1. Welcome to TechPowerUp Forums, Guest! Please check out our forum guidelines for info related to our community.

How to Benchmark a HDD & SSD ?

Discussion in 'Storage' started by wolf2009, May 20, 2009.

  1. wolf2009 Guest

    How does one go about testing a HDD to show real world testing ?

    I know you can run HDTach and HDTune and show synthetic benches. But how would one do a real world comparison ?

    I'm trying to do a real world test, but the results turn out to be different while copying real files to a hard drive. Sometimes the result ends up at 180s, while in others it is 150s. Could Superfetch be causing this ?

    Using Vista x64 SP1, gigabyte x38-ds4, WD 2500AAKS
     
  2. crazy pyro

    crazy pyro New Member

    Joined:
    Jun 28, 2008
    Messages:
    1,662 (0.66/day)
    Thanks Received:
    125
    Location:
    Newcastle
    Copy a large number of small files across while timing it.
    Copy a small number of large files across while timing it.
    Copy a mixture of file sizes across while timing it.
     
  3. mlee49

    mlee49

    Joined:
    Dec 27, 2007
    Messages:
    8,501 (3.17/day)
    Thanks Received:
    2,108
    Location:
    Kansas City
    Read Write speeds can be timed for file transfer

    Access times can be monitored by random file look up or boot times.
     
  4. fire2havoc New Member

    Joined:
    May 12, 2009
    Messages:
    133 (0.06/day)
    Thanks Received:
    9
    HD Tune Pro.
     
  5. wolf2009 Guest

    I appreciate your taking the time to make a post, but please read the OP
     
  6. Mussels

    Mussels Moderprator Staff Member

    Joined:
    Oct 6, 2004
    Messages:
    42,921 (11.11/day)
    Thanks Received:
    10,213
    hard drives are faster at one end of the drive than the other, it messes with RW results.

    My 1TB's are 110MB/s on the innter platter and 70MB/s on the outer - if i defrag alphabetically its easy to tell what files will be slower than the rest.
     
  7. fire2havoc New Member

    Joined:
    May 12, 2009
    Messages:
    133 (0.06/day)
    Thanks Received:
    9
    Aside from copying actual files (which is also not always consistent throughout), benchmarks are the easiest way to test.
     

Currently Active Users Viewing This Thread: 1 (0 members and 1 guest)

Share This Page