Memory performance
We'll begin by measuring the memory subsystem performance of these solutions. These synthetic tests don't track closely with real-world application performance, but are enlightening anyhow.

Notice that I've included a graphic above the benchmark results. That's a snapshot of the CPU utilization indicator in Windows Task Manager, which helps illustrate how much the application takes advantage of up to four CPU cores, when they're available. I've included these Task Manager graphics whenever possible throughout our results.

The X2 6000+ achieves higher memory throughput with lower latencies thanks to its built-in, on-die memory controller and dual channels of DDR2 800MHz memory. The Core 2 Duo E6700, meanwhile, hits a bottleneck in the form of its 1066MHz front-side bus; its chipset-based memory controller lies beyond that bus. That limits the E6700 to about 5.6GB/s of memory bandwidth, though it, too, has a dual-channel DDR2 800MHz memory subsystem.

These system architecture differences contribute to the E6700's higher memory access latencies compared to the X2 6000+, although the gap is bridged somewhat by the Core 2 Duo's so-called memory disambiguation feature, which moves memory loads ahead of stores in certain situations.

As our use of the term "memory disambiguation" shows, we like to scare off newbies here at TR in order to drive advertising revenues down. (It's all about the tax write-off.) As part of that effort, here's a look at some 3D graphs showing access latencies when going to L1 cache (yellow), L2 cache (orange), and main memory (er, burnt sienna).

The advantage in access latencies for the X2 6000+ is consistent and pronounced, and it grows at larger step sizes. This advantage is offset, however, by the E6700's larger 4MB L2 cache. Whether the E6700 or the X2 6000+ has the upper hand in memory-constrained scenarios will depend on the application and its memory access patterns.

These are interesting things to know, but their bearing on overall performance is mixed. Let's move now to some real-world tests to see how these two very different CPU and system architectures compete.