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.

Page 2 of 5
  1. 2010/10/21
    Christer

    Christer Geek Member Staff Thread Starter

    Joined:
    2002/12/17
    Messages:
    6,585
    Likes Received:
    74
    I bought it in mid august and going by the BIOS F2 date, 2010/06/07, revison 2.1 should have been released. Had it arrived in Sweden, well, I don't know but I'll ask at the shop.

    That's how it looked when I opened the case. The numbering of the slots are as in the manual, the first two are blue (DDR3_1 and DDR3_2) the other two are white (DDR3_3 and DDR3_4). The modules were seated in the blue slots.

    I shifted them to DDR3_1 and DDR3_3, the blue and the white closest to the processor.

    When seated in DDR3_1 and DDR3_2, Memtest86+ v4.10 reported 4094 MB at 4601 MB/s.

    When seated in DDR3_1 and DDR3_3, Memtest86+ v4.10 reported 4094 MB at 4209 MB/s.

    This indicates to me that the modules now work in "single channel mode ", right?

    (There's no indication whatsoever of which "mode" the memory is working in. The POST screen says "DDR3 1333 MHz Unganged" in both seatings and you can only change between "Ganged" and "Unganged ".)

    I actually believed (and still do) that this difference in module positions between 2.0 and 2.1 was beyond logic. Why would Gigabyte make such a radical change between revisions?

    Comparing the GA-P55A-UD4 which was used for a computer that I built for a friend, the slot numbering is different but the modules end up in the same physical position as for the 2.1 revision of my motherboard.

    Am I wrong in drawing the conclusion (by the transfer rates in Memtest86+ v4.10) that the modules are now working in "single channel mode "?

    It's soon bed time here in Sweden and I'll let the modules stay in DDR3_1 and DDR3_3. If a cold start tomorrow morning, running Memtest86+ v4.10 produces no errors, then what? One of the memory controllers on the processor?
     
    Christer,
    #21
  2. 2010/10/21
    Christer

    Christer Geek Member Staff Thread Starter

    Joined:
    2002/12/17
    Messages:
    6,585
    Likes Received:
    74
    This guy seems to have an GA-870A-UD3 (rev. 2.0):

    VERY IMPORTANT INFORMATION REGARDING AMD AM3 CPU's and RAM SPEEDS

     
    Christer,
    #22


  3. to hide this advert.

  4. 2010/10/21
    Bill

    Bill SuperGeek WindowsBBS Team Member

    Joined:
    2002/01/11
    Messages:
    3,368
    Likes Received:
    411
    While, I am still confused, but I think you are right and they are in single mode, but to be sure, download and run CPUz then check under the Memory tab. It will tell you if running Single, Dual, or Triple (for boards that support that).

    Then if me, I would RMA the board, explaining the RAM runs fine in single channel but not dual. Then hopefully, you will get the latest version.

    BTW, I don't see that RAM on the QVL (Memory Support List). That does not mean it does not work, but it does mean Gigabyte has not tested that RAM with that board. As for that comment on the Gigabyte forum, I'm not sure that is related to this, but in any case it does suggest a problem with boards.
     
    Bill,
    #23
  5. 2010/10/21
    Christer

    Christer Geek Member Staff Thread Starter

    Joined:
    2002/12/17
    Messages:
    6,585
    Likes Received:
    74
    I'll download and run CPUz tomorrow, thanks for the hint.

    Yes and if that doesn't help, RMA the processor for a failing memory controller.

    No but as they say:

    They are on Corsair's "list" of RAM tested with the Phenom II.
     
    Christer,
    #24
  6. 2010/10/22
    Christer

    Christer Geek Member Staff Thread Starter

    Joined:
    2002/12/17
    Messages:
    6,585
    Likes Received:
    74
    If you compare the pictures of the motherboards, the colour coding for the slots has been shifted and it's clear that the redesign actually has been that radical. Amazing! To me, that indicates serious problems with rev. 2.0.

