Memory performance
Memory subsystem performance doesn't always track with real-world applications, but it's a good place to start with integrated graphics chipsets that cannibalize a portion of system memory and therefore bandwidth.

These results aren't terribly surprising. AMD's integrated memory controller is renowned for providing plenty of bandwidth with low access latencies, and the Pentium E2180's relatively slow 800MHz front-side bus speed certainly doesn't help matters for the G45 Express. Note, also, that our 780G platform has a significant advantage in the bandwidth test, likely due to its use of dedicated "Sideport" graphics memory, which lessens the burden on system memory.

Motherboards don't always cooperate when you populate all of their DIMM slots, and our Asus GeForce 8300 board wouldn't boot with a four-DIMM config, even after backing off on memory timings. We've alerted Asus to the issue, but don't have a resolution yet.

The relative position of the G45 Express doesn't really change when we load it up with memory. It still offers far less memory bandwidth with higher access latencies than the 780G.

The following latency graphs are a little indulgent, so I won't be offended if you skip them. They show access latencies across multiple block and step sizes, painting a fuller picture of memory controller performance with each chipset. I've arranged the graphs in order of highest latency to lowest. Yellow represents L1 cache, light orange is L2, and dark orange is main memory.

Obviously, the G45 Express can't compete with the low access latencies provided by AMD's on-die memory controller.