Lies, Damn Lies, and Battery-Life Statistics

theodp writes "What if automakers measured gas mileage by rolling their cars downhill with their engines idling? They might, Newsweek's Daniel Lyons suggests, if they took inspiration from the MobileMark 2007 notebook battery-life benchmark test, the creation of a consortium called BAPCo, whose members are — surprise — computer makers and other tech companies. Laptops score big numbers, Lyons explains, because they're tested with screens dimmed to 20%-30% of full brightness, Wi-Fi turned off, and the main processor chip running at 7.5% of capacity. Professional reviewers see company-generated battery-life claims as a joke. 'The rule of thumb is that in real-world use you get about 50 percent of rated battery life,' says a Gizmodo associate editor. Leading the call for reform is the not-necessarily-altruistic AMD, who gripes that MM07 was created in Intel's labs and rigged so Intel chips would outscore AMD chips, which draw more power when idle."

Comments

Lenovo here

[SchizoStatic]

I have a cheap lenovo from last year and if I am on 50% brightness with wifi on and just browsing web with some videos I can go 3 hours. The battery is rated for 3. *shrug*

Re:Apple Don't

[wnknisely]

For what it's worth, I'm seeing numbers comparable to these on my new Macbook pro. Perhaps Apple is using a different benchmark than the one in the article above?

Re:Hate to sound like a Mac whore

[OrangeTide]

I got my Macbook because it makes a pretty decent Linux box (although I now dual-boot OSX because Apple has slowly wormed its way into my life), and the lowest end Macbook is pretty affordable. Comparable in price and specs to a mid-range laptop. And yes it does get good battery life (the battery life is better in OSX than in Linux though)

If you use computers the way I do, the you can really get those 4 hours that those cheap PC laptops claim. I use it in the dark(so screen dimmed), no WiFi, command-line only running a text editor for Vi. Occasionally kicking on the CPU to compile a .c to a .o and link the .o's (using 'make' saves a tremendous amount of battery life). My Macbook got a little over 5.5 hours for me on a flight before it had to shut down(just writing code and reading man pages). If I would have sprung for the Macbook Pro it would have been more like 7-8 hours though.

Already have that

[SuperKendall]

If someone comes out with a spare-battery-attached-to-a-magsafe-connector for those die-hards who absolutely *need* it, angels may sing in the treetops

This is exactly why I don't understand the fuss over non-removable batteries. You get better battery life, and if you need extra power you've always been able to buy external battery packs. They have cable that attach to the Magsafe connector. You can get them in a range of sizes, including sizes that are not much larger than a spare battery would have been anyway...

Similarly there are tons of external packs for smaller devices like the iPhone/iPod (or anything usb charged).

I also have not often found the need for an second battery in a laptop if I can get at least three to four hours out of it. Basically the only time is an international flight, and for that the external batteries are perfect. Heck, until it broke the Solio solar powered recharger I had could even recharge itself in-flight as long as I was at a window!!

Re:Apple Don't

[speculatrix]

there's no point in ordering a *spare* battery when you buy a laptop; ordering a *second* battery because you're travelling a lot might make sense. this is because LiIon batteries age, and you can only slow down the ageing by keeping them 40% charged in a moderately cool place.

LiIon batteries also lose capacity after a number of charge/discharge cycles, the number varies according to how deep the cycles are and the temperatures you reach in the process. When the battery packs are particularly expensive - think Prius and other electric cars - the battery management electronics are crucial to protecting the investment, and the batteries are never run anywhere near flat and particularly are never fully charge to prevent some cells overcharging, and the charge/discharge controlled carefully to prevent temperature rises leading to premature ageing.

thus, in conclusion, keep your laptop battery cool, don't thrash its battery, don't deep cycle it, buy a good brand so you can buy a new battery *in*the*future*not*for*storage*.

Re:Bad analogy

[neBelcnU]

WRONG.
No matter whether closed-loop or open-loop, modern FI always injects the amount of gas most closely matching the volume of air taken in.

Closed-loop (using the oxygen sensor in the exhaust) is totally adaptive, if there's unused O2, add more fuel. Open-loop (no O2 sensor, or 02 sensor not hot enough or not working or...) thet FI computer consults a table based on all the usual factors: Mass-air flow, engine temp, throttle position.

In either case, the opinion that your FI turns off the flow of fuel when you let off the gas is wrong. While the FI can indeed cut the fuel, it does so only under extremely rare conditions. I've actually never been able to make it happen on my car.

Oh, and the reasons? When you shut the fuel off, you run the risk of a lean mixture, which is both damaging to the engine (burning holes thu pistons) and super hi rush of NOx. That's why your throttle plate doesn't snap fully to the idle position, that slight loiter is to allow the system to "balance out" before returning to idle.

Damn, wasted my mod points to write this.

Re:Hate to sound like a Mac whore

[OrangeTide]

Since I presume that you really mean "VIM", then I have an information which might be interesting to you.

No, I use nvi.

Also writes to .swp on Vim have an impact if the file is opened O_SYNC, else it goes to cache and is rarely written. I just installed and tried Vim and the harddrive indeed does stop (I can hear it). Linux will indeed eventually wake up and commit the cache to disk. But it does take a while, but it's totally tunable. Takes about 7 minutes(and 13 seconds) for a 16K text file on my system to be committed to disk after it has been "written" 8 times (timed with stop watch, nothing super accurate)

strace output shows that I am correct (opens swp file with O_RDWR|O_CREAT|O_EXCL|O_LARGEFILE). as well as checking the source in http://vim.cvs.sourceforge.net/viewvc/vim/vim7/src/

In other words, your conjecture is wrong.

Re:Apple Don't

[moosesocks]

All of these things are true for Lithium Ion batteries.

However, Apple's current line of notebooks use Lithium Polymer instead of Lithium Ion. Additionally, the charging circuitry is supposedly much more sophisticated.

I'm not sure if the modes of failure for commerical Li-Pol packs are well-known at this point, and I have no reason to believe that they'd be the same as those for Li-Ion.