Excel Macro running time on Windows 7 Pro GPT vs MBR performance


  1. Posts : 94
    Windows 7 Pro 64-bit
       #1

    Excel Macro running time on Windows 7 Pro GPT vs MBR performance


    A while ago, I swapped a Windows 7 Pro MBR onto my HP 800 G1 using Paragon. It's my 1st Windows 7 MBR desktop. (I had been using Windows 7 Pro GPT). I did test my Excel macro running time and compared it on MBR vs on GPT.

    test condition: Microsoft Office2010 on Windows 7 x64. running same macro script and same data on same HP 800 G1 computer, three times to average.

    • running time on GPT format: 13-minutes (every time almost same, three times)
    • running time on MBR format: 20-21 minutes (every time slightly different, three times)


    Note. The MBR PC is by Paragon swap, not clean-installed. while the GPT one is by clean-install. the test is tedious because I needed to restore OS on same HP 800 G1 one-by-one. during my retest on the MBR PC, I later run CCleaner to clean up the OS. re-running but didn't see significantly different on test time.

    The running time hints, the performance on GPT is superior to that on MBR, for my excel macro running on Windows 7 Pro x64.

    Before my test, I did google Microsoft web and various forums (quora, tenforums, tomshardware , , ,). they claim "Neither one is faster.". " there should be no performance differences."

    my testtime summary is obviously NOT supporting their claim.

    Any thoughts on the longer running time on Windows 7 x64 MBR?
    I would try my best to tune up the computer so that the performance similar to the GPT one. that's interesting.
    If no solution, my choice will be GPT than MBR according to my test on performance.
    Last edited by siliconbeaver; 2 Weeks Ago at 12:50.
      My Computer


  2. Posts : 6,379
    Windows 7 Ultimate x64
       #2

    Could be differences in cluster size for the file system chosen. Also, you need to keep all things in a control for testing. Having several variables will affect the outcome.

    For testing, temp burn Windows 7 in MBR and Windows 7 in GPT to the same USB drive using NTFS with a cluster size of 4096. Test in each version. If you have the SATA hard drive to spare do that instead. But you need to format the entire drive with each test.

    Is your computer native UEFI? That may be why the increase in speed as it's more suitable in GPT partition schemes. I'm not a hard drive expert so I'm just assuming that's the case.
      My Computer


  3. Posts : 94
    Windows 7 Pro 64-bit
    Thread Starter
       #3

    Thanks for your response,

    the HP computer was shipped with GPT. so it's native UEFI
    Its hdd is WD 500 GB SATA III.

    the OS (with Office2010) MBR was installed by swap using Paragon, NOT clean-install. this could be a problem. before its swap, the hdd formatted to MBR from GPT.
    (I was expecting if working, i.e. similar PC performance, I could save one Office2010 License)
      My Computer


  4. Posts : 375
    Windows 7/8.1/10 multiboot
       #4

    they claim "Neither one is faster.". " there should be no performance differences." [...] my testtime summary is obviously NOT supporting their claim.
    I wouldn't call that obvious at all. As F22Simpilot pointed out, a proper A/B test requires you control all other variables, but:

    The MBR PC is by Paragon swap, not clean-installed. while the GPT one is by clean-install.
    That's a significant difference.

    Even something as mundane as a clean install followed by last month's cumulative update then this month's cumulative update, can be different versus a clean install followed by only this month's cumulative update. In theory they're supposed to result in the same thing, but the former may end up with superfluous files and registry entries that Windows has to ignore or work around.

    Extrapolate that out to a well-used installation transferred by Paragon (or any other method), and it's not uncommon for the old installation to be substantially slower. Even a variety of tools and cleanup attempts on your old installation will seldom get everything. (For illustration, what's your MBR-installation's occupied space vs. your GPT-installation? If it's significantly larger, that should be a clue.)

    That's why some of us old-timers recommend periodically wiping and starting over with a new clean install. (Personally, I do a "rolling clean install" annually.) A clean reinstall will often work wonders on a 5- or 10-year old computer, using the same hardware and same OS. I believe too many people throw out an old computer because it got too slow, when a simple wipe/clean install would have been satisfactory.

    As for the difference between BIOS/MBR and UEFI/GPT, it will make zero difference once the OS has booted. They use different methods to find and launch the OS partition (and there can be differences in the boot time it takes for those processes), but once Windows has booted, BIOS/UEFI/MBR/GPT are all out of the loop and no longer relevant.

    Even the partitions are identical. Some people incorrectly believe "GPT partitions" are somehow different from "MBR partitions". They are not. GPT and MBR are not types of partitions, they are types of partition tables. The partitions themselves use the same, identical partition structure and partition boot sector. (I'm talking about NTFS here.)

    Likewise, UEFI and BIOS are not types of partitions, they are merely differing methods of finding the aforementioned partition boot sector.

    Once Windows has booted, it uses a sector pointer to each partition. It does not refer back to the MBR or GPT any longer. So once booted, there are no speed differences in how a GPT-style disk works vs. a MBR-style disk.
      My Computer


  5. Posts : 94
    Windows 7 Pro 64-bit
    Thread Starter
       #5

    Thanks,

    my next trial would be a clean-install. wait for holidays oncoming so that can have more time.
    If possible, will test "no cumulative update" vs. "more cumulative update".
    my guess is, the more cumulative update, the slower performance.
      My Computer


  6. Posts : 94
    Windows 7 Pro 64-bit
    Thread Starter
       #6

    Update

    yes,
    test time on MBR clean install is similar to that on GPT clean install, 13-minutes.

    This weekend, I did MBR clean-install, simply for the test, Windows 7 Pro +Office. no other apps, even didn't install any HP PC drivers (no video/audio/nic/USB3, , ,).

    It's true. there should be no performance differences.

    for best computer performance, do clean install.
    Last edited by siliconbeaver; 1 Week Ago at 03:13.
      My Computer


  7. Posts : 375
    Windows 7/8.1/10 multiboot
       #7

    Excellent. Now that you've seen the value of a clean install, this would be a good time to implement my "rolling clean install" technique to keep your system running top notch in the future.
      My Computer


  8. Posts : 94
    Windows 7 Pro 64-bit
    Thread Starter
       #8

    clean install, annual image backup, , , it's feasible.

    Thanks,

    bookmarked,
      My Computer


 

  Related Discussions
Our Sites
Site Links
About Us
Windows 7 Forums is an independent web site and has not been authorized, sponsored, or otherwise approved by Microsoft Corporation. "Windows 7" and related materials are trademarks of Microsoft Corp.

Designer Media Ltd
All times are GMT -5. The time now is 18:09.
Find Us