GeForce 6 series
Sixth-generation Nvidia GPUs with Shader Model 3.0 and SLI.
The GeForce 6 series, internally known as NV40, marked Nvidia's sixth generation of graphics processors. Released on April 14, 2004, this lineup brought three major features: PureVideo for video post-processing, Scalable Link Interface (SLI) for linking two cards, and support for Shader Model 3.0, which met the Microsoft DirectX 9.0c and OpenGL 2.0 standards.
SLI allowed two identical GeForce 6 cards (models 6500 and above) to work together, with driver software splitting the workload. This feature was limited to cards using the PCI-Express bus. Nvidia's PureVideo technology combined a dedicated video processing core with software to decode H.264, VC-1, WMV, and MPEG-2 video while reducing CPU load. Nvidia was the first to deliver Shader Model 3.0 in its GPUs, which extended Shader Model 2.0 with full 32-bit floating-point precision, dynamic branching, greater efficiency, and longer shader programs. Game developers quickly adopted SM3.0 because existing SM2.0/2.0A/2.0B shaders were easy to convert, and it boosted performance across the entire GeForce 6 line.
Not all models offered the same PureVideo capabilities; some lacked acceleration for WMV9 or H.264. Additionally, motherboards with certain VIA or SIS chipsets paired with an AMD Athlon XP processor had compatibility issues with the GeForce 6600 and 6800 GPUs. These problems—freezing, artifacts, reboots—made gaming and 3D applications nearly impossible, but only affected Direct3D-based programs, not OpenGL.
The first family in the GeForce 6 line was the 6800 series, aimed at high-performance gaming. The flagship GeForce 6800 Ultra (NV40) had 16 pixel pipelines and was two to two-and-a-half times faster than Nvidia's previous top card, the GeForce FX 5950 Ultra. It packed four times the pixel pipelines, twice the texture units, and a much improved pixel-shader architecture, yet it was built on the same 130-nanometer process as the FX 5950 and consumed slightly less power. A later revision, NV48, used a 110-nanometer process.
Initial 6800 models used the AGP bus. Nvidia later added PCI Express support, often through an AGP-PCIe bridge chip. For the 6800 GT and 6800 Ultra, Nvidia created the NV45 chip, which shared the same core as the NV40 but embedded the bridge on the package.
Quick Facts
- Codename
- NV4x
- Created
- April 14, 2004
- Model
- GeForce nForce series
- Entry
- 6100 / 6150 / 6200 / 6500
- Midrange
- 6600 / 6700
- Highend
- 6800
Facts from the source article.
Lore & Background
The GeForce 6 series introduced the Scalable Link Interface (SLI), allowing two GeForce 6 cards of the same type to be connected in tandem, with the driver software balancing the workload between them. SLI-capability was limited to select members of the GeForce 6 family (6500 and above) and was only available for cards utilizing the PCI-Express bus. Nvidia PureVideo technology combined a dedicated video processing core and software to decode H.264, VC-1, WMV, and MPEG-2 videos with reduced CPU utilization. Nvidia was the first to deliver Shader Model 3.0 (SM3) capability in its GPUs, which extended SM2 with standard FP32 precision, dynamic branching, increased efficiency, and longer shader lengths. SM3 was quickly adopted by game developers because it was simple to convert existing SM2.0/2.0A/2.0B shaders to version 3.0, offering noticeable performance improvements across the entire GeForce 6 line.
The first family in the product line, the 6800 series, catered to the high-performance gaming market. The 16 pixel pipeline GeForce 6800 Ultra (NV40) was 2 to 2.5 times faster than Nvidia's previous top-line product (the GeForce FX 5950 Ultra), packed four times the number of pixel pipelines, twice the number of texture units, and added a much improved pixel-shader architecture. The 6800 Ultra was fabricated on the same (IBM) 130 nanometer process node as the FX 5950 and consumed slightly less power. NV48 was a later version of NV40 manufactured in 110 nanometer process. Initial 6800 members were designed for the AGP bus; Nvidia added support for PCI Express via an AGP-PCIe bridge chip. The use of a bridge chip allowed Nvidia to release a full complement of PCIe graphics cards without redesigning them for the PCIe interface.
