Intel is talking about improvements it has made to transistor scaling for the 10nm process node, and claims that its version of 10nm will increase transistor density by 2.7x rather than doubling it.
On the face of it, three years between process shrinks, rather than the traditional two years, would appear to end Moore's Law. But Intel claims that's not so. The company says that the 14nm and 10nm process shrinks in particular more than doubled the transistor density. At 10nm, for example, the company names a couple of techniques that are enabling this "hyperscaling." Each logic cell (an arrangement of transistors to form a specific logic gate, such as a NAND gate or a flip flop) is surrounded by dummy gates: spacers to isolate one cell from its neighbor. Traditionally, two dummy gates have been used at the boundary of each cell; at 10nm, Intel is reducing this to a single dummy gate, thereby reducing the space occupied by each cell and allowing them to be packed more tightly.
Each gate has a number of contacts used to join them to the metal layers of the chip. Traditionally, the contact was offset from the gate. At 10nm, Intel is stacking the contacts on top of the gates, which it calls "contact over active gate." Again, this reduces the space each gate takes, increasing the transistor density.
Intel proposes a new metric for measuring transistor density:
Intel wants to describe processes in terms of millions of logic transistors per square millimeter, calculated using a 3:2 mix of NAND cells and scan flip flop cells. Using this metric, the company's 22nm process managed 15.3 megatransistors per millimeter squared (MTr/mm2). The current 14nm process is 37.5MTr/mm2, and at 10nm, the company will hit 100.8MTr/mm2. Competing 14nm/16nm processes only offer around 28MTr/mm2, and Intel estimates that competing 10nm processes will come in at around 50MTr/mm2.
See also: the International Roadmap for Devices and Systems.
A number of stories here have covered the advancement to 10nm chips: Samsung: Exynos, TSMC: MediaTech Helio X30 for example. A reoccuring comment in the discussions is if 10nm from Samsung is equivalent to 10nm for TSMC or Intel.
Intel's Mark Bohr discussed the difficulty of comparing process nodes during Manufacturing Day, specifically proposing to move the industry to transistor density as a comparative metric. Surprisingly enough, Intel claims their 10nm process is roughly twice as dense of the competition. Intel is not the only ones frustrated by comparing process nodes, as this recent article tries to compare current "14nm" nodes between the major vendors.
To further confuse the discussion is new 22nm processes: Global Foundries 22nm FD-SOI and Intel's just announced 22FFL process, both targeting energy efficient devices. GF's is in high volume manufacturing already while Intel's is just announced, but further cement's Intel's delve into foundry work.
These topics are largely covered by EETimes' summary of Intel's recent announcements
(Score: 3, Interesting) by kaszz on Thursday March 30 2017, @07:37PM
The price point will have a huge impact whether this sells or not in volumes large enough to motivate the R&D+fab investment. The transistors per area unit makes sense. But there may be caveats in interference, capacitance, heat dissipation and a lot of other issues when packing it tighter. I suppose the feature will measure the performance in logic switches per volume.
Will be interesting to finding out the GHz of this new CPU. Supposedly there are 100 GHz transistors in the lab but designing a chip with them would just overheat. If this new technology just will enable more cores per chip then it may be cheaper to have more chips instead at the price of inter-core latency.