DFI LanParty UT X58-T3eH8 LGA 1366 Motherboard Review

Memory Benchmarks

Everest Ultimate is the most useful tool for any and all bench markers or overclockers. With the ability to read most voltage, temperature, and fan sensors on almost every motherboard available, Everest provides the ability to customize the outputs in a number of forms for display on your desktop. In addition to this, the memory benchmarking provides a useful tool of measuring the changes to your memory sub-system when tweaking to measure the differences.

What is interesting about these results is that despite the drastic differences in CPU frequency, memory frequency and timings, all of our results are fairly similar. There also doesn't seem to be much of a difference, if any at all, between the 2x2GB and 3x2GB setup. This is perhaps an indication that there won't be much in the way of a difference with the rest of the benchmarks since Everest has traditionally been a could gauge of actual memory bandwidth.

And as soon as I praise Everest for its ability to gauge performance, it throws us one of these curveballs. Clearly Everest is having trouble calculating or measuring the memory latency as there is no way that a dual channel setup has a much lower latency than the triple channel configuration. At the same time, if this is true, it should show up in SPi 32M results since the memory bandwidth was so similar. At least one would think this to be the case.

SiSoft Sandra is a popular and well used benchmark in the industry but not really a friend of serious benchmarkers. The results SiSoft Sandra produces have been suspect at times basing the numbers it comes up with on system specs and not actual testing. The latest version of Sandra seems to be one of the few programs that appear to calculate memory bandwidth consistently so we decided to include it in today’s benchmarks. Like we have always said with SiSoft Sandra though, take these results for what they are and nothing more.

My love for Sandra just continues to grow...can you see the sarcasm in the printing? Just like Everest latency, there seems to be a problem with SiSoft Sandra and the dual channel configuration. Memory bandwidth drops off considerably while in Everest it holds a steady line going from three to tow sticks. I guess we will have to rely on ScienceMark to provide the tie breaker, and ScienceMark is what we would tend to trust anyway as Everest is getting more and more skeptical with its results on i7.

In an exact opposite response to the Everest bandwidth and latency results, Sandra has gone on to conclude that latency between the two and three stick setups is the same with the disparity being in bandwidth. Take the results for what they are worth and let’s move on.

ScienceMark is an almost ancient benchmark utility at this point in time and hasn't seen an update in a long time. It is, however, still a favorite for accurately calculating bandwidth on even the newest chipsets.

With ScienceMark looking more and more like the better program for measuring average memory bandwidth over Everest and SiSoft Sandra, we can see how our three setups scale. They are all very similar in bandwidth but the 2x3GB setup does seem to be slightly behind the 3x2GB setup and the stock configuration is just a hair behind both of those. One thing is for certain, the 6-7-6 timing set we have gone with for our setup definitely has some strength in bandwidth results, let's see hot that relates to program performance. SuperPi 32M should be an interesting set of results given these bandwidth and latency numbers.