Ever thought about building a Mac to your exact requirements? Don't pick up your screwdriver until you've read about our attempt to build a Mac - and the pitfalls we encountered.
How well does it work?
But the real question is, just how does the £490 FrankenMac compete with Apple's hardware? Overall, quite well. To get a sense of just how well, I ran a few tests here, though I plan to ship the whole machine off for an official run through our Speedmark test suite.
For my unofficial testing, I used a few old favourites - Xbench for overall benchmarking, Cinebench for graphics, and a quick blast through the standard Quake 3 timed demo to give a sense for raw gaming performance. I ran the tests on both the FrankenMac and my Mac Pro (a 2.66GHz quad-core Xeon with 4GB of RAM and the ATI X1900xt video card).
Using Xbench, the FrankenMac beat the Mac Pro on the overall score, 149 to 143. Within the individual tests, the Mac Pro was faster at the CPU, thread, and OpenGL tests, while the FrankenMac was better at the memory, Quartz graphics, user interface, and disk tests.
Overall, there's not much difference in the two machines' Xbench results and keep in mind that the Mac Pro is more than twice as expensive as the FrankenMac.
The Cinebench results reflected the faster processors in the Mac Pro, helping it win the CPU-rendering tests, the Mac Pro won the single-CPU test by about 10 percent, and the multiple-CPU test by about 3 percent. The OpenGL rendering results were basically a dead heat, separated by less than a single percentage point.
The Quake3 benchmark test was fun, both machines blast through this aging game's demo mode amazingly quickly. At the end of the test, though, the FrankenMac beat the Mac Pro, scoring 756 frames per second (fps) to the Mac Pro's 655fps. I also ripped a portion of a DVD using HandBrake, and found the Mac Pro to be about 10 percent faster than the FrankenMac, which is the difference in the clock speed of the two machines' processors.
Based on my unofficial results, it appears that my home-built machine compares quite favourably with a much more expensive Mac Pro. We'll have to wait for the official Speedmark benchmark, though, to see how it does on a broader and more-demanding set of tests.
NEXT PAGE: The pitfalls of DIY Macs
- OS X-free Mac clone vs Apple Mac Pro
- Building my own Mac
- Booting up our home-made Mac
- How our DIY Mac performed compared to the Mac Pro
- The pitfalls of DIY Macs
- Mac clone vs Apple Mac: our expert verdict