1. You are viewing our forum as a guest. For full access please Register. WindowsBBS.com is completely free, paid for by advertisers and donations.

Resolved RAM or motherboard problem?

Discussion in 'PC Hardware' started by Christer, 2010/10/20.

  1. 2010/11/13
    Christer

    Christer Geek Member Staff Thread Starter

    Joined:
    2002/12/17
    Messages:
    6,560
    Likes Received:
    71
    I saw no extra stand-off or foreign object. The motherboard is not as wide as the standard ATX boards. It doesn't reach the stand-offs closest to the drives but is secured by six stand-offs and s*c*r*e*w*s. I have fitted two rubber stand-offs (self adhesive) to the chassie to prevent the board from flexing when I fiddle with the RAM-modules. I've checked and double checked and am convinced that all connections are good.
     
  2. 2010/11/13
    Bill

    Bill SuperGeek WindowsBBS Team Member

    Joined:
    2002/01/11
    Messages:
    3,332
    Likes Received:
    389
    Since cases are designed to support 1000s of motherboards, it is common for there to be many more case mounting spot than there are motherboard mounting holes.

    You mentioned the CPU, board and RAM - so it was different PSUs, drives, perhaps graphics card. Other than that, there is nothing on a case that would stop a system from working - again, assuming the front panel connections were connected correctly.
     

  3. to hide this advert.

  4. 2010/11/13
    Christer

    Christer Geek Member Staff Thread Starter

    Joined:
    2002/12/17
    Messages:
    6,560
    Likes Received:
    71
    Different PSU and drives (same brand though, Hitachi) but the same graphics card (Gigabyte GV-R545SC-1GI).
     
  5. 2010/11/14
    Christer

    Christer Geek Member Staff Thread Starter

    Joined:
    2002/12/17
    Messages:
    6,560
    Likes Received:
    71
    Referring to tests on rev.2.0:

    I don't know if that was coincidence or not. It's possible that the cool down time had been too short. After a "cold start" without errors, I was happy and didn't repeat the test. (This takes time ... :( ... !)

    On rev.2.1 Memtest86+ v4.10 returns errors at cold start with the modules in slots 1 + 3 as well as in slots 2 + 4 (dual channel).

    Next, I'll test single channel in slots 1 + 2 and 3 + 4.
     
  6. 2010/11/14
    Christer

    Christer Geek Member Staff Thread Starter

    Joined:
    2002/12/17
    Messages:
    6,560
    Likes Received:
    71
    In a previous post, I asked about the Command Rate. Right now, the modules are in slots 1+2, single channel and I noted that CPU-Z reports the Command Rate as 2T. It seems like, with the BIOS setting in "auto ", one module per channel > 1T, two modules per channel > 2T but in BIOS, it still shows up as "-- ". The "--" is consistent, at least and that mystery is solved.
     
  7. 2010/11/15
    Rockster2U

    Rockster2U Geek Member

    Joined:
    2002/04/01
    Messages:
    3,181
    Likes Received:
    9
    In true Christer Style, the quest for knowledge (and a solution) continues. And, while all readers should appreciate your methodology, it brings me to a couple of quick questions.

    First, why not borrow the memory from the flawless "sister machine" and try it with your R2.1 motherboard. It may bring you to a conclusion that not all memory is created equally. Admittedly, manufacturers have pretty rigid AQL guidelines but it won't be the first time that there is a noticeable variance in "identical" modules and the more machines one builds, the more understandable this becomes.

    Secondly, why the refusal to test an increased memory voltage? This is a pretty common procedure and when applied in moderation, its a safe practice. This is quite different from "overclocking ".

    I read your thread over at the Crucial Forums and I also found some similar issues with earlier ASUS boards and that same memory. As a general guide to some of the questions you have raised, tuck this in your back pocket.

    A long time ago I referenced in detail some of the issues I encountered with some XMS modules that defied logic. Yes, I was working on a pretty serious overclock but the differences between "identical" memory modules was an eye opener.

    I will continue to follow your efforts with interest and wish you good luck in finding a solution.

    Regards,
    ;)
     
    Last edited: 2010/11/15
  8. 2010/11/15
    Christer

    Christer Geek Member Staff Thread Starter

    Joined:
    2002/12/17
    Messages:
    6,560
    Likes Received:
    71
    Rockster,

    Well, I've modified the saying "If it works, don't fix it" to "If it works, don't touch it ". I haven't run Memtest86+ v4.10 on that build and maybe it's a blue screen waiting to happen.