    GA-870A-UD3 (rev. 2.0)

    GA-870A-UD3 (rev. 2.1)

    CPU-Z confirmed that the modules are working in "single channel mode ". Memtest86+ v4.10 reported errors.

    I will do one more test to rule out problems with the processor and that will be to run the modules in DDR3_2 + DDR3_4 on the other memory controller.
     
    Christer,
    #25
  7. 2010/10/22
    Christer

    Christer Geek Member Staff Thread Starter

    Joined:
    2002/12/17
    Messages:
    6,585
    Likes Received:
    74
    I checked that too and you probably mixed it up:

    GA-870A-UD3 (rev. 2.0) - Overview: Dual Channel DDR3 1866+ MHz

    * DDR3 1866+ is supported with combination of AM3 CPU and qualified memory modules, please refer "Memory Support List" for detail memory support information.

    GA-870A-UD3 (rev. 2.0) - Specification:

    1. 4 x 1.5V DDR3 DIMM sockets supporting up to 16 GB of system memory
    (Note 1)
    2. Dual channel memory architecture
    3. Support for DDR3 1866(OC)(Note 2)/1333/1066 MHz memory modules

    (Note 2) To reach DDR3 1866MHz or above, you must install two memory modules and install them in the DDR3_3 and DDR3_4 memory sockets.

    GA-870A-UD3 (rev. 2.1) - Overview: Dual Channel DDR3 2000+ MHz

    My comment - no limitations noted

    GA-870A-UD3 (rev. 2.1) - Specification:

    1. 4 x 1.5V DDR3 DIMM sockets supporting up to 16 GB of system memory
    (Note 1)
    2. Dual channel memory architecture
    3. Support for DDR3 2000(OC)/1333/1066 MHz memory modules

    My comment - no limitations noted

    This indicates to me that I should get a rev. 2.1 even if I have no intensions whatsoever to overclock anything.
     
    Christer,
    #26
  8. 2010/10/22
    Bill

    Bill SuperGeek WindowsBBS Team Member

    Joined:
    2002/01/11
    Messages:
    3,368
    Likes Received:
    411
    If for no other reason than it is newer, and hopefully has corrected any problems found in the previous version.
     
    Bill,
    #27
  9. 2010/10/22
    Christer

    Christer Geek Member Staff Thread Starter

    Joined:
    2002/12/17
    Messages:
    6,585
    Likes Received:
    74
    Maybe we'll know the answer or at least have an indication by Sunday evening. My friend's (the guy with the GA-P55A-UD4) son wanted to upgrade his system. I told him what I had bought and he bought the same motherboard, graphics card, processor and memory. I had him check his motherboard and it's the GA-870A-UD3 (rev. 2.1). I'll assemble it on Sunday and my bet is that it will work but we'll see.

    It also confirms that the shop can get the "good one ". I have already talked to them and we agree that if the last test (with the modules in DDR3_2 + DDR3_4 on the other memory controller) also produces memory errors running Memtest86+ v4.10, it confirms that the memory controllers on the processor are not to blame but either DDR3_1 or DDR3_2 on the motherboard since one of them is always involved in failure but not DDR3_3 and DDR3_4. Then the motherboard will get RMA'd.
     
    Christer,
    #28
  10. 2010/10/22
    Bill

    Bill SuperGeek WindowsBBS Team Member

    Joined:
    2002/01/11
    Messages:
    3,368
    Likes Received:
    411
    Sounds like a plan. Keep us posted.
     
    Bill,
    #29
  11. 2010/10/23
    Christer

    Christer Geek Member Staff Thread Starter

    Joined:
    2002/12/17
    Messages:
    6,585
    Likes Received:
    74
    This morning, the "last" test ended with errors running Memtest86+ v4.10. The motherboard will get RMA'd and replaced by a GA-870A-UD3 (rev. 2.1).

    More next week!
     
