Featured Articles

AMD sheds light on stacked DRAM APUs

AMD sheds light on stacked DRAM APUs

AMD is fast tracking stacked DRAM deployment and a new presentation leaked by the company  points to APUs with stacked DRAM,…

More...
Nvidia officially launches the 8-inch Shield Tablet

Nvidia officially launches the 8-inch Shield Tablet

As expected and reported earlier, Nvidia has now officially announced its newest Shield device, the new 8-inch Shield Tablet. While the…

More...
Intel launches new mobile Haswell and Bay Trail parts

Intel launches new mobile Haswell and Bay Trail parts

Intel has introduced seven new Haswell mobile parts and four Bay Trail SoC chips, but most of them are merely clock…

More...
Aerocool Dead Silence reviewed

Aerocool Dead Silence reviewed

Aerocool is well known for its gamer cases with aggressive styling. However, the Dead Silence chassis offers consumers a new choice,…

More...
AMD A8-7600 Kaveri APU reviewed

AMD A8-7600 Kaveri APU reviewed

Today we'll take a closer look at AMD's A8-7600 APU Kaveri APU, more specifically we'll examine the GPU performance you can…

More...
Frontpage Slideshow | Copyright © 2006-2010 orks, a business unit of Nuevvo Webware Ltd.
Sunday, 15 September 2013 11:24

Intel Chief Engineer talks Haswell’s 5-year journey

Written by Fuad Abazovic

How Intel came to care about low power

As we all know, some of the key fundamentals of CPU design take years to develop. Intel’s new generation every second year is small miracle in itself, as the development cycles is a bit longer. Haswell that launched this June at 22nm was actually not drafted to be such an efficient mobile processor because work on Haswell started years ago.

Rani N. Borkar, Vice President General Manager, Intel Architecture Development Group mention that five years ago, when her group started the long journey of developing Haswell, the original plan was not to make it a great mobile chip. She was the leading development of Haswell and had big involvement in Bay Trail development, making her very qualified to explain the journey to 22nm products.

Five years ago Intel cared mostly about the performance of the core, not so much about power efficiency, but this was about to change. Intel recognized the need to make lower TDP on chips and necessity to make the transition in what Rani calls changing Intel’s mobile culture.

This is when the journey started. Four generations of Intel’s Core architecture got better and better, eventually culminating in Haswell as the crown jewel in Intel’s current portfolio. Intel went back to the drawing boards and made necessary optimisations and changes in order to lower TDP, create great performance that won’t kill battery life making Haswell one of the most scalable cores in the company’s history.

Haswell works scales from 84W TDP in a higher end desktop or business machines down to 17W in ultrabooks and even 4.5W in tablets that should be shortly announced. This is not all as with Broadwell transistors go almost 30 percent smaller and Intel showcased at the IDF keynote that we can see an additional 30 saving compared to Haswell. The future of Intel is high performance with turbo support, but the emphasis has clearly shifted to scalable designs, so it can tweak a CPU in order to save as much power as possible, while at the same time using the same core for high-end desktops. This is the current path.

This is also leading to another interesting trend – a lot more overlap. Haswell is going down to 4.5W, while some new Atoms will end up with Celeron and Pentium branding. To some extent the same trend applies to AMD, as some of its Jaguar-based products will replace some entry level Richland chips.

blog comments powered by Disqus

 

Facebook activity

Latest Commented Articles

Recent Comments