    I've read a few "customer reviews" and it seems like there's a fair share of failing pairs of modules, not only this particular pair but in general. The thing is that you see all the failures but only a fraction of the successes.

    I'll get to that later. If they work with increased voltage, I regard that as faulty RAM and a cause to RMA.

    Thanks for the link, I'll download and read!
     
  9. 2010/11/16
    Christer

    Christer Geek Member Staff Thread Starter

    Joined:
    2002/12/17
    Messages:
    6,560
    Likes Received:
    71
    To summarize, with F3 BIOS:

    Dual channel in slots 1+3 and slots 2+4 > cold start errors, running Memtest86+ v4.10, showing up during the first pass after a few minutes.

    Single channel in slots 1+2 and slots 3+4 > no cold start errors, running Memtest86+ v4.10 for several passes/hours.

    I was considering increasing the voltage and repeating the dual channel tests but had concluded that the modules simply aren't matched to work in dual channel when I had a new message from Gigabyte GTS. I must say that I'm impressed by their support:

    That supports my conclusion (they don't mention increasing the voltage) but I'll update the BIOS to F5a and run the dual channel tests. Kind of "back to square one" with fingers crossed.
     
  10. 2010/11/16
    Bill

    Bill SuperGeek WindowsBBS Team Member

    Joined:
    2002/01/11
    Messages:
    3,332
    Likes Received:
    389
    Yeah, if true about them trying to duplicate your problem, it seems like they are trying. I think I would try to get my hands on 4 different sticks of RAM and see what happens.
     
  11. 2010/11/16
    Christer

    Christer Geek Member Staff Thread Starter

    Joined:
    2002/12/17
    Messages:
    6,560
    Likes Received:
    71
    Well, it's still only two but I agree. My most recent (I nearly wrote "last" ... :rolleyes: ...) post at Corsair:

    If it still doesn't work with BIOS F5a, I'll ask Gigabyte GTS which version the RAM modules are that they used. My bet is that they differ from my version.
     
  12. 2010/11/16
    Christer

    Christer Geek Member Staff Thread Starter

    Joined:
    2002/12/17
    Messages:
    6,560
    Likes Received:
    71
    I had the opportunity to ask so I didn't wait. Gigabyte GTS used ver3.13 in their tests.
     
  13. 2010/11/17
    Christer

    Christer Geek Member Staff Thread Starter

    Joined:
    2002/12/17
    Messages:
    6,560
    Likes Received:
    71
    BIOS F5a was not the solution. In dual channel, slots 1+3, Memtest86+ v4.10 returned errors at cold start. Reboot > still errors. Power off + Restart > no errors.

    I will run the modules, one at a time, in slot 1 and slot 3 to see what happens. I don't think I will but if I do get errors with one or the other, that module is bad.

    No matter what happens, since the same hardware (motherboard, processor and RAM) work in dual channel in my friends computer and in the one Gigabyte use for their supportive tests, I will RMA the modules via the shop where I bought them.
     
  14. 2010/11/17
    Hugh Jarss

    Hugh Jarss Inactive

    Joined:
    2002/07/22
    Messages:
    908
    Likes Received:
    6
    Hi Christer :)

    one last thing you could maybe try before going through the aggro of RMAing everything

    ...could you borrow the PSU from your friend's kit and try it in your box?

    I know that it seems very unlikely as the problem seems to depend upon which slots you put the RAM in etc...

    but try looking at it the other way round for a moment. What's been a common factor through all your woes? And what's different between your setup(s) and the ones of your friend with the same kit, and Gigabyte's, which work OK?

    PSU problems can manifest in really barmy ways*; and it would fit very well with the difference between starting from cold, and restarting (a common factor through all your tests).

    (*though if this does turn out to be PSU related, it'll "take the biscuit" for barminess. But if so, it will likely be the +5SB that's the culprit)

    But the biggest reason to try is to avoid the grief of RMAing everything, the new kit arrives, and then you find yourself stuck in the same situation once again! Plus, think of the "egg on face" you might save.

    Rather than trying to find a way of making it work, choose the configuration which is most likely to fail! That way you'll get to the answer quickest. All that you need is one failure with a different PSU, and that would (dis)prove the point.

    best wishes, HJ
     
  15. 2010/11/17
    Christer

    Christer Geek Member Staff Thread Starter

    Joined:
    2002/12/17
    Messages:
    6,560
    Likes Received:
    71
    Hi Hugh!

