Intel talk about releasing a low-power processor that halves the CPU power requirements and all of a sudden there’s a second coming.
Even Robert Scoble seems to be celebrating although I’m not sure if he knows why. I think he’s just agreeing with Engadget. The comments in his blog are a better read. Scoble said the VW demo was ‘awesome.’ If he was to take a look the picture below, he’ll see that it it was the application that was awesome, not the hardware. Thats a Samsung Q1 there!
Is all this talk just link bait or has everyone gone mad?
The fact is that halving the power requirements of the Pentium-M processor from 5W to 2.5 is not going to make much difference at all. Halving the average power-requirements from about 2W to 1W is going to make even less difference.
The average ultra mobile PC is using 12watts of power so it’s easy to see how much more battery life you’ll get by reducing that figure by a maximum 2.5watts
Well done to Intel for working hard on lowering the power requirements for CPUs but 1) VIA already produce a chip and chipset that’s smaller and more efficient and 2) It’s not all about the CPU.
Take screen technology for example. The average 7″ ultra mobile PC screen with cold cathode back lighting takes something like 30% of the total power drain under normal operation. By lowering the backlight on my ultra mobile PC by 50% I can save the same 2.5W. Sony and Fujitsu (P1610d) are moving in the right direction here with LED backlit screens but for Origami-based units these need to be changed too. Then there’s Wifi. Some manufacturers are still putting in cheap USB-based wifi modules that take 2-3Watts of power. These need to be reduced right down to 1W or less parts. GPUs need to be more efficient too. If you want to run Vista with any sort of aero-glass feature, add another 1watt on to your power drain. And finally, batteries. UMPCs in 2006 all came with 25-30W/hr batteries. A step-change here is going to be the most important development for UMPCs. We need 40w/hr batteries at 250g before it all comes together nicely. It will take time before that sort of technology reaches mass-market pricing levels.
So here’s a warning to all readers, don’t expect a 2nd generation of power efficiency with UMPCs in 2007 unless they combine efficient CPUs with technology improvements across the board. Its not all about the CPU. Microsoft know that and that’s why they publicised the UMPC2007 specification back in July.
This article gets my award for being the most uninformative about the ultra mobile PC 2007 news. The PC World techlog ‘as far as I can tell’ is a classic zero-info post. Stick with Carrypad and I’ll try and bring you the honest, real world and non-baited headlines about the ultra mobile PC market.
Steve / Chippy.
Well done, Steve! Your pragmatic analysis is the reason I keep coming back to Carrypad. Keep it up.
Spot on Steve! This is exactly the same sort of marketing tactic that Intel had used for years in the past on the desktop side to help cover their ineptitude and inability to produce a decent processor that could competed with the offerings of AMD. Back then they just tried to distract everyone with the clock speed race as they tried fruitlessly to develop a chip that could rival the performace of the AMDs. Then there was the failing ITANIUM and their decisions to abandon the 64-bit game and focus on multicore 32-bit architecture in the hopes that no one would notice. Now, they know they can’t compete with VIA and they can’t do much better than they have with their Embedded processors, so they do the whole thing again with power consumption statistics.
Misdirection!
Well I guess if I had run out of ideas after 1995, then I’d have to do something to get attention as I peddled my overpriced antiques while watching my big Manufacturers’s contracts expire and I try tohold on to the slipping reputation I once held and the name the world knows. After all, they have only survived because of the contracts they have had with major brands , Dell, Gateway et al. and because of their marketing department. Alas parlour tricks and gimmicks won’t keep you alive forever.
My $0.02,
diJenerate