Windows 10 MEMORY_MANAGEMENT cramping my style

doctorcheese

New Member
Hi,

Been getting the BSOD a few times a week since installing new GPU and RAM. Happens under stress 99% of the time.

Windows & all drivers should be pretty up to date, I did a clean wipe of the old GPU drivers before installing the newest. RAM is set to XMP 2.0 settings

Already ran 12+hrs of memtest86, chkdsk, SFC, and windows memory diagnostic, oddly no errors picked up anywhere.

I'm already past what I understand about all this, and would love some help from someone in the know.
PH
 

Attachments

  • W7F_05-02-2019 (2).zip
    657.1 KB · Views: 475
Have you done any over clocking on the ram?

Does setting them to XMP 2.0 settings count as overclocking? If not, then no.

I do have the processor bumped up a bit, and honestly I've just left it there, as temps have been ok.
  • R5 1600 set to 3700mhz @ 1.3 V
 
Whenever a system starts blue screening it's always prudent that in the first instance any overclocks be turned off and the system returned to default values.

See how the system runs.

If all is well then try your overclock again but with slightly different values.
 
Whenever a system starts blue screening it's always prudent that in the first instance any overclocks be turned off and the system returned to default values.

See how the system runs.

If all is well then try your overclock again but with slightly different values.

OK, thanks! I went back to factory settings on the processor. For the heck of it, I re-seated the ram and gave it a little air-can burst. Will see how it does.

Anything else it might be? From the most cursory, clueless investigation, it seems like I'm dumpin' 0x403 in parameter 1 of the bug check, which corresponds to "The page table and PFNs are out of sync . This is probably a hardware error, especially if parameters 3 & 4 differ by only a single bit. "
 
I took a closer look at your dump files and in the driver section came across this:
ssgdio64.sys Fri Sep 9 19:40:54 2005
Searches didn't bring up a deal of info apart from it being related to a ATI driver?

Try running what's known as the DDU or display driver uninstaller. This little handy app removes the gpu driver under safe mode and leaves your system clean and ready for the new gpu driver.
( I always disconnect from the net too just so bloody windows update doesn't sneak a driver install on, this actually happened one time so since then I disconnect.)

You can download the DDU here:
Display Driver Uninstaller (DDU) V18.0.0.8 Released. - Wagnardsoft Forum

Remember, it's best used in safe mode, oh and run it as administrator too.. :)

Once your back in Windows re-install the latest gpu driver for your set up.
 
I took a closer look at your dump files and in the driver section came across this:
ssgdio64.sys Fri Sep 9 19:40:54 2005
Searches didn't bring up a deal of info apart from it being related to a ATI driver?

Try running what's known as the DDU or display driver uninstaller. This little handy app removes the gpu driver under safe mode and leaves your system clean and ready for the new gpu driver.
( I always disconnect from the net too just so bloody windows update doesn't sneak a driver install on, this actually happened one time so since then I disconnect.)

You can download the DDU here:
Display Driver Uninstaller (DDU) V18.0.0.8 Released. - Wagnardsoft Forum

Remember, it's best used in safe mode, oh and run it as administrator too.. :)

Once your back in Windows re-install the latest gpu driver for your set up.

So, I set the CPU back to factory speed, and it was playing fine for a day or so, [edit: actually a week or so] then I just got ye olde memory management BSOD again.

I just did a full DDU/driver reinstall at the beginning of the month, but I'll give it another try anyways, as I see there's a newer version of my GPU driver.

Why would there be an ATI driver from 2005? Compatibility help bundled with an older game via GOG maybe?
 
Hmm... Well I guess the best thing would be to upload the dump file when you get chance and we'll take a look see.

Why that driver was there is anyone's guess but it probably came with something else like you suggest.

You did the best thing by using the DDU again followed by the new driver install
 
Hi!

Just got another two Mem mgmt BSODs, so I figured I'd try posting my most recent dump file. Here it is! CPU is still at stock, and this is after a DDU/driver reinstall.

Thanks very much for the help and feedback!
 

Attachments

  • W7F_01-03-2019zippo.zip
    683.8 KB · Views: 393
Been getting the BSOD a few times a week since installing new GPU and RAM. Happens under stress 99% of the time.
Your dump files don't reveal a great deal other than one was bugcheck 1a and the other was too corrupted to open.

Can you post what gpu you fitted and also what spec psu your using.

Testing with memtest86 sometimes still won't reveal a bad stick. It could be that memtest86 needs to be run longer but another way to check for a bad stick of ram is to run on one stick at a time. You may find that one stick blue screens whilst the other doesn't.

Setting xmp is actually overclocking the ram to it's stated speed quoted by the manufacturer. Try running without xmp and see if makes a difference.

If the ram is a replacement, do you still have the old sticks? If so try popping them back in and see if the bsod still occurs. If it doesn't you know it's probably the new ram at fault.

Stress test the gpu using furmark:
FurMark > Home
 
My board, cpu, ram, etc.. aren't new... as my stats will show. But I made sure that my (2) two pairs of ram on equal colour slots. Even though I have (4) 4-gigs sticks... the same/identical are on (slots 1&3, and the same are on (slots 2& 4)

My system rocks with my quad-core & 16-gigs of ram. I have no need to build a new system for years to come.
 
It's actually a 6 core with 12 threads, as you're on a Ryzen and having ram issues, try setting the V Soc volts in the bios to 1.15v I suggest this as you're running 4 sticks. Channel A/B volts for both channels to 0.700v instead of the 0.600 it's at default.
 
The op is running two sticks of 8GB. The post above may have confused the issue.
 
Last edited:
I'd still try the vsoc and if that doesn't help the a/b channel volts as it is known to stabilise xmp settings.
 
Can you post what gpu you fitted and also what spec psu your using.
1070ti. PSU is a corsair gold 550W.
I'll stress test the GPU, try running memtest for longer, then one stick at a time, and give things a spin without the xmp settings.

I'd still try the vsoc and if that doesn't help the a/b channel volts as it is known to stabilise xmp settings.
Never played around with vsoc. What should I try if I'm only using two sticks?

More data!
I just crashed when quitting a game and got a new BSOD, PFN_LIST_CORRUPT this time! Here's the dump file:
 

Attachments

  • dump 3-4-19.zip
    680.8 KB · Views: 397
Hmm... I'm wondering if your pagefile has become corrupted.

Pagefiles can also become fragmented and the best way to deal with issues with the pagefile is to simply disable it, reboot and then re-enable.

This creates a brand new pagefile hopefully with any issues resolved.
 
Never played around with vsoc. What should I try if I'm only using two sticks?

Try setting Vsoc to 1.1v to start and if that is of no help try 1.15v but under no circumstances go over 1.2v
 
Pagefiles can also become fragmented and the best way to deal with issues with the pagefile is to simply disable it, reboot and then re-enable.
OK. I've never played with the pagefile, so I'm just going off of googled directions. wish me luck?

Try setting Vsoc to 1.1v to start and if that is of no help try 1.15v but under no circumstances go over 1.2v
ok thanks, will give it a try. thanks!
 
Back
Top