I have read some posts regarding the Machine Check Event reported is a fatal memory hierarchy error (WMIxWDM) errors. There have been no resolution except for Microsoft telling you to upgrade your PC to the latest BIOS...which was done and didn't work.

I have a gigabyte K8U 939 MOBO with the AMD x2 3800 Dual Processor.
A GeForce 5500 Dual Video card and 2 (1 gig) CORSAIR Chips - Running Dual DDR Enabled.

Any idea why these Events 120 and 123 keep coming up? The system is a dual boot system and these errors only appear running XP Pro x64 not XP Pro.

And yes, I have updated the AMD drivers, tested and replaced the memory as well.

Any help would be appreciated.
Thanks
Mike

Overclocked machine?

Nope, all settings in BIOS as "Safe" settings No overclocking at all.

@mgrande............After all that fuss and muss, it turned out my CPU was bad. (I even got a new mobo, memory, etc.)--It may have been my fault for letting it get up to 77c for a few moments. (but that's another story). It works fine in 32 bit. In 64 however, it was crash-city. All I got was the ever popular yellow error sign--at the slightest thing too-- i.e. the activation notice trying to start, etc. I yanked it out, and put in another new CPU--problem was gone, no errors. Everything loaded up just like it's suppose to. It was the CPU in my case...see if you can borrow one for a few minutes to test with.

wow CPU?!? I am going to follow up with AMD 2morrow

Thanks!

Saw similar post in NT/2000/2003 forum, so I'll post the same response here:

A machine check error is an error in the processor that is detected by the OS and is correctable (if it wasn't, your machine would bugcheck/bluescreen). This usually signifies that your processor may need to be repaired/replaced (hopefully under warranty), as these are hardware errors happening in the processor. Since these only occur when running x64 versions of Windows, it would stand to reason that these errors are occurring in one of the additional registers addressable only by 64bit software. You can disable this error on Server 2003, but since it is a serious error, you should try to replace that hardware under warranty while you still can.

912310: How to disable the corrected machine check on a Microsoft Windows Server 2003-based computer
http://support.microsoft.com/default.aspx?...kb;EN-US;912310

Well, replaced the CPU today and still get the error! Just getting Event 120: Machine Check Event reported is a fatal memory hierarchy error.
Trasaction Type: 3
Memory Hierarchy Level: 3
Request Type: 13
Address: 18446735140177573855

Don't know what else to do! Any help would be appreciated!

Perhaps check the vcore settings of the motherboard to the processor, make sure your BIOS and motherboard are compatible with the processor, and lastly run the system through some sort of hardware diagnostics checker. Not sure what else to tell you - you already know what a machine check error is, so perhaps AMD or your motherboard manufacturer (or again, the vendor of your machine) could give you some insight as to why this keeps occurring?

I had that problem, found a solution that worked for me. It was my nVidia drivers. Had to downgrade to a lower version (don't remember which version, I can check tomorrow).

EDIT:
I checked the version of the nVidia drivers that fixed this problem on my install. I'm using version 6.6.9.6

Huh, well I'm having very similar issues and also on a Gigabyte motherboard. I'm running a GA-K8U-939 with an opteron 165 processor and an nvidia AGP graphics card. I occasionally get the BSOD with the MACHINE_CHECK_ERROR message. I've had this setup for about 6 weeks and have received the error only twice. I'll try downgrading the graphics drivers and will let you know if that fixes things for me as well. May take me a while to report back since I get the BSOD so infrequently.

Updated link here: http://support.microsoft.com/kb/912310/en-us

Thanks for this...I have been looking for a work around for an annoying error similar to this I keep getting:and
I tested my ram with MemTest86+, and it yielded nothing. People have also said it could be the AMD Athlon 64 CPU itself. Another person from another board also mentioned about going back to an older version on the nvidia drivers. This system is not overclocked or anything. The CPU however is a packaged bundle with an nForce3 board with the geForce FX nVidia video component.

I get no BSODs though...nothing except the annoying popup on event ID 26, and then the warning icon in the taskbar. I also notice some sluggish video performance when watching videos. What I do not understand is that even though this is an x64 system, still a lot of the windows components like WMP, Outlook Express, etcetera, are still not native 64 bit--they heavily reside in the Program Files (x86) folder. In a clean installation, the subfolders in the x86 program files greatly outnumbers the ones in the regular program files folder (64 bit).

If Microsoft made a 64 bit OS, almost everything should have been rewritten to 64 bit...at least the basic window components and accessories. I do like the idea that there are two versions of IE, and the fact that we still can run most x86 programs via the WOW64 of course.

Hello, I'm new here.

Got the same error on my setup



I ran the Prime 95 torture test for a medium duration and nothing found.
In normal conditions my system runs fine without any problem.

I get this error only if I wake my system from stand-by, but nothing dissapear and there is no misbehavior in resuming. I can resume Windows X64 without any problem. But often I get this message as a warning. Not always!

I discovered that this behavior in Windows X64 is inherent with Vista PCI to ISA bridge missing when booting up. This results in the lost of the graphic interface Aero so graphic drivers reinstall themself immediately, but a restart is required ans Windows X64 fix this problem on-line. Windows X64 never forgot drivers and the PCI to Isa bridge runs fine.

I also discover that a cold boot results in a 3 time reboot before BIOS information turns on screen. The third time the machine always boot up and never fails, but it's some strange behavior. I guess it's a bug in the BIOS.

It is not always the CPU. Now my CPU ran at stock speeds, but it handles a 20% overclock without doubt.

But the mainquestion of this story is:

Could it be confirmed we face a BIOS problem by design or?

System:

Asus A8R-MVP with BIOS 0503
AMD64+ 3800+ Venice s939
4x1024 MB DDR400
2 IDE harddisks
ATi X1950pro GPU
High range 500 watt PSU
22 inch TFT running @ 1680x1050



At 0008, the letter is always "A "

That same system is still up and many, many kind of these errors later, not any hazzle.

Strange

I've found the sollution for my system and maybe it is usefull to you all.

Many users uses serial ATA instead of IDE and I still use an IDE setup with 2 harddisks and a fast 16x DVD writer. Nothing about it, or?

I've maintained my system by cleaning out all the dust in the chase and te system works like never before. What happend?


This is the old situation:
PRI-IDE Second harddisk for dumpfiles, images and the 1st line back-up (out of 3, The USB key is the second line and a back-up on DVD also remains)
SEC-IDE The burner as slave and the main harddrive as master

Far from ideal and I thought the difference between the location of the devices don't matter, but the master should be the last on the cable.

New situation:
PRI-IDE Main harddisk (no install needed)
SEC-IDE The burner as Master and the dumpdisk as slave.

No WDMxWMI error seen after this action and the system boots up properly (It did, but my system restarts itself twice before BIOS information is seen. It is solved)


So it whas the controller what creates this error in my cirumstance.
Figure out of everything is set perfect on the IO controllers!