Ortho and errors on Blend test

Hi
I have been observing a weird problem with my new mobo. I have 8G of PNY DDR3 1333MHz RAM (4x2G). When I put all the sticks into the mobo, and run tests using memtest86/86+ and other memory tests from the UltimateBootCD for more than 24 hours without a single problem. I also used Windows Memory test Tools with the same result - clean. However, when I run windows, I get blue screens with MEMORY_MANAGEMENT problem being reported all the time.
I used also MemTest from http://hcidesign.com/memtest/support.html, which generates plenty of read/write errors under windows. I am very much confused so to what can be the potential problem of an error.
I run Ortho stress test. CPU only tests are clean after 5+ hours of testing. RAM only tests are clear after 5+ hours. When I try a BLEND type test, I get error within 10 seconds from the start.
Memory runs at nominal timings, (9/9/9/24) with upped voltage (1.6V versus 1.5 in automode).
Any suggestions on what I should be looking at ?
Thanks
Marek
14 answers Last reply
More about ortho errors blend test
  1. full system spec please, and are you overclocking? It's common in a lot of motherboards need a small voltage bump to the NB when all slots are populated.
  2. System: Windows 7 x64
    Mobo: GA-P55A-UD4P
    Memory: 4x2G PNY DDR3 1333MHz, running 1333MHz at 9/9/9/20 (stock timing) at 1.6V (stock voltage)
    HDD: 1TB Caviar Black
    GPU: PNY GF460

    CPU & memory running stock voltage and automatic settings

    Thanks

    Marek
  3. So you set the ram to it's rated voltage and main timings? If not give that a try, if still no joy, you can safely add .5 volts to the memory as long as you stay under 1.65.
  4. Correct, however, I tested the voltage range between 1.3 and 1.67 V already and it does not seem to impact the observed instability problems.
    I tested voltage range between 1.3 and 1.67V, with memory clock between 800 and 1333 MHz. I could not find a single combination that would work in a stable manner.
    The best part of it is there is no memory corruption using Memtest86/86+ tools without running Windows ...
  5. If you set the proper voltage and timings and your not overclocking, it might be time to start testing the modals for defects. It's a pain in the arse but I would test each stick individually. You might have a bad stick or two, can't hurt to try. And just to clarify, all other settings are default or auto correct?
  6. I spent already 2 weeks testing them one by one. Same result. No errors shown in Memtest86/86+, errors and blue-screens when running Windows. I am out of ideas. If nothing else can be done, I will return memory sticks today and replace with a different brand. Sick of PNY
  7. Just did something really stupid: disabled the CPU EIST / C1E / C3/6/7. memory is stable under Windows. Any explanation for that ??????
  8. This thread might shed some light on the subject.

    http://www.tomshardware.com/forum/266033-30-ud3r-random-reboots

    Side note, have you tried +.5 on the MCH voltage?
  9. Lame question: MCH?
  10. memory controller hub
  11. Sorry, no such option in my BIOS :(
    Now testing whether 8G will work fine with this configuration, 4G seems to have no problems passing burn in tests. Will let it run for 1 hour and come back with any feedback
    Thanks for the time
  12. NP

    No MCH/ICH setting? Should be under advanced voltage settings
  13. Thanks. Found it (was in strange greenish colour and not really visible). It turns out that 3 banks out of 4 have errors. Only one is stable.
    Will return the memory tomorrow and ask for a different brand.
    Thank you for your time
  14. NP GL
Ask a new question

Read More

Memory Overclocking