PureVideo functionality varied by model, with some models lacking WMV9 and/or H.264 acceleration. The first GeForce products to offer PureVideo, the AGP GeForce 6800/GT/Ultra, failed to support all of PureVideo's advertised features. Media player software with support for WMV-acceleration did not become available until several months after the 6800's introduction. User and web reports showed little if any difference between PureVideo enabled GeForces and non-Purevideo cards. In late 2005, an update to Nvidia's website confirmed that WMV-acceleration is not available on the AGP 6800.
Reader's Guide
The GeForce 6 series marked a significant generational leap for Nvidia, introducing Shader Model 3.0 support, which became a standard for DirectX 9.0c games and was quickly adopted by developers due to its compatibility with existing shader code and performance benefits. The series also brought SLI multi-GPU technology to the consumer market, though it was limited to PCI-Express cards and higher-end models. PureVideo technology represented Nvidia's push into hardware-accelerated video decoding, though early AGP 6800 models had notable gaps in support, particularly for WMV9 acceleration, which was later confirmed as unavailable. The 6800 series, as the flagship family, demonstrated substantial performance improvements over the previous generation, with the 6800 Ultra being 2 to 2.5 times faster than the GeForce FX 5950 Ultra while using similar manufacturing technology. The use of an AGP-PCIe bridge chip allowed Nvidia to transition to PCI-Express without redesigning GPUs, though it initially raised concerns about bandwidth utilization; benchmarks showed that AGP 8× was sufficient for most contemporary games, making the additional PCIe bandwidth largely superfluous. The series also saw compatibility issues with some VIA and SIS chipsets and AMD Athlon XP processors, particularly with GeForce 6600 and 6800 GPUs in Direct3D applications, causing freezing, artifacts, and reboots. The GeForce 6 series established many features that would become standard in subsequent generations, including Shader Model 3.0, SLI, and integrated video processing.
Did You Know?
- The GeForce 6 series was the first to deliver Shader Model 3.0 capability in GPUs.
- SLI-capability was limited to GeForce 6 cards 6500 and above and only available on PCI-Express bus.
- The AGP GeForce 6800/GT/Ultra failed to support all of PureVideo's advertised features, and WMV-acceleration was later confirmed unavailable on the AGP 6800.
- The GeForce 6800 Ultra was fabricated on the same 130 nanometer process as the GeForce FX 5950 Ultra but consumed slightly less power.
Frequently Asked Questions
What is the GeForce 6 series?
The GeForce 6 series was Nvidia's sixth-generation consumer GPU family, built on the NV40 architecture and released on April 14, 2004. It was the first Nvidia lineup to ship with native Shader Model 3.0 support.
What key technologies did the GeForce 6 series introduce?
The lineup debuted three headline features: PureVideo for hardware-accelerated video post-processing, SLI for pairing two cards, and full Shader Model 3.0 compliance. Together these set the generation clearly apart from the previous GeForce 5 family.
How did SLI work on GeForce 6 cards?
SLI allowed two identical GeForce 6 cards (model 6500 or higher) to share rendering work, with the driver splitting the frame load between them. The setup required both cards to occupy PCI-Express slots and used a small bridge connector to keep them in sync.
What API standards did the GeForce 6 series support?
The cards were the first Nvidia GPUs to fully meet the DirectX 9.0c and OpenGL 2.0 specifications, made possible by their Shader Model 3.0 implementation. This made them the go-to choice for titles that demanded those newer graphics APIs.
Why is the GeForce 6 series considered a turning point in GPU history?
It was the first generation where Nvidia offered consumer multi-GPU scaling out of the box, shifting the market away from single-board performance races. Its PureVideo pipeline also laid the groundwork for the hardware video-decoding focus that later became standard across the industry.
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
