Jump to content


Photo

Windows 8.1 BSOD?

BSOD

  • Please log in to reply
40 replies to this topic

#1 revert

revert

    Apprentice

  • Atomican
  • 98 posts

Posted 15 August 2016 - 07:47 PM

G'day

 

My PC just blue screened and idiotically is set to auto restart and i didn't catch the codes or anything just mentioned memory something, I reacted way too slowly.

Regardless.

 

 

How do i go about reading my dump files? And decipher them

 

And also why did it make me reset my overclock that I have had for over a year? is this just a precaution after a BSOD?

 

Slightly confused.

 

*Edit*

 

So I've hunted about, 

 

It was error 0x0000001a

parameters 00000000`00041284, 00000000`0d3ee000, 00000000`0000565a, fffff580`10804000

 

from google it seems the general idea is an issue with RAM.

But my PC is running fine atm. Will do a memtest and report back.


Edited by revert, 15 August 2016 - 08:00 PM.

"From the Ghastly Eyrie I can see to the ends of the world, and from this vantage point I declare with utter certainty that this one is in the bag!" 

#2 Rybags

Rybags

    Immortal

  • Super Hero
  • 34,829 posts

Posted 15 August 2016 - 08:13 PM

Manage -> Event Viewer>Windows Logs - System should contain the problem entries.  At the least e.g. with power loss or other unexpected failure the subsequent boot will have an entry saying there wasn't a clean shutdown preceding it.

 

If you're overclocking, have a dodgy piece of hardware or a buggy driver then often the error will be practically meaningless.  But at least in such cases acts as a symptom to something not being right.



#3 revert

revert

    Apprentice

  • Atomican
  • 98 posts

Posted 15 August 2016 - 09:58 PM

Okay. 

 

There are a few here. I don't really know what they mean or how to interpret them.

 

1 was critical Kernal Power?

 

The others at the first BSOD where it failed to restart several times was MEMORYDMP

 

Think I'll look at doing a memtest.


"From the Ghastly Eyrie I can see to the ends of the world, and from this vantage point I declare with utter certainty that this one is in the bag!" 

#4 Rybags

Rybags

    Immortal

  • Super Hero
  • 34,829 posts

Posted 15 August 2016 - 10:11 PM

It's usually easiest to find messages relating to the next boot then work backwards.

 

And yep, the auto restart thing is ridiculous, shouldn't ever be used on a home system or machine you actually perform work on.

I set mine to not produce dumps, you'll likely get nothing of use, they're generally something you'd send to a vendor.  Just the logging of critical errors is enough.

 

Memtest and stuff to stress the machine from inside Windows is a good idea.  Also a good idea to check the Ram timings in the Bios, compare to what CPU-z reports the modules capable of doing.



#5 revert

revert

    Apprentice

  • Atomican
  • 98 posts

Posted 15 August 2016 - 10:42 PM

Kk. So running memtest. Ram is definitely faulty.
Timings were fine I set those prior to my overclock, which has been reverted to base clock after that crashes.
Ill have to replace the ram. Would replacing the ram require me to reformat? Ive always reformated after any hdd change just out of good measure. But is it actually required for ram?
"From the Ghastly Eyrie I can see to the ends of the world, and from this vantage point I declare with utter certainty that this one is in the bag!" 

#6 Rybags

Rybags

    Immortal

  • Super Hero
  • 34,829 posts

Posted 15 August 2016 - 10:46 PM

No action needed for Ram replacement.  Just before doing so though, go into the Bios and make sure the settings for stuff like timings and voltage are on auto.  Then do the tweak/overclock stuff later.

 

I'd try your tests with stock speeds before replacing the Ram, it might just be that you're stretching it to far.



#7 revert

revert

    Apprentice

  • Atomican
  • 98 posts

Posted 15 August 2016 - 11:44 PM

It was clocked for a full year and running fine? Think age is playing a factor?
"From the Ghastly Eyrie I can see to the ends of the world, and from this vantage point I declare with utter certainty that this one is in the bag!" 

#8 Rybags

Rybags

    Immortal

  • Super Hero
  • 34,829 posts

Posted 15 August 2016 - 11:56 PM

Overclock doesn't necessarily mean the Ram was running faster or on tighter timings than spec.

You can even run stock speed or underclock the CPU and run the Ram faster.

