At this year’s edition of the International Solid-State Circuits Conference, which takes place these days in San Francisco tech companies in the industry are expected to captured everybody’s attention with detailed information about the emerging technologies. The attendees expect to hear the latest news in a wide range of interests from integrated digital radio to optical interconnects. However, one chipmakers didn’t wait until the event’s debut to make big announcements.
Intel’s researchers announced two new technologies targeting energy efficiency in a prelude to the International Solid-State Circuits Conference. Justin Rattner, Intel CTO explained that the new ultra low voltage processor developed by Intel is capable of operating at 280mV. The prototype chip presented has a dynamic range reaching from 3MHz to 915MHz.
In addition to a new circuitry, Intel announced in a press brief, that the company also developed a new design method for floating point processing. This will allow the processors to vary in precision of its calculations dynamically which means a boost in energy efficiency that is close to sevenfold over traditional implementations. Gareth Halfacree from BitTech managed to get his hands on Intel presentation and found out that details about the technology. “The variable-precision floating point fused-multiply add unit hits 52 to 162 gigaflops per watt at 1.45GHz. The technology will be used in Intel’s future graphics products, the company confirmed, along with a vertex and pixel shading lighting accelerator capable of processing 2.05 gigavertices a second while drawing just 151mW,” writes Gareth Halfacree.
However, at the International Solid-State Circuits Conference Intel is expected to host over twenty presentations, demonstrations and forum sessions regarding the company’s new technologies. The giant chipmaker seems dedicated to focus on the products with applications in supercomputing, but also managed to release on the market a few technologies for the end-user.