IntelBurnTest vs. SP 2004

Once in a while, I find myself with poor concentration, and perhaps a little bit of a lack of ambition. That just comes with the territory of writing so often. But, what also comes with that as a result is some strange ideas for either a piece of content or a news post. On Monday, I was having one such “blah” day (the “case of the Monday’s” really does exist!), so the idea of comparing the effectiveness of IntelBurnTest came to mind.

Partial results can be seen in the graph below, and no, that’s not a measurement of my concentration level that day, but rather the range of temperatures exhibited during a half-hour run of IntelBurnTest 1.9. For those who might be unaware, IntelBurnTest is a recent CPU stress-testing tool that carries huge claims of being the most effective method of testing a CPU for stability. After these personal tests, I believe those claims.

Although I’ve used SP 2004 to a great extent over the course of the past few years, it hit me one day just how linear its operation is. Judging by the results posted in the respective thread, the temperatures simply didn’t budge… it just seems to pummel the CPU with the exact same algorithms constantly, and that’s no way to test for true stability, right?

After testing, I’m confident IntelBurnTest is probably the best tool we’re going to find ourselves with right now, unless AMD or Intel ever chooses to release their own stress-testing tools, but that’s not likely. Why IntelBurnTest impressed me is because it showed lots of variation, and it managed to heat up our Core 2 Extreme QX9770 19°C hotter than SP 2004. That’s huge! I’m definitely confident in leaving that in our virtual toolbox for the time-being. It is winter, after all. If a benchmarking rig can dual as a space-heater, bring it on.

I think the results speak for themselves. At its peak, the CPU was pushed hard enough to hit 106°C, and 85°C on Core #0. That is a full 19°C hotter than SP 2004 on the CPU, and 18°C hotter on Core #0. What’s interesting, though, is that while the Small FFT test doesn’t fluctuate in its design at all, no matter how long it runs (it just runs with higher byte sizes), IntelBurnTest pushes more variability at the CPU, as shown in the above graph.

Source: IntelBurnTest vs. SP 2004