PC Makers Still Not Being Totally Truthful With Battery Life Claims
For whatever reason, AMD has taken a strong, vocal stance against shoddy battery life claims. They've had it with PC manufacturers pumping out notebook after notebook with crazy claims, ones that they can't possibly deliver. Take any laptop of your own, for example. Has it ever fully lived up to the claims on the box? Have you ever considered returning it because it didn't? The answer to both questions is probably "no" considering that you know very well that any machine you received in return would have the same problem: lofty promises, lackluster delivery.
AMD has been calling for manufacturers to settle on a standardized testing process for battery life in order to give the public a better idea of exactly how long one machine will last compared to another. As it stands, the consumer has no real idea how a PC maker tests battery life, and thus, they can't accurately compare one machine to another. You can pretty much bet that a Gateway with five hours of claimed battery life and a Dell with five hours of claimed battery life will actually run out of gas at different points, even if they're both tested the same way at home.
Moreover, the company has also pushed for a new way of displaying battery life. Pat likens the "single battery life figure" you see on boxes to buying a car with only a city MPG figure; the latter would never fly, so why does the former? Pat also took the time to really examine some of the newer back-to-school circulars floating about while paying particularly close attention to the verbiage used to describe battery life. Below are his observations in full:
He points out that in all of his research, he still found just one single battery life figure being advertised, as if a computer drains similarly while watching a DVD or running a screensaver. He also found that Apple notebooks "never list battery life," though we will say that it's easy to find that battery life figure on the company's website, and considering that the are only sold in a few places, anywhere you go will present you with intelligent agents to fill you in. Still, no listing at all on the box is ludicrous.
- 23% increase over the prior two weeks in the number of SKUs advertised with battery life. (34 to 42 SKUs) 23 SKUs I observed advertised battery life or inference to it during the week of 8/10/09, and 19 the week of 8/17/09. (See raw data at very end of blog.)
- 2X increase in the number of ways battery life is being advertised.
- Four different ways observed from weeks of 7/27 and 8/3:
- “up to X hours, Y minutes”
- “up to X hours” (no minutes)
- “X+ hours” (no “up to”)
- “X cell battery for longer performance” (adds concept of battery cell)
- Five NEW ways battery life was described or inferred to the consumer weeks of 8/10 and 8/17 in addition to those listed above:
- “up to X hours of battery life that will last in class all day” (adds idea of all day computing)
- “X cell battery” (no information on what this means)
- “X cell battery will give you up to X hours of battery life”
- “X cell Li-ion battery” (no battery life claim and introduces battery type)
- Graphic with icons, small description, but no data.
Real-world battery tests that we've conducted
He closes things out by restating that a single battery life test is needed across the industry. Getting PC makers to buy in, however, remains a tall task. After all, each maker wants their machine to seem like the best, and by running their own tests under their own supervision, they can basically reach any conclusion they want. What do you think of all this? Do you just automatically adjust whatever value you see down by 10 - 30% to get a more realistic view? Do you agree that folks should be required to do that in order to get a solid grasp on battery life?