Memory Stress Testing: DFI LANParty UT nF3 250Gb

The memory stress test is very basic, as it simply tests the ability of the DFI nF3 250Gb to operate at its officially supported memory frequency (400MHz DDR), at the lowest memory timings that OCZ PC3200 Platinum Rev. 2 Modules will support. Because of the numerous questions that have been asked about the number of DIMMs on Socket 754 boards, Memory Stress Testing was conducted by running 1 DIMM, 2 DIMMs, and all 3 DIMMs at DDR400. All DIMMs used for stress testing were 512MB double-sided (or double-bank) memory.

Stable DDR400 Timings - 1 DIMMs
(1/3 DIMMs populated - Single-Channel mode)
Clock Speed: 200MHz
CAS Latency: 1.5
RAS to CAS Delay: 2T
RAS Precharge: 10T*
Precharge Delay: 2T
Command Rate: 1T
*Several memory tests have shown that memory performs fastest on the nVidia nForce chipsets at a TRas (RAS Precharge) settings in the 9 to 13 range. We ran our own Memory Bandwidth tests with memtest86 with TRas settings from 5 to 15 at a wide range of different memory speeds. The best bandwidth was consistently at 9 to 11 at every speed, with TRas 10 always in the best range at every speed. The performance improvement at TRas 10 was only 2 to 4% over TRas 5 and 6 depending on the speed, but the performance advantage was consistent across all tests. All benchmarks were run at a TRas setting of 10.

We have not tested a Socket 754 board that offered a CAS 1.5 setting. The OCZ memory ran without problems with 1 DIMM at 1.5-2-2-10 settings.


Stable DDR400 Timings - 2 DIMMs
(2/3 DIMMs populated - Single-Channel mode)
Clock Speed: 200MHz
CAS Latency: 2.0
RAS to CAS Delay: 2T
RAS Precharge: 10T*
Precharge Delay: 2T
Command Rate: 1T
*Several memory tests have shown that memory performs fastest on the nVidia nForce chipsets at a TRas (RAS Precharge) settings in the 9 to 13 range. We ran our own Memory Bandwidth tests with memtest86 with TRas settings from 5 to 15 at a wide range of different memory speeds. The best bandwidth was consistently at 9 to 11 at every speed, with TRas 10 always in the best range at every speed. The performance improvement at TRas 10 was only 2 to 4% over TRas 5 and 6 depending on the speed, but the performance advantage was consistent across all tests. All benchmarks were run at a TRas setting of 10.

All earlier 754 benchmarks were run with 2 DIMMs, and this is the configuration that was used for all standard benchmark tests. The DFI LANParty UT was completely stable with 2 DIMMs at the rated timings of 2-2-2-10 at default speed.


Filling all three available memory slots is more strenuous on the memory subsystem than testing 2 DIMMs on a motherboard. We were very pleased to find that 3 DS DIMMs (1.5GB) of memory worked fine at timings almost the same as the aggressive timings that we used for 2 DIMMs. The DFI is one of the few Socket 754 boards that we have tested which can handle 3 DS DIMMs. You do need to use slightly slower timings with 3 DIMMs, but the 2-2-3-10 is very close to the rated 2-2-2-10 timings of the memory which we used for testing.

Stable DDR400 Timings - 3 DIMMs
(3/3 DIMMs populated - Single-Channel mode)
Clock Speed: 200MHz
CAS Latency: 2.0
RAS to CAS Delay: 3T
RAS Precharge: 10T*
Precharge Delay: 2T
Command Rate: 2T
*Several memory tests have shown that memory performs fastest on the nVidia nForce chipsets at a TRas (RAS Precharge) settings in the 9 to 13 range. We ran our own Memory Bandwidth tests with memtest86 with TRas settings from 5 to 15 at a wide range of different memory speeds. The best bandwidth was consistently at 9 to 11 at every speed, with TRas 10 always in the best range at every speed. The performance improvement at TRas 10 was only 2 to 4% over TRas 5 and 6 depending on the speed, but the performance advantage was consistent across all tests. All benchmarks were run at a TRas setting of 10.

The boot screen and Sisoft Sandra 2004 both reported a Command Rate of 2T when all 3 DIMM slots were filled.

FSB Overclocking: DFI LANParty UT nF3 250Gb Performance Test Configuration: DFI LANParty UT nF3 250Gb
Comments Locked

54 Comments

View All Comments

  • Bozo Galora - Wednesday, September 8, 2004 - link

    #33 - yeah whats up with that
    best perf with one stick, but only 512

    and BTW:ZZF had them $135, went out the door fast
  • ukDave - Wednesday, September 8, 2004 - link

    Also, i've just been checking DFI's site and it says it only supports 512MB DIMMs, bit of a bummer cos i was thinking of getting a single 1GB stick, sux0r.
  • ukDave - Wednesday, September 8, 2004 - link

    Great article Wesley. Lucky for you guys in the states DFI's RMA process is actually good, here in Europe (UK) we have to deal with DFI NL (Netherlands), and they tbh are crap. So bad in-fact some retailers are debating whether to carry future boards from DFI. I certainly hope they do though, as this board sure looks like a belter, and it WILL be replacing my Asus K8V-SE Deluxe which has no PCI lock (although its still a great board).
  • cnq - Wednesday, September 8, 2004 - link

    For everyone looking for "754 overclocking for dummies" guide: it's already been written. Check out

    http://www.dugu9tweaks.net/guides/a64oc/index.html
  • 8NP4iN - Wednesday, September 8, 2004 - link

    /me wants oveclocking for dummies guide 2 :]
  • sprockkets - Wednesday, September 8, 2004 - link

    No UV reactive green like the other boards? :(
  • JBT - Wednesday, September 8, 2004 - link

    Wow looks like a sweet board maybe I can pick up one of these instead of my weak OCing 8KDA3J
  • Wesley Fink - Wednesday, September 8, 2004 - link

    #23-#26 -
    You are correct. Since the board was sent without the usual manual, I assumed the SATA ports between the AGP slot and CPU socket were SATA 1 and 2. I now have the manual file, and in fact those ports are identified as SATA 3 and 4.

    The 250GB SATA drive was connected to SATA 3 for the overclocking tests.
  • alex1971 - Wednesday, September 8, 2004 - link

    S-ATA 3+4 are OK for HTT > 240 Mhz

    S-ATA 1+2 finished at HTT > 240 Mhz because of the external PHY !!!!
  • alex1971 - Wednesday, September 8, 2004 - link

Log in

Don't have an account? Sign up now