MSI 845PE Max2-FIR (845PE)

MSI 845PE Max2-FIR: Stress Testing

The MSI 845PE Max2-FIR did not have a lot of stress testing potential, very similar to the AOpen AX4PE Max. As a result of these facts, we weren't able to run as many stress tests as we usually do, although we added more brands of memory to spice up the possible problems that might arise with stress testing. We still managed to test this board in several different areas and configurations, including:

1. Chipset and motherboard stress testing was conducted by running the FSB at 160MHz.
2. Memory stress testing was conducted by running RAM at 333MHz with one bank filled as well as as many as was possible from that point on with the most aggressive timings available.

The MSI 845PE Max2-FIR performed like a champ in our numerous FSB stress tests running at 160MHz FSB. The Max2-FIR was able to run exactly one full day (24hrs) of Prime95 torture tests under these conditions. Ideally we would have to run months of Prime95 torture tests in addition to general applications and benchmarks to better guage the stability and reliability of the MSI 845PE Max2-FIR (we would have to run these tests on dozens of these boards to ensure our board isn't an exception to the rule). Unfortunately, that type of scenario is obviously quite unrealistic and beyond even our resources. So, in addition to the successful Prime95 stress tests, we reran our entire benchmark suite several times over at 160MHz FSB in addition to general application usage (such as data compression, Word, IE, Outlook, etc.). All of our stress tests proved successful at 160MHz FSB.

This first scenario is very basic, all 845PE motherboards should (and have so far) be able to run memory at 333MHz with just one DIMM installed, preferably at aggressive memory timings. Here were our results in this conservative DDR333 situation with one Corsair DIMM running at 333MHz with the most aggressive timings possible:

These are great timings, and exactly what you should expect with this type of memory and chipset. It should be noted that high quality memory is a must to achieve these timings, especially if you're going to be running more than one DIMM at a time.

After completing a very simple DDR333 run with one DIMM installed, we moved on to the most strenuous situation possible at stock memory speeds with this particular motherboard, which was 2/3 banks filled with memory running at 333MHz. We were able to achieve the following DRAM settings:

These are without a doubt very good timings and the majority of users will be happy with this combination of memory performance at 333MHz. However, we're still disappointed that we couldn't run three sticks of memory at 333MHz, as there are quite a few users out there that like to use as much memory as possible in their systems. MSI needs to work on this aspect of the Max2-FIR, especially since we've seen more than a couple of 845PE boards run three DIMMs at 333MHz with similar DRAM timings.

To ensure these timings were stable we ran our usual array of stress tests. We ran Prime95 torture tests for hours at a time at the timings listed in the charts above. In fact, the MSI 845PE Max2-FIR was able to withstand an impressive 48 hours of Prime95 torture tests. During the time this board was being stressed, we also performed general tasks like running Word, Outlook, IE, and some data compression here and there. Our Super Pi tests also proved successful, as the most strenuous runs didn't faze the Max2-FIR.

Due to the constrained memory options and adjustments available in the Max2-FIR's BIOS, we were forced to run additional memory tests on this motherboard. These memory tests are exactly the same ones that were run in our previous 845PE review, which featured the AOpen AX4PE Max. So as before, we decided to install multiple types of memory from different manufacturers. Samsung, Kingston, Twinmos, Mushkin, and Corsair memory were all tested at the frequency of 333MHz on the MSI 845PE Max2-FIR. Due to the massive amount of different memory modules used, we were only able to run Prime95 for a handful of hours at a time. Still, we're fairly happy to report that after hours of testing, all memory modules except Mushkin's passed our Prime95 torture tests. For some odd reason, there seemed to be a problem with running two Mushkin modules (normally rated at CL2.5 DDR400) at 333MHz. We couldn't seem to run both Mushkin modules at CAS 2-2-2-5 timings, even though this was continuously repeatable with four different brands of DDR333 or better memory. Perhaps this issue is only regulated to this particular MSI motherboard, as we haven't encountered a problem with Mushkin memory with any other 845PE motherboard.