Intel rebrands their Broadwell Y line as Core M

Scorpus

Posts: 2,162   +239
Staff member

During Intel’s Computex 2014 keynote, the company announced a ‘new’ line of CPUs designed for low-power tablets and similar mobile devices. The new line, called Core M, isn’t unfamiliar to those keeping up with Intel’s roadmap: it essentially refers to their upcoming Broadwell Y parts, which are manufactured using a 14nm process.

Core M parts will be found in thin-and-light x86-powered tablets, including Asus’ recently announced Transformer Book T300 Chi, towards the end of 2014. Intel showed off a reference tablet as thin as 7.2mm with a Core M part inside, which was completely fanless due to the CPU’s low power consumption (somewhere around 10W).

Intel claims Core M is the most power efficient processor they’ve ever made, which will allow thin and light x86 devices similar to Intel’s reference design to be produced. If low-power Haswell Y parts are anything to go by, Core M parts should easily outperform the best ARM chips available. How Core M stacks up against ARM on a price and battery consumption level is a different story.

For those who aren’t aware, Broadwell is Intel’s 14nm die shrink of their Haswell microarchitecture. Each die shrink provides a better level of energy efficiency, and Intel can typically also squeeze out some more performance through various optimizations.

I’ve taken a look at the T300 Chi from Asus, which is powered by a Broadwell-based Core M CPU, at their press event during Computex, and the impressive tablet goes to show how Intel’s new chips will bring about better x86 tablet designs.

Permalink to story.

 
What does "Blazing Fast Performance" do? I thought that the last one, that they sold to me, had that.
 
They should only be allowed to benchmark these products based on how much watts it costs to perform a particular task. The actual task duration does not matter, as long as it is completed within a particular user experience tolerance.
 
They should only be allowed to benchmark these products based on how much watts it costs to perform a particular task. The actual task duration does not matter, as long as it is completed within a particular user experience tolerance.

Do you want to account for how many times you had to punch the keyboard to get this idea formulated? And how will it compare to the time that you simply wasted? If any good, that's your personal benchmark of efficiency ;) And the effect is the same - nobody really cares.
 
Do you want to account for how many times you had to punch the keyboard to get this idea formulated? And how will it compare to the time that you simply wasted? If any good, that's your personal benchmark of efficiency
prekesh-wink.png
And the effect is the same - nobody really cares.

No. For example, the benchmark should should play a 720p video @ 25fps for 2 hours and measure how much power was consumed. Another should should render a 3D scene @ 60fps for 2 hours and measure how much power was consumed. If it can't do the 3D scene @ 60fps, then it is no use for such a task, which is fine if all you want to do is play FaceBook all day.
 
Back