 

What were the speeds and settings and how did they compare to what the SPD specifies?



#9 revert

revert

    Apprentice

  • Atomican
  • 98 posts

Posted 15 August 2016 - 11:58 PM

Im beyond confused. Did mem tests on both sticks individually. No errors. Re did tests in different slots no errors. Went to re run tests eith both sticks and memtest says saything about smp? And fails to run tests.....
"From the Ghastly Eyrie I can see to the ends of the world, and from this vantage point I declare with utter certainty that this one is in the bag!" 

#10 Master_Scythe

Master_Scythe

    Titan

  • Hero
  • 19,966 posts
  • Location:QLD

Posted 16 August 2016 - 09:07 AM

Im beyond confused. Did mem tests on both sticks individually. No errors. Re did tests in different slots no errors. Went to re run tests eith both sticks and memtest says saything about smp? And fails to run tests.....

 

You're assuming RAM because the system did a memory dump? thats unrelated.

 

Give this guy a shot:

http://www.majorgeek...screenview.html

 

Hopefully it'll tell you what file caused the error, and then you can update\replace the driver related to it.

 

If its something like IRQ not Less or Equal, then that's usually RAM.

 

If the sticks test overnight, individually fine, then they're not faulty.

Normally overnight willget you about 10 passes, while thats not very many for memtest, its usually enough for a home system.

If you can leave it a full 24 hours per stick you'll be 90% sure they're OK.

 

Took me over a week once to find the error on a stick.....

 

However, so far nothing you've posted actually suggests RAM.


Edited by Master_Scythe, 16 August 2016 - 09:07 AM.

Wherever you go in life, watch out for Scythe, the tackling IT support guy.

"I don't care what race you are, not one f*cking bit, if you want to be seen as a good people, you go in there and you f*ck up the people who (unofficially) represent you in a negative light!"


#11 Nich...

Nich...

    Professional Tart

  • Mod
  • 43,072 posts
  • Location:Mexico

Posted 16 August 2016 - 11:55 AM

http://www.nirsoft.n...creen_view.html

 

I get a lot of BSODs on my system.  The errors are usually memory related, but memtest has shown they seem to be ok.  They're all basically kernal faults, tho', so short of reinstalling my windows, I won't know for sure.

 

BSV is a lot better than just looking at event logs, tho'.  Assuming you let it make dump files.


"I think it is a sad reflection on our civilization that while we can and do measure the temperature in the atmosphere of Venus we do not know what goes on inside our soufflés" -- Nicholas Kurti

#12 revert

revert

    Apprentice

  • Atomican
  • 98 posts

Posted 16 August 2016 - 01:28 PM

Yeah, I pretty much just jumped to the conclusion, I am not a smart man in regards to computers.

 

I don't really know how to read the bluescreen viewer.

 

bug string: MEMORY_MANAGEMENT

bug check code: 0x0000001a

parameter:00000000`00041284

parameter:00000000`0d3ee000

parameter:00000000`0000565a

parameter:fffff580`10804000

 

nothing in cause by driver or file fields...


"From the Ghastly Eyrie I can see to the ends of the world, and from this vantage point I declare with utter certainty that this one is in the bag!" 

#13 Master_Scythe

Master_Scythe

    Titan

  • Hero
  • 19,966 posts
  • Location:QLD

Posted 16 August 2016 - 02:03 PM

Despite me trying to calm you, that error (0x0000001a) is related to the kernel memory managment.

In other words, the base of the system the hands memory where it needs to go.

 

USUALLY it is a faulty stick.

 

You don't happen to have any "speed up my computer" type applications I hope? They'll cause things like this as they try and 'take over' ram control.

 

Test each stick for a full 24hours. If no errors, then you start to expect a slot going bad, or even a memory controller on your CPU if you've overclocked\overvolted it.

Because from what you've mentioned, it sounds like its having trouble when it has to address 4 lanes of memory (2 on each stick) not just 2.......


OH, also:

 

http://forums.atomic...n10-still-free/

 

Get on it boy! 8.1... whyyyyyy


Wherever you go in life, watch out for Scythe, the tackling IT support guy.

"I don't care what race you are, not one f*cking bit, if you want to be seen as a good people, you go in there and you f*ck up the people who (unofficially) represent you in a negative light!"


