The Notebook Review forums were hosted by TechTarget, who shut down them down on January 31, 2022. This static read-only archive was pulled by NBR forum users between January 20 and January 31, 2022, in an effort to make sure that the valuable technical information that had been posted on the forums is preserved. For current discussions, many NBR forum users moved over to NotebookTalk.net after the shutdown.
Problems? See this thread at archive.org.

    Weird RAM (?) issue - Can anyone help/advize me?

    Discussion in 'Alienware 17 and M17x' started by Cosmiccc, Oct 20, 2012.

  1. Cosmiccc

    Cosmiccc Newbie

    Reputations:
    0
    Messages:
    4
    Likes Received:
    0
    Trophy Points:
    5
    Hello Guys,


    Mid Septembre I got me my first brandnew M17x R4. Among other upgrades, I purchased 2 x 8GB Corsair Vengeance (model n° CMSX16GX3M2A1600C10) to replace the stock 6GB RAM which came with the machine.

    As adviced throughout the net and this forum, I checked, after installing the memory in the DIMM1 and 2 sockets, if the RAM-sticks were OK. At first I chose to do this with MemTest for Windows ( MemTest: RAM reliability tester for Windows). Pretty quickly I got a bunch of error messages. Due to my (very) limited IT-knowledge, I thought this could be caused by MemTest itself (start up the test software multiple times at once if CPU has multiple cores etc). So I switched to Memtest86+ v.2.40. After burning the ISO-file and rebooting, I let the test run on the 16GB of newly purchased RAM during the whole night (approx. 9,5 hours and for nearly 5 passes). No errors were detected by Memtest86+. So consequently I thought all was well and continued to install SSD, do clean install, install games and other software.

    I was using the machine for about 3 weeks when suddenly 2 days ago, I noticed several programs (Crusader Kings 2, Microsoft Outlook and even Windows Explorer) crashing to the desktop out of the blue. A couple of hours later I started getting BSoD's with the following information ( sorry it's in Dutch but the most important stuff, the codes, should be clear):

    Probleemhandtekening:
    Gebeurtenisnaam van probleem: BlueScreen
    Versie van besturingssysteem: 6.1.7601.2.1.0.768.3
    Landinstelling-id: 2067

    Aanvullende informatie over dit probleem:
    BCCode: 1a
    BCP1: 0000000000041284
    BCP2: 00000001F92A2001
    BCP3: 000000000017A96F
    BCP4: FFFFF70001080000
    OS Version: 6_1_7601
    Service Pack: 1_0
    Product: 768_1​


    Some research on the net seemed to point out mostly to memory issues. At first I performed a complete check with AlienAutopsy. During the memory check, I again got the same BSoD and the test stopped and the laptop rebooted.
    The BSoD's kept increasing in frequency from that point on.

    Secondly, I ran the Dell/Alienware diagnostics tool which can be accessed via the boot-options at startup. During the first testprocedure I got a message saying "Memory issue resolved". Afterwards the diagnostics tool performed the more thorough test but didn't find any hardware problems.

    As time went by, it became even impossible to start up Windows properly. A boot in safe mode was possible sometimes.

    Testing with Memtest86+ (1,5 hours, about 2 passes) didn't reveal anything at this point.

    Nevertheless, I was getting more and more suspicious about the Corsair RAM-modules (also remembering the error messages from the MemTest for Windows I ran a couple weeks back), so I reinstalled the stock 6 GB RAM and rebooted. Everything ran perfectly normal without crashing software or BSoD's. Then, I installed only 1 strip of 8GB Corsair memory. AlienAutopsy and MemTest for Windows ran fine. I'm using the laptop now with this memory and I do not encounter any problems.

    Every time I reinstall the (probably) faulty module, I immediately get startup-problems, crashes and BSoD's.

    Is it possible that testing the faulty RAM-module outside the Windows-environment (Memtest86+ and Dell diagnostics tool) gives no errors while testing the same piece of hardware in Windows (AlienAutopsy, MemTest for Windows and just using the comp) does result in all kind of mayhem?

    Has anyone already encountered this (I thought Memtest86+ always revealed any problem)? Do you think it is OK to RMA the faulty stick at this point or should I do additional testing (in this case, how)?

    Thanks in advance for your thoughts and sorry for the wall of text!
     
  2. LinkRS

    LinkRS Notebook Consultant

    Reputations:
    56
    Messages:
    295
    Likes Received:
    22
    Trophy Points:
    31
    Hi Cosmicc,

    It sounds like you have pretty much isolated it to one of your sticks of RAM. SInce you determined that the system crashes only with that stick of RAM, that is most likely the issue. The question is if the RAM is faulty or just having compatibility issues (which would show the same symptoms). The only thing I would suggest to try before RMA, is perhpas try the faulty stick in different sockets, see if you get the same results in different slots (there are 4 slots in the M17xR4). Several years ago I had purchased a new motherboard with some corsair RAM, and expereinced similar problems, and I could mitigate the issue for a while by putting the faulty DIMM into the second or third slot (it was a tri-channel board). This ended up just allowing me to run longer before crashes, as the faulty memory addressess would not be used immediately at boot. This happend to be Corsair memory as well, and the resultng solution was to exchange the memory. It is not uncommon to get faulty modules. Corsair is normally pretty good with RMAs too. Good luck!

    Good luck!
     
  3. Special_K

    Special_K Notebook Guru

    Reputations:
    0
    Messages:
    73
    Likes Received:
    0
    Trophy Points:
    15
    Had the same issue, I RMA'ed it and all my problems were solved :).

    Best of luck!
     
  4. James D

    James D Notebook Prophet

    Reputations:
    2,314
    Messages:
    4,901
    Likes Received:
    1,132
    Trophy Points:
    231
    at first you should use the latest memtest86+ it is. Latest is v4.2 Always use memtest written to USB stick.
    Take 1, 2 or 4GB USB flash drive and use this memtest86+ http://www.memtest.org/download/4.20/memtest86+-4.20.usb.installer.zip

    It is often that Corsair puts 2 sticks where 1 of them is much worse than another one. I got the same with 2 kits of 2x4GB 1600. Each kit had 1 stick which is well-overclockable while other sticks were barely overclockable.

    Also I heard that Corsair Vengeance is badly compatible to R3 Alienware. (not r4 as yourth).

    Anyway nowadays the only RAM I sugges to buy is Samsung. It is the best SODIMM RAM I would say. Even though stock clocks are average they are very easy overclockable.


    Guy here with the same laptop. He has NO problem with OCed Samsung RAM
    http://forum.notebookreview.com/har...rmarket-upgrades/692638-overclocking-ram.html
     
  5. Cosmiccc

    Cosmiccc Newbie

    Reputations:
    0
    Messages:
    4
    Likes Received:
    0
    Trophy Points:
    5
    Thx guys for the replies. Just for the sake of it, I've installed the suspected RAM-stick into another laptop (an HP dv6-1350) and this wouldn't even boot.

    So I guess I can be pretty sure that the RAM is the cause for these crahes and BSoD's. I'll RAM it and see if another stick can do the trick.
     
  6. Alienware-L_Porras

    Alienware-L_Porras Company Representative

    Reputations:
    3,658
    Messages:
    6,874
    Likes Received:
    969
    Trophy Points:
    281
    Just get a replacement for the faulty one and you should be ok, it is totally possible.