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. We utilize a RAM disk as our "server", since memory removes almost all overhead from the serving end. We turn off disk caching on the USB and Firewire side by setting up the drives for "quick disconnect" so our results are consistent.

We use 2GB of fast 3-3-3-8 system memory set up as a 450MB RAM disk and 1550MB of system memory. Our standard file is the SPECviewPerf install file, which measures 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 utilizing our internal Windows based timing program. 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.

USB Performance


Possibly the most interesting finding in our Firewire and USB throughput tests is the outstanding performance of an external hard drive connected to Firewire 800. Our benchmarks show Firewire 800 is up to 48% faster than a drive connected to the more common Firewire 400, and about 32% faster than the fastest USB 2.0 solution.

The AOpen board offers the Agere based IEEE 1394 Firewire option that provides very competitive performance with the more common TI chipset solutions. The Intel USB 2.0 performance offers 22% better performance than the ULi based controller in throughput.

Ethernet Performance

The current motherboard test suite includes LAN performance measurements. All of these boards utilize PCI or PCI Express based 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 Intel motherboards.

We set up one machine as the server; in this test, an Intel system with an Intel CSA Gigabit LAN connection. Intel CSA has a reputation for providing fast throughput and is a logical choice for our Gigabit LAN server.

On the server side, we used the following Command Line as suggested by the VIA whitepaper on LAN testing:

Ntttcpr -m 4,0,‹server IP› -a 4 -l 256000 -n 30000

On the client side (the motherboard under test), we used the following Command Line:

Ntttcps -m 4,0,‹client IP› -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


AOpen's choice of the Marvell 88E8053 10/100/1000Mb/s Ethernet controller is an excellent addition to this board. The performance is basically equal to that of the same Marvell 88E8053 controller utilized on the Asus board. The lower CPU utilization rate on the Marvel 88E8001 PCI controller is directly attributable to the lower throughput numbers generated in our tests.

Disk Controller Performance Audio Performance
Comments Locked

81 Comments

View All Comments

  • Myrandex - Thursday, May 4, 2006 - link

    Believe it or not, people can run 64bit apps for other reaons from 4GB memory...such as increased performance? I knwo not many apps right now show increased perofrmance with 64bit mode, but some do and I believe this will go up when more developers start programing for it with more intensity and all that.
    Jason
  • peternelson - Thursday, May 4, 2006 - link

    For me this is the showstopper, and the reason I much prefer Turion at present.

    Core Solo/ Core Duo are 32 bit only. That makes them quite rarely disadvantaged - even the new Celerons can do 64 bit EM64T.

    This board would be of much more interest if the review said "THIS BOARD IS MEROM-READY". As it doesn't, I assume it isn't. The review here might have highlighted this important issue which will be crucial deciding factor for some potential purchasers, especially considering the price.

    I see Conroe and Merom systems quickly filling the market need this board is targetted at.
  • Gary Key - Friday, May 5, 2006 - link


    quote:

    Core Solo/ Core Duo are 32 bit only. That makes them quite rarely disadvantaged - even the new Celerons can do 64 bit EM64T.


    Only disadvantaged for those utilizing 64-bit applications and operation systems, but for 95% of the marketplace they are fine. Not that I do not think it is an issue, but one that is way overblown in today's marketplace.
    Probably in a couple of years it will be completely different but for now, 32-bit applications are still king of the hill (market share wise). This upsets me as I have to believe the entire migration path should have been completed by next year but the foot dragging between Microsoft and Intel negated AMD's advantage in this area for far too long. With that said, Merom takes care of the Yonah 64-bit issue in the mobile sector for Intel in a couple of months. It will be very interesting to see how it compares to Turion 64 X2, it will be good for all of us to have choices in the mobile sector.


    quote:

    This board would be of much more interest if the review said "THIS BOARD IS MEROM-READY". As it doesn't, I assume it isn't. The review here might have highlighted this important issue which will be crucial deciding factor for some potential purchasers, especially considering the price.


    We cannot quote on whether this board is Merom approved or not until qualification testing is complete. Will a Merom T7400 work in this board with the bios 1.3b, yes it will, and the initial results show a great deal of improvement over Yonah in certain benchmarks. Will Merom be fully supported by Intel and AOpen on this board, we do not know yet. AOpen is waiting a production spin of Merom before trying to the qualify it on this board and even then Intel might force everyone to Crestline for official support. We are still trying to figure out if Merom is going to work with 945GM as stated a couple of months ago. The entire situation is so up in the air right now that any statement has the potential to be wrong so being quiet is the best possible option. :)
  • johnsonx - Thursday, May 4, 2006 - link

    from the spring idf article on AT:

    quote:

    The other beauty of Merom is that it is 100% compatible with existing Yonah designs, meaning that all Core Duo notebooks today should be able to accept a Merom processor with at most a BIOS update.


    Now that may not be a 100% promise, but until Merom is actually released, there's no way to be 100% certain any given board will work with it.
  • Viditor - Thursday, May 4, 2006 - link

    quote:

    You plan on running apps that require greater than 4GB of ram in the next couple years?

    You mean 4GB or greater...at 4GB, a 32bit processor will be using PAE already.
    And yes, I find that 4GB is a great help on a number of my apps...
  • Questar - Thursday, May 4, 2006 - link

    Mobile/SFF apps?

    Name them.
  • Viditor - Thursday, May 4, 2006 - link

    quote:

    Name them

    Photoshop CS2, Premiere Pro, and Mental Ray...
  • Questar - Thursday, May 4, 2006 - link

    I'm a heavy Premiere user, and I can tell you it's a 32 bit app.
  • Chadder007 - Thursday, May 4, 2006 - link

    You can still run it under a 64 bit OS and have access to the 4gb or greater amount of ram.
  • defter - Thursday, May 4, 2006 - link

    No you can't. If you have an 32bit software then it can only address 4GB of memory.

Log in

Don't have an account? Sign up now