Firewire and USB Performance

After looking at many options for Firewire and USB testing, we finally determined that an external USB 2.0, Firewire 400, and Firewire 800 hard disk might be a sensible way to look at USB and Firewire throughput.

Our first efforts at testing with an IDE or SATA drive as the "server" yielded very inconsistent results, since Windows XP sets up cache schemes to improve performance. Finally, we decided to try a RAM disk as our "server", since memory removed almost all overhead from the serving end. We also managed to turn off disk caching on the USB and Firewire side by setting up the drives for "quick disconnect" and our results were then consistent over many test runs.

We used 1GB of fast 3-2-2-8 system memory set up as a 450MB RAM disk and 550MB of system memory. Our standard file is the SPECviewPerf install file, which is 432,533,504 bytes (412.4961MB). After copying this file to our RAM disk, we measured the time for writing from the RAM disk to our external USB 2.0, Firewire 400, or Firewire 800 drive using a Windows timing program written for AnandTech by our own Jason Clark. The copy times in seconds were then converted into Megabits per second (Mb) to provide a convenient means of comparing throughput. Higher rates therefore mean better performance.

Firewire and USB Performance

Possibly the most striking finding in our Firewire and USB throughput tests is the performance of an external hard drive connected to Firewire 800. Firewire 800 matters and should be a standard option on this board. Our benchmarks show Firewire 800 is up to 46% faster than a drive connected to the more common Firewire 400, and about 29% faster than USB 2.0.

Our test is just one of many throughput tests, but in this benchmark, it is clear that the VIA Firewire 400 chip is faster than TI's 1394a chip. The NVIDIA nForce4 USB 2.0 controller is slightly faster than Intel’s solution. Unlike the other boards in this price range, the Asus P5N32-SLI Deluxe does not offer a Firewire 800 option.

Disk Controller Performance Ethernet Performance
Comments Locked

70 Comments

View All Comments

  • Beenthere - Thursday, October 27, 2005 - link

    Why would ANYONE spend the coin to buy an Intel based SLI system when you can buy an AMD system for the same price that will out-perform the Intel system and be upgradable for years???

    The only reason I can see for anyone buying any Intel product at this time is if they are stuck with an Intel system already and they desire to upgrade to a faster chip if they can find one to fit whatever socket Mobo they have. Otherwise I see no logical reason whatsoever for even considering an obsolete, under-performing Intel product.
  • Shintai - Thursday, October 27, 2005 - link

    You ask the wrong question. Why would anyone buy a SLI/CrossFire system at all.
  • bob661 - Thursday, October 27, 2005 - link

    quote:

    You ask the wrong question.
    He can ask any question he wants. There are no wrong questions.
    quote:

    Why would anyone buy a SLI/CrossFire system at all.
    Because one can. :)
  • Ricky Ling - Thursday, October 27, 2005 - link

    I only afraid on the future upgradeability on this board because according to ASUS Support FAQ (as this news has not been revealed on the Internet), this board does not support future Pentium 4 that based on 65nm technology, namely Cedar Mill (Single core replacing Prescott) and Presler (Dual core replacing Smithfield)

    So how is testing done the Presler sample processor cause I thought Anand got holding of 1 Presler sample...Pls double check at the following :
    http://support.asus.com/faq/faq_right_second_detai...">http://support.asus.com/faq/faq_right_s...P5N32-SL...

    As Presler start hitting OEM already....we need more info on this issue???

  • Gary Key - Sunday, October 30, 2005 - link

    quote:

    As Presler start hitting OEM already....we need more info on this issue???


    The initial information I have from Asus this weekend is the board will support the 65nm CPU range with a bios update and the CPU steppings must be B1 or above.
  • Gary Key - Thursday, October 27, 2005 - link

    Hello,

    I had already asked for public clarification about this issue from Asus after visiting the website last week. I believe the entire FAQ has not been properly updated as it still states the Pentium 820 will only work in single core mode which is no longer the case with this chipset revision. We have not had an issue testing the Presler or Cedar Mill CPUs with current board designs from different suppliers.

    I will respond once I have further information.

    Thank you.
  • Chuckles - Thursday, October 27, 2005 - link

    How's the clearance on the right-most PCIe x1 slot? It looks like a card put in there would be perilously close to the northbridge heatsink.
  • Gary Key - Thursday, October 27, 2005 - link

    Hi,

    The D-Link DGE-560T PCIe network adaptor card fit fine and it is the longest card I tried in the x1 slot. I would agree if a card had several logic chips placed in the right area on the back of the board then clearance could be an issue with the heatsink. However, the few peripheral PCIe cards that I have fit fine.
  • erwos - Thursday, October 27, 2005 - link

    I wouldn't have minded seeing the difference between bridged SLI and bridgeless SLI on both SLI Dual x16 and SLI Dual x8 motherboards.

    My gut feeling is that nVidia is going to be phasing out the bridge in the near future, if they can get the same performance without it using the PCIe bus.

    -Erwos
  • DigitalFreak - Thursday, October 27, 2005 - link

    "ASUS CPU Lock Free (BIOS setting to unlock multiplier locked CPUs)"

    Does this mean that all multipliers (up and down) are available again?

Log in

Don't have an account? Sign up now