Did Moore's Law just get a new lease on life?
In 2015, Intel admitted that it was taking a bit longer than two years to roll out their next generation of chips, ostensibly no longer keeping up with Moore's Law, the generation’s-old concept that more or less promises chip technology will double our processing power every two years.
However, a new breakthrough from IBM, announced this week in Paris, could help revitalize the circuit industry and transform our digital devices.
The good news comes by way of 50-year-old technology: Phase-change memory. It's the concept that certain crystalline materials can exhibit reversible resistance properties in a phase change — the change from a solid to, say, a fluid.
In 2000, as tech companies like Intel were grappling with the very real limits of Moore's Law, it turned to Phase-change memory (PCM) as a possible solution.
PCM is especially attractive to tech companies because it's a memory technology that appears to share the best qualities of DRAM and Flash memory, while avoiding their significant deficits. Unlike Flash, PCM doesn't need a separate step to switch information from, say, a 1 to a 0 (if we're talking basic bits) and it's nonvolatile, meaning that, unlike RAM, it doesn't need a power supply to retain information.
Make it affordable
The key to making memory cost effective is size and bits. Putting more bits in a cell is critical. Standard Flash can store two bits per cell (Triple Level Cell Flash can do 3 bits). IBM's breakthrough puts, for the first time, 3 bits per PCM cell. The previous record for PCM was one bit per cell. The result is TLC PCM (Triple Level Cell Phase-change memory).
IBM Research scientists presented their findings the IEEE International Memory Workshop.
“Reaching 3 bits per cell is a significant milestone because at this density the cost of PCM will be significantly less than DRAM and closer to flash,” said Dr. Haris Pozidis, co-author of the paper and manager of non-volatile memory research at IBM Research in Zurich.
With traditional circuits, the laws of physics will dictate just how small a chip can get while still increasing circuit density. With PCM, the volume of material used to make it shrinks along with the size of each cell, which makes it possible to keep adding cells in an even tinier space. This could help PCM-based circuits rewrite the rules of Moore's Law.
Chip manufacturer Intel applauded IBM's breakthrough, but was quick to point out that it has its own nonvolatile memory technology: 3D Xpoint. "We believe that innovation in memory is great for the industry, and we’re excited about the performance and density benefits of our 3D Xpoint technology, which was announced last year and is currently in production," said an Intel spokesperson.
As for what's next, an IBM Research spokesperson told Mashable that it is now looking for a licensing partner. "The technology challenges are all finally addressed — of course new challenges may arise during mass production, but those aren't expected to be significant. Assuming we can license the technology in a relatively short time-frame, TLC PCM can be available as early as next year. Whether the applications are for the enterprise or consumer market will depend on the partner."
[UPDATE: Added Intel's comments ON MAY 18, 11:55 AM ET]