Latest MacBook has a long life but shows signs of aging. (March 7th, 2008)
Product Manufacturer: Apple
Price: $1,299 (2.4GHz, 2GB RAM, 160GB, Superdrive)
- Stunning battery life that beats Apple's estimates.
- Faster than a model just a few months old; rivals Mac desktops.
- 2GB of RAM finally enough to use out of the box.
- Best value for dollar in the MacBook lineup.
- Design quickly getting old, prone to cosmetic damage.
- Sorely needs a dedicated graphics option; X3100 too slow in some cases.
- LCD is unnecessarily low-quality compared to the MacBook Air.
performance: subjective experience
Ever since introducing the MacBook, Apple has focused on processor speed over graphics, and that remains truer than ever for the early 2008 refresh. The 2.4GHz Core 2 Duo chosen by Apple for its higher-end 13-inch MacBooks is not only the same as for the base MacBook Pro but just 100MHz away from the top-end version. That's an important consideration when raw CPU performance is more important, as there will be very little to gain from the $700 difference between this system and the base Pro model, let alone the $1,200 extra necessary for the 2.5GHz version.
I saw -- or perhaps, didn't see -- this narrow difference in day-to-day use. Anything chiefly processor-driven, such as watching a 1080p QuickTime movie or music encoding, is effectively seamless. It's helped in no small part by an overdue increase to 2GB of RAM on all but the least expensive MacBook. This is the first model that doesn't immediately require a memory upgrade to run more than basic apps smoothly.
Having said this, graphics performance is merely adequate, not stunning. Apple continues to deny even the $1,499 model basic dedicated graphics from ATI or NVIDIA, and the Intel GMA X3100 graphics that are left to the system seem unusually slow considering that many Dell, Gateway, and HP systems at the same price have at least a GeForce 8400M GS to offload some of the 3D work, even with the same processor choice. Cost simply isn't an excuse.
Older games such as Warcraft III play acceptably well, but even they can be hampered in very hectic action. The lack of dedicated memory also leaves the MacBook's video prone to stuttering in some extreme cases for just the operating system itself, such as viewing many open apps in Expose or attaching a 20-inch display or larger and spanning the desktop (lid-closed mode is faster). Such performance is only just acceptable for the price and should really change for the next update.
performance: objective tests
I ran three core Mac-specific tests, each of which tests the full scope of the processor and (in the cases of Cinebench and Xbench) graphics performance as well. Gaming tests were not going to be an option: as most 3D games are extremely dependent on graphics performance, the basic GMA X3100 simply wasn't an option. All tests were run against a 2.4GHz iMac, which uses the previous-generation Core 2 Duo processor but also shares the same 2GB of RAM.
The tests are somewhat surprising. The iMac takes a dramatic lead whenever the video chipset is the most important metric, such as in the Cinebench OpenGL shading test or the Xbench user interface test (a difference which skews the overall results in the iMac's favor). However, the MacBook is often near or sometimes well ahead of the iMac in raw performance, even with obstacles such as a slow notebook hard drive and having to share memory with the Intel graphics. In the Xbench thread test, which checks multi-CPU tasks, the MacBook was a clear 37 percent faster. While Apple is likely to replace this iMac with a new model soon, it can't be ignored that an $1,800 desktop can sometimes be outrun by a $1,300 notebook in pure CPU tasks.
For its latest round of processors, Intel shifted much of its attention from raw performance to battery life, and Apple has followed suit: a change to the company's testing methods now trumpets battery life as it exists in a real-world browsing test with wireless enabled, rather than a theoretical (and often unrealistic) peak.
I saw this in practice. Using Apple's recommended settings of normal power usage, wireless turned on, and half screen brightness, I netted better than expected battery life in the exact same conditions. The test system lasted for 4 hours and 54 minutes, or about 4.9 hours, of average web browsing. This is not only longer than Apple's official claims but leaps ahead of the previous model; most users of the 2GHz and 2.2GHz MacBooks from the fall often net between 4 and 4.5 hours of similar use, sometimes netting less.
The importance of that battery metric can't be overstated. The battery performance is as good or better than the extended batteries for many rival notebooks and is enough to last many students and workers for an entire day. Longevity can still drop quickly if the system is pushed to its energy limits, as with playing back a DVD or processor-intensive tasks, but for many users the need to carry a second battery is that much lower.