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 would 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 in this particular test.

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 at this time. 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.

Disk Controller Performance Ethernet Performance
Comments Locked

20 Comments

View All Comments

  • Dfere - Thursday, May 18, 2006 - link

    Just bought this and I am having a problem even getting this to post. Fans boot and HD starts to boot. No video image, monitor resets and then blanks.

    I stripped out unnecesary components- still no joy.

    I have an Thermaltake XaserIII 480 with adapter and EZ 4 plug molex connected.
    SATA 1 Port using WD 80 gig HD
    Pentium D 805 w/ stock heatsink.
    Inno3d 6200 TC video in 1st PCI 16 slot.

    2X1 Gig PNY memory in Black series of slots.

    I will be trying to use only one stock of memory, and then will be taking MB out of case and trying to boot on non-conducitve surface. Anyone else got advice?
  • TechJunkie - Friday, March 3, 2006 - link

    I've had it now for almost a week and have started noticing problems. Many forums are now littered with complaints, mainly aimed at the Marvell issue.

    I mistakenly had set up my system the first time with the Intel chipset and using "Standard IDE"... system worked fine other than the issue with the Marvell. Benchmarks ran fine and 3DMark06 looked great.

    I then reinstalled the OS, this time doing the PITA F6 procedure and set up the drives on the Intel controller using RAID (but didn't actually set up a RAID, simply wanted the AHCI with the ability to in the future set up RAID).

    This morning I tried to run the 3DMark06 benchmark and while it ran to completion, the graphics screen seemed all corrupted throughout...I'm thinking power problem but it didn't have this issue last go around. Can setting up the drives differently have caused this from a power perspective or is this just one of the possibly many things wrong with this board?

    I am wondering if I should RMA it back to zipzoom and get the Intel board, which has always been my choice, except this time the extra SATAII ports provided by the Marvell persuaded me to give ASUS a shot...ironic, isn't it!

    In your opinion, based on what you know, will the issues with this board be resolved by BIOS/driver updates? It is a PITA to take my rig apart to swap out the motherboard but I will bite the bullet and get the Intel 975X board if this board can't be fixed via BIOS/driver updates.
  • medic91b2 - Monday, February 6, 2006 - link

    I've had this m/b for a few week's and I have to say that it has many bug's in it. As far as i know, no one has gotten the marvell sata driver's to work and on the intel side people have gotten many mixed result's. I gave up trying to make a raid on this board. Asus support is nowere to be found, I think if any one want's to rate it then yhey should test all of the components of the board
  • Gary Key - Saturday, February 11, 2006 - link

    quote:

    I've had this m/b for a few week's and I have to say that it has many bug's in it. As far as i know, no one has gotten the marvell sata driver's to work and on the intel side people have gotten many mixed result's. I gave up trying to make a raid on this board. Asus support is nowere to be found, I think if any one want's to rate it then yhey should test all of the components of the board


    Email me and I will provide you with an updated bios that solves the USB issue and voltage spikes. I still have not had an issue with the Marvell RAID setup and two Raptors.
  • Zucker2k - Wednesday, February 15, 2006 - link

    Gary,
    Could you please e-mail me the updated bios? email: qommonsense@gmail.com. Thanks.
    Richard.
  • Joepublic2 - Wednesday, January 18, 2006 - link

    As far as the motherboard goes, it looks pretty good. NB/SB are passively cooled, the layout is good and the MOSFETs are heatsinked (I've heard of problems with the voltage regulators on cheaper intel boards failing, often because they got too hot). I like how Anandtech's motherboard articles are very critical and mention things that other sites seem to glaze over, such as if the board's voltage regulator is three, four or eight phase.

    quote:

    considering this processors penchant for electricity


    should read "processor's"
  • Gary Key - Thursday, January 19, 2006 - link

    quote:

    should read "processor's"


    It is corrected. I had it marked on the final edit but missed it before publication, excuse the Homer moment please. :-) It is a good board but we felt like Asus did not go all the way on it and should have considering the price point. Although the board is not in retail yet we were somewhat disappointed with the bios also. Asus is working on a new release at this time and we hope to be able to test it in the next roundup.
  • ghg - Monday, January 23, 2006 - link

    Hi

    When may we expect the ANANDTECH 975x shootout ?

    BR

    Gary G.
  • Gary Key - Monday, January 23, 2006 - link

    quote:

    When may we expect the ANANDTECH 975x shootout ?


    We should be finished in about three weeks. We have two additional boards in house and another two on their way. We will also have an overclocking shoot out with the Presler and Cedar Mill chips.
  • ghg - Friday, March 3, 2006 - link

    Hi

    Any news about the 975x shootout ?

    BR

    Gary G.

Log in

Don't have an account? Sign up now