Intel has developed an entirely new manufacturing process for the next generation of mobile phone chips.
Recognising that power leakage was becoming an issue, the chip giant has created a second 65nm manufacturing process, said Mark Bohr, a senior fellow and director of Intel's process architecture and implementation.
That process will significantly reduce power leakage in chips designed for devices that do not require as much performance as PCs or servers, he said.
The process is basically the same as the main 65nm process used to build upcoming chips such as Yonah, a dual-core version of the Pentium M notebook processor that is expected to arrive early next year. However, it uses a few different manufacturing techniques designed to contain leakage at the expense of transistor switching speeds.
Intel is getting ready to introduce its first chips built on the 65 nanometer process - itself a generational advance over the company's current chip-making.
Power consumption, and the related problem of dissipation, were central to Intel's development of its 65nm manufacturing technology. Transistors are now so small at the 90nm generation - currently used to build Intel's most advanced chips - that the electrical power used to run the transistors can leak out of the chip as heat.
This problem was exacerbated by Intel's marketing and design strategy in previous years, which called for chips with faster and faster clock speeds. More power is required to raise a chip's clock speed, or the number of instructions it completes in a second.
There are a variety of techniques to help prevent this but they all reduce the performance of transistors. But not every chip requires the same amount of performance as a PC processor, Bohr said. The new process will provide Intel's chip designers with the means to create low-power chips for phones, handhelds, or devices that haven't even reached the drawing board, he said.
Intel plans to have the low-power manufacturing process up and running in about a year's time.