with noticeable performance differences.
The GT200a and GT200b GPUs in the GeForce 200 series (excluding the GTS 250 and GTS 240, which use older G92b chips) support double precision for GPGPU tasks. They also deliver better geometry shading performance. The GT200 die contains 1.4 billion transistors on a 576 mm² surface, built on a 65 nm process. It ranks as the seventh largest commercial GPU ever made and the fifth largest CMOS-logic chip fabricated at TSMC. While the GeForce 400 series surpassed the GT200 in transistor count, the original GT200 die remains larger than the GF100 die. It is also bigger than the Kepler-based GK210 GPU in the Tesla K80, which has 7.1 billion transistors on a 561 mm² die at 28 nm. The Ampere GA100 is currently the largest commercial GPU, at 826 mm² with 54.2 billion transistors.
On October 12, 2009, Nvidia announced and released the GeForce 210 (GT218 GPU) and GeForce GT 220 (GT216 GPU), which were previously OEM-only. The GeForce GT 240 (GT215 GPU) followed on November 17, 2009. These 40 nm GPUs include the new PureVideo HD VP4 decoder hardware, whereas older GeForce 8 and 9 series chips only had VP2 or VP3 (G98). They also support Compute Capability 1.2, compared to the 1.1 support in earlier GeForce 8 and 9 cards. All GT21x GPUs contain an integrated audio processor and support eight-channel LPCM output over HDMI.
All models in the GeForce 200 series support Coverage Sample Anti-Aliasing, Angle-Independent Anisotropic Filtering, and 240-bit OpenEXR HDR. Compute Capability varies: 1.1 for the G92-based GTS 250, 1.2 for GT215, GT216, and GT218 GPUs, and 1.3 (with double precision) for GT200a and GT200b GPUs only.
Quick Facts
- Codename
- GT200
- Architecture
- Tesla
- Created
- June 16, 2008
- Model
- GeForce series
- Transistors
- 505M 55nm (G94b)
Facts from the source article.
Lore & Background
The GeForce 200 series introduced Nvidia's second generation of the Tesla microarchitecture, the first major update to it since the GeForce 8 series. The GeForce GTX 280 and GTX 260 are based on the same processor core; during manufacturing, chips were binned and separated through defect testing. Those that failed to meet the GTX 280 specification were re-tested and binned as GTX 260, which has 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, resulting in two GTX 260 cards in production with non-trivial performance differences.
The GT200 GPU 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 fabricated at the TSMC foundry. The GeForce 400 series later superseded the GT200 chips in transistor count, but the original GT200 dies still exceed the GF100 die size. The GT200 is larger than 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. The Ampere GA100 is currently the largest commercial GPU ever fabricated at 826 mm² with 54.2 billion transistors.
In late 2009, Nvidia announced and released the previously OEM-only GeForce 210 (GT218 GPU) and GeForce GT 220 (GT216 GPU), as well as the GeForce GT 240 (GT215 GPU). These 40 nm GPUs feature the new PureVideo HD VP4 decoder hardware, whereas older GeForce 8 and 9 GPUs only have PureVideo HD VP2 or VP3. They also support Compute Capability 1.2, while older GeForce 8 and 9 GPUs only supported Compute Capability 1.1. All GT21x GPUs contain an audio processor and support eight-channel LPCM output through HDMI.
Reader's Guide
The GeForce 200 series marked a significant step in Nvidia's unified shader architecture, being the first major update since the GeForce 8 series. The GT200 GPU's large die size and transistor count made it one of the largest commercial GPUs ever constructed at the time, though later architectures surpassed it in transistor count. The series introduced double precision support for GPGPU applications on GT200a/b GPUs, excluding the older G92b-based GTS 250 and GTS 240. The GT200 GPUs also improved performance in geometry shading. The later 40 nm GT21x GPUs brought updated video decoding hardware and Compute Capability 1.2, along with integrated audio processing for HDMI output. Nvidia ceased driver support for the GeForce 200 series on April 1, 2016, with final drivers released for Windows XP, Vista, 7, 8, 8.1, 10, and Linux 64-bit.
Did You Know?
- The GeForce GTX 280 and GTX 260 are based on the same processor core, with GTX 260 chips being binned from chips that failed GTX 280 specifications.
- The GT200 GPU consists of 1.4 billion transistors on a 576 mm² die, making it the seventh largest commercial GPU ever constructed.
- The GeForce 210, GT 220, and GT 240 were released on October 12 and November 17, 2009, featuring 40 nm GPUs with PureVideo HD VP4 and Compute Capability 1.2.
More in Nvidia and AMD Graphics Processors 1-24
Spotted an error? Know more?
Reader corrections go straight into our review queue. Suggest an edit · How this site is sourced