Other World Computing has a that it says is a first for Intel Core 2 Duo MacBook laptops. Apple doesn't offer this option. The same $340 kit - a set of one 1 GB and one 2 GB PC5300 DDR2 SO-DIMM modules - also works with the Core 2 Duo iMac and the 15-inch, 2.16 GHz Core 2 Duo MacBook Pro. The other two standard MacBook Pro models include 2 GB as two 1 GB modules, and can be upgraded by swapping one of those modules for a 2 GB unit for $260 from Other World Computing.
Apple doesn't offer a 3 GB build-to-order configuration for either its MacBook or iMac models, although the 1 GB stock MacBook Pro can be upgraded to 3 GB for $750, while the 2 GB stock MacBook Pro models can be upgraded to 3 GB for $575.
Other World Computing of between $44 and $60 for 1 GB of memory, depending on the Mac model it was pulled out of and the configuration (as two 512 MB modules or one 1 GB module).
Apple pairing identically sized RAM modules for all its Intel-based Macs. The Mac Pro and Xserve require paired modules, but the company emphasizes the benefits for those models that use system RAM for video operations instead of dedicated RAM for graphics purposes - the Mac mini, MacBook, and iMac. The Intel Core 2 Duo MacBook Pro models have no Apple recommendation for paired memory; the original Intel Core Duo models did, just to add to the confusion.
With paired memory, the processor can access RAM at up to twice the speed of unpaired memory, which is especially important for video output. What that means for actual performance can be determined only through real-world benchmark testing.
It's possible that increasing memory to the maximum 3 GB may trump the increased performance of paired memory, because additional RAM can prevent an operating system from moving data back and forth between RAM and hard disk-based swap files as it pages data and program pieces in and out. The more RAM, to some extent, the less time the computer spends performing relatively glacial hard disk operations.
We recommend using TidBITS Contributing Editor Matt Neuburg's freeware program to see whether or not you currently rely on disk swapping enough that additional RAM might improve overall application performance.
Oddly, OWC's with varying amounts of RAM and varying tests doesn't seem to show that more memory produces substantially different results. However, the benchmarks they used look at sets of operations or program functions, rather than a typical Mac user's array of different programs in real-world usage scenarios.