Skip to main content

Put down that wallet before buying a PC — Intel just teased its eighth-gen CPU

Intel is such a tease. After spending for what seemed like forever talking about its seventh-generation “Kaby Lake” desktop processors, the company finally revealed the chips in January during the CES 2017 tech convention. Now, just mere weeks later, Intel is already teasing us with a new dangling carrot: Eighth-generation Intel Core processors.

The tease showed up on Intel’s official Twitter account revealing that its eighth-generation CPUs are on the 2017 roadmap for the second half of the year. However, the real meat of the story behind the social tease is that Intel “mentioned” its upcoming Core i7-8000 processor series during its investors meeting on Thursday. The chips will be based on Intel’s 14nm plus processing technology it used with the seventh-generation chips.

Recommended Videos

If you’re scratching your head with this one, you’re not alone. Intel was expected to launch three 14nm+ processor families in the second half of 2017 and into the first quarter of 2018: Kaby Lake-X, Skylake-X, and Coffee Lake. The supposed high-end Skylake-X and Kaby Lake-X processors are actually expected to launch during Gamescom 2017 in August supporting Intel’s flagship X299 platform for enthusiasts.

Skylake-X and Kaby Lake-X will be based on the 14nm plus process node and marketed under the Core i7-7000 K-Series brand. Skylake-X will have a maximum core count of 10 and Kaby Lake-X will have a maximum core count of four. At least four SKUs will be made available at launch.

That leaves Coffee Lake, which was expected to arrive in the first quarter of 2018. These will be six-core mainstream processors based on the 14nm plus process node, and split up into Coffee Lake-U (ultra-low power), Coffee Lake-X (enthusiast desktop), Coffee Lake-S (mainstream desktop), and Coffee Lake-H (notebook) SKUs. They reportedly not only improve on the Kaby Lake design, but include “architectural enhancements.”

That’s supposedly everything Intel is expected to roll out from its 14nm plus process technology. The 2-in-1 device Intel showcased during its CES 2017 keynote was a Cannon Lake processor based on the company’s 10nm process technology. Cannon Lake chips are expected to arrive in 2-in-1s and similar products during the back half of 2017.

So what Intel is likely teasing on Twitter is Coffee Lake. That’s just a guess, and Intel wasn’t willing to talk about the eighth-generation lineup outside what it revealed through Twitter. Intel told us that the new chips were only “briefly mentioned” during Thursday’s investor meeting. Sigh.

Ultimately, customer will see 20-30 percent better performance with the eighth-generation Core i7-8000 Series chips than they did with the Skylake family. That’s due to a 10 percent performance increase when moving from Skylake to Kaby Lake, and a 15 percent performance increase when moving from Kaby Lake to Coffee Lake.

Since Intel won’t spill the official goods behind its Twitter post, we’ll have to wait for the Computex and/or CeBit tech conventions later this year. We already know Intel isn’t sweating AMD’s Ryzen invasion, and this is probably one of the reasons why.

Kevin Parrish
Former Digital Trends Contributor
Kevin started taking PCs apart in the 90s when Quake was on the way and his PC lacked the required components. Since then…
Intel Alder Lake processors might outperform previous gen by over 50%
An engineer smiles as she holds up two processor chips.

Update: Intel's new line of Alder Lake processors have now been officially launched at the company's Innovation event. Intel's claims include the 50% boost in multi-core performance over the previous generation.

Today, new CPU-Z benchmarks emerged, showcasing the performance of the upcoming Intel Core i5-11600K, as part of Intel's 12th-gen Alder Lake processors. Designed as an affordable mid-tier option, this CPU outperformed the previous generation by miles -- and this applies to chips from both Intel and AMD models from the past.

Read more
Intel Arc GPU users lose Deep Link features as support ends without notice
The back of the Intel Arc B580 graphics card.

Intel has quietly discontinued its Deep Link technology, the suite of features designed to enhance collaboration between its CPUs and GPUs. Notably, the confirmation did not come through an official announcement, but via a developer comment on a public GitHub thread, where an Intel representative acknowledged that Deep Link is “no longer actively maintained.”

Launched in 2020 alongside Intel’s push into discrete graphics, Deep Link aimed to improve performance and efficiency in systems combining Intel 11th, 12th, or 13th generation processors with Intel Arc GPUs. It bundled several features like Dynamic Power Share which redirected power between the CPU and GPU based on load, Hyper Encode that enabled multi-engine video encoding, and Stream Assist for offloading media tasks to the GPU during live streaming.

Read more
AMD CPUs should support CUDIMM memory soon, but not this generation
Official product render of the G.Skill Trident Z5 Neo memory for AMD.

AMD processors can't make full use of CUDIMM memory just yet, but it may well do before the end of this socket. In a recent interview with DigitalTrends, AMD's product management lead for gaming and workstations, Sourabh Dhir, told us that there was no reason that AM5 couldn't support CUDIMM, but wouldn't be draw on a timeline of when we might see it.

Considering we expect AM5 to be AMD's flagship CPU socket for the next couple of generations at least, that probably means we don't have long to wait for the added memory speed support.

Read more