Blue Screen

[Terryan]

Hello

 

I assembled a new PC.

Here is my (big) problem

 

Each time I turn on the computer after having let it off a night long, I have instability leading to a Blue Screen with a 'Memory Management' message. :mad:

 

I have to shut down the computer, make a clear CMOS and, after that, the computer works perfectly well (no instability at all, whatever the program I use (microsoft program, network games...):D:

 

I know that I should change some parameters in the BIOS but I don't know what and how to proceed (capturing data and save them... forever). Could you please help me to resolve that problem ? I would be very greatful :praise:

[Wired]

Could be how the OS handled the memory and not the actual memory itself. Checked the logs?

[Terryan]

I think it's probably a voltage issue.

 

All bios memory parameters are set to 1.50v. My memory voltage is 1.65v. Should I set all parameters in Bios to 1.65 ?

 

Moreover there is DRAM voltage, DRAM termination, CH A data vref.......

 

I'm a little confused with all these features

[Chazzy84]

Interestingly enough, I have the exact same issue with my TW3X4G1333C9A kit. I've narrowed it down a little more, though.

 

When running memtest on a fresh boot (in other words, after it's been off long enough for the RAM to have cooled to ambient temperature) then there are a staggering amount of errors.

 

If I delay the boot (either by sitting in BIOS for a while, or, as you stated you need to do, reset the BIOS settings) long enough for the RAM to get to approximately 3-4 degrees above ambient temperature and run memtest, there isn't a single error, and I can boot normally without any hardware issues.

 

This stuff is 1.5V RAM, and, even if I wanted to, I can't change the voltage because of my rather... limited... board.

 

I've never seen a reverse-heat-related problem before! That is to say, I've never seen this kind of problem outside of a flyback transformer for an old CRT monitor...

 

Edit: The only thing I can figure this to be, at the moment, would be some ripple noise from my power supply causing the 1.5V to spike or drop low until the caps in the PSU warm up and the waveform stabilizes. I'm going to do some waveform tests soon to see if I can figure this one out. The more I think about it, the more likely this seems to be the case, and quite likely your problem, too. If you don't feel as bored and testing-inclined as I am, replace the PSU and let us know. Save me from digging out the old scopes!

 

Edit 2: The PSU didn't seem to be giving any more voltage fluctuation than I would expect a two year old PSU to do (+5V and +3.3V only had a variance of a hundredth of a volt in either direction.) So I removed the RAM. I then kept the system powered up with no memory (I was willing to risk the CPU stress for heat-related testing.) Putting the cooled memory in to the system and booting caused the same error we've both described. I do believe you should change the voltage to 1.65V if you have the capability (my EVGA H55V does not.) Aside from observing the 3 degree above room temperature reading I mentioned earlier, I have no new light to shed on this. I'm using the system I did these tests on right now, and, warmed up, it's rock-solid stable.

[Terryan]

So I have the problem again and I'm a little bit desesperate

 

Here is the WhoCrashed analysis

 

On Wed 24/11/2010 19:56:57 GMT your computer crashed

crash dump file: C:\Windows\Minidump\112410-20280-01.dmp

This was probably caused by the following module: ntoskrnl.exe (nt+0x70740)

Bugcheck code: 0x1A (0x41790, 0xFFFFFA8000E6D240, 0xFFFF, 0x0)

Error: MEMORY_MANAGEMENT

file path: C:\Windows\system32\ntoskrnl.exe

product: Microsoft® Windows® Operating System

company: Microsoft Corporation

description: NT Kernel & System

Bug check description: This indicates that a severe memory management error occurred.

This might be a case of memory corruption. More often memory corruption happens because of software errors in buggy drivers, not because of faulty RAM modules.

The crash took place in the Windows kernel. Possibly this problem is caused by another driver which cannot be identified at this time.

 

 

I really think that it is not a hardware issue... After half an hour, my PC becomes very 'stable' and I can use all programs you can imagine without any other crash

 

I do think that this problem is at the beginning of the boot, caused by windows 7 64bits which wants to do something with the Corsair memory and this ntoskrnl.exe file

 

Is there somebody who has had the same issue with the same config ?

[mikeinak]

Interestingly enough, I have the exact same issue with my TW3X4G1333C9A kit. I've narrowed it down a little more, though.

 

When running memtest on a fresh boot (in other words, after it's been off long enough for the RAM to have cooled to ambient temperature) then there are a staggering amount of errors.

