System Builder Marathon, Q4 2012: $2,000 Performance PC

Benchmark Results: 3DMark And PCMark

3DMark is primarily a graphics test, and we’re not surprised to see our current PC’s CrossFire configuration beat the single GeForce GTX 670 in last quarter's build. Even overclocking can't overcome the performance delta. Might we see our value gamble pay off after all?

PCMark puts quite a bit of emphasis on storage performance, despite its purpose as a general system metric. Naturally then, we expected both Mushkin SSD-equipped builds to generate similar scores.

By now we're used to seeing this suite use only four cores. So, the fact that our Ivy Bridge-based Core i7-3770K manages to slide past a Core i7-3930K isn't particularly surprising.

Remember that last quarter's baseline score was measured using Intel's RSTe driver, which improves data security at the expense of performance. We shifted over to Microsoft's default AHCI driver for that machine's overclocked configuration.

Overclocking can make software-based drive controllers perform better, but the conditioning each drive received is a more likely reason for the overclocked Q4 build’s greater storage performance. Although we're dealing with a consistent drive image and the exact same SSD, even letting each machine sit for different amounts of time can affect the outcome of storage subsystem testing. 

Create a new thread in the UK Article comments forum about this subject
This thread is closed for comments
1 comment
Comment from the forums
    Your comment
  • Steveymoo
    I was getting ready to whine about power consumption with those AMD cards, but it looks like they're not so bad! I'd be interested to know how loud/whiny this system gets, and how it would perform with 2 gtx 680s instead.

    Also, I find it weird that those crossfire GPUs fall flat on their face at 2560 resolution in some games. Is there some kind of crossfire/gpu memory issue going on there?