New add-on batteries from Dell and HP kept notebooks going and going in InfoWorld's tests. Download our benchmark to see what these extenders could do for you.
Many competing factors enter into the choice of a specific laptop: weight, size, price, battery life, expandability, and so on. Of these, battery life is rarely a primary concern until the system is actually purchased. The unspoken expectation is that standard batteries will get 3 to 4 hours of useful life from a single charge, while batteries with more cells can add a couple of hours of working time to this number.
This general rule, which has been valid for several years, is being challenged by new battery technology that can significantly extend productive time on today's notebooks. The newly extended window has an important consequence for a purchasing factor that is material to all users: the laptop's weight. If a battery can carry a system through an entire business day and beyond, then there is no longer any need to lug the power supply and cables in a roller bag or briefcase. This benefit is slightly tempered by the added weight of the batteries themselves.
This article discusses several extended-life battery models available on standard notebooks and the kind of performance you can reasonably expect. The good news is that some batteries can carry you for well past 15 hours of work on a single charge.
Systems and batteries
To perform this evaluation, InfoWorld approached Dell and Hewlett-Packard to provide us with notebooks that represent typical professional models and to supply their corresponding long-life batteries. We specifically asked that they not make any extraordinary configuration changes to the systems to extend battery life for our tests. Ultimately, I examined a Dell Latitude E4300 with a "battery slice" and HP EliteBook 6930p laptop with two models of long-life batteries.
It's important to note that we asked for models with substantially different configurations (processor, operating system, capacity, etc.) so that we could obtain multiple data points. As a result, the numbers should not be used for comparing models, because there is no established baseline in common. Rather, the numbers should provide insight into what kind of battery life you can expect for recent laptop models.
The benchmark jungle
Industry benchmarks for laptop battery usage are a mixed bag. Different vendors use different measures, including JEITA (Japan Electronics and Information Technology Industries Association), Mobile Mark, and combinations of the two. Typically, these benchmarks involve measuring battery usage in a near-hibernation state, then at various levels of activity, and averaging the results. The problem is that users have no way of knowing the extent to which this average corresponds to their own usage profile. As a result, consumers end up viewing the numbers in the same way they do U.S. mileage figures for automobiles: as a generous overestimate that cannot be reproduced except in laboratories.
To avoid this mess, I decided to write a simple Java program that would simulate a light load. The program, InfoCountdown, runs continuously, writing the current and elapsed times to the screen and to the disk every 10 seconds. This generates a regular, intermittent disk and screen I/O operation. I configured the notebooks so that the screen was permanently on, the screen saver was disabled, and hibernation was turned off. Hence, the timings should represent the maximum battery life for very light but continuous work.
I encourage you to download the InfoCountdown benchmark (the .zip archive includes both the executable and the source code) and try it on your own system. (To run it from the Windows command prompt, type in java -jar InfoCountdown.jar; to stop it, type in Control-c or kill the corresponding Java process in Task Manager.) To determine how much to adjust my results for your specific workload profile, run the benchmark on your notebook (from full charge and using the same conditions I did) and compare these results with the amount of work time you obtain from a single charge. Then adjust the battery results from this article by the same margin, and you'll have a good idea of what to expect from these products.
To my surprise, the results from my benchmark closely tracked the advertised numbers for the batteries. Initially, I was confused that such minor activity could equal the industry benchmarks for working life until I realized that the industry tests include large blocks of time in which the computer is doing nothing and is in a low-power, near-hibernating state. So when you read manufacturers' numbers, understand that, say, a 5-hour battery does not imply the ability to do 5 straight hours of work. It's a liberal estimate.
Batteries and their duration
Dell's Latitude E4300 has an external battery option, called a "slice," that complements the laptop's existing 6-cell power source. Officially billed as a 6-cell, lithium-ion battery, the slice is a thin, flat panel that fits onto the entire underside of the notebook. Just 3/8-inch thick (1 cm), it weighs 22 ounces (620 grams). For the fully charged combined 12 cells, Windows Vista predicted 9 hours, 57 minutes of usage. On my benchmark, I registered 9 hours, 13 minutes.
According to Dell, if the E4300's hardware configuration were modified (reducing RAM, disabling various drivers, and removing some components such as the optical drive), it would score 12:56 on the Mobile Mark 2007. But with the more normal shipping configuration, a very respectable 9 hours is a practical expectation. At a retail price of $230, the slice is a little costly, but it is by far the least obtrusive and most convenient of the external batteries I looked at.
HP's EliteBook 6930p has two battery options: an Extended Life battery and an Ultra Capacity model. The Extended Life model, at $159 and weighing 1 pound (454 grams), is an ungainly product that adds an inch to system height. It has a C shape that plugs into the underside of the notebook. However, it frequently came loose during regular movement of the notebook. Its 8 cells, when added to the notebook's 6-cell battery, were projected by Windows Vista to deliver 13:46 from one charge. In fact, they delivered slightly more on my benchmark: 14:05. For $159, this is an impressive extension of battery life.
However, I would recommend skipping the Extended Life battery in favor of HP's 12-cell Ultra Capacity model, which costs only $30 more -- provided the 28.6 ounces (810 grams) it adds to the laptop's weight is not a problem. The Ultra Capacity battery is shaped as a wedge and adds approximately 1 inch of slope to the back end of the EliteBook. It fits more snugly than its Extended Life counterpart. HP's marketing literature claims that the Ultra Capacity battery provides 4 hours more than the Extended Life model. In my tests, it did better than that, coming in at a whopping 20 hours, 24 minutes for an astonishing performance!
These results show convincingly that battery technology has advanced substantially in recent years -- so much so that mobile workers really can comfortably and reliably leave the power cords and bricks in their hotel rooms when visiting clients or local offices. Just imagine, you could work all the way through a long international flight without worrying about finding plugs. Then again, perhaps a 15- or 20-hour workday is not such a good thing...
Windows 7 is suddenly telling users it isn't genuine -- and it has nothing to do with Windows being...
Last Tuesday's MS14-066 causes some servers to inexplicably hang, AWS or IIS to break, and Microsoft...
These strong alternatives to the popular languages are gaining steam -- and may be the perfect fit for...
Sponsored by Nuage Networks
Sponsored by Fibre Channel Industry Association
There's only one explicit Apple Watch MDM policy, but more controls than you might realize
Don't want your home address or other personal info published to the world? This weekend, take an hour...
Little languages abound to bring your code to the Web with surprising ease and few compromises
Apple’s ‘my way or the highway’ upgrade policy for OS X really needs to change