News
News Categories

Intel’s 8th-generation processors will remain on 14nm for the fourth time in a row

By Koh Wanzi - on 13 Feb 2017, 10:42am

Intel’s 8th-generation processors will remain on 14nm for the fourth time in a row

Image Source: Intel

If ever there was any doubt that Intel’s tick-tock cadence of upgrades was gone, it’s vanished now. At Intel’s annual Investor Day in California last week, the company confirmed that its 8th-generation microarchitecture will remain on the 14nm node, the fourth generation of chips to do so. 

This move is significant because the 7th-generation Kaby Lake processors are supposed to be the “optimization” stage in Intel’s new Process-Architecture-Optimization strategy, which means that the next generation should be a “Process” stage that sees a move to a smaller node. 

As it turns out, that isn’t going to be the case, and Intel’s dragging out of the 14nm node is a telling sign of the persistent difficulties the company is facing in delivering process shrinks (despite the recent announcement of a US$7 billion investment in a factory that will produce 7nm chips). Still, the company is claiming that it will maintain the 15 per cent performance increase that Kaby Lake offered over Skylake. 

Intel 8th generation 14nm

Furthermore, rumors suggest that the 8th-generation chips may focus more on the low-power U- and Y-series chips featured in thin and light notebooks, similar to how Kaby Lake debuted on those processors first last year.

Another interesting thing is the characterization from Dr. Venkata ‘Murthy’ Renduchintala, VP and GM of Client and IoT, of process technology going forward as “fluid based on segment”, which may suggest a mix of process nodes for different segments in the same generation. However, there’s no saying if this will mean a split between consumer and enterprise products, or a division within the consumer segment itself.

That said, Intel did say that it would be prioritizing data centers for the next process node. According to AnandTech, this will be achieved by using multiple small dies on a single package, as opposed to a multi-chip package in previous multi-core designs.

But despite all that, Intel is apparently still confident in its foundry business. In fact, the company expects to have a three-year process node advantage even when competitors like Samsung and TSMC roll out 10nm.

Source: Intel via AnandTech

Join HWZ's Telegram channel here and catch all the latest tech news!
Our articles may contain affiliate links. If you buy through these links, we may earn a small commission.