Ethernet Performance

The new motherboard test suite, launched earlier this year, includes LAN performance measurements. The test procedure that we use was first described in a VIA white paper.

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 to be 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:
Nttcps - m 4,0, -a 4 - l 256000 - n 30000
On the client side (the motherboard under test), we used the following Command Line:
Nttcpr - m 4,0, -a 4 - l 256000 - n 30000
At the conclusion of the test, we captured the throughput and CPU utilization figures from the client screen.

Ethernet Throughput

Ethernet Overhead

Earlier reviews have already measured the performance advantage of PCI Express Ethernet compared to PCI Ethernet in Gigabit LAN performance. The Sapphire PURE Innovation uses the Marvel 88E8052 Ethernet on the PCIe bus. The Marvel used by Sapphire competes very well with other PCIe Gigabit Ethernet showing a throughput in the 950 Mb/sec range. CPU utilization was a very high 51% in a field where every Gigabit Ethernet controller on the PCIe bus measures high CPU utilization of 35% to 50%.

CPU utilization measurements for Gigabit Ethernet can be somewhat misleading, since they measure the percentage of CPU tie-up at sustained Gigabit receiving. In reality, it would be very rare that you would actually see sustained Gigabit transmission levels on your PC. For that reason, you should put the measured CPU utilization of Gigabit PCIe LAN in perspective. It will almost always be much lower than what we have measured.

Firewire and USB Performance Audio Performance
Comments Locked

52 Comments

View All Comments

  • RobFDB - Saturday, July 30, 2005 - link

    Guess you missed where i said "(with the exception of MSI)". Learn to read mate before you go posting.
  • RobFDB - Friday, July 29, 2005 - link

    ATI and Sapphire should be congratulated for bringing the AC880 to AMD users. We had it good with Soundstorm but since then onboard audio as gone back several steps (with the exception of MSI). Its good that AMD users are being given the option to have quality onboard audio.
  • bob661 - Friday, July 29, 2005 - link

    This what impresses me the most about these boards is this codec support. I still won't buy an ATI chipset until the third or fourth version comes out (you guys can test it for me) but impressive features and performance nonetheless.
  • jab98 - Friday, July 29, 2005 - link

    *codec
  • erwos - Friday, July 29, 2005 - link

    "[AMD] Enthusiast" is written with a capital E in the article, and it should not be, since it's not a proper noun. Please fix this error, because it looks grossly unprofessional to anyone with a reasonable command of the written word.
  • RobFDB - Friday, July 29, 2005 - link

    Really though, get over it. It doesnt matter in the slightest if we're being honest here. Anyway back to more important matters.

    I'm really happy that ATI have managed to bring a top performing board aimed at enthusiasts to market. I was also extremely impressed to see Sapphire implement 4v for the RAM. One issue that i'd like to see investigated is wether the cold boot issue that affects DFI NF4 boards using OCZ VX mem @ high voltages affects the Sapphire board too. Aside from that this is a very impressive showing from ATI. One last thing. I have a x850XT PE and i'm not sure if that can be used as a slave card when ATI bring out the R520. If so that would make a very attractive upgrade.
  • rjm55 - Friday, July 29, 2005 - link

    The X850XT PE works fine as a slave with the X850 Master Card. In demos at Computex, ATI was showing an X850 Master with an X850XT PE slave.
  • Jojo7 - Friday, July 29, 2005 - link

    This isn't exactly true. Ati distributed a special driver that SIMULATED crossfire. The actual cards were really just 2 identical x850xtpe's. Though, one probably had an altered bios to simulate a master card.

    Read it for yourself: http://anandtech.com/weblog/default.aspx?bid=231">http://anandtech.com/weblog/default.aspx?bid=231
  • dlamblin - Friday, July 29, 2005 - link

    Did I miss the mention in the article? Is this an ATX or an mATX board. I'm guessing the former, but it wouldn't be out of place to list the fact along side the rest.
  • erwos - Friday, July 29, 2005 - link

    It's ATX. If it has more than four slots, it's too big to fit the mATX standard.

Log in

Don't have an account? Sign up now