RAM test: Rember vs Apple Utility

Discussion in 'MacBook Pro' started by skiltrip, Jan 21, 2012.

  1. skiltrip macrumors 68030

    May 6, 2010
    New York
    I have been having some crashes with memory heavy drum samplers on my MacBook Pro. So I did some tests on my RAM (G.Skill 2x4GB, had it about a year, never gave me any problems that I've noticed until now).

    When using Rember to test, I consistently get a data mismatch error pretty early on in the test. Happens every single time (Rember results are pasted below). However, when I do a test using Apple's hardware check utility that comes on the Applications disk in the box, it finds no problems. Which should i trust. Like I said, the Rember results are 100% consistent. I did 3 tests in row with the same results. Again, Rember data is below...

    Memtest version 4.22 (64-bit)
    Copyright (C) 2004 Charles Cazabon
    Copyright (C) 2004-2008 Tony Scaminaci (Macintosh port)
    Licensed under the GNU General Public License version 2 only

    Mac OS X 10.6.8 (10K549) running in multiuser mode
    Memory Page Size: 4096
    System has 8 Intel core(s) with SSE
    Requested memory: 6498MB (6814281728 bytes)
    Available memory: 6498MB (6814281728 bytes)
    Allocated memory: 6498MB (6814281728 bytes) at local address 0x0000000101000000
    Attempting memory lock... locked successfully
    Partitioning memory into 2 comparison buffers...
    Buffer A: 3249MB (3407140864 bytes) starts at local address 0x0000000101000000
    Buffer B: 3249MB (3407140864 bytes) starts at local address 0x00000001cc14d800

    Running 255 test sequences... (CTRL-C to quit)

    Test sequence 1 of 255:

    Running tests on full 6498MB region...
    Stuck Address : setting 1 of 16 ok
    Linear PRN : setting 1 of 16 ok
    Running comparison tests using 3249MB buffers...
    Random Value :  ok
    Compare XOR :  ok
    Compare SUB :  ok
    Compare MUL :  ok
    Compare DIV :  ok
    Compare OR :  ok
    Compare AND :  ok
    Sequential Increment:  ok
    Solid Bits : setting 1 of 64 ok
    Block Sequential : setting 1 of 256 ok
    Checkerboard : setting 1 of 64

    FAILURE! Data mismatch at local BUFA address 0x000000018e4e8cf0, BUFB address 0x00000002596364f0
    BUFA Data: 0x555d555555555555, BUFB Data: 0x5555555555555555

    Bit Spread : setting 1 of 128
  2. bill-p macrumors 68000

    Jul 23, 2011
    I'd trust Rember over Apple's test. Also had some G.Skill RAM failing on me earlier in another laptop (not Macbook) even though it was running fine for several months. I think it's something with G.Skill... even though their RAM prices are typically lower, the quality and durability might not be up there.
  3. skiltrip thread starter macrumors 68030

    May 6, 2010
    New York
    Thanks. I ordered 8gb of OWC. So when it comes, I'll run rember on it and hopefully it'll be error free. Then I'll RMA the G.Skill using the lifetime warranty, and eBay whatever they send as a replacement.
  4. bill-p macrumors 68000

    Jul 23, 2011
    To be fair, it might have been bad luck on my part, since I did run those G.Skill RAM in a laptop that was hot almost 24/7. But since you are running into the same issue, I'm starting to think it might have been the RAM.

    Never thought about RMA though, since it was just a single stick.
  5. Ledgem macrumors 65816


    Jan 18, 2008
    Hawaii, USA
    I prefer to use Memtest86+ over Rember. Unless they've changed something since I last used it, Rember runs with the OS open, meaning that it can't get at all of the RAM. With Memtest, you reboot the computer and it loads directly into Memtest. Memtest on the Mac doesn't run perfectly, as it won't accept keyboard input (not sure if they've fixed that in newer versions), but it'll still run the tests for an infinite number of times until you stop it.

    Just a note to anyone wanting to test their memory: a single error isn't anything to worry about; you should allow the tests to run multiple times, and if the error seems to be repeatable (or if there are multiple errors frequently coming up), then it indicates a problem.

    By way of example, when I first bought my computer I would frequently get kernel panics when I did RAM-intensive tasks. I ran Memtest, and it became clear that one of my Apple-supplied RAM sticks was faulty. I've since swapped RAM sticks around. While testing a new RAM module, one test popped up with an error, but it was not repeatable; I've been using that RAM stick for a year or two now, and have put it through RAM-intensive tasks, and have not had any issues with system stability.

Share This Page