Intel Claims Cannonlake Core i7-8000 Series 15 Percent Faster Than Kaby Lake, Launches 2H 2017 [Updated]
Update, 2/12/17 -
This is a bigger lift in performance than what you might get going from Skylake to Kaby Lake in a best case scenario, and Kaby Lake's kicker is partly due to a simple clock speed bump. Intel's claim here is that Cannonlake will deliver a performance boost over Kaby Lake that is at least the same, though generally it will be higher than 15 percent. That is especially good news for anyone using a processor that is older than Kaby Lake, such as Skylake or Broadwell, as the performance gain will be more pronounced.
Intel made the performance claim at its annual investor day on Thursday. The 15 percent jump over Kaby Lake is based on results from SysMark, an application-based benchmark that evaluates performance in a number of areas, including office productivity, media creation, and data and financial analysis. If you're hoping to see how your system fares by running SysMark at home, you'll have to wait—Intel did not provide any specific numbers.
Cannonlake represents Intel's move to a 10nm manufacturing process. It has been a somewhat tumultuous road getting there, as Cannonlake was originally scheduled to arrive in early 2017. However, technical challenges and poor chip yields ultimately caused Intel to push Cannonlake back to the end of 2017 or early 2018, which led to Intel's decision to release Kaby Lake, a tick-tock disrupting architecture that is the third-consecutive 14nm chip family.
Intel might be feeling the pressure to solve whatever manufacturing issues might remain in the 10nm arena. AMD is weeks away from releasing Ryzen, a new-generation processor line that is supposed to make AMD competitive with Intel again at the high end. Early indications suggest that Ryzen is the real deal, while a recent price leak shows AMD aiming to undercut Intel in the marketplace by a wide margin.