: News
: Reviews
: Editorials

: Archives
: About
: Contact
: Advertising
: Privacy
: Links
: Forums

BFG Tech 590 SLI Motherboard Review

BFGRAN590S… or Something

February 26, 2007

By Josh Walrath

Results

SiSoft Sandra XI

            This latest version supports all Windows operating systems, and adds a few new wrinkles to some of the benchmarks.  The 64 bit version was used.

            Both boards show essentially even performance with this benchmark.  Neither is clocked higher than the other.  The Multimedia benchmark testing iSSE2 performance showed essentially the same performance for both boards.

            Here we see the first real difference between the boards.  While the DIMMS are set at 4:4:3:8 timings, we see that Asus is probably using some tighter hidden timings than the BFG board.

The BFG 590SLI is very close in size to the Asus M2N32-SLI.  It is unfortunate that BFG was not able to replace the cooling with something similar to Asus.

            For the next networking tests, I used two different computers on my network to take a look at Gig-E performance.  Test System 1 is based on the MSI K9A Platinum which uses a Realtek PCI-E Gig-E chip running Windows Vista Business 32 bit.  The second system is the Biostar T-Force 550 which uses the integrated NVIDIA Gig-E controller running Windows XP 64.  Test System 1 is located on the same Gig-E switch as the motherboard tested, while Test System 2 is on another Gig-E switch uplinked to the primary switch (one extra hop).

            The Realtek controller on the MSI K9A is not the world’s greatest.  What is interesting though is the latency difference between the Asus and BFG boards.  For being on the same switch, it is quite a bit slower in terms of latency with the BFG.  With such a small overall time latency, differences in the speed of the onboard clock could account for differences in the results.  Still, it does appear as though the BFG implementation is slightly slower.

            Here we see a lot more parity between the two, even with the extra jump to the uplinked Gig-E switch.  It is interesting to see the latency for the BFG board is the same for Test System 2 as it is for Test System 1.  The nForce 550 Gig-E controller does appear to be a much more robust unit than the Realtek chip on the MSI board as it is about 10 MB/sec faster.

In this view we see a dual slot graphics card covering the 1X PCI-E slot.  The 4X slot is still usable.

            In the SiSoft drive index tests, I chose random read performance instead of sequential.  This should give a good idea of general applications performance versus the typical sequential read speeds of HD Tach.

            We see that both SATA controllers have the same level of performance with this particular drive.

NVIDIA did a nice job in spacing the SATA ports, as all of them are exposed even when using a dual slot cooler graphics card.

HD Tach

            While the sequential reads and writes of HD Tach may not directly reflect desktop usage, it does test a few other factors to hard drive performance that SiSoft skips.

            There is a slight difference in burst speed between the boards, but nothing major.  Overall the controllers are evenly matched, which is not a surprise since they use the same Forceware 9.35 drivers and 590 SLI chipset.  The CPU percentage is nice and low as well.

Next:  More Results

Copyright 1999-2007 PenStar Systems, LLC.