Nvidia is to phase out its GeForce3 line of graphics chips entirely as it makes room for the new GeForce4, the company said on Wednesday. Instead, its low-end product will become the GeForce2 MX, complementing the midrange GeForce4 MX and the high-end GeForce4 Ti.
Sources close to the company have also suggested that Nvidia will move to a fully programmable architecture with its next chip design. Such a move could mean a much shorter lag time between when new features are introduced into a graphics chip and when they appear in new games. Current chips have their functions fixed in hardware.
The GeForce4 is much faster than its predecessor, last year's GeForce3 Ti, offering more than double the speed on certain features. Nvidia excecutives said GeForce chips are improving at a rate of "Moore's Law cubed", with power doubling every six months. Moore's Law states that the power of a processor doubles every eighteen months. The downside of this, however, is that other components such as memory may end up forming a bottleneck within the graphics card itself, Nvidia said.
Even so, Nvidia estimates that 3D moving images rendered on consumer graphics cards will not reach the same quality as filmed images for another 10 years.
Executives spoke at the Atomium in Brussels on Wednesday for the European launch of GeForce4.
Nvidia did not explain why it would be dropping the GeForce3 line, but it is likely that the company does not want its budget products competing too closely with the midrange and high-end lines. GeForce4 MX is a stripped-down version of GeForce4 Ti, lacking some high-end features like hardware vertex shaders, instead relying on software for vertex shading. Vertex shading is a technique for more realistic rendering of gradations in 3D objects.
As a result, GeForce4 MX does not offer the same massive performance boost over GeForce3 that is to be found in the GeForce4 Ti line. GeForce4 MX does, however, include some of GeForce4 Ti's new features like multiple display support and hardware-based anti-aliasing.
The GeForce4 chips uses a dedicated hardware engine to carry out the processing work of anti-aliasing, which smoothes the jagged edges of 3D images, Nvidia said. The dedicated hardware means the load on other components of the processor, such as the pixel shaders, is reduced. For users, that means the performance cost of turning on such features is lower; for example, the top-of-the-line GeForce4 Ti 4600 with the basic "Quincunx" level of anti-aliasing turned on runs faster than the GeForce3 Ti 500 without anti-aliasing.
The next generation of Nvidia chips will offer a feature to make game developers' lives easier in the form of fully programmable hardware, according to sources. This means that unlike with current chips, which implement features in hardware, developers will be able to specify which functions the chip carries out using a firmware image.
Game developers would be able to test out new chip features even as Nvidia is developing them, meaning a far shorter lead time before those features find their way into new games. It currently takes about 12 months from the time a new graphics chip is launched before new games can fully take advantage of it.
A fully programmable architecture would also make it easier to build multi-processor graphics cards, as one GPU (graphics processing unit) could be programmed to carry out half of the chip's functions while another GPU took care of the remaining functions.
To find out more about the computers and hardware that these chips are being used in, see ZDNet UK's Hardware News Section.
Have your say instantly, and see what others have said. Go to the Chips Central Forum.
Let the Chips Central editor know what you think by email. And sign up for the weekly Chips Central newsletter.