 

If I delay the boot (either by sitting in BIOS for a while, or, as you stated you need to do, reset the BIOS settings) long enough for the RAM to get to approximately 3-4 degrees above ambient temperature and run memtest, there isn't a single error, and I can boot normally without any hardware issues.

.

 

I'm having the exact same problem. I have a CMP6GX3M3A1600C8 kit with a gigabyte GA-X58A-UD3R. Is this a bad module? After warming up, memtest shows no errors. Over the next couple of days, I'm going try to test each one individually and see what happens.

[Chazzy84]

I can pretty much guarantee that you both need to replace your memory modules. You'll find that they only work properly when warmed up, and that when you test them individually, they'll pass, but they'll fail as a pair. Essentially, something is keeping them from running their SPD timings properly. I've replaced my kit, and the machine has been completely stable for well over a week, now.

[maural]

@ chazzy 84

please better explain what do you mean with "replace" simply go to pc store and have ram changed ?

to all:

have pretty similar problem described above: multiple crashes with different os i.e. xp32, w7 32, w7 64 all happening when starting machine in the morning or next to several hours stop. this is probably due to ram modules warming upn need so therefore tomorrow morning will leave bios screen up for 5 minutes prior to load os and will post results.

also please someone let me know if issue could be related to voltage setting: mine is auto 1,5v while read 1,65 could fix probl though risky for mb&cpu

many thanks in advance

[maural]

Switched on pc -as planned- keeping bios screen some 10 mins prior loading os..to receive a couple of blue screen. Then made scan disk..to get blue screen once again. Reboot again to receive a perfect/stable system.

3 questions:

1. Ram warm up is not the problem ?

2. 10 mins warm up not enough ?

3. Is problem related to voltage setting 1,5v ?

Thanks in advance for your comment.

 

dmp files details below..

==================================================

Dump File : Mini121810-01.dmp

Crash Time : 18/12/2010 6.42.24

Bug Check String : KERNEL_MODE_EXCEPTION_NOT_HANDLED

Bug Check Code : 0x1000008e

Parameter 1 : 0xc0000005

Parameter 2 : 0x805beb34

Parameter 3 : 0xa9e32ae8

Parameter 4 : 0x00000000

Caused By Driver : ntoskrnl.exe

Caused By Address : ntoskrnl.exe+e7b34

File Description : Sistema e kernel NT

Product Name : Sistema operativo Microsoft® Windows®

Company : Microsoft Corporation

Processor : 32-bit

Computer Name :

Full Path : C:\WINDOWS\Minidump\Mini121810-01.dmp

Processors Count : 8

Major Version : 15

Minor Version : 2600

 

 

==================================================

Dump File : Mini121810-02.dmp

Crash Time : 18/12/2010 6.46.57

Bug Check String : MEMORY_MANAGEMENT

Bug Check Code : 0x0000001a

Parameter 1 : 0x00041284

Parameter 2 : 0x0007f001

Parameter 3 : 0x00000017

Parameter 4 : 0xc0883000

Caused By Driver : ntoskrnl.exe

Caused By Address : ntoskrnl.exe+22f43

File Description : Sistema e kernel NT

Product Name : Sistema operativo Microsoft® Windows®

Company : Microsoft Corporation

Processor : 32-bit

Full Path : C:\WINDOWS\Minidump\Mini121810-02.dmp

Processors Count : 8

Major Version : 15

Minor Version : 2600

==================================================

 

Dump File : 121710-27362-01.dmp

Crash Time : 12/17/2010 1:22:40 PM

Bug Check String : MEMORY_MANAGEMENT

Bug Check Code : 0x0000001a

Parameter 1 : 00000000`00001236

Parameter 2 : fffffa80`059d8830

Parameter 3 : fffffa80`059d8860

Parameter 4 : 00000000`0009c581

Caused By Driver : rdyboost.sys

Caused By Address : rdyboost.sys+4eb2

Processor : x64

Full Path : C:\Windows\Minidump\121710-27362-01.dmp

Processors Count : 8

Major Version : 15

Minor Version : 7601

[Chazzy84]

Sorry for the delayed response. Yes, you need to have your RAM modules replaced entirely, ie. new modules. Your problems will NOT get better, and will only get worse with time. It is my understanding the series of modules that we are in possession have a history of these errors.

[Wired]

It is my understanding the series of modules that we are in possession have a history of these errors.

Not true. Got anything to back up that statement?

[maural]

While waiting for Chazzy84 statement backup, could you please reply to my questions ?

1. Can Ram warm up solve the problem, or deftly no relation at all ?

2. Can problem be related to voltage setting 1,5v someway, though Terryan negative experience ?

many thanks in advance