Hard drive reads okay, writes slow

Discussion in 'Cakewalk' started by Steven Bell, Dec 30, 2003.

  1. Steven Bell

    Steven Bell Guest

    Here are the Catena dskbench results. Note the initial "write" time!!!

    Steven

    K:\Temp>dskbench
    DskBench 2.12
    (c) 1998, SESA, J.M.Catena (, www.sesa.es)
    Timer Check = 1000 (should be near 1000)
    CPU Check = 0.63 % (should be near 50.00 %)
    CPU index (relative to Pro 200 MHz) = 8.701599
    Open = 0 ms
    Write = 26297 ms, 9.73 MB/s, CPU = 0.11 %
    Flush = 16 ms
    Rewin = 0 ms
    Read = 9031 ms, 28.35 MB/s, CPU = 0.56 %
    Close = 0 ms
    BlockSize = 131072, MB/s = 9.21, Tracks = 109.55, CPU = 0.04 %
    BlockSize = 65536, MB/s = 6.54, Tracks = 77.73, CPU = 0.09 %
    BlockSize = 32768, MB/s = 5.03, Tracks = 59.75, CPU = 0.30 %
    BlockSize = 16384, MB/s = 3.79, Tracks = 45.11, CPU = 0.57 %
    BlockSize = 8192, MB/s = 3.07, Tracks = 36.46, CPU = 0.98 %
    BlockSize = 4096, MB/s = 2.40, Tracks = 28.54, CPU = 1.60 %
     
    Steven Bell, Dec 30, 2003
    #1
    1. Advertisements

  2. Steven Bell

    Kevin Perry Guest

    More to the point that access to audio data is fairly random (as in, if you're
    reading 16 waves, the head is jumping around anyway), so statistically, you're
    better off not defragmenting. There was a great article a few years back on
    StorageReview (I think) where a random access file server was tested before and
    after defragmenting, and the fragmented drive was faster because of the random
    nature of the access - the same applies to the way we pull audio from drives.

    --

    Kevin Perry
    The Mountain Grill
    http://www.mountaingrill.co.uk/

    "It is the business of the future to be dangerous;
    and it is among the merits of science that it equips the future for its duties."
     
    Kevin Perry, Dec 31, 2003
    #2
    1. Advertisements

  3. Steven Bell

    Steven Bell Guest

    How about defrag before a live 16-track recording, but not during post?

    Steven
     
    Steven Bell, Dec 31, 2003
    #3
  4. Steven Bell

    C Hudson Guest

    hmm, I doubt that for audio though. I remember several examples wher a file
    stuttered and gaged during play back until the wave drive was defragged.
    From as far back as PA 8.
    That tests results do not explain why it worked for me. hmm....
     
    C Hudson, Jan 3, 2004
    #4
  5. Steven Bell

    Jose Catena Guest

    Perhaps the disk has write verify enabled. Some come this way from factory.
    If so, DL an utility from the manufacturer to switch it off.
    In some cases, write verification is switched off automatically after a
    certain period of use.
    Software write verification can also make it happen, be sure the OS is not
    doing it.
    Some antivirus software can also slow down writes. Don't install resident
    file monitors in a DAW.
    There is yet another possible reason for slow write speeds: when a fast disk
    is operated at lower ATA speeds than it was designed for, this may happen.
     
    Jose Catena, Jan 3, 2004
    #5
  6. Steven Bell

    Jose Catena Guest

    Hey, that's our Jose Catena! Small world.

    Our? What do you mean saying "our"? I'm a cheap guy, but you didn't pay the
    fee yet! <g>
     
    Jose Catena, Jan 3, 2004
    #6
  7. Steven Bell

    Steven Bell Guest

    Thanks, Jose'.

    I don't know how to see if a disk has "write verify" enabled. I did not
    install any special software when I installed these.

    I have not seen a write verification switch in Windows XP, either. Do you
    know where to look?

    I'm using Norton AV 2003 with everything disabled (AFAIK) other than email
    checking.
    I suppose I can manually stop all of the NAV services and re-test.

    According to BIOS setup the drives are set up for DMA mode 5. It also lists
    a PIO mode in BIOS so it's hard to tell which one is used by the OS. Do you
    know a way to verify which mode a disk is in (within Windows)?

    Excellent ideas, BTW. I'm going to keep them around in case this happens to
    someone else.

    Steven
     
    Steven Bell, Jan 3, 2004
    #7
  8. Steven Bell

    Jose Catena Guest

    What's the disc's brand and model?
    The first thing you need is the manufacturer utility to check/disable write
    verify.
    XP does not have write verify switch AFAIK.
    To see the ATA mode being used, see the properties of the disk controller in
    device manager, but surely it will be DMA5.
     
    Jose Catena, Jan 3, 2004
    #8
  9. Steven Bell

    Alan Le Guest

    I don't know how to see if a disk has "write verify" enabled. I did not
    Man... I would've never thought to check that, and it has even happened to
    me before (WD drive)! Write verify absolutely kills performance. There's
    probably a utility available from the manufacturer, and it likely needs to
    be run from DOS.
    It should tell you in the properties for the IDE channel in Device Manager.

    Alan
     
    Alan Le, Jan 3, 2004
    #9
  10. Steven Bell

    Steven Bell Guest

    They are Seagate Barracuda's ST380011A and ST380021A (Master and Slave).
    I'll take a look for that utility.

    Thanks again, Jose',

    Steven
     
    Steven Bell, Jan 3, 2004
    #10
    1. Advertisements

Ask a Question

Want to reply to this thread or ask your own question?

You'll need to choose a username for the site, which only take a couple of moments (here). After that, you can post your question and our members will help you out.