The display was mentioned as the largest typical consumer of power although if you are always compiling or you don't have a standard display then the processor will consume more. Higher DPI, better color saturation, thinner backlights and of coarse larger, brighter screens usually also increase the the display consumption but as with any product the question becomes what trade off do you want to make. Getting the job done quickly is usually better then running slowly because this is a time * power consumption problem. I think this has been proven in the case of some of the newer faster WIFI protocols, although there maybe a few exceptions to the do it as fast as possible rule. SSDs when they are designed correctly are also solving the hard drive power consumption issue assuming you are willing to pay more for fewer GB. As most Engineers already know, lowering the voltage and having efficient software are 2 of the better ways to lower power consumption in all usage models.
This article is not only a good introduction, but it also includes pointers at some solutions. The largest consumers of power are the processors, with memory access taking a large part of that. So one solution is code that runs with less memory access, and runs for less time. Returning to sleep mode whenever possible can reduce wasted energy without impairing operation, and running some portions of the code at slower clock speeds could also help. In fact, slowing the edge speed can reduce energy consumption, but the means to reduce edge speed when the clock is slowed is usually unavailable. Sectioning a system so that only the portions needed for a task is a great idea, but it certainly adds complexity and a whole additional level for error introduction. Sectioning has been presented a few times, but it may not be as widespread as it needs to be, yet.
"Chips consume power."
Technically, chips "consume" energy. Power is a measure of the rate of energy consumption.
The term power is often misused in this fashion where people say "power" when they really mean "energy". Perhaps it's because everyone gets a monthly bill from the power company, but if you check, your electricity is actually measured in kilowatt hours (power * time = energy).
This is a nice intro to designing for low power, and thanks for pointing out that power is a system design issue that includes software as well as hardware.
Looking forward to the next segment in this series.
Replay available now: A handful of emerging network technologies are competing to be the preferred wide-area connection for the Internet of Things. All claim lower costs and power use than cellular but none have wide deployment yet. Listen in as proponents of leading contenders make their case to be the metro or national IoT network of the future. Rick Merritt, EE Times Silicon Valley Bureau Chief, moderators this discussion. Join in and ask his guests questions.