facebook rss twitter

Review: MSI 648 MAX Motherboard

by Tarinder Sandhu on 25 October 2002, 00:00

Tags: MSI

Quick Link: HEXUS.net/qanm

Add to My Vault: x

Benchmarks II

From our SANDRA analysis we can glean that the SiS 648 DDR chipset is better of the two at fulfilling the P4's bandwidth requirements. Let's see how this affects our first practical benchmark.

Pifast simply calculates the constant Pi to the desired number of decimal places. I've chosen 10 million using the fastest method possible. Memory bandwidth has historically played a major part in this benchmark. Not only does pure bandwidth matter, the timings matter too.

Remember that the CPU speed is held constant at ~ 2800MHz, only the memory solutions change. This benchmark is all about getting data to the CPU as quickly as possible; bandwidth matters greatly. The MSI's DDR-400 time would have been a couple of seconds faster had I been able to use the same timings I did at DDR333. Running at CL2 'Ultra', 1T timings was not possible, however. The i845E, ravaged by a meagre 2.1GB/s of bandwidth, limps home in last place.

Next we'll turn our attention to MP3 encoding. We're benchmarking by encoding a 638MB custom WAV file (Moby's Play album, incidentally) into 192kb/s MP3 using the LAME 3.92 encoder and Razor-Lame 1.15 front-end.

MP3 encoding is generally insensitive to changes in bandwidth; that is largely shown here. Why are the MAX's times a little faster than the RAMBUS and i845E setup ?. The answer lies in the actual clock speed of the processor on the MAX 648. Setting it at 133FSB, we get a speed of 2815MHz, enough for it to steal a march on the other two more conservatively clocked platforms.

I'm now an active SETI-runner. SETI thrives on lots of memory bandwidth delivered by an efficient memory controller. I'm running the OcUK SETI benchmark. One advantage in this benchmark is its ability to display results to within 1/1000th of a second, I've rounded the results up to the nearest second for the sake of brevity. You can download it here. Simply unzip and click on the runbench.bat and wait.

We're looking at a difference of 23 minutes from fastest to slowest. The difference is simply explained by the fact that SETI is an absolute lover of real bandwidth. The Pentium 4 is starved by the limitations of DDR266 memory. RAMBUS reigns supreme, although the MAX648 isn't that far behind in DDR-400 mode.