It's taken us awhile to get around to publishing a review of the new 2013 MacBook Air for one simple reason: We've been a bit stymied by how to go about testing it. On the outside, the Air looks barely different from its predecessors. But on the inside, it packs some features that are so new that it's actually difficult to evaluate them. There are essentially four things worth knowing about Apple's latest generation of ultrathin notebooks. The first is that they look exactly the same as last year's model—we mean exactly the same. The only physical difference in the hardware is the dual mics on the left-hand side (Apple uses them for noise cancellation). Design consistency is hardly a point against the Air, since it seems to be the Platonic ideal for super-compact laptop design. Rigid and lightweight without feeling flimsy—and nicely weighted so that the top opens without lifting the bottom—the Air is perfect for moving from room to room but also disappears in a bag, making it ideal for travel.
The second thing worth knowing is that MacBook Airs are now using Intel's new Haswell processors. That makes a big difference, though not necessarily in the way you'd think. When it comes to raw processing speed, the new Haswell chips are not necessarily that much faster than the last generation of chip. Our $1099, 13-inch test model Air came equipped with an i5-4250U rated at 1.30 GHz. We ran Primate Labs' Geekbench benchmarking suite and got a score of 6806—a modest bump up from Primate Labs' own score for last year's MacBook Air with similar specs, 6665.
That brings us to the third thing worth knowing: Those Haswell processors are not sprinters, but endurance runners. The new Intel architecture is designed to provide what Apple calls "all-day battery life" during typical usage. Anecdotally, we can assure you that this feels valid. When used for Web surfing, music streaming, and occasional video, the new Air can go all day without a recharge. Apple's official claims for the 13-inch Air are 12 hours typical usage and 10 hours of iTunes movie playback. The previous generation was rated for only "up to 7 hours." We asked Apple for their battery testing protocol, and got back this:
"The wireless web test we use measures battery life by wirelessly browsing 25 popular websites with display brightness set to 75 percent. The HD movie playback test measures battery life by playing back HD 720p content with display brightness set to 75 percent."
That sounds fair, but we like to test worst-case scenarios. For the PM battery test, we turn all the battery protection features off and lock the screen brightness at 100 percent. Then we use Endless YouTube (a video clip looper) to continuously stream a 720p video clip over and over and over (we've now seen the trailer for Dreamworks' Epic so many times that we couldn't possibly enjoy the movie if we wanted to). In our test, the Air's battery still lasted an impressive 5 hours, 15 minutes. (For reference, we recently ran the same test on a Toshiba KiraBook, another ultrabook with an impressive battery, and got precisely 4 hours.) So we're pretty comfortable saying the new Air has the longest-lasting battery we've seen on a notebook.
The last thing worth knowing about the Air is actually the thing that we had the most trouble figuring out how to test. These new MacBook Airs are the first laptops to ship with new 802.11ac antennas. This is the newest Wi-Fi standard, and it comes with some pretty astounding claims. For example, 802.11ac is supposed to be capable of gigabit-plus data transfer speeds over local area networks (LANs), and Apple claims its 802.11ac equipment can reach theoretical speeds of 1.3 gigabits per second (Gbps). That's not such a big deal for Internet traffic. But for the Air, which has no internal optical drive and limited hard drive space, having a fast connection to other drives on the network is a huge bonus.
Our problem is: When you only have one 802.11ac device, how do you test data transfer speeds? In theory, the AC wireless standard should even be faster than hard-wired gigabit Ethernet equipment. For our test, then, we relied upon the supposition that in real-world use, 802.11ac equipment wasn't going to outpace hard-wired Ethernet gear, so we used an Apple Airport Extreme 802.11ac router ($199), to move some files to the Air from a gigabit Ethernet Western Digital NAS drive. It turns out that the MacBook Air's AC Wi-Fi is blazing fast by wireless standards, but still not as fast as a hard-wired connection.
When we hooked up the Air directly to the router via a Thunderbolt Ethernet cable, we moved a 1.5-GB movie file in 14.4 seconds. That's a data rate of 106 megabytes per second, which translates roughly to 0.88 Gbps (remember there's a difference between bytes moved and bit rate). That's about what we'd expect from gigabit Ethernet. Over the 802.11ac wireless connection, that same file took 57.8 seconds. That works out to a data rate of 26.6 megabytes per second, or about 0.22 Gbps—well below the theoretical maximum, but still way faster than a traditional 802.11n connection. We then moved the same file over the Airport Extreme's 802.11n channel. It took 2 minutes, 17 seconds—a data rate of 11.2 megabytes per second, or 0.09 Gbps. That's tortoise-like by comparison.
(It's worth mentioning. If you go out and invest in one of these 802.11ac routers, keep in mind that they really spawn two networks—one over 2.4 GHz spectrum and one over 5 GHz. Only the 5-GHz channel supports 802.11ac.)
The MacBook Air may not look like a huge improvement over its predecessors on the outside. But everything on the inside is faster, better, and longer-lasting.
0 comments:
Post a Comment