What to Expect From Intel’s Kaby Lake
For the past decade Intel has been creating new CPUs based on its “tick-tock” model, where each tick represents a whole new process technology (starting with 65nm, back in 2006), and each tock involves a new architecture built on an existing process.
Intel is tweaking the architecture and refining the process technology
With Kaby Lake, Intel is moving away from tick-tock, and embracing a new philosophy: process-architecture-optimization. The first two correspond to tick-tock, but that third “optimization” stage is new, and Kaby Lake will be our first look at how it plays out. “Optimization” means Intel is tweaking the architecture and refining the process technology.
On the architecture front, the major update is an enhanced video decoder block that will handle 4K HEVC encoding/decoding. The block is a fixed function implementation that supports the MainlO profile, which is increasingly important for things such as 4KVP9 YouTube videos. Using a fixed function design sacrifices flexibility for improved speed and efficiency, and Intel claims Kaby Lake laptops will be able to provide up to 9.5 hours of 4K video playback. Intel also states that encoding a one-hour video into 4K HEVC can be done in 12 minutes. Kaby Lake will also feature improvements to the Speed Shift tech, introduced with Skylake, allowing the CPU to more quickly enter and exit higher clock speed states. It’s the old “hurry up and go slow” mentality, letting the CPU complete a task as quickly as possible, so it can then return to a low-power idle state.
Outside these architectural changes, Kaby Lake appears to be largely the same as Skylake, with one qualification: Intel is tweaking the Unm process, with a refined fin profile (the FinFET part of the process) it’s calling 14nm+. Intel says Unm+ will be less transistor-dense than Unm, with a larger fin pitch and reduced channel strain. So, at roughly the same power requirements, Intel should be able to tack on a few extra hundred MHz in clock speed. For standard workloads, Intel claims Kaby Lake ultraportable and two-in-one laptops will be up to 19 percent faster than Skylake equivalents.
If you’re reading between the lines, many of the above changes obviously mean a lot for laptops. Like the past several CPU releases. Kaby Lake isn’t designed as a desktop-first architecture. It should already be available in laptops by the time you read this, but desktops are slated to launch in early 2017, probably coinciding with the annual Consumer Electronics Show. And, recently, we were able to find out some of the names and core specs for the upcoming desktop parts.
The Core i7-7700K will sit at the top of the stack, with a base clock of 4.2GHz—200MHz higher than the 6700K—and the maximum turbo will probably be 4.4-4.6GHz. Hopefully, it will be able to hit 5GHz with overclocking, but we ll need actual hardware to test. The Core i5-7600K will be the usual quad-core minus Hyper-Threading, with a base clock of 3.8GHz. That’s the highest base clock for a Core ¡5 Intel has ever released, and I expect it will be the first time Core i5 has officially broken the 4.0GHz barrier—not that we haven’t been doing that and much more with overclocking for years. Also apparently breaking the 4.0GHz barrier will be the Core ¡3-7300, clocked at an even 4.0GHz.
Along with the desktop CPUs come new chipsets: the Z270, H270, B250, Q270, and Q250. It’s too early to say what exactly these new chipsets will introduce, although we do know they’ll use the same LGA1151 sockets. These new boards should work with Skylake, and current Z170 motherboards should work with Kaby Lake after a BIOS update too.
Jarred Walton has been a PC and gaming enthusiast for over 30 years.
Do I really need a…Kaby Lake processor?
What does it do?
The seventh generation of Intel’s modem PC processor range, nicknamed Kaby Lake, was released on 3 January 2017. Its the next step in the x86 CPU architecture, following Skylake, and these are the chips you’ll see at the heart of most PCs this year.
Why would I want it?
New chips generally aim to offer better performance for the same price, and Kaby Lake is no exception. It has faster clock speeds and – because the number of instructions per clock cycle is the same as in Skylake chips – can do more work in the same amount of time. It also supports more PCI Express ‘lanes’ for expansion, although this is only useful to high-end users.
What’s the catch?
Kaby Lake really doesn’t change very much. Its based on the same 14-nanometre (nm) process as Skylake, meaning its transistors are the same size. Intel’s planned move to a more efficient 10nm had to be pushed back to the middle of next year, when Cannonlake processors should appear.
But underwhelming innovation isn’t Kaby Lake’s only problem. It’s also the first CPU series with no official support for versions of Windows before 10, and some users still prefer 7 or 8. And although some of the new ¡3 and ¡5 processors bring performance gains, top-end i7 desktop PCs see little or no benefit. Meanwhile, many Kaby Lake laptops are coming up short in battery-life tests.
So can I do without it?
So far, the answer seems to be yes, but if you buy a new PC you’ll get it anyway. We’re hoping for better from Cannonlake.