Benchmarks
For comparison, we have put the Mesh system up against the outrageous Alienware Aurora 7500, which costs over £1000 more, and the
bit-tech reference rig, as sourced for us by the excellent guys at
Udiggit, which includes a Athlon 64 X2 3800+, Gainward GeForce 7800 GT, 2GB of OCZ RAM and is housed in a
Shuttle SN25P.
Unfortunately for the Mesh system, it was configured with the 2T command rate, despite having two 1GB modules installed. Due to the way that the Athlon 64 memory controller works, when four modules are installed it's only possible to use the 2T command rate. However, since the system has two modules, you would have thought that Mesh would configure the system with the 1T command rate for better performance.
The two FX-60s roundly clout the X2 3800+ in the SuperPi test, although we were disappointed to see that the overall system stress test of MP3 encoding saw the Mesh lacking ever so slightly. We feel that the MP3 encoding test was affected by the poor-performing memory.
These results just go to show that putting the best components together in one box is a guarantee of fast performance: setup and configuration can be just as important.
High quality gaming
Although the Viewsonic monitor included is capable of 1680x1050, we wanted to push the boat and and see where the horsepower of this system really ranks, so we cranked up the resolutions a little.
The Alienware system had a couple of 512MB GTXs in, so is faster at higher resolutions, as you might expect. With just a single GT, the
bit-tech rig can't really keep up, but gives an interesting reference for single-card versus SLI performance against the Mesh system's dual GTs. Oddly, we couldn't get the Mesh system to work properly in Counter-Strike, no matter what we did. It's performance was sub-par and we don't know why. We'll be doing some further testing and will bring you an update on the matter next week.
As a reference, here's some indication of the results you'll get running at the native resolution of the included ViewSonic monitor.
Want to comment? Please log in.