Asus P5N32-SLI: Dual x16 - What Dreams Are Made On . . .
by Gary Key on October 27, 2005 12:05 AM EST- Posted in
- Motherboards
Ethernet Performance
The current motherboard test suite includes LAN performance measurements. All of these boards utilize PCI Express controllers with the only difference being the supplier of the core logic.
The Windows 2000 Driver Development Kit (DDK) includes a useful LAN testing utility called NTttcp. We used the NTttcp tool to test Ethernet throughput and the CPU utilization of the various Ethernet Controllers used on the nForce4 Ultra motherboards.
We set up one machine as the server; in this case, an Intel box with an Intel CSA Gigabit LAN connection. Intel CSA has a reputation for providing fast throughput and this seemed a reasonable choice to serve our Gigabit LAN clients.
At the server side, we used the following Command Line as suggested by the VIA whitepaper on LAN testing:
All standard Ethernet tests were performed with standard frames and the NVIDIA Active Armor suite disabled unless otherwise noted. Gigabit Ethernet supports Jumbo frames as well and will provide a further reduction in CPU overhead.
We added a further test scenario in which ActiveArmor was enabled on the Asus P5N32-SLI Deluxe board via the new 6.82 platform driver set. The throughput numbers increased by almost 3% while the CPU utilization dropped by over 7% with the standard settings enabled in the application. We also enabled Jumbo frames with ActiveArmor resulting in an increase in throughput numbers by 5% while the CPU utilization dropped by over 12%. We will be adding this benchmark testing to our standard test suite along with providing more information regarding the NVIDIA firewall solution in the near future. I highly recommend that if you have a NF4 based board, you should utilize ActiveArmor in its current configuration.
The current motherboard test suite includes LAN performance measurements. All of these boards utilize PCI Express controllers with the only difference being the supplier of the core logic.
The Windows 2000 Driver Development Kit (DDK) includes a useful LAN testing utility called NTttcp. We used the NTttcp tool to test Ethernet throughput and the CPU utilization of the various Ethernet Controllers used on the nForce4 Ultra motherboards.
We set up one machine as the server; in this case, an Intel box with an Intel CSA Gigabit LAN connection. Intel CSA has a reputation for providing fast throughput and this seemed a reasonable choice to serve our Gigabit LAN clients.
At the server side, we used the following Command Line as suggested by the VIA whitepaper on LAN testing:
Ntttcps - m 4 ,0, -a 4 - l 256000 - n 30000On the client side (the motherboard under test), we used the following Command Line:
Ntttcpr - m 4 ,0, -a 4 - l 256000 - n 30000At the conclusion of the test, we captured the throughput and CPU utilization figures from the client screen.
All standard Ethernet tests were performed with standard frames and the NVIDIA Active Armor suite disabled unless otherwise noted. Gigabit Ethernet supports Jumbo frames as well and will provide a further reduction in CPU overhead.
We added a further test scenario in which ActiveArmor was enabled on the Asus P5N32-SLI Deluxe board via the new 6.82 platform driver set. The throughput numbers increased by almost 3% while the CPU utilization dropped by over 7% with the standard settings enabled in the application. We also enabled Jumbo frames with ActiveArmor resulting in an increase in throughput numbers by 5% while the CPU utilization dropped by over 12%. We will be adding this benchmark testing to our standard test suite along with providing more information regarding the NVIDIA firewall solution in the near future. I highly recommend that if you have a NF4 based board, you should utilize ActiveArmor in its current configuration.
Facebook
Twitter
RSS
70 Comments
View All Comments
Wesley Fink - Thursday, October 27, 2005 - link
Those accustomed to looking at AMD Athlon64 Performance Scores are used to seeing numbers almost the same across motherboards because the memory controller is on the CPU. The Intel Memory Controller is in the chipset and performance varies much more depending on the chipset and the quality of the motherboard design.That is one reason we often test Latency in Intel MB tests. If you look at the Latency test results in this review you will see a fairly wide variation across the tested chipsets and motherboards for the Intel CPU. Athlon64 Latency tests would all be virtually the same with the memory controller a part of the processor.
toyota - Thursday, October 27, 2005 - link
how can motherboards make that much difference in gaming?? in the Doom3 benchmarks they range from 63 to 95 fps! i dont understand benchmarks like that and nobody else ever makes a comment. am i missing something?Gary Key - Thursday, October 27, 2005 - link
Good Day.....I revised the article statement about the Doom3 scores. I left it out on the final copy by mistake. We are still investigating the differences as an upcoming article from Randi on another nF4 Intel SLI board has scores higher than the numbers I have reported by a fair margin again. In fact, I will be testing the Abit board once it arrives with an disk image from my previous tests.
Due to the memory controller not being on the CPU (current Athlon64 family design) the Intel based motherboard design makes a great deal of difference not only from a chipset choice but also from how well a board manufacturer designs and implements the supporting components and bios.
Thank you.
xsilver - Thursday, October 27, 2005 - link
the 63fps is on another chipset, the chipset affects memory, hdd performance directly and everything else indirectly..the 95fps actually looks like an anomaly -- and AT member will have to confirm that (SLI setup in the NI8?)
so in fact the numbers are actually 75.3-79.4fps which is an acceptable range for the same chipset
many people forget the mobo is the heart of the system, it pays to get a good one :)
TransientBen - Thursday, October 27, 2005 - link
Mixing philosophy, classical literature and computer hardware reviews? Gotta love it. Though it's difficult to not get caught up in the philosophy and then question, "Is it worth $200 for a motherboard or a plane ticket to a new place?"There is so little time, afterall.
Always been a big Asus fan. Have one of the first (original slot a) Athlon boards still up and running after all these years - rock solid - and, more recently, a Z33A laptop that's blown me away with it's quality. I look forward to the inclusion of many of these features on future AMD boards.
noac - Saturday, November 12, 2005 - link
Hi, Im reading my manual and it says:DIMM_A1 (yellow), DIMM_A2 (black), DIMM_B1 (yellow), DIMM_B2 (black).
Channel A = DIMM_A1 and DIMM_A2
Channel B = DIMM_B1 and DIMM_B2
For dual-channel configuration, the total size of memory module(s) intalled per channel must be the same (DIMM_A1 + DIMM_A2, DIMM_B1 + DIMM_B2).
Anandtech:
Asus did an excellent job with the color coordination of the various peripheral slots and connectors. The DIMM module slots' color coordination is correct for dual channel setup.
My question which way is it? Im I getting the manual wrong? How to I palce my two mems for dualchannel?
Gary Key - Wednesday, November 16, 2005 - link
Email me if you have any issues or further questions please.Gary Key - Wednesday, November 16, 2005 - link
Sorry about the late reply...You place the memory in the two yellow DIMM slots for dual channel.
Gary Key - Thursday, October 27, 2005 - link
Depends on the time and place in my book. :-> However, considering where I could go for $200 at this point in time I will take the board. I also believe Dickens is queued up for the next article.
I had been concerned about Asus the past couple of releases as I honestly thought Intel had passed them on the high end side (useable features, stability, throw in Abit for performance) with their 925x and 955x boards until this gem landed on my doorstep.
I think the AMD version of this board should be equally adept and we should find out shortly. ;->
xsilver - Thursday, October 27, 2005 - link
I can see asus and the other mobo companies making this refresh right after/before christmas and then obviously another refresh for M2 socket