#14 revert

revert

    Apprentice

  • Atomican
  • 98 posts

Posted 16 August 2016 - 04:53 PM

Haha! Yeah, right. So pretty much I had done a bluescreen viewer and googled the outcome and came to the assumption of a faulty stick and proceeded to memtest with them both in and it had errors, so then I tested 1 stick at a time and didn't get any errors, tested the next stick, no errors. Although clearly I wasn't testing them for long enough as it was probably a few hours per stick, so I was lead to believe I might have a faulty slot and I've proceeded to test them for a few hours each stick in each slot to no avail it wouldn't error, until I placed them back in their rightful places and it failed to boot and even failed to memtest had something come up about SMP.
 

Which is probably why I seemed worked up.

 

My overclock is just a simple multiplier and voltage.

I can't remember the voltage of the top of my head but it was very minor and has been stable for about a year.

 

So now, I'll test each stick for 24hrs is their respected slots and see what I can find.

 

I have read that. Just lazy. I as delayed to get windows 8.1 from 7 and barely have gotten comfortable.

 

Anyway, As always I appreciate the help and I'll report back with my finding in a few days of testing.

 

The ram is $60 anyway. So I've already ordered it. Worst case scenario i have 16gb of fully functional ram and a dead ram slot!


"From the Ghastly Eyrie I can see to the ends of the world, and from this vantage point I declare with utter certainty that this one is in the bag!" 

#15 Master_Scythe

Master_Scythe

    Titan

  • Hero
  • 19,966 posts
  • Location:QLD

Posted 17 August 2016 - 09:08 AM

 

My overclock is just a simple multiplier and voltage.

 

What voltage was on the memory controller?

 

Its rare but if each stick is OK. and each slot is OK. We have to start thinking about the memory controller on-chip.

 

Its especially risky if it was left on "AUTO" while you bumped other voltages....


Edited by Master_Scythe, 17 August 2016 - 09:08 AM.

Wherever you go in life, watch out for Scythe, the tackling IT support guy.

"I don't care what race you are, not one f*cking bit, if you want to be seen as a good people, you go in there and you f*ck up the people who (unofficially) represent you in a negative light!"


#16 revert

revert

    Apprentice

  • Atomican
  • 98 posts

Posted 17 August 2016 - 01:22 PM

Probably was auto. I only changed a few settings and the multiplier and vcore voltage iirc. Was my first shot at overclocking.
"From the Ghastly Eyrie I can see to the ends of the world, and from this vantage point I declare with utter certainty that this one is in the bag!" 

#17 Master_Scythe

Master_Scythe

    Titan

  • Hero
  • 19,966 posts
  • Location:QLD

Posted 17 August 2016 - 02:28 PM

Lesson learn't about Overclocking then; first step is to find out what all voltages are STOCK, and LOCK THEM where possible.

When they can't be locked, simply bump them up by the smallest increment possible (which will also lock them, just a few mv higher).

 

Hopefully its the RAM, but if each stick is passing... I'm not holding my breath....


Wherever you go in life, watch out for Scythe, the tackling IT support guy.

"I don't care what race you are, not one f*cking bit, if you want to be seen as a good people, you go in there and you f*ck up the people who (unofficially) represent you in a negative light!"


#18 Rybags

Rybags

    Immortal

  • Super Hero
  • 34,829 posts

Posted 17 August 2016 - 02:32 PM

The other thing is, you can have a totally stable system with just 1 or 2 sticks of Ram installed then have it become rocky once you put 4 in.



#19 revert

revert

    Apprentice

  • Atomican
  • 98 posts

Posted 17 August 2016 - 04:58 PM

So I should have locked my RAM at their stock voltage to avoid them being bumped up passed recommended by the AUTO setting?

I havent done tests longer than an hour yet. Been busy.

So as it stands I need to thoroughly test each stick. Then each slot.

Correct me if I am wrong.
"From the Ghastly Eyrie I can see to the ends of the world, and from this vantage point I declare with utter certainty that this one is in the bag!" 

#20 Rybags

Rybags

    Immortal

  • Super Hero
  • 34,829 posts

Posted 17 August 2016 - 05:12 PM

Probably.  That's the problem with Bios settings.  In some cases Auto just means "use the default", in others it means that the setting is auto-adjusted to what it thinks should be right.  In most cases that's OK but with voltages you need to be careful about it.






0 user(s) are reading this topic

0 members, 0 guests, 0 anonymous users