It is a newspaper's duty to print the news and raise hell - William Storey, Chicago Times
|
|
|
ARM chief calls for low-drain wireless
THE POWER DRAWN by wireless links is far too high and the industry has a "duty" to cut it, ARM president Tudor Brown has warned.
Speaking in a keynote at The Future of Wireless conference in Cambridge today, he quoted figures from wireless chip designer Atheros indicating that WiFi on a laptop is only between one to two per cent efficient, which means that it draws between two and eight watts for transmit powers between 20 and 200 milliwatts.
He quoted a 2007 consultation paper as saying, "Arguably what is needed are wireless access systems that can support multimedia service data rates at two to three orders of magnitude lower transmission power than is currently used."
Brown said that the issue is not simply about extending battery life by reducing consumption. He pointed out that the world's billion mobile phones and 100 million laptops translate into an equivalent number of batteries.
The volume of handsets' batteries alone amount to 20 cubic metres, the size of a fair-sized hall, and laptops account for 30 cubic metres. "Imagine how much [toxic] Lithium is in all that lot," he said.
The inefficiency also accounts for a lot of wasted electrical power, with radio base stations accounting for around twice the consumption of other network equipment.
"Between us we have a duty to drive down the power we are using... we need to do a lot better," Brown said.
He said nothing about how wireless consumption can be reduced, but other speakers at the conference spoke about femtocells, small footprint base stations that can reduce transmit power requirements.
Brown, in response to a question, said the need to cut power consumption extends to other areas of computing.
He warned that if Intel keeps pushing out Xeon servers that are generally used at only a fraction of their capacity, the energy consumption of server farms will exceed that of aircraft travel by 2020.
Currently servers themselves account for around 2.5 percent of the US energy bill, and cooling them accounts for as much again. µ
Share this:
Share
DavidC1 has jumped to misleading conclusions about Anandtech’s interesting figures. 3g is operating at between 10 and 100 times the range of Wifi - and remember signal drop-off is proportional to the square of the distance, requiring transmit power to be commensurately higher. In practice, I understand, signal degradation is worse than indicated by the inverse square law. And 3g generally operates in a far less friendly environment than that of Wifi. So in lasting only 33 percent less battery time than Wifi, 3g is actually doing well. Tudor Brown’s valid point was that Wifi could do its job on less power.
A better comparison is between Wifi and 3g talking to a femtocell, which provides a level playing field. Femto specialist Uniquisys reckons in this situation 3g transmit power is around 200 times less than Wifi – see my more recent article at
http://www.theinquirer.net/inquirer/feature/1720377/a-whispering-wireless-revolution
which riffs on Brown’s other point -that one way to make communication more efficient is to change network topology.
http://www.anandtech.com/show/3794/the-iphone-4-review/13
Right, that is why WiFi beats all other wireless standards used in cellphones that use ARM processors for battery life.
The ARM chief is clearly not bright.
Wow, I never thought of it that way. But I think this guy forgets the power needed to transcode the signal, encode and decode security, maintain signal and a whole slew of other tasks other than the actual transmittion of signal. Still, It'd be nice to get far more battery with WiFi turned on.