Solved

[AMD] How to test CPU (especially IMC) and can CPU damage RAM ?

I some months ago built a gaming rig which was initially a budget machine and then I replaced almost everything beside CPU to convert it into gaming rig; during this I started getting BSODs (once a day in start and then it remains stable for that day).

Here I am trying to give as much detail as possible about my build so don't mind and skip it if you want. The replacment process was:

-- Bought a budget machine with a Phenom II X6 1090T CPU, 4GB DDR3 1333MHz and cheap 750W WinPower PSU
-- Was facing system powering issue with that PSU thus replaced it with Thermaltake XT 775W
-- Then bought Radeon 5770 Vapor-X; which I replaced with Radeon 6970 2GB with-in a month
-- Due to that size of 6970, I had to replace my casing (from normal cheap one to CM 690-II Advanced)
-- Then found that the motherboard (GA-880GM-UD2H) has very very bad layout thus covering 3 out of 5 SATA ports _and_ remaining 2-ports when plugged-in was pushing graphics-card to a side
-- Replaced motherboard (with Asus M4A89GTD Pro) and then I started getting BSODs
-- I thought its RAM thus I changed my RAM (from Mushkin enhanced 1333MHz to G.Skill 1600MHz 7-8-7-24 4x2GB modules)
-- Also bought CPU cooler CM V6GT and played with some overclocking
-- After lot of troubleshooting (including memtest86+ where I found one issue with RAM but later memtest86+ tests where fine) I found that the motherboard has issues thus I returned it; and bought new one (GA-990FXA-UD3)

-- With new motherboard system was fine for 5-days then when I start playing game (Crysis) on this fully replaced system (beside CPU); I faced some random application crashing, game lockups (pressing ctrl+alt+del recovers my system but game crashes) and occasional BSODs with SYSTEM_SERVICE_EXCEPTION and poiting towards MEMORY

-- I tested everything like my overclocking, RAM settings, when to everything stock and then did RAM settings manually. 2-days ago I again did memtest86+ v4.20 on my individual RAM modules in different slots _and_ found that one of my RAM module is faulty (have some bit-errors in different locations); yesterday I replaced that one with new one and in 7-passes of memtest86+ that new RAM is fine

-- So last night I plugged-in all my RAM modules (4x2GB) and was assuming that my system will work perfectly fine now. BUT while loading Crysis after giving some load on system/RAM (like running Chrome-Browser (15 tab pages), firefox (with 15 or so pages) and some random applications including Prime95 blend for 15-mintues); I _again_ _got_ _one_ _process_ _crash_ error "Activation Process crash or something similar) and then game started and was played for 30-mins without any issues (altougth was facing some hangs for split-second occasionally which was not present before).

Now my questions is: Why I got that application crash issue again ? How-to TEST CPU's full functionality especially Internal Memory Controller (IMC) ? And is this CPU is damanging my RAM ? Because now if this crash thing is present then seems like its due to CPU because thats the only thing left in my system since day-one of building this machine.

It will be good if I can get some suggestions soon.

Thanks in advance,

P.S. I will monitor for further issues and will report _but_ now I am really annoyed. I already spent 80% of my system-usage-time in past couple of months in fixing hardware rather than doing some gaming/software stuff on it.

My current system config is:

Motherboard - GA-990FXA-UD3
CPU - Phenom II X6 1090T CPU @ Stock; Cooler is: Coolermaster V6GT CPU cooler
RAM - 4 x 2GB G.Skill 1600MHz DDR3 7-8-7-24 @1.61v
GPU - Asus Radeon HD 6970 2GB GDDR5
Storage - 500GB WD 7200rpm (16MB cache) and 1TB WD 5400rpm (64MB cache)
Case - Coolermaster CM-690 II Advanced
PSU - Thermaltake XT-775W
Monitor - AOC 2436Vwa (1080p, 60Hz)
6 answers Last reply Best Answer
More about test especially damage
  1. You might have to bump up your ram voltage (especially since you have 4 sticks) a couple of notches do it in 0.05V increments up to max about 0.2volts over std and see if the issue disappears.
  2. rolli59 said:
    You might have to bump up your ram voltage (especially since you have 4 sticks) a couple of notches do it in 0.05V increments up to max about 0.2volts over std and see if the issue disappears.


    I already set my ram 0.1v above the rated which is 1.6v and I am using 1.61v. Although on G.Skill forum I posted this some weeks ago and got response that 1.65v is also safe to set. I will try increasing this a bit more like up-to 1.62v.

    Also I noted another thing that in my BIOS System Health Status page; RAM voltage is 1.596v (not 1.61v which I set). So is this can be a issue ? Like my motherboard is not providing enough voltage to RAM ?
  3. Best answer
    Settings and readings always have a little difference, your ram should be able to handle up to about 1.8 volts without trouble. I would set it at 1.65 then test try again if still not fixed with 1.7 and so on.
  4. Thanks for replying quickly.

    Will try this asap when reach home from job. Also how-to check maximum voltage settings for my DDR3 ? I am afraid of increasing voltage too much.
  5. rolli59 said:
    Settings and readings always have a little difference, your ram should be able to handle up to about 1.8 volts without trouble. I would set it at 1.65 then test try again if still not fixed with 1.7 and so on.


    This helped me a lot. From past 2-days I haven't saw any application or system crash, and just have to increase it 1.63v and seems good.

    Thanks
  6. Best answer selected by fslateef.
Ask a new question

Read More

CPUs RAM