Random Freezing again.

[Penfold7]

Hmmm interesting. Will get PSU load tested…will post up speccy results when I can get my hands on system. Again today no issues, but my underlying suspicion is something is inherently not right.

[Penfold7]

Here is the speccy report, and Ive attached a screenshot of temps, as the system has been on gaming all afternoon. No freezes as yet.

http://speccy.piriform.com/results/LISI0XVonRzHt5LqUPdTNwo

Let me know if you see anything potentially dodgy. Its the most basic install of Windows I could do, just windows and updates plus essential drivers. Nothing else installed, using default edge browser.

[phillpower2]

Some strange voltage readings showing in Speccy & Speedfan, the latter is showing high CPU usage as well which can be caused by unstable voltage.

Something odd about the BIOS version as well, there are different revisions of the Gigabyte 970A-DS3P but none of them have a version FD.

Please restart your computer and check the temperatures/voltages in the BIOS, no screenshot is required just make a note of the temperatures, the DRAM voltage, the +3.3V, +5V and the +12V rails on the PSU and post them with your next reply.

NB: BIOS voltage readings.

The readings are not conclusive in the BIOS as the computer is under the least amount of load, if they are higher or lower than what they should be though it most often suggests a PSU problem.

[Penfold7]

Hi @phillpower
Thanks for reply… I will check the temps and voltages in the BIOS as you suggest. I’ve attached a screenshot of the Gigabyte support page for the Rev2.0 which is what he has. It appears this is the only Rev with the FD Bios. I checked this when this first started to see if the BIOS was up to date.
Ive also no idea where there is a CANON printer driver on Speccy… I do have a Canon printer on the Network at home its a Canon MG5750, but I havnt installed the drivers for this, as I was trying to keep a relatively clean install of windows till we get to the bottom of this.

Again thanks for the support and time, its much appreciated.

Chris

[Penfold7]

Hi. Its just frozen again, as I got home from work. Same as before, no warning, just suddenly freezes, and a forced reset needed. Nothing obvious in the event logs, just a Kernel Power Error, due to the system unexpectedly restarted. Its started to take ages to reboot after one of these forced resets as well. Seems to be OK after a controlled restart. Anyhow, these are the BIOS temps and voltages.
CPU Temp - 29degrees
System Temp 32degrees
CPU Fan 1486rpm

VCore 1.356v
DRam Volts 1.476v
Profile DR voltage 1.5v
ProfileVTT voltage 1.05v

+3.3v 3.185v
+5v 5.010v
+12v 12.096v

Hope this helps with the troubleshooting. My son says it does this only on a Saturday and not any other days since the HDD replacement. Seems like a total coincidence to me, as its been well used on other days. Out of curiosity, could the BIOS battery be an issue?? It seems to keep to time fine etc, but the time is set from internet servers…

[phillpower2]

Chris, hand on heart they were not the same results as when I checked just after midnight when I got home or as when I have just checked again, FD matches your Speccy report whereas I was getting F1, weird.

Kernel Power Error is a generic error that Windows reports in the Event Viewer when Windows does not get shut down as it should be, it is of no help to man nor beast tbh as it basically means that Windows was working away happily and then someone pulled the plug and the power went out.

Hope this helps with the troubleshooting. My son says it does this only on a Saturday

Odd, lets have a look at a MTB log, the PSU does need to be sorted but an issue like this only happening on a Saturday :unsure:, I`m most likely older than yourself and can remember Stars on Sunday, that program would do it to the best of us :LOL:

Download MiniToolBox and save the file to the Desktop.

Close the browser and run the tool, check the following options;

List last 10 Event Viewer Errors
List Installed Programs
List Devices (Only Problems)
List Users, Partitions and Memory size

Click on Go.

Post the resulting log in your next reply for us if you will.

[Penfold7]

Chris, hand on heart they were not the same results as when I checked just after midnight when I got home or as when I have just checked again, FD matches your Speccy report whereas I was getting F1, weird.

Wasnt doubting you, I just recalled I had checked for a BIOS update…
I will post a MTB log later on. I am 49, but don’t remember stars on Sunday, Just songs of praise!!

The Saturday issues is odd, likely coincidence… What were your thoughts on the voltages??

Will keep you updated…

[phillpower2]

I am 49, but don’t remember stars on Sunday, Just songs of praise!!