    Christer,
    #30
  12. 2010/10/23
    Bill

    Bill SuperGeek WindowsBBS Team Member

    Joined:
    2002/01/11
    Messages:
    3,368
    Likes Received:
    411
    Well, I hate to see problems with my preferred brand of motherboards, but until man can create perfection, there will always be some imperfections. Even Rolls Royce has recalls.
     
    Bill,
    #31
  13. 2010/10/23
    Christer

    Christer Geek Member Staff Thread Starter

    Joined:
    2002/12/17
    Messages:
    6,585
    Likes Received:
    74
    ... :confused: ... Rolls who? ... :p ...

    Well, anyway, I too am a bit disappointed but "you know what" happens. If it works well with rev. 2.1, then my confidence is restored.
     
    Christer,
    #32
  14. 2010/10/24
    Christer

    Christer Geek Member Staff Thread Starter

    Joined:
    2002/12/17
    Messages:
    6,585
    Likes Received:
    74
    I've forgotten to comment on this:

    I no longer think it's a driver issue.

    If I, when getting errors in Memtest86+ v4.10, hit the reset button and let the test restart, I keep getting errors.

    If I shut down on the power switch and restart the test, no more errors (when warm).

    If I, when getting errors in Memtest86+ v4.10, hit the reset button and let Windows 7 start, I get all kinds of bluescreens.

    If I shut down on the power switch and start Windows 7, no more bluescreens (when warm).

    I now conclude that it's a memory problem.
     
    Christer,
    #33
  15. 2010/10/24
    Christer

    Christer Geek Member Staff Thread Starter

    Joined:
    2002/12/17
    Messages:
    6,585
    Likes Received:
    74
    No problems installing Windows 7-64 Professional, getting it updated and activated. No "cold tests" yet, but I don't expect a call tomorrow morning (turned the phone off ... :D ...).
     
    Christer,
    #34
  16. 2010/11/08
    Christer

    Christer Geek Member Staff Thread Starter

    Joined:
    2002/12/17
    Messages:
    6,585
    Likes Received:
    74
    I had no call that morning and have now fitted rev2.1 in my own computer. No cold start yet but I noticed one "detail" in the BIOS:

    Under "Advanced BIOS Features" > "CPU core Control" is set to auto and below that setting, all four cores are showing as enabled.

    I'm not totally sure, only 99.99 %, that on rev.2.0, only two cores were showing. I should have reacted on that but I didn't since all four cores were showing in Task Manager. It seems like something was seriously wrong with that MB.
     
    Christer,
    #35
  17. 2010/11/09
    Christer

    Christer Geek Member Staff Thread Starter

    Joined:
    2002/12/17
    Messages:
    6,585
    Likes Received:
    74
    First cold start with rev.2.1 this morning and I didn't even get my first cup of coffee ... :( ... before it went wrong. The same symptoms as with rev.2.0 > a restart and the memory failures remain but if I shut down, wait a minute and start, the memory failures are no more. I have not yet run Windows 7 and don't know if the blue screening persists.

    I'm still puzzled about the number of cores showing as enabled on rev.2.0 but now I'm not so certain that I'm remembering it correctly.

    So, what's next?

    Memory Modules?

    Processor?

    Power Supply?

    ... and of course ... :confused: ... can Memtest86+ v4.10 be relied upon ...
     
    Christer,
    #36
  18. 2010/11/09
    Bill

    Bill SuperGeek WindowsBBS Team Member

    Joined:
    2002/01/11
    Messages:
    3,368
    Likes Received:
    411
    Wow, that's too bad. I would certainly want to ensure you had good power. I would swap in a known good supply and start with just one stick of RAM.
     
    Bill,
    #37
  19. 2010/11/09
    Christer

    Christer Geek Member Staff Thread Starter

    Joined:
    2002/12/17
    Messages:
    6,585
    Likes Received:
    74
    I haven't noticed anyting with the power supply and I have no other to install. The modules should run at 1.5 V and the BIOS reports the voltage as 1.504 V. (Early on in this thread, Arie suggested upping the voltage but that shouldn't be neccessary on standrad memory modules running at "auto" settings.)

