Thinkpads Forum

I wonder just how accurate the battery meter is. I am running Vista Business in an X61s (type 7668-a19). As the battery meter drops, the remaining time INCREASES. At near-full power (ninety-something percent, can't remember now), the timer said 2:08, and as I type, the percentage says 89%, but the time is 2:21. My wi-fi and bluetooth is disabled, but I'm using a USB WiMAX modem.

I've finished the battery a couple of times, and while I did get the impression it took a while, I feel like I could get a few more minutes. I'm using the four-cell battery, but I can't tell whether it's the slimline or enhanced capability -- it has two half-cylindrical shaped bulges underneath if that helps. Much thanks in advance.

Does it only go up initially (say from 100-98% capacity, the time goes up) and then drops? Because it's probably averaging power consumption and the time remaining at 98% is a more accurate prediction (has more data to extrapolate from).

Vista does this on my Dell (still waiting for my X61t...). When I unplug, it'll give 3:20, then it'll rise a bit to 3:30 or so at 98%, then the time will decrease normally.

carsncars is right.

Directly from lenovo:

Remaining time is calculated based on the current remaining capacity divided by the average current battery consumption (wattage). Depending on the current battery wattage consumption, remaining time calculated can change. For example, with the system running at the Maximum Performance setting for the processor, the system consumes more battery power, so the remaining time is shorter. A good analogy would be driving a car. When you drive faster, you use more gasoline; slower uses less. If you speed up and slow down frequently, it is more difficult to estimate how much driving time you have left.

What is posted above is also important when you consider that your first time checking the battery is probably shortly after you boot or come out of sleep mode, a relatively processor- and hard drive-intense period. Using that level of power consumption as a gauge will create what is likely a shorter runtime estimate, and when your usage stabilizes and you're underway your life estimate will probably increase.

In the end, I take it that short of taking a stopwatch and actually doing something, there's no way to get an accurate reading of the battery life?

About carsncars's comment, I'll re-try this from a full charge and pay closer attention to the meter. Stay tuned!