Question Random errors and blue screens- win 7

[spdfreak]

So, for the past few days, I've been getting random blue screens and reboots- some are system errors, today it was an IRQ error and a few others. DiskChk found one error and fixed it. I ran memtest and no errors with the memory. I suspect maybe the SSD is crapping out. Any suggestions for what to look for?

Ryzen 7 1700X
Gigabyte AB350 Gaming 3
16 GB Corsair 3200 running at 2997
ADATA SU800 512 SSD
Radeon HD6850 vid card
several other storage drives

 

[DAPUNISHER]

Check all your drives health https://crystalmark.info/en/download/

Do you have the latest UEFI for that mobo? That HD6850 is out of place in that system too, just sayin'.

 

[UsandThem]

In addition to the drive health utility above, you can try Bluescreenview to see if it can shed any additional info on the blue screens:

https://www.nirsoft.net/utils/blue_screen_view.html

 

[mxnerd]

It's possible that PSU is causing problem. Either not enough power or unstable.

Also check file system using sfc /scannow command.

 

[spdfreak]

Check all your drives health https://crystalmark.info/en/download/

Do you have the latest UEFI for that mobo? That HD6850 is out of place in that system too, just sayin'.

I used the ADATA drive utility and it found no problems, but I've had other SSD's that showed good and were bad. Latest bios was installed a few weeks ago. This problem is pretty sudden, though- in the past few days. Previous to that no problems. I did move the memory to the other 2 slots as a test. PSU is a Seasonic 620W and only a few months old.
Yeah, the 6850 is pretty old. Still waiting for mid-range vid card prices to drop into reasonable territory. Maybe I'll just buy a used RX480. They are about 100.00 on ebay now.

Ran SFC and it did not find any problems...

 

[mxnerd]

PSU is a Seasonic 620W and only a few months old.
Ran SFC and it did not find any problems...

Since you have multiple drives, try reconnecting the drives with different SATA connectors. Or if you have another PSU handy, try swapping the PSU and see.

 

[Ketchup]

Are the dumps happening under heavy loads or light? Doing anything specific when they happen? Can you share the dump files with something and send us a link?

 

[spdfreak]

No, they are random. Sometimes I will come back to the office and the computer will have rebooted. But, after moving the memory sticks to the other slots, it has not done it today. Can I just attach the mini dump file to a post? There are a few older ones in the folder but 3 or 4 day since 12-27.

 

[Ketchup]

I can't say 100% for the most recent forum updates, but I doubt you can share without a host site.

I will say that I have experienced a memory slot go bad before, but in all cases the boards were several years old.

Since you are running a Ryzen 7 with Windows 7, which isn't "officially" supported, you may want to run driver verifier to make sure it isn't just a software conflict.

https://www.reviversoft.com/blog/2012/09/irql-not-less-or-equal/

Any reason you aren't running 10 on this machine?

 

[VirtualLarry]

I've had Windows 10 Pro 64-bit 1809 on my Ryzen R5 1600 PC (with Sapphire RX 560 2GB, Adata SP550 240GB SSD, Seagate 4TB 5400RPM HDD, 4x8GB Team Vulcan DDR4-2400 RAM), spontaneously reboot on me. But it's weird, it's not like a BSOD, and it automatically boots back up. But it doesn't appear to be a Windows Update thing. It just happens out of the blue, and no updates are reported in Reliability Monitor.

I found out that I needed to run my DRAM voltage slightly higher than 1.2V (nominal) for things to be stable, and I used to be overclocked to 3.80Ghz (on 120mm AIO WC), but I've removed the OC for now. Seems the latency is better without OC? I know, strange, right?

 

[spdfreak]

I can't say 100% for the most recent forum updates, but I doubt you can share without a host site.

I will say that I have experienced a memory slot go bad before, but in all cases the boards were several years old.

Since you are running a Ryzen 7 with Windows 7, which isn't "officially" supported, you may want to run driver verifier to make sure it isn't just a software conflict.

https://www.reviversoft.com/blog/2012/09/irql-not-less-or-equal/

Any reason you aren't running 10 on this machine?

I'm going to add another SSD with Win10 at some point but I still use Media Center to record OTA TV shows I want to edit/watch later.

It could be a driver- the vid card driver is from 2016 but is the latest available. The strange thing is that it has only started happening in the past few days with no hardware changes to the system at all.

 

[spdfreak]

I've had Windows 10 Pro 64-bit 1809 on my Ryzen R5 1600 PC (with Sapphire RX 560 2GB, Adata SP550 240GB SSD, Seagate 4TB 5400RPM HDD, 4x8GB Team Vulcan DDR4-2400 RAM), spontaneously reboot on me. But it's weird, it's not like a BSOD, and it automatically boots back up. But it doesn't appear to be a Windows Update thing. It just happens out of the blue, and no updates are reported in Reliability Monitor.

I found out that I needed to run my DRAM voltage slightly higher than 1.2V (nominal) for things to be stable, and I used to be overclocked to 3.80Ghz (on 120mm AIO WC), but I've removed the OC for now. Seems the latency is better without OC? I know, strange, right?

This is blue screen stuff... but could be memory voltage. I will try dropping the speed back down and see if it is more stable although it has been good all day since I changed slots. What voltage should I set it at to give it a little extra headroom?

 

[spdfreak]

Update- so far no further errors or blue screens since moving the memory sticks. So, hopefully, moving the memory to the other slots has fixed the problem. I guess I could move them back and see if the problem reoccurs but I'm going to wait a few more days before I try that.

 

[VirtualLarry]

AM4 boards ARE picky about which slot(s) the RAM is in. Did you look in the manual? It should tell you which ones to put them in, if you're not filling all of the slots.

 

[spdfreak]

AM4 boards ARE picky about which slot(s) the RAM is in. Did you look in the manual? It should tell you which ones to put them in, if you're not filling all of the slots.

Yeah, I noticed even the 300 series Intel boards have recommended population order. These are now in the opposite of what the manual recommends but it seems to have fixed the problem, so probably a slot went bad.

 

[spdfreak]

Well, I still get the random blue screen error like memory management about once a week so I dropped the memory back to default speed from the 3000 it was rated for. Is it advisable to bump the memory voltage a bit to stabilize it at 3000? If so, what is a good starting point?