Lies, damn lies, and energy benchmarks
Analyst Opinion - When I was at a vendor event recently, the guys there were making a big deal about all of the power saving features their parts had achieved, indicating energy savings at the component level now exceed 70% over previous generations. These were desktop PC parts, and when I asked about the power savings at the wall I got the usual “we have no control over the inefficiencies of power supplies” answer, and then it struck me: With all those wonderful power saving parts at the component level, how many of them were completely eaten up by massively inefficient power supplies? The reality is this: People buying notebook and desktop PCs aren’t given anything but an occasional EnergyStar sticker (which doesn’t appear to mean much) to help them make energy smart choices. I believe it's well past time for something more.
On laptops, things are generally worse than on PCs. While we aren't generally given any idea of how much power they use, we are informed about battery life -- which is typically measured by turning all the radios off (Wi-Fi, Bluetooth), putting the system into an extreme low power mode, dimming the display to a level below an acceptable point for ambient lighting, and then taking a reading. This typically results in a promised battery life that runs about 2x the actual observed battery life (in fact, it's a good rule of thumb to take the battery life promised by a notebook maker and cut it in half).
I should point out that at least Apple measures their battery life with the Wi-Fi radio turned on -- making them more accurate than most. But it's obvious we still need more.
Why we need better power metrics
According to Verdiem and Lawrence Berkeley National Labs, better power management for every 10,000 computers can result in power savings of about 2 million KWH, which equates to the energy equivalent of 274 cars emitting greenhouse gases for one year. Applied across the one billion plus computers now in use, it's as if the savings took a full 27 million cars off the road -- about half the total number of cars produced in 2009.
I think we'd all like to make a contribution to energy savings, particularly because energy is expensive, but right now we can't due to factors beyond our control. For laptops, this is more than just a budget issue. When buying a laptop with a DVD drive, for example, there is the idea that a consumer might actually use it for watching DVDs when it's not plugged-in to the wall socket. But most systems simply don't have enough battery life to watch a full-length DVD movie, let alone do much else. The one laptop I have with a Blu-ray optical drive won’t even make it through one average length movie on battery life alone.
No EnergyGuide for Desktops
For desktop and laptop power users, we need the kind of energy graph that's used for appliances. Some kind of simple chart showing consumers the range of what systems like this typically fall in, and where this specific device lives. In the United States this type of chart for appliances is called an EnergyGuide. It tells the buyer what the annual expense will be, allowing each consumer to decide if paying a little more up front is worth it for more energy-efficient savings over time.
They do this for refrigerators and stoves, so why can’t we have the same data available for our PC purchases? It would help interested consumers make similar decision when buying PCs today.
Greater Granularity needed for Notebooks
For laptops we need a little more, and I thought we could take some cues from the smartphone industry, which seems to provide more granular metrics. Unfortunately, even with their greater granularity, in my experience you can’t always trust them.
Smartphones typically provide several metrics. For instance, the iPhone has a talk time of 5 hours on 3G vs. 10 hours on 2G. It has a standby time of 300 hours, Internet use time of 5 hours, video playback of 7 hours, and audio playback of 24 hours (according to Apple). While this is all very granular, it isn’t very accurate based on real world observations. And Apple isn’t alone in these errors as it appears across the cell phone spaces.
While these numbers are more granular, the results can’t be trusted. AMD’s Pat Moorhead has posted a detailed comparison of just how (in)accurate the cell phone guys are.
I still think granularity is good though, but we need better consistency and we need to be able to trust the results.
Wrapping Up: We need better energy metrics for PCs
People really want to be able to make smarter choices and to use their products more successfully while saving on energy costs, and even helping to save the planet. The need surely exists for tools which enable consumers to make those smarter decisions.
It's time for notebooks and PCs to be sold with metrics that not only help us buy smarter, but also reward those vendors which put in the extra effort toward making their machines more energy efficient.
We all have to live on this planet, and in this day and age that means making smarter energy decisions. It would be nice if somehow we could bring the necessary information to consumers so that the realities of shopping could actually live up to the ideals we aspire to.
Rob Enderle is one of the last Inquiry Analysts. Inquiry Analysts are paid to stay up to date on current events and identify trends and either explain the trends or make suggestions, tactical and strategic, on how to best take advantage of them. Currently he provides his services to most of the major technology and media companies.
The opinions expressed in this commentary are solely those of the writer.