Solved BSOD : Frequent and random stop messages

[richc46]

Try with clean boot
http://support.microsoft.com/kb/929135

 

[lengchai]

tried that ... still did not work.

I noticed my RAM sticks are DDR2 5300.

Here is what I tried next and the last thing I am trying seems to be working so far ...
a) Replace 1x 1GB RAM stick from another machine speed=6400 no crashes (only used it for ~1-2 hrs)
b) Replace 1x 1GB RAM stick from another machine speed=5300 - still crashed (seemed more stable ?)
c) Replaced both RAM sticks from another machine speed=5300 - no crash as yet.

So my guess is that it is hardware.

My only hypothesis is that with both RAM sticks, either the current is too high or some thermal/voltage spike is occuring taking my notebook out. I can only assume that this does not happen with 1 RAM stick as the current/voltage deviations may be within the mobo's limits. When Memtest86 is used to check out the RAMs, I guess the other parts of the mobo are idle, hence the RAMs can take more current than allowed ... hence not showing any errors.

I guess the current/voltage deviation with the 6400 RAM stick must be different or the newer ones are better designed/more efficient allowing my original 5300 RAM to take just that bit more current ... hence no crashes.

In any case ... my original RAM sticks are "dodgy" I would say.

Will report back after a few more days of "stress" testing ...

 

[lengchai]

I am happy to report that after the RAM swap, my vista has been stable for a whole day. Nothing crashed at all.

The conclusion is :-
a) If there are random BSODs at random times and various causes, e.g. ntkrpamp.exe, ntkrnlpa.exe, klif.sys, ntfs.sys, atikndag.sys, iastor.sys etc ... then it may be hardware.
b) Even if memtest86 does not throw up any errors, the RAMs may still be at fault.
c) Another useful debug method may be to just swap out the RAMs to see if the fault happens again.

In my case, the RAMs seem to be working on their own (even pass memtest86), but as part of a complete system, they may be drawing too much current or causing voltage spikes that eventually took out windows.

Thanks again to richc46 for being patient and providing guidance throughout this "ordeal" !

Hopefully this is my last post !

 

[richc46]

I agree with you completely. Many different codes and causes often indicate hardware. Memory is considered hardware in this situation.Glad that we found the solution.

 

[lengchai]

now how do i close this thread or mark it as solved ?

 

[richc46]

Since we want you to come back and be a member to both learn and to have fun, this is how its done
Click on the triangle upper right of the screen. When the box comes out just type solved and it will be closed.

 

[lengchai]

will do ... =)