Comments on: First 802.11n handset spotted in the wild – what took so long? http://www.wirevolution.com/2009/12/24/first-80211n-handset-spotted-in-the-wild-what-took-so-long/ Mobile Unified Communications Fri, 28 Oct 2011 16:47:54 +0000 hourly 1 http://wordpress.org/?v=3.3.1 By: Michael http://www.wirevolution.com/2009/12/24/first-80211n-handset-spotted-in-the-wild-what-took-so-long/comment-page-1/#comment-7772 Michael Fri, 08 Jan 2010 15:11:51 +0000 http://www.wirevolution.com/?p=1080#comment-7772 Modern power management techniques involve switching off the radio between packets; it is worth powering down for just a few microseconds. The radio is only powered up while bits are being transmitted or received, so if you are thinking of the radio being on for a second, that on-time is distributed over ten times as much clock-time in an 802.11n environment as an 802.11g environment. So you are right in the sense that a second of radio-on time will consume more power with 802.11n, but for streaming a movie that lasts two hours, the 802.11n radio will only be consuming power for a tenth of the time that the 802.11g radio would, and the battery life would actually increase, according to McFarland. Modern power management techniques involve switching off the radio between packets; it is worth powering down for just a few microseconds. The radio is only powered up while bits are being transmitted or received, so if you are thinking of the radio being on for a second, that on-time is distributed over ten times as much clock-time in an 802.11n environment as an 802.11g environment. So you are right in the sense that a second of radio-on time will consume more power with 802.11n, but for streaming a movie that lasts two hours, the 802.11n radio will only be consuming power for a tenth of the time that the 802.11g radio would, and the battery life would actually increase, according to McFarland.

]]> By: Rob http://www.wirevolution.com/2009/12/24/first-80211n-handset-spotted-in-the-wild-what-took-so-long/comment-page-1/#comment-7770 Rob Fri, 08 Jan 2010 10:23:55 +0000 http://www.wirevolution.com/?p=1080#comment-7770 Uhm ok actually, what is happening here: 22MBps at 140mW of POWER: If we assume our transmission lasts for one second we get 22MBit for 140mJ which gives us an energy consumption of 6.4nJ per bit. and a total consumption of 140mJ from the battery. So there we are. as above. Talking about 802.11n: 200MBps at 1W of POWER: If we assume the transmission lasts for 1 second (as above) then we get 200MBits for 1 Joule and each bit cost us 5nJ. so each bit is cheaper yes. BUT we sent 200 million of them, not 22 Million. and so our battery sucked up a massive 1 Joule of energy to send it all. Hence: our battery died approximately 5-6x faster. So yes, you get your data faster. And if downloading the same file should do it with less total energy consumption. BUT you will take much less time to do it on 802.11n... and your battery WILL die much faster since the average power consumption of the device WILL go up 5-6X ... :-) Uhm ok actually, what is happening here:

22MBps at 140mW of POWER: If we assume our transmission lasts for one second we get 22MBit for 140mJ which gives us an energy consumption of 6.4nJ per bit. and a total consumption of 140mJ from the battery.
So there we are. as above.

Talking about 802.11n:

200MBps at 1W of POWER: If we assume the transmission lasts for 1 second (as above) then we get 200MBits for 1 Joule and each bit cost us 5nJ. so each bit is cheaper yes. BUT we sent 200 million of them, not 22 Million. and so our battery sucked up a massive 1 Joule of energy to send it all.
Hence: our battery died approximately 5-6x faster.

So yes, you get your data faster. And if downloading the same file should do it with less total energy consumption. BUT you will take much less time to do it on 802.11n… and your battery WILL die much faster since the average power consumption of the device WILL go up 5-6X … :-)

]]> By: Bob http://www.wirevolution.com/2009/12/24/first-80211n-handset-spotted-in-the-wild-what-took-so-long/comment-page-1/#comment-7758 Bob Wed, 06 Jan 2010 07:43:52 +0000 http://www.wirevolution.com/?p=1080#comment-7758 The anecdotal reports may very well be correct. What is likely is that the control to power down the chipset is based on a timer, i.e., after a certain period of inactivity, power down some or all of the chipset. This timer is probably not modified in the hack enabling 11n. So, the .11 chipset power consumption increases when 11n is enabled and the time that chipset is active does not change, even though the bit rate is 10x greater. This would cause the observed shorter battery life. The anecdotal reports may very well be correct. What is likely is that the control to power down the chipset is based on a timer, i.e., after a certain period of inactivity, power down some or all of the chipset. This timer is probably not modified in the hack enabling 11n. So, the .11 chipset power consumption increases when 11n is enabled and the time that chipset is active does not change, even though the bit rate is 10x greater. This would cause the observed shorter battery life.

]]> By: Phil http://www.wirevolution.com/2009/12/24/first-80211n-handset-spotted-in-the-wild-what-took-so-long/comment-page-1/#comment-7663 Phil Sat, 26 Dec 2009 23:43:04 +0000 http://www.wirevolution.com/?p=1080#comment-7663 The anecdotal reports of increased consumption may come about because the experience of web surfing 9x faster is so much more pleasurable that (say) 50% more pages are visited, more than offsetting the nJ/bit gains. The anecdotal reports of increased consumption may come about because the experience of web surfing 9x faster is so much more pleasurable that (say) 50% more pages are visited, more than offsetting the nJ/bit gains.

]]>