Disk Controller Performance

With the variety of disk drive benchmarks available, we needed a means of comparing the true performance of the wide selection of controllers. The logical choice was Anand's storage benchmark first described in Q2 2004 Desktop Hard Drive Comparison: WD Raptor vs. the World. The iPeak test was designed to measure "pure" hard disk performance, and in this case, we kept the hard drive as consistent as possible while varying the hard drive controller. The idea is to measure the performance of a hard drive controller with a consistent hard drive.

We played back Anand's raw files that recorded I/O operations when running a real world benchmark - the entire Winstone 2004 suite. Intel's iPEAK utility was then used to play back the trace file of all IO operations that took place during a single run of Business Winstone 2004 and MCC Winstone 2004. To try to isolate performance differences to the controllers that we were testing, we used the Maxtor MaXLine III 7L300S0 300GB 7200 RPM SATA drive in all tests . The drive was formatted before each test run and a composite average of 5 tests on each controller interface was tabulated in order to ensure consistency in the benchmark.

iPeak gives a mean service time in milliseconds; in other words, the average time that each drive took to fulfill each IO operation. In order to make the data more understandable, we report the scores as an average number of IO operations per second so that higher scores translate into better performance. This number is meaningless as far as hard disk performance is concerned, as it is just the number of IO operations completed in a second. However, the scores are useful for comparing "pure" performance of the storage controllers in this case.

iPeak Business Winstone Hard Disk I/O

iPeak MM Content Creation Hard Disk I/O

The performance patterns hold steady across both Multimedia Content IO and Business IO, with the ULi based disk controllers providing the fastest IO operations followed by the on-board NVIDIA nForce4 SATA controllers. I n particular is the excellent performance generated by the ULi IDE controller logic while the SATA performance is up to 12% better when compared to the nForce4 chipset.

CrossFire Performance Firewire and USB Performance
Comments Locked

42 Comments

View All Comments

  • CrystalBay - Friday, March 10, 2006 - link

    sorry, ;)
  • Witchfire - Friday, March 10, 2006 - link

    I have to disagree that the reviewers do not follow up on comments or complaints from users. Gary has been communicating with me regularly about my concerns with the AT8 I've been having, and has been very helpful. I'm no eleite member, and haven't even managed triple digit postings, yet he took the time & trouble to listen to my concerns, attempt to recreate them on his mobo, and lend assistance.

    Thanks, Gary, your help and professionalism have been greatly appreciated.
  • bthjf1 - Friday, March 10, 2006 - link

    The problem with all these reviews is that they don't go back to check what's is going one with the board they just tested with normal users. A number of peoples (one being me) have been arguing (with their helpdesk) and waiting for 3 months now for Abit to provide a proper support for the intel Presller core on the AW8 motherboard series (i955) or proper support for different type of memory. All I can say Abit support is very very poor and I will strongly advice anybody thinking of purchasing a board from Abit to look elsewhere. Just poppin's into http://forum.abit-usa.com/showthread.php?t=102711">http://forum.abit-usa.com/showthread.php?t=102711 and have a read ...
  • Gary Key - Friday, March 10, 2006 - link

    quote:

    The problem with all these reviews is that they don't go back to check what's is going one with the board they just tested with normal users.


    I cannot comment on other review sites, but I can assure you that Wes and I do go back and look at the history of the motherboards tested and take comments from users very seriously. I know I spend at least 10 hours a week assisting users with their issues and at times being an arbitrator for the customer with the supplier. We both spend a similar amount of hours each week if not more working directly with the suppliers on issues and trying to ensure problems are solved or at least corrected in the next product design. While we are not always successful, I do believe you would be surprised at the number of issues that are solved quickly. I will bring up your AW8 issue with Abit this weekend. :)
  • bthjf1 - Saturday, March 11, 2006 - link

    Thanks for any help you can provide. The comment was not direct to Anandtech but to a couple of other review sites which didn't really care (ie: not my problem). The concern with Abit is since the takeover from USI the support have not been great ~ total of communication. I've been reading Anandtech for many years (too long !) and my post was more to get some help with Abit . They will listen to Anandtech since you can reach millions but not necessary a single user ;-)

    Jon
  • Plasmoid - Friday, March 10, 2006 - link

    I have to agree with you, Abit need to sort out their support.

    The benchmarks prove that this is a great motherboard, interestingly it outperforms all its rivals at stock just like the AN8, but the bios lets it down (interestingly just like the AN8)
    If they can sort out these bios issues fast it should be a fantastic value motherboard. There were problems with the temperature reporting on the AN8 and incompatibility with XI-FI cards from creative that took much to long to address though. Im sad to say this kind of thing is what you can get with Abit. Hopefully it is a short term problem that they have had over the last 6 months.

    All bad things said im really happy with my abit motherboard, and the OCGuru certainly is a godsend for overclockers. Havent seen such things as completly dynamic fan speed control in any other motherboard.
  • Patrese - Friday, March 10, 2006 - link

    Am I the only one who finds this kind of positioning of the IDE connectors crappy? Most people who are buying PCs now are using SATA HDs, so the connectors should be exactly where they are, on the low-right of the board. But the IDEs are used for CD-RW/DVD-RWs drives mostly, and would be much better suited to the top right of the board IMO. I got a good-old AN7 and I just hate tbe IDE cables crossing my case from top to bottom just to reach the connnectors...

    Besides that, great review and promising mobo, once the BIOS is fixed. It's great to see Abit back!
  • n7 - Friday, March 10, 2006 - link

    I think his comments were a tad harsh, but i am one of those users who read the Asus A8R-MVP review here, & got pretty excited.
    I actually starting spreading the praises on forums i visit of this great value mobo.

    Then i got the mobo, & actually, i was quite happy with it.
    For a few weeks, that is.

    Then i started discovering all sorts of issues, the worst being the ethernet hanging/crashing, something which many others have also had issues..

    I now hear there's a Marvell driver directly from their site that supposedly fixed the problem, but i've already bought an ethernet card, so i don't really care to try it, since both the driver from the CD & the driver from Windows update didn't work properly.

    Another issue i've encountered is wildly fluctuating vcore. (1.31-1.41V, for example! - with overvoltage disabled)
    Others have also reported this. I am using software to measure this, so it's possible that it is incorrectly reading it, but i'm guessing not.

    Another problem is that even smaller overclocks aren't stable in games for me, whereas i had much higher OCs stable with my previous Neo2.
    I cannot figure out why yet, but it seems to be either the fluctuating vcore, or it's been mentioned that the RAM isn't even stable even with a divider when using 1T.

    It's one thing for OCs to vary between mobos, but going from 2.55 GHz to 2.2 GHz with the same hardware? No way.

    I am going to test things this wknd with the RAM @ 2T, & see if that fixes the stability issues, but one shouldn't have to run RAM @ 2T on any good motherboard in the first place.

    Anyway, what i'm trying to say is that while some many not have had trouble, a lot of people have, which is why there's negativity surround that A8R-MVP reviews.

    As far as i am concerned, it seems like Asus used us who bought the A8R-MVP as beta testers so they could release the A8R32-MVP with the issues fixed...

  • Zoomer - Tuesday, March 21, 2006 - link

    I get a similar problem with vCore on their A8N-E. Perhaps its due to Asus's choice of voltage regulator?
  • Pete84 - Friday, March 10, 2006 - link

    With conroe ~6 months away, I wonder how badly AMD's FX and high end sales will suffer. Who is going to spend out for a brand new system when it will be destroyed by Conroe?

Log in

Don't have an account? Sign up now