Intel's 8th-gen Core CPUs could boost laptop performance by 40 percent
the first time, Intel’s upcoming eighth-generation Core CPUs will feature quad-core processors inside of its chips aimed at ultrabooks, offering performance as much as 40 percent faster than in the previous generation, the company said.
Intel hasn’t disclosed the prices of its four new 15-watt 8th-gen Core U-series chips, though the company revealed Monday how fast they’ll run: The slowest Core i5-8250U will run at 1.6GHz, with a boost clock of 3.4GHz; the fastest Core i7-8650U will run at 1.9GHz, boosting up to 4.2GHz.
All four U-series chips include four cores and eight threads. PCs using the new 8th-gen Core chips should begin shipping soon, Intel said, with about 80 new system designs ready to go by the 2017 holiday season.
The far more interesting story, at least for enthusiasts, will be how Intel has loosened the definition of “generation,” as a slowing Moore’s Law has thrown off the company's legendary tick-tock manufacturing process. These 8th-generation CPUs are not the upcoming Coffee Lake, as previously expected. Intel characterizes its new chips as a "Kaby Lake refresh," referring to the current Kaby Lake chips as "prior generation" parts. Here’s another surprise: both Coffee Lake and the upcoming 10-nm Cannon Lake chips will all be part of Intel’s eighth-generation Core branding.
Confused? Let’s sum up: Intel’s eighth-generation Core chips will include three separate chip architectures and two process technologies, all under a single brand name. Whew!
What this means for you: Intel has muddied its brands before: In 2013, Intel began taking traditional P brands like Celeron and Pentium,
throwing them on top of non-PC Atom chips. The
current Core i9 family includes both Skylake-X and Kaby Lake-X chips. But we’ve never seen a scenario where Intel apparently has thrown branding to the winds. More than ever, savvy PC buyers and builders will have to research which upcoming parts are going into their builds.
Optimizations yield solid 8th-gen Core performance
It’s important to note two key points about Intel’s new eighth-generation chips. First, that a combination of process technology, design, and optimizations yielded the 40-percent improvement. Second, that performance boost is more than double what
Intel promised with the launch of the Kaby Lake chip, which saw increases of between 12 and 19 percent over the Skylake generation.
Here are the details of Intel’s new U-series chips:
Intel would have seen a 25 percent generation-to-generation boost with its 8th-gen, U-series chips simply by increasing the number of cores from two to a total of four, according to Karen Regis, director of mobile platform marketing for Intel. Tightening some of the manufacturing parameters allowed for higher clock speeds, she said.
Though rival AMD hasn’t disclosed its plans for its mobile “Raven Ridge” chips, AMD’s low-power Ryzen 3 chips use four cores and (only) four threads. Applying more cores to multithreaded applications has been one of the ways AMD has touted a performance advantage.
As it normally does, Intel demonstrated the hypothetical performance of the eighth-gen Core chips in a number of ways. Intel’s new U-series chips will create a 4K video in three minutes on PowerDirector Ultra HD (HEVC), 14.5 times faster than a five-year-old Core i5 Ivy Bridge PC. In Adobe Lightroom, the new chips will be 28 percent faster than 7th-gen Kaby Lake parts, or 2.3 times faster than a 5-year-old PC. Intel’s own content tests show the 8th-gen U-series parts will be 48 percent faster in slideshow creation than its 7th-gen predecessors. (According to information provided by an Intel spokesman, Intel's new U-series chips use Intel's Turbo Boost Max Technology 3.0, boosting not one but two cores at the maximum rate.)
In several cases, Intel made its generational comparisons versus the Core i7-7500U, the Kaby Lake chip launched a year ago--apparently carving out the “Kaby Lake refresh” chips as a half- or semi-generational family. PC makers PCWorld has spoken to have begun referring to the family as "Kaby Lake-R." Theoretically, that would mean that Broadwell, Skylake, Kaby Lake, Kaby Lake-R, and probably Coffee Lake would all be based on 14-nm technology—five separate families of parts manufactured on a single process, versus the once-traditional two.
Officially, Intel’s new U-series chips are "based on the Kaby Lake architecture and 14nm+ process technology with additional design improvements and new optimizations," a spokeswoman said. "We’ve redefined and reimagined what 8th-generation will mean," Regis added.
[...]
What the future holds for 8th-gen Core
Tweaking an established design and process has paid off in terms of availability. Consumer desktops based on the new 8th-generation Core designs will ship a bit earlier this year than they did with the 7th-generation devices, Walker said. Early next year, Intel will ship 8th-gen parts for commercial clients, both desktop and mobile, then gaming notebooks. Intel plans fanless PCs using the 4.5W Y-series near the beginning of 2018, he added. The company is also talking about "other options purpose-built for different segments," Walker said, but the company hasn’t disclosed what exactly those will be. "It’s a fairly consistent timeframe and rollout,” Walker said.
http://www.pcworld.com/article/3217...pu-price-release-date-specs-features-faq.html