Scratch Drive question

Discussion in 'Digital Video' started by 4God, Nov 2, 2012.

  1. macrumors 68000

    4God

    Joined:
    Apr 5, 2005
    Location:
    My Mac
    #1
    Hi all, so I edit using FCPX on an early 2011 15" MacBook Pro. I have an SSD for boot drive and moved the original HDD to the optical bay. I'm using that HDD (750gig 7200rpm) as a "scratch" drive for FCPX. I also have an external FW800 drive that I use mainly for storage. I'm in a position that requires me to purchase another drive for storage and/or scratch but was wondering about drive speed and connection speed.

    I ran XBench just for kicks and below are the results of the external and the internal scratch.

    External Drive:
    Results 58.51
    System Info
    Xbench Version 1.3
    System Version 10.7.5 (11G63)
    Physical RAM 16384 MB
    Model MacBookPro8,2
    Drive Type Ext Hard Disk
    Disk Test 58.51
    Sequential 63.39
    Uncached Write 114.33 70.20 MB/sec [4K blocks]
    Uncached Write 87.54 49.53 MB/sec [256K blocks]
    Uncached Read 35.27 10.32 MB/sec [4K blocks]
    Uncached Read 68.59 34.48 MB/sec [256K blocks]
    Random 54.34
    Uncached Write 24.20 2.56 MB/sec [4K blocks]
    Uncached Write 150.78 48.27 MB/sec [256K blocks]
    Uncached Read 70.06 0.50 MB/sec [4K blocks]
    Uncached Read 87.75 16.28 MB/sec [256K blocks]

    Internal Drive:
    Results 29.92
    System Info
    Xbench Version 1.3
    System Version 10.7.5 (11G63)
    Physical RAM 16384 MB
    Model MacBookPro8,2
    Drive Type ST9750420AS
    Disk Test 29.92
    Sequential 122.99
    Uncached Write 114.97 70.59 MB/sec [4K blocks]
    Uncached Write 132.72 75.09 MB/sec [256K blocks]
    Uncached Read 99.15 29.02 MB/sec [4K blocks]
    Uncached Read 161.19 81.01 MB/sec [256K blocks]
    Random 17.03
    Uncached Write 4.95 0.52 MB/sec [4K blocks]
    Uncached Write 84.94 27.19 MB/sec [256K blocks]
    Uncached Read 74.47 0.53 MB/sec [4K blocks]
    Uncached Read 128.96 23.93 MB/sec [256K blocks]

    Based on these results, should I consider not using the internal drive as a scratch drive? Any other recommendations?
     
  2. macrumors 65816

    matteusclement

    Joined:
    Jan 26, 2008
    Location:
    victoria
    #2
    everything here says the internal is better. what leads you to think otherwise?
    keep in mind that the optical drive usually runs on a sata1 speed and that can slow it down a bit.
     
  3. thread starter macrumors 68000

    4God

    Joined:
    Apr 5, 2005
    Location:
    My Mac
    #3
    I guess the Random speeds suggest otherwise. Thankfully, my optical is SATAII.
     
  4. macrumors 65816

    matteusclement

    Joined:
    Jan 26, 2008
    Location:
    victoria
    #4
    I did the same thing with my old macbook. It was awesome.
     
  5. macrumors member

    Joined:
    Apr 12, 2011
    Location:
    Athens, OH
    #5
    My 2 cents.

    I use the same setup for my MBP. I am not exactly sure what you are asking but here is my 2 cents.

    What options are you contemplating? A different internal drive? A better external? As I mentioned I have the same setup and can not notice a difference between my SSD and HDD in FCP X. Yes things like copying files are faster on the SSD but for actually working in FCP X the difference is negligible. Forget the numbers they are not always right. What do your experiences tell you? Is your external drive 3.5 inch? Has the internal drive been in use for a while? A full format on the internal might improve the results. Often external drives have a larger cache size. Have you compared the specs for each model?

    1920 × 1080 in ProRes 4:2:2 with PCM audio requires approximately 16.77 MB/s or 59 GB per hour. Video for the most part should not require tons of random R/W's and more likely it will be R/W's in larger chunks. Obviously the CPU/GPU have a greater impact on the video side. From my own experiences with Apple and the optical drive replacement, it does not always do what you expect. For example my SATA III SSD would negotiate at SATA II speeds in the CD bay but when I put it in the factory HD location it would negotiate at the full SATA III speed, even though both are SATA III capable.

    Now for your question, "Based on these results, should I consider not using the internal drive as a scratch drive? Any other recommendations?"

    Again forget what the numbers say, are you satisfied with the performance in FCP X? Yes you can get better results in the test with other drives but what are your needs? My planned upgrade path is a replacement for the drive in the CD bay with a 960GB OWC Mercury Electra SATA II SSD its current price is $1,117. I don't have high hopes for getting the full SATA III from my CD bay drive so I am ok with a SATA II drive, plus SATA III SSD's in the 1TB variety are well over $2000. I am however very happy with the performance of my current configuration and until I start shooting RAW video I not in a hurry to upgrade.

    For external drives there are many possibilities out there but thunderbolt will be the best for performance. We have a 12 TB Promise Pegasus R6 RAID array using Thunderbolt at work, and it is the fastest drive I have ever used. Personally I am thinking about getting a Drobo Mini if the price would come down a little bit and it gets favorable reviews. If you don't know the Drobo Mini holds four 2.5 inch HD drives and one mSATA drive. It can be configured in a RAID or JBOD using the mSATA for a cache. As the price of SSD's come down you could build an array made entirely of SSD's, put that in your XBench tests and smoke it. :)
     

    Attached Files:

Share This Page