    Since I built his system a few years ago and upgraded it a few weeks ago ... :cool: ... I believe I could insist on borrowing his PSU but borrowing his RAM modules would be the quickest way to find out if my RAM modules are bad. Maybe he'll accept being without his computer for a few days?

    When still on BIOS F3:
    - in single channel, slots 1+2 or slots 3+4, no cold start errors
    - in dual channel, slots 1+3 or slots 2+4, cold start errors
    - in single channel with two modules, all slots seem to work fine
    - in dual channel, well, I don't know what happens
    - the above doesn't indicate problems with the motherboard or PSU
    - the above indicate a problem with either the CPU (memory controller) or RAM (not matched for dual channel)
    - my bet is the RAM since browsing the web has confirmed that I'm not alone having this issue and replacing the RAM has cured the problem in many/most/all (?) cases

    I haven't done a "full" test since upgrading the BIOS to F5a, only dual channel in slots 1+3 which returned errors. Over at Corsair's, they have suggested testing a single module at a time in slots 1 and 3. That's what I'll do next but I don't expect any failures. If I'm wrong, one or both modules is/are proved/en bad.

    My friend's setup - identical but original BIOS F2
    Gigabyte's setup - in their own words: "with exact same hardware as you "

    Bill has also suggested testing the PSU but that's the most difficult thing to do. I think I'd rather take som egg on my face, RMA-ing a good RAM-kit, than the work involved in taking two PSU's out from two chassies, swapping his into my chassie and finally put everything back with its rightful owner. Am I being lazy? Yes ... ;) ... but I admit it!

    Thanks for your input!
     
  16. 2010/11/17
    Bill

    Bill SuperGeek WindowsBBS Team Member

    Joined:
    2002/01/11
    Messages:
    3,332
    Likes Received:
    389
    Well, I said testing it "conclusively" is difficult but a good plug in tester like the FrozenCPU Ultimate PSU Tester is almost as good. Still, swapping in a known good PSU is not too much of a technical challenge and can conclusively determine if the old PSU bad.

    When troubleshooting electronics I always try to start at the beginning - and that is with power.
     
  17. 2010/11/17
    Christer

    Christer Geek Member Staff Thread Starter

    Joined:
    2002/12/17
    Messages:
    6,560
    Likes Received:
    71
    There's nothing old in my new build. My friends PSU is "old" ... :rolleyes: ... eight months or so (I don't remember when I replaced his failing one)!
     
  18. 2010/11/17
    Bill

    Bill SuperGeek WindowsBBS Team Member

    Joined:
    2002/01/11
    Messages:
    3,332
    Likes Received:
    389
    I did not mean old in the sense of age. I meant the one untested, or potentially suspect.
     
  19. 2010/11/17
    Christer

    Christer Geek Member Staff Thread Starter

    Joined:
    2002/12/17
    Messages:
    6,560
    Likes Received:
    71
    Okey, I understand but it would be easier to swap the RAM-modules between the two. If his modules have no cold start problems in my build, then my modules are not "properly matched" for dual channel.

    (There has not been a trace of problems with my friend's build and Memtest86+ v4.10 has never been run on that one.)
     
  20. 2010/11/17
    Bill

    Bill SuperGeek WindowsBBS Team Member

    Joined:
    2002/01/11
    Messages:
    3,332
    Likes Received:
    389
    Sure it's easier to swap RAM. But the point is if the PSU is providing dirty (too much ripple), unstable, or out of tolerance voltages, you can swap RAM modules all day long and still have dirty, unstable, or out of tolerance voltages. If you got water in your gas tank, swapping out spark plugs ain't gonna make a difference.

    Now I am not convinced you have a bad power supply, but I am certainly not convinced you don't either. A bad power supply, as HJ suggests, can present some very strange symptoms. They don't always just stop working. So I think you need to get it professionally tested or swap in another - if for no other reason than to eliminate it from the equation.
     
  21. 2010/11/17
    Christer

    Christer Geek Member Staff Thread Starter

    Joined:
    2002/12/17
    Messages:
    6,560
    Likes Received:
    71
    If my friends RAM-modules do not result in cold start problems in my build, doesn't that indicate that my modules are bad and the PSU probably good?
     

Share This Page

  1. This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
    By continuing to use this site, you are consenting to our use of cookies.