Weird HDD transfer-speed graph.

Discussion in 'OT Technology' started by deusexaethera, Mar 3, 2010.

  1. deusexaethera

    deusexaethera OT Supporter

    Joined:
    Jan 27, 2005
    Messages:
    19,712
    Likes Received:
    0
    Help me understand what's going on here:

    [​IMG]

    Normally these graphs start high and roll-off towards the end because hard drives "start" on the outer edge of the platter and work their way inwards, and the angular velocity of the platter decreases as the reader arm moves inwards. In theory, if a hard drive were designed to "start" at the inner edge and work its way to the outer edge, I would expect to see a graph that rolls-off to the left instead of to the right. (horizontally-inverted.) But this graph is vertically-inverted -- it doesn't roll-off, it rolls up. What the hell would cause that?
     
  2. kimsland

    kimsland New Member

    Joined:
    Feb 19, 2010
    Messages:
    174
    Likes Received:
    0
    Seagate Barracuda ATA IV ST380021A

    Capacity: 80 GB
    Speed: 7200 rpm
    Average Seek, Read Time: 9.5 ms
    Cylinders: 1023
    Heads: 256
    Sectors: 63
    Internal data transfer rate: 555 Mbits/sec max
    I/O data-transfer rate: 100 Mbytes/sec max
    Cache buffer: 2 Mbytes
    Ambient operational temperature: 0° to 60°C

    An 80-conductor 40-pin cable is required to run Ultra DMA mode 3, 4, and 5.

    Note. If you are using a 40-pin 80-conductor cable, attach the blue connector to the motherboard, the black connector to the master drive, and the gray connector to the slave.

    [​IMG]

    Information found here: http://www.seagate.com/support/disc/manuals/ata/100129212b.pdf

    -----------

    The drive is spinning up to maximum speed as the current is stabilized
    The drive is also extremely old and outdated. ie Slow
    Recommendation: Update to SATA if your your Motherboard supports it.
     
  3. deusexaethera

    deusexaethera OT Supporter

    Joined:
    Jan 27, 2005
    Messages:
    19,712
    Likes Received:
    0
    It's slow because I'm feeding it through a USB adaptor. It's still ATA100 on the inside. The weird thing is all the other drives I tested had a perfectly flat throughput line with a little roll-off at the very end, which is what I'd expect to see from a normal hard drive with data being fed through a slower connection than it was designed to use. This one hard drive is behaving differently than all the rest, including others from the same make, model, and batch.
     
  4. kimsland

    kimsland New Member

    Joined:
    Feb 19, 2010
    Messages:
    174
    Likes Received:
    0
    Even though, this could be because its holding different data and amounts of data. It could be a different filesystem too etc etc.
    I don't think you would get the same graphs on two same hard drives unless they were both brand new and of the same age, and physically jumpered the same, Master or Cable Select. Plus same conditions and USB port and cable etc etc.

    Plus it could just be a dieing hard drive, you could run the Seatools hard drive diagnostic test on it; but you would need to plug it in internally for this
     
  5. Graham

    Graham OT Supporter

    Joined:
    Jun 8, 2008
    Messages:
    21,197
    Likes Received:
    363
    Location:
    This close to 88mph
    ok, that is definitely strange :eek3:
     
  6. deusexaethera

    deusexaethera OT Supporter

    Joined:
    Jan 27, 2005
    Messages:
    19,712
    Likes Received:
    0
    Well, the graphs on the other hard drives aren't identical, no, but they do exhibit the sloping curve (connected internally) or the flat line with a little roll-off at the end (connected through USB) on a large scale. The little peaks and dips are different, but this one is WAY different. It's not a quirk of the program I'm using either, because I tested it with a couple other benchmarks and their graphs showed the same general upward slope.

    Point of note: the plastic shroud around the IDE connector and power connector is a different color on this one than on the rest. I think it might have a different driver board even though it's got the same batch number than a couple of the other drives. Weirdness.
     
  7. kimsland

    kimsland New Member

    Joined:
    Feb 19, 2010
    Messages:
    174
    Likes Received:
    0
    Oh ok they do that. They bring out two "same" hard drives, with different boards. So obviously they are not the same. Like revision 2 or something.

    I'm thinking the internally connected Seatools test would be good.
    But 80Gig ATA, you've got to expect that it could be just too old and slowly dieing
     

Share This Page