The Test

Our hard drive test bed is designed to shift the bottlenecks, as much as possible, onto the hard drive, but while still within reason. To accomplish that purpose, our test bed is configured as follows:

Intel Pentium 4 Extreme Edition 3.4GHz
Intel D875PBZ Motherboard
ATI Radeon 9800 Pro (128MB)
Creative Labs Audigy
Ultra ATA/100 or Serial ATA 150 cables were used where appropriate

The important drivers used are as follows:

Intel Chipset INF 5.1.1002
ATI Catalyst 4.5
Windows XP Service Pack 1 (no further updates were installed)

What's important to point out is that although we could have outfitted our test bed with 256MB of memory, we wanted to avoid over-exaggerating the performance impact of the hard drive. After all, if your system is swapping to disk a lot, you should be considering a memory upgrade before or in tandem with a hard drive upgrade.

The tests we run are as follows:

Business Winstone IPEAK - a playback test of all of the IO operations that occur within Business Winstone 2004.

Content Creation IPEAK - a playback test of all of the IO operations that occur within Multimedia Content Creation Winstone 2004.

Business Winstone 2004 - the official Business Winstone 2004 test suite.

Multimedia Content Creation Winstone 2004 - the official Multimedia Content Creation Winstone 2004 test suite.

SYSMark 2004 - the official SYSMark 2004 test suite.

Far Cry Level Load Test - a timed test of loading a level in Far Cry.

Unreal Tournament 2004 Level Load Test - a timed test of loading a level in Unreal Tournament 2004.

More details about each individual test will appear in the section of the review dedicated to that particular test.

The Contenders Pure Hard Disk Performance


View All Comments

  • Athlonite - Saturday, March 24, 2007 - link

    both raptor drives are SATA and all the other drives are Pata i know for a fact the seagate baracuda 7200.7 comes in a sata II form coz i have two of them in raid 0 i'd pit the against your single raptor any day of the week as i said like comaring apples with peas Reply
  • peufeu - Monday, May 09, 2005 - link

    Stop benchmarking copies of 1 MB files !
    Linux, with reiserfs4 :
    My crap laptop harddrive does about 16 MB/second raw bandwidth. It does 15 MB/second reading 20 KBytes files. Not that bad.
    I'd like to see the raptor benchmarked with reiser4. I'm pretty sure it can sustain at least 80% of its peak bandwidth with 1kbyte files...
  • jferdina - Sunday, August 22, 2004 - link

    here is a link:
  • jferdina - Sunday, August 22, 2004 - link

    I want to WARN you all of Seagate Barracuda 7200.7

    I just bought one for 1 day and I am terribly disappointed.

    Unlike the seagate barracuda 7200.7 that was tested by AnandTech, the newer version is EXTREMELY noisy. It is the most noisiest HD that I have ever had.

    The reason is Seagate got into a lawsuit problem with Convolve. Apparently Seagate steals their technology, and at least for now they have to remove it. The technology is 'Automatic Acoustic Management' (AAM).

    But for those of you who are tricked, like me, there is some information that may give a hope, AAM is still there but they set it in "performance" mode by default on current firmware. But claimed they had firmware 3.16 and could change the setting.

    So MAYBE IF they can settle the lawsuit, Seagate would provide firmware upgrade to solve this problem.

    Anandtech, please change your review about Seagate since the information is already old because it is very frustrating for customers that wishes to find an accurate info, gets the completely opposite result from what informed.
  • Fietsventje - Wednesday, June 16, 2004 - link

    I didn't read all of the comments, but I would like to say that I, like some others here, would like to see what impact a RAID0-array has on these benchmarks. Could finally resolve the ever-lasting discussion (at least, for myself) about the influence of a RAID-0-setup to general performance.

  • skyce - Monday, June 14, 2004 - link

    I would really like to see a comparison of one 74GB SATA Raptor 10k to two new Raptors in Raid0. I'm building a system over the next couple months and would like to have this drive in my system, but am somewhat reluctant to fork over another $200 for a second drive for Raid0, as Wesley suggested in his High End System Guide (May 26th). Reply
  • HelzBelz - Monday, June 14, 2004 - link

    ... and also, it just so happens that I've just received the exact same hardware (see previous post), but for 2 different systems (upgrades for other people).

    i.e. 2x 80GB ATA WD 8Meg w/ Highpoint 370 PCI RAID Card, and one 74GB SATA Raptor 10k...

    Perhaps we could then compare results !?



  • HelzBelz - Monday, June 14, 2004 - link

    Perhaps this has already been suggested, but here it goes:

    - What about a RAID0 test of some pair of less expensive drives VS a Single Raptor ?

    i.e. For "about the same money", I've often been asked: "Which of the abobe is better / faster ?"

    For example, one could compare the performance of a "2x WD 80GB 8MB cache RAID0" setup, versus a Single 74GB Raptor 10k drive; since either way, you're paying about the same total price...

    Just a thought,


  • artifex - Sunday, June 13, 2004 - link

    I'd like to get a MTBF comparison, not just some data on warranties, too. Or better yet, it would be cool for Anandtech to actually set up a system to run the test drives continuously until they start failing.

    Why? Our local Fry's often has drives like the PATA version of a Hitachi 200GB on sale for $50 after rebates, but if they burn up or crash in half the time as a $100 drive, I'd much rather get the $100 drive. This is particularly important for applications like adding drives to PVRs, where it's not easy to remap around known bad sectors after they're found, and at least one manufacturer's low level formatting utility won't work with Nforce-based IDE, so remapping at the lower level is out also.

    I'm also hoping for thermal comparisons; in small form factor enclosures this is just as important as noise.
  • MadAd - Saturday, June 12, 2004 - link

    IMO you are missing a very important 'real world' test that has always interested me on machines through years.

    Test: Tester starts stopwatch as power button us pressed. Tester stops stopwatch at the point that the desktop appears ready for use...... Record Time Taken. Thats it!

    The funny thing is youve probably not realised how much you run this 'test' and its a valid one insofar as everyone needs to boot up at some point - just please, do us a favor, time it and chart it? :)


Log in

Don't have an account? Sign up now