    When I run Windows Memory Diagnostic (from a floppy, not from within Windows 7) I get the message "The memory map contained ranges that extend above four gigabytes." with the options to quit or continue but the latter means that all memory will not be tested.

    WMD has never reported any fauilures but Memtest86+ v4.10 always reports the failures above 4 GB (which WMD can not test). So back to the quoted question: Can Memtest86+ v4.10 be relied upon? Were the blue screens on rev.2.0 coincidence and has nothing to do with the "alleged" memory failures?

    I have no "good power supply" to swap in but I'll do tests with a single module only.
     
    Christer,
    #38
  20. 2010/11/09
    Bill

    Bill SuperGeek WindowsBBS Team Member

    Joined:
    2002/01/11
    Messages:
    3,368
    Likes Received:
    411
    You normally don't unless they fry, pop, and stink up the room - that's why power problems are often difficult to troubleshoot. Here's my canned text on testing PSUs.
    To properly and conclusively test a power supply unit (PSU), it must be tested under various realistic "loads" then analyzed for excessive ripple and other anomalies. This is done by a qualified technician using an oscilloscope or power analyzer - sophisticated (and expensive) electronic test equipment requiring special training to operate, and a basic knowledge of electronics theory to understand the results. Therefore, conclusively testing a power supply is done in properly equipped electronic repair facilities.

    Fortunately, there are other options that are almost as good. I keep a FrozenCPU Ultimate PSU Tester in my tool bag when I am "in the field" and don't have a good spare power supply to swap in. While not a certain test, they are better than nothing. The advantage of this model is that it has an LCD readout of the voltage. With an actual voltage readout, you have a better chance of detecting a "failing" PSU, or one barely within specified ATX Form Factor Standard tolerances. Lesser models use LEDs to indicate the voltage is just within some "range ". These are less informative, considerably cheaper, but still useful for detecting PSUs that have already "failed ". Newegg has several testers to choose from. All these testers contain a "dummy load" to fool the PSU into thinking it is connected to a motherboard, and therefore allows the PSU to power on, if able, without being attached to a motherboard - great for testing fans, but again, it is not a true load or suitable for conclusive testing.

    Note the required voltage tolerance ranges:
    [​IMG]
    [size= "1"]NOTE: Disregard the -5VDC reading. It is no longer used.[/size]​

    Swapping in a known good supply is a tried and true method of troubleshooting used for years, even by pros. If you have access to a suitably sized, spare power supply, carefully remove the suspect supply and replace it with the known good one, and see if the problem goes away.

    I do not recommend using a multimeter to test power supplies. To do it properly, that is, under a realistic load, the voltages on all the pins must be measured while the PSU is attached to the motherboard and the computer powered on. This requires poking (with some considerable force) two hard and sharp, highly conductive meter probes into the main power connector, deep in the heart of the computer. One tiny slip can destroy the motherboard, and everything plugged into it. It is not worth the risk considering most multimeters, like plug-in testers, do not measure, or reveal any unwanted and potentially disruptive AC components to the DC voltages.

    The voltages can be checked in the BIOS Setup Menus of most motherboards but they do not reveal ripple or other anomalies either. And the Setup Menu places very little demands on system resource so, like the temperature readings found in BIOS Setup Menus, they may not reflect values obtained when the computer is processing demanding tasks.

    And remember, anything that plugs into the wall can kill. Do not open the power supply's case unless you are a qualified electronics technician. There are NO user serviceable parts inside a power supply. ​
     
    Bill,
    #39
  21. 2010/11/09
    Christer

    Christer Geek Member Staff Thread Starter

    Joined:
    2002/12/17
    Messages:
    6,585
    Likes Received:
    74
    Thanks for that, Bill! A bit out of my "reach" but maybe the shop where I bought the components will be able to assist.

    I have posted a question at the Corsair Support Forums regarding the compatibility of the modules.
     
    Christer,
    #40
Page 2 of 5

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.
    Accept Learn More...