While I’ve done my own share of mobile device battery testing, Steve Paine has easily done more. From smartphones to MIDs and UMPCs to netbooks, Steve has tested down to the milliwatt over the past few years. Today he observed that on the CPU side of the house, the power difference between ARM and x86 is drastically reduced over what it was. If you’ve been following the progress of Intel’s Atom platform, that’s no surprise. And it doesn’t take an engineering degree to know that larger backlit displays can consume more power that most other device components. So what’s the “sweet spot” for a device display to effectively cancel out the power efficiency of ARM over x86? Here’s what Steve says:
“When you get to screen sizes of 4” and above, something happens that levels the playing field for Intel somewhat. Their CPU platforms (*1) don’t idle down very well but in a typical ‘internet-connected’ scenario on one of these ‘smart’ devices, that becomes almost insignificant as the screen backlight adds such a huge load to the platform that when combined with Wifi, 3G, BT, GPS and audio, the CPU is just 10% of the total load. Swapping Intel out for ARM would save you just 5-10% battery life in an ‘active’ scenario.”
Steve’s point is rather timely, considering all of the ARM-powered devices we peeped at the Consumer Electronics Show. Many of them offered displays well over 4″, with some in the netbook-like 10″ range. It makes you wonder if pairing a low-power Atom (s intc) chip with Moblin or other form of Linux might make for a better experience than an ARM device running Android or a custom Linux distro. Put another way: if you could get potentially more processing power but not pay a power premium, would you?
Of course, display technologies are bound to mature. In fact, our video demo of the Pixel Qi display on a Notion Ink prototype tells me that this whole situation of power hungry displays is due for a refresh in the near future. But until then, Steve may have a pretty good point. Thoughts?