One thing you have to love about Apple are their benchmarks. When they used to tout the PowerPC series of chips, they had benchmarks showing how the G4 and G5 could topple anything that AMD or Intel were producing. Then came the transition to Intel, and those benchmarks magically reversed themselves, perhaps in small part due to the major improvements brought about by Intel's “Core” architecture.
So what does Apple do for benchmarks when it upgrades one of its product lines from an “older” Intel platform to a new one? Make ridiculous benchmark comparisons of course! Here are a few of our favorites:
“In an iPhoto common application task -- which involves importing 100 photos into an existing iPhoto library containing 2000 photos and then exporting those photos as a web page, a movie file, and preparing them for iDVD -- the new 2.2GHz MacBook was 35 percent faster than the original 2.0GHz Intel Core Duo model, while the 2.16GHz MacBook introduced in May was just 25 percent faster.
A similar iTunes application test -- which measures the speed of importing a song from the hard drive to the iTunes library, encoding a video clip for iPod, and encoding 60 minutes of music and burning it to a CD -- showed the new 2.2GHz MacBook to have a 6 percent edge over previous-generation 2.16GHz MacBook.”
There are a few problems with this test that make it hard to really believe. First is that we're comparing what looks like a T2500 CPU with half the cache of the modern Core 2 Duo CPU, which leads us right to the second problem: there's a 10% difference in clock speed despite the fact that Apple offers the new MacBook with 2.0 GHz CPUs.
The third factor here is going to be the hard drive. The past year has seen a large increase in portable hard drive size and performance. Given that these tests all seem to involve importing/exporting, there is the concern that the hard drive is probably playing a more central role in these scores than the CPU or FSB.
Regardless of why or how these numbers came to be, it is clear that the new MacBook is a decent upgrade from the last generation, but don't hold your breath if you're expecting to see 20%-40% increases unless you're doing something that is disk intensive.