SATA; Or, an Acronym for How Computing Should Be

Review: SATA hardware

One of the most important and valuable innovations to hit the computing market in the past few years is the SATA interface. That stands for "Serial ATA," yes, that's serial as opposed to parallel, and no, I have no clue what "ATA" means. SATA is a way for the motherboard to interact with your disk drives. That probably doesn't sound terribly exciting or significant. In fact, even if you know a decent bit about computers, you may have no clue what SATA means or why it's such a great idea. I didn't until a few weeks ago when I bought a new hard drive and tried to install it.

The great thing about SATA is the fact that a SATA device has exactly two cables at all times: a power cable, which goes into the power supply, and a data cable, which goes into the motherboard. If you're thinking "Well, isn't that the way drives should work?" then the answer is yes, that's exactly how they should work. That's nominally how non-SATA drives work too, but the amazing part about SATA drives is that they do not have jumpers. If you've got any familiarity with old hard drives, there's a good chance you're preparing a celebratory offering to your god(s) right now. If not, allow me to delve into the horrific nightmare-scape that is IDE.

See, the old standard for hard drive interfacing is (sometimes) called PATA, but much more commonly called IDE. I don't know what IDE means either. Confused yet? It gets worse. SATA devices are built on the basis that one drive goes into one port. You've got SATA 0, SATA 1, etc. and that's it. (If the numbering convention doesn't make any sense, good, I don't understand either why computers insist on counting from 0 instead of 1.) IDE devices allow you to daisy-chain; that is, a single IDE cable has two slots that plug into hard drives and one that goes into the motherboard. So how does a single IDE controller handle two IDE devices? With a ghastly construction called master/slave configuration.

Each IDE controller can only support one master and one slave. So for every pair of drives connected by the same IDE cable, you have to put one drive on "master" and one on "slave" to get either of them to work. This is much easier said than done, however, as to be able to change from master to slave, one has to take a small (we're talking 1/8 of an inch) piece of circuitry off one set of pins from the back of the hard drive and put it onto another set of pins on the back of a drive. Somewhere along the line, it was decided that this was probably not the best way to do things, so new drives began to support the CSEL (Cable Select) paradigm. If you put both drives' jumpers onto CSEL, the drives will function, and the computer assigns master/slave automatically.

This still means jumpers, though.

So for anyone who has ever dealt with new drive installation and configuration, SATA represents a breakthrough in technological convenience. This breakthrough will never be noticed by the average computer user, and it may not even be apparent to a more advanced user. But if you ever get to the point where you need to use it, I guarantee you will appreciate SATA.