HOW WE CONFIGURED THE TEST SYSTEMS:

The first thing we did when setting up our test systems, was enter the system BIOS and set the memory timings for optimal performance.  With both the Shuttle SB61G2 and Abit IS7 test beds, this meant setting our memory to run synchronously with the processor FSB at 200MHz, with the timings set to 2-5-2-2.  After configuring the BIOS, we booted the systems, formatted the hard drives and installed a fresh copy of Windows XP with Service Pack 1.  When our Windows XP installation was complete, we hit the Windows Update site and installed all of the critical updates and service packs currently available (Except for Windows Media Player 9 and the updates associated with Windows Messenger).  We then disabled Auto-Updating, Hibernation, and System Restore, and set up a 768MB permanent page file. On both test systems we set the visual effects to "best performance" in system performance to limit any effects these settings would have on the benchmarks.  All of the necessary drivers were installed, and then we defragged the hard drive.  Lastly, we installed all of the necessary benchmarking software, de-fragmented the hard drive again, and ran our tests.

The HotHardware Test Systems David vs. Goliath?

SiSoft SANDRA 2003 Benchmarks Lets Get The Party Started

SANDRA (the System ANalyzer, Diagnostic and Reporting Assistant) is an information and diagnostic utility developed by SiSoftware. Beyond benchmarking, SANDRA also provides a host of other information about your hardware and operating system. We began our testing with three of the built-in tests that partially comprise the core of the SANDRA 2002 suite, CPU, Multimedia and Memory benchmarks. We ran these tests with an Intel Pentium 4 2.40C CPU on the SB61G2, using the integrated Intel Extreme Graphics 2 configured with a 16MB buffer, and again with a Radeon 9500 Pro installed.  We compared its performance to the Abit IS7, another Springdale-based board, to see how the FB61 fares versus a top-of-the-line, full sized motherboard.

In each of the Sandra tests, the SB61G2 put up impressive numbers, and usually matched or beat the Abit IS7 when a Radeon 9500 was installed.  Where the SB61G2 faltered in the tests, was when the on-board Intel Extreme Graphics 2 was enabled.  The Floating Point CPU test was off by approximately 33 percent, and both memory scores dropped off by 200 points.  This is a result of the integrated graphics engine sharing internal memory bandwidth, leaving less available for processing and other system tasks.  Still, the hit was smaller than what we had originally expected.    

PCMark2002 Benchmarks More Synthetic Action

Next we ran some tests with Futuremark's PCMark2002 benchmarking suite.  Like other synthetic benchmarks, it's difficult to translate PCMark2002 scores into "real world" performance.  However, because it is very easy to run, and produces repeatable, comparable results, PCMark2002 has become a staple here in the H.H. labs.  We ran PCMark2002's CPU, Memory, and Hard Drive performance modules on both of the systems. 

There's not much to comment on the CPU performance.  Each board produced the same exact score when paired with the Radeon 9500 Pro.  Without it, the SB61G2 tested only four points off this mark, which is completely negligible.  The Memory performance module showed the expected drop-off in performance with the on-board graphics enabled.  Without it, the two boards are right on par with each other.  The last test shows the Abit IS7 with a slight lead over the SB61G2, regardless of graphics solution.  With PCMark 2002 used as a comparison, these two boards are shown as equals.

So far, so good, but what about gaming?