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. To refresh your memory, 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 difference 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 Multimedia Content Creation Hard Disk I/O

It is interesting that the performance patterns hold steady across both Multimedia Content IO and Business IO, with the on-board NVIDIA nForce4 SATA 2 still providing the fastest IO, followed closely by the Intel ICH7R and Silicon Image 3132 SATA 2 controllers.

SLI Gaming Performance & Overclocking Performance Firewire and USB Performance
POST A COMMENT

70 Comments

View All Comments

  • scottyfree111 - Friday, December 16, 2005 - link

    Gary, I got one of the first production runs of this board and I smelled something burning on the first post. I got it to work, but mobo temps were 46-48C. my friend got the same board a few weeks later in an almost identical setup (P43.46EE, dual Quadro FX-3400's, PC P&C 850 PSU) and got 28C mobo, and 29C (idle on the CPUw/Zalman 9500 cooler). So I RMA'd it, and am hoping for similar results. It should be back any day now and I'm hoping they replaced it. Asus forums list many unhappy campers having all sorts of Bios issues. Which Bios are you using now(212 maybe?) Also, I was disappointed that there are only 5 internal sata ports since I use 9 hard drives. I've been looking into using a Sil Image 5X1 port multiplier with a 4 bay esata enclosure from either Addonics or Cool Drives to get a RAID 0X4 on the Sil 3132 with my 4 WD740 Raptors for video capture/editing. Do you know anything about setting up an esata RAID? Thanks Reply
  • MadAd - Friday, November 04, 2005 - link

    Thanks for the reply Gary, dont bust a gut though, 1920x1440 is really good enough :)

    Im just welcoming the recognition that people do use high res to do normal things like play games on (shock horror).

    Ive always found it strange that Anandtech reports on so much cutting edge hardware in the cpu, mobo and video areas, the bread and butter of a good review site, but seemed to ignore people with high end monitors in test results - at least you are putting that right now, thank you again.
    Reply
  • MadAd - Wednesday, November 02, 2005 - link

    Hooray! a reviewer that benchmarks games up to high end video resolutions! Thank you thank you!!

    My own resolution of 1920x1200 wasnt mentioned but I got the picture from what you did cover, thank you again- The other Anand reviewers seem to think theres no life after 1600x1200. (but there is :¬) )

    Nice one
    Reply
  • Gary Key - Wednesday, November 02, 2005 - link

    Hi,

    The review results will depend upon the monitor's capability during testing (several of my monitor's only go to 1600x1200 or 1280x1024). Also, please realize that a large majority of readers have 19" LCDs that only support up to 1280x1024. This is why we standardized on this resolution recently. However, we will start providing higher resolution benchmarks depending upon the hardware being reviewed (performance oriented boards) and time allowances. This is based upon the fact that the next step up for most buyers will be the 1600x1200 ~ 1920x1200 capable monitors.

    The higher end monitor's I have for testing do not support the 1920x1200 format. They support the 1920x1440 and up resolutions. However, that should change in the next month or so hopefully. ;-)

    I appreciate the comments and hopefully we can get you and others some 1920x1200 benchmarks in the near future.

    Thanks!
    Reply
  • H0witzer - Saturday, October 29, 2005 - link

    As the owner of a P5ND2-SLI-Deluxe I have to say the C19 chipset gets WAY too hot. Not to mention the horrible stability of the BIOS. There has been Many complaints from ASUS owners about these issues. Another issue as of late has been power supplies blowing. I personally experience a 3 day old 500W Aspire power supply blowing, and after flashing the BIOS the entire system dying.

    I wonder about the stability of the NVIDIA chipset and how it will hold up in the long run. I would have liked to see a comparison against the P5WD2-Premium.
    Reply
  • Gary Key - Saturday, October 29, 2005 - link

    I have read some of the horror stories about the P5ND2-SLI-Deluxe and agree about the thermal issues among others with the A2 stepping of the C19. The A3 and up steppings are better in this regard and the new board layout is significantly more robust than the P5ND2 series. I believe Asus realized their errors from several product releases lately (I went to Intel boards after an earlier issue with Asus) and are in the process of correcting them. This board has restored my faith in their new product lineup. This bios we tested with (0047) has been wonderful from both a stability and performance viewpoint. I personally use OCZ or PC Power and Cooling power supplies and have never had an issue with any board, good or bad. I did not have the P5WD2 to test against but I do have the P5LD2 now. :-> Reply
  • H0witzer - Sunday, October 30, 2005 - link

    Hey Gary,

    Thanks for taking the time to write back. Based on your article and statements here I am going to contact my online retailer who is processing my RMA and tell them to just issue a credit and I will purchase the P5N32-SLI instead. Don't worry i won't hold you responsible for my decision :-)

    I just figured it couldn't be any worse than the last board. Hopefully they fixed the things that were issues with the P5ND2-SLI Deluxe.

    One last question, when will this be available, have you heard anything yet? All the online retailers are saying Pre-Order.
    Reply
  • Gary Key - Sunday, October 30, 2005 - link

    Good Day,

    I bought a couple of boards from NewEgg last week but I noticed this morning they are out of them. I will send them a message and see when their next batch is in. When you receive the board please email me and we can discuss specifics on the board and bios. I really think you will be pleased with this board. If not, please post your issues. I would appreciate another viewpoint or opinion on the board.
    Reply
  • H0witzer - Wednesday, November 02, 2005 - link

    Hi Gary,

    Any word on why the boards all disapeared? Seems that No one has them anymore.
    Reply
  • Gary Key - Wednesday, November 02, 2005 - link

    Good Day,

    We spoke with Asus tonight. The second production run of boards was slightly delayed due to a couple of minor hardware changes on the board that should not affect performance. They changed resistors in a couple of areas apparently. However, Asus will provide us a new board for testing to ensure results are indicative of the review board results. The next production run of boards should arrive within 7 business days according to our sources. I will post an update to the article if there are any changes with this new release.

    Thanks....
    Reply

Log in

Don't have an account? Sign up now