Hi guys. Let’s talk about some recurring misconceptions.
System shutdown. OS X, BSD, Mach, blah blah blah, are designed to be systems that never need to be restarted in the course of their normal operation. Why do we restart our computers, then?
– To swap out libraries that are in use by the system. Actually, we’ve gotten pretty good at doing this while the computer is still running, but OS X likes to play safe. Linux is a lot better at this.
– If the kernel panics (enters a state from which it cannot safely recover). Shit happens.
– That’s all.
That’s right, folks, those are the only things that require you to restart your computer. Everything else you need to do, can be done without restarting. Now, if your system is spewing a megabyte a second to syslog, maybe it’s going to be easier to just restart than to try and get to a console to order launchd around. But you don’t NEED to restart for that, and as @tonedef mentioned, OS X is pretty good with memory management.
—
Batteries. Discharging your battery and recharging it again does not help it. Battery technology has moved quickly, and we’ve burned through several different battery chemistries in the recent past:
– Lead-acid batteries: did we use these in computers? I forget. I’m too lazy to look it up.
– Nickel-cadmium and nickel-metal-hydride (you may know them as NiCd and NiMH) cells were next up. These types of batteries DID have a “memory”, and they COULD be conditioned by fully depleting and recharging the battery pack. These cells were still in use in laptops until fairly recently, and we still use them for our rechargeable AA/AAA cells. This is where I think the “deplete your battery and it’ll be happy” myth comes from.
– Lithium-ion and, more recently, lithium-polymer cells. These cells power the nicest, shiniest notebooks. They have high energy capacities, although they can have issues with high-current discharge. Unfortunately, these cells are both finicky and highly dangerous if mishandled. I’m going to break out of my nice clean bulleted list for a moment to say this:
Lithium-ion and lithium-polymer cells do not benefit from repeated charge and discharge cycles; in fact, this practice will decrease their useful lifespan.
These cells are rated for the number of cycles they can handle; that is, the number of times they can be fully discharged and fully recharged. You don’t have to bring your battery down to 0% to tick the “cycle counter”: if you discharge and recharge halfway, twice, that’s one cycle. I know that Apple estimates a battery lifespan of 300 cycles (excepting the new MacBook Pro 17” Unibody, which uses magic battery juju). After that, the cell is not expected to retain a charge reliably.
Now, some of you snippity folk out there may point me to the Apple KB article which recommends discharging and recharging your battery therapeutically. Yes, I know Apple says that. The reason Apple says that is because your battery is more than just a bunch of Li-Ion or LiPo cells strung together. Remember how I mentioned that lithium cells are both finicky and dangerous? Remember the Sony battery fiasco, when we had Dells turning into fire-plumes left and right? That’s what I mean. Li* cells ALWAYS retain a charge, even when your computer cries and shuts down. This is because if any individual cell is allowed to deplete past about 3.2V, it is more or less ruined. If any individual cell is allowed to charge past about 4.2V, it can puff and rupture. If it ruptures, you’re mixing lithium and water. Anybody remember what happens then?
Anyways, the battery you put in your computer has a little bitty microcontroller whose job it is to maintain the battery voltage within nominal ranges. Sometimes its measurements can become offset, making it cut off the battery power—to “save” the battery from destruction—before it’s really necessary. Depleting and recharging the battery will recalibrate that microcontroller, but it costs you a cycle.
That’s also why you shouldn’t leave fully-charged batteries in the heat, by the way—as temperature increases, the voltage of the cell increases. If a cell is already at 4.2V, that increase can cause failure. Same with extreme cold and fully-depleted cells.