Hands on with Windows 8 CP: Battery life test
Tablets and Ultrabooks are on the rise and Windows-on-ARM (WOA) tablets are on the horizon. For Windows 8 to succeed in today's -- and tomorrow's -- market for both low- and high-powered portable devices, Microsoft needed to get rid of its "fat" Windows without losing functionality. The goal was to not just increase responsiveness, but also to improve battery life, which is obviously one of the core criteria for everyone shopping for a new laptop or tablet. How does Windows 8 fare? We've got answers":
[ FREE DOWNLOAD: Windows 8 Deep Dive Report | Hands on with Windows 8 CP: New and improved built-in tools ]
Microsoft's promise: Less energy consumption
Redmond pushed out almost a dozen blog posts that either demonstrate their battery life/power consumption improvements or show off how new their new features (e.g. USB 3.0 support, Connected Standby, Live Tiles, Metro Apps) are built with saving power in mind.
Here are the highlights:
- All background Metro-style apps will be suspended; only the foreground app consumes resources and thus battery life. If an app is not on screen and actively used, it shouldn't drain battery. Only a set of apps will be allowed to use background activity (e.g. for music playback or printing).
- Improved idle usage of the OS.
- On-system drivers, such as the USB host controllers and keyboard drivers, put the device into low-powered modes sooner.
- Memory deduplication mechanisms to reduce RAM workload.
Battery life benchmark: Windows 8 consumer preview
"We think of power as a critical system resource, just like CPU utilization, hard disk activity, or memory consumption." explains Pat Stemen, a Program Manager on the Windows Kernel team, on the B8 blog.
So how power-friendly is the new Windows exactly? We wanted to put Windows 8 CP up for a test run on a variety of laptops and see how it stacks up against Windows 7 SP1. Here's the hardware I used:
Laptop A -- Acer Aspire 7551G. A Run of the mill 17-inch laptop with a Phenom II X4 CPU and 4 GB of RAM. Good for multimedia, but not particularly well-suited in the battery life department.
Lapotp B -- Samsung NC 10. One of the better 2008-era netbooks. The usual yada-yada specs (Atom N270 1.6 GHz, 1 GB RAM). Low performance, extremely good battery life.
Laptop C -- 13" MacBook Air (2011) equipped with a 1.8 GHz Core i7, 4 GB of RAM and a 256 GB SSD drive.
* I didn't test on an ultrabook, but the MacBook Air running Windows matches the most current ultrabooks in terms of specs, battery power and form factor.
All three represent very different laptops, targeted at very different audiences. Evaluating these devices should give a good glimpse of what to expect on the battery life front.
I used PCMark 7's "Lightweight" test, which simulates typical workloads such as adding music to Media Player, browsing the web, copying files and scanning for malware while leaving a couple of seconds of idle time in between. I looped the test until the battery of the device ran dry. Second, I just let the laptop sit there, do nothing and just "die". While that's not a particular realistic scenario, it demonstrates idle usage which is very important for battery life: Remember, you're not constantly pushing your CPU to its limit -- while you're reading something, sitting in a meeting or just waiting in between presentation slides, your laptop needs to go into idle as fast as possible and stay there until you do something. It is a good basis for judging power management.
Since third party software usually impacts battery life, I installed only the very basic programs such as Office 2010, TeamViewer, SnagIT, Adobe Reader and Flash on both machines. In all cases, I used the "Balanced" power saver mode and cranked up brightness to about 50-70%, while leaving Wi-Fi on.
I repeated each test run exactly three times to avoid discrepancies. In each and every case I got pretty much same results (+/- 5 minutes). Results are measured in minutes:
Microsoft has done its homework. Idle performance of Windows 8 seems to best that of Windows 7 SP1 in every case. This is especially true on the MacBook Air: Here, idle battery life increased by 51 minutes. Even under the PCMark 7 workload, it managed to squeeze 24 minutes of additional battery life out of the machine. Our trusty old Samsung NC 10 ran for 40 more minutes in "idle" while it gave me another 20 minutes under typical workload. While that's not a lot, it may just mean the difference between wrapping up a movie or work on an airplane and having to stare at a blank screen for the rest of the flight.
Overall: While there are no major leaps, these are all solid improvements across the board.