Pretty much the same thing so that would do it :sleep:

Sorry missed mentioning the voltages, they are circa what they should be but as previously mentioned the readings are not conclusive in the BIOS as the computer is under the least amount of load, running Speccy or Speedfan looks to be enough to raise certain voltages and cause high CPU use.

[Penfold7]

So what we are saying is, that the problem is likely to be intermittent issues with the PSU, likely causing the GPU to not get the power it needs and stall the system. Odd that it takes a forced reset to cure, and not to just leave it… I still have a gut feeling around the CPU as this all seemed to coincide with the CPU running hot with the supplied cooler, an Arctic cooler. It was getting to 75ish degrees. The CPU is still under warranty, and it was a refurbished, not new… Anything else I could do??

[phillpower2]

Been doing this stuff for years and that is how I come to have an in depth knowledge of hardware behaviour and PSUs in particular, for some reason you seem to be ruling out the PSU as the cause so below are a couple of things to explain why that is a bad idea.

The PSU is the one bad component in your build, made up of thin wires and weak capacitors and would definitely cause the GPU to heat up when it was struggling for power.

The PSU when independently tested was found to only be good enough to power a computer that had no add on video card only integrated video like that used by a notebook.

GPUs like the one that you have need a Gold efficiency rated PSU so that they are guaranteed to get good, clean, stable and efficiently delivered power whenever they are put under load, you get none of these from brands like Aerocool, hence the reason what they are on the avoid and lemon lists in my sig.

You ran Prime to test the CPU and the PC did not turn off, 99.99% rules out the CPU and Arctic cooler as Prime would definitely find out a bad CPU or its cooler.

Running Memtest caused no problems.

You had to turn off Furmark after ten minutes, if a PSU and/or GPU cannot do at least fifteen to twenty minutes test one or both are bad.

Not yet posted is the MTB log.

[Penfold7]

Hi. Please don’t think I am dismissing anything you have said…that’s not the case, I appreciate your help. Sorry about my log, I ran it last night and was sure I posted it… Will post it up in a Mo. I was just worried the high temps had damaged the CPU. I will just go and buy a replacement PSU. No point in testing the one I have if it’s not going to perform anyway… Out of the ones I listed, anyone’s good or bad, or any others you would recommend. I normally get most of my pc stuff from scan computers. As an aside, his previous build only had an athlon x3 and a radeon r7, that needed no PSU power.

[Penfold7]

MTB Log.

[phillpower2]

As was suspected we are seeing;

Error: (07/24/2021 03:40:39 PM) (Source: EventLog) (User: )
Description: The previous system shutdown at 15:30:14 on 24/07/2021 was unexpected.

Which means that the PC suddenly lost all power and was in no way related to the CPU having a thermal issue, side note, if it were you would get warning beeps before the PC suddenly turned off.

There are problems with Windows as well, the error below is a known bug that was meant to have been fixed by an update way back in version 2004 of Windows, the tool is only meant to be run on an SSD and not a mechanical HDD and that is why it throws up the error.

Description: re-trimSystem ReservedThe operation requested is not supported by the hardware backing the volume. (0x8900002A)

I know that Gold efficiency rated PSUs were being endorsed by myself but thinking outside of the box for a minute, if there are no plans to upgrade the GPU any time soon, being an older GPU you may not want to go the expense of a Gold efficiency rated PSU, if the case, go for some extra power from a proven brand of PSU, XFX recommend a 550W for their card that you have so if going this route the PSU here if you do want to go the Gold efficiency route the PSU here Scan do have it but are asking almost £80 + £5.48 for P&P.

[Penfold7]

Hi. There are no plans to upgrade in the near future. The system is perfectly adequate for his needs currently. Happy to go with whichever psu. Is seasonic a good brand then, even though its as cheap as the one I’ve got. Happy to pay less of course, but also happy to pay more… Just want the problems to go away. Its running happily at the minute… Maybe because its not a saturday. Abit concerned re window’s as this was a fresh install from a newly created disk of 21h1, on a new hdd. Also no ideas where the canon drivers have Come from, as I haven’t installed them yet…

[Penfold7]

Apologies if I’m being dim, but if pc lost all power wouldn’t it just turn off rather than freeze up with the display frozen on the monitor…