by Anton Shilov
10/28/2009 | 04:06 PM
Nvidia Corp. prepares to release a new GeForce 300M series of graphics processing units (GPUs) designed for notebooks, a listing at the company’s web-site revealed. Specifications of the new graphics chips are unclear at this point.
Based on a list of chips located at Nvidia’s own web-site, the company preps GeForce GTS 360M, GT 335M, GT 330M, 310M and 305M products. The lineup seems to cover performance-mainstream, mainstream and entry-level segments of the mobile market. The list has been discovered by a member of NotebookReview web-site's forums.
Obviously, GeForce 300M name suggests that the new family of chips is based on the Fermi (NV60/G300/GT300/GF100) architecture; hence, it should support DirectX 11 and other advanced features. However, considering the fact that GeForce 200M line is based on the GeForce 9800M/G92 design and has nothing to do with G200/NV55 architecture, the GeForce 300M may have nothing to do with Fermi. In fact, so far Nvidia G200/NV55 architecture has not reached the mobile segment. At present Nvidia’s most advanced mobile graphics chips are made using 40nm process technology and support DirectX 10.1 application programming interface.
Even though the details about Nvidia’s next-generation Fermi architecture are quite sketchy, it is rather obvious that the new chips will be quite complex and simple reduction of execution units may not be enough to trim power consumption to the level required for mobile computers. The chief scientist of Nvidia claims that Fermi architecture is quite modular and Nvidia may sacrifice some parts in order to reduce consumption of energy.
“We are paying a bit of a compute tax in that we launched a part where a lot of the consumer compute applications haven’t really taken hold yet. But over time as more consumer computer applications are developed that take advantage of our compute (consumer) features, I think it's going to give us a big leg up. [...] We're not talking about other (chips) at this point in time but you can imagine that we can scale this part by having fewer than the 512 cores and by having these cores have fewer of the features, for example less double-precision,” William Dally, chief scientist at Nvidia, said in an interview earlier this month.
Nvidia did not comment on the news-story.