by Anton Shilov
04/12/2013 | 05:51 PM
While Nvidia Corp. reveals some things about its next-generation products, it does not want to disclose the whole picture and it clearly does not provide any precise timeframes. While without providing further details, at the annual investor day conference this week the company did disclose some of its expectations for its next-gen graphics and compute architecture code-named Maxwell.
What we do know about the Maxwell family of chips so far from the official sources is that they will integrate general-purpose Denver ARMv8-compatible cores in addition to graphics stream processors and that they will be able to support unified virtual memory technology with microprocessors from Intel or AMD, a rather big deal for many applications. It is also logical to expect higher horsepower in general, which will boost video games, the main driver for Nvidia’s GeForce business.
Officially, Nvidia expects three major things from the Maxwell architecture: improved graphics capabilities, simplified programmability (probably thanks to ARMv8 and unified memory enhancements) as well as energy-efficiency.
“Number one for Maxwell, that is likely something that we are doing that breaks new ground in visual capability, something that is even more beautiful. […] Number two, it is likely that Maxwell breaks new ground in programmability, ease of programmability, because we want to expand the general purpose nature of the processor without sacrificing its speedup relative to a microprocessor. […] The last thing, the energy efficiency of Maxwell, it is going to crush Kepler. […] We know exactly how to measure it now and we know what it means to be good,” said Jen-Hsun Huang, chief executive officer of Nvidia, during a Q&A session at the company’s investor day conference.
The first GeForce consumer-class products based on Maxwell architecture are expected to emerge sometimes in 2014. Later on, Nvidia will release Maxwell-powered Tesla compute accelerator cards as well as Quadro professional graphics cards. Eventually, Maxwell architecture will be used for mobile application processors that belong to Parker family of Tegra products.