Are these normal CPU Voltages?

[Long Silog]

My PC is hard freezing under heavy load (e.g. Playing AAA games or needy games like PUBG). I'm told it's my PSU but I would really like to confirm before buying a new one. Also often my rig would turn on by itself when I put it to sleep and even when I shut it down. I would have to turn of AVR.

https://imgur.com/0HGx9w2

OS: Windows 10

MOBO: Gigabyte B85m HD3-A (previously Asus Z97 Deluxe)

CPU: Intel Core i7-4790k

GPU: Sapphire R9 390x Tri-x OC 8gb

RAM: G.Skill Ripjaws X series Dual 16gb 1866

HDD:1 x 128gb Samsung 840 Evo1 x 2tb WD Caviar Black1 x 3tb WD Caviar Green

Liquid cooling: Swiftech H240x

PSU: XFX Pro series 1250w 80 plus gold

 

[Long Silog]

Reason I mentioned that is because many ppl actually do check that, upgrade bios/drivers to the newest, still have issues somewhere, so decide to reinstall windows. Unfortunately, this also wipes out any and all mobo drivers, and many don't see it, because windows has generic drivers it prefers to use, which still cause errors with legacy hardware looking for specific drivers and they are no longer there. It's hard to guage anything you may have done and any specific order things are done in.

Hard freezes resulting in black screens are almost always gpu failures. Doubtful it's the psu, since the gpu uses the 12v rail and if that fails the pc shuts off, fans turn off etc. Full, instant shutdowns. Personally, my money would be on a VRM failure in the gpu. The VRM's/vram is not temp reported in almost all video cards, only the processor (gpu) is, so it's quite common for a card to see temps in the 60's yet the vrms/vram is exceeding 90°C. If the vram is the issue, common affects are artifacts on screen, leading me to suspect it's the VRM's instead, which can cause gpu shutdowns, black screens etc.

Yeah. I also read that if PSU is the one failing the system should turn off but that's not the case. I can go from day to night when browsing the web, watching movies, listening to music, etc. but only minutes to an hour max when playing games. Also I plugged the HDMI cable form my GPU to my MOBO but the hard freeze still occurred but no black screen, there was a display.

 

[Karadjgne]

Which eliminates the gpu as the issue as the freeze still happens when the monitor is hooked to the igpu.
So.
Not psu.
Not gpu.
Not mobo.
Only things left is storage and cpu and ram.
Storage is possible cause if it is hdd and motor is failing or if ssd controller is failing under heavy usage, like gaming.
Ram is possible if it's overheating or there's an issue with the memory controller. Which is part of the cpu. Normally you'd get a bsod though.

Is Windows event viewer showing any critical errors other than the forced shutdown?

 

[Long Silog]

Which eliminates the gpu as the issue as the freeze still happens when the monitor is hooked to the igpu.
So.
Not psu.
Not gpu.
Not mobo.
Only things left is storage and cpu and ram.
Storage is possible cause if it is hdd and motor is failing or if ssd controller is failing under heavy usage, like gaming.
Ram is possible if it's overheating or there's an issue with the memory controller. Which is part of the cpu. Normally you'd get a bsod though.

Is Windows event viewer showing any critical errors other than the forced shutdown?

Nothing critical besides the forced shutdown but there's an error. Windows event viewer only shows DCOM errors and from what I've read it's about permissions, I think? Event ID 10016

-Clean OS install
-Updated drivers
-Rolled back drivers
-Ran Intel Diagnostic Tool
-Ran memtest86 and Windows memory diagnostic
-Changed HDMI cable from GPU to MB and hard freeze occurred without the black screen. (Can see the display)
-CHKDSK
-Reset BIOS to default
-Changed game location to other hard drive/SSD
-Fresh install of Steam and PUBG (Game I frequent)
-CPU and GPU temp: Idle: Below 50 Gaming: Below 70
-Nothing is overclocked.
-Tested PSU with multimeter (All positive and negative 3.3v, 5v, and 12v are in good range) Is it still possible it's the culprit?

 

[Karadjgne]

You've done pretty much everything I can think of. I've never heard of a partial rail failure on a single rail psu, although that's possible on multi-rails where the pcie is separate from other areas. But, since the freeze happens when using the igpu instead, that precludes the draw on the pcie as all the power comes from the cpu, a fraction of what the gpu can pull from that psu.

So as far as I can see, it's 1 of 2 things. Either the psu at the connection for the wires leading to the 20+4, inside or outside the casing, or it's damage caused by the prior bent pins to a part of the cpu that's only really taxed during heavy usage.

Swapping out the cpu or swapping out the psu.

 

[Darkbreeze]

I say it's the CPU, damaged caused by the previous shorting of at least one or more circuits due to the bent pins. Which is pretty much what I've been suspecting all along every since he mentioned the bent pins on the previous motherboard.

It's a lot less likely for the PSU to have been damaged, due to it's own protections. The CPU, has no such protections from a direct short or crossed circuit.

 

[Long Silog]

You've done pretty much everything I can think of. I've never heard of a partial rail failure on a single rail psu, although that's possible on multi-rails where the pcie is separate from other areas. But, since the freeze happens when using the igpu instead, that precludes the draw on the pcie as all the power comes from the cpu, a fraction of what the gpu can pull from that psu.

So as far as I can see, it's 1 of 2 things. Either the psu at the connection for the wires leading to the 20+4, inside or outside the casing, or it's damage caused by the prior bent pins to a part of the cpu that's only really taxed during heavy usage.

Swapping out the cpu or swapping out the psu.

Yeah. I think the only way to test it is to swap out the parts but if ever it's the CPU then I think it's time to buy a new setup.

 

[Long Silog]

I say it's the CPU, damaged caused by the previous shorting of at least one or more circuits due to the bent pins. Which is pretty much what I've been suspecting all along every since he mentioned the bent pins on the previous motherboard.

It's a lot less likely for the PSU to have been damaged, due to it's own protections. The CPU, has no such protections from a direct short or crossed circuit.

I think I'll buy a PSU but if it doesn't solve the problem then I'll just need to assemble a new rig. LGA 1150 CPU and Motherboard are very hard to come by since they're phasing it out.

 

[Darkbreeze]

That's a good idea, because if it IS the PSU, problem solved, and if it's NOT, it's not a terrible loss. It's always good to have an extra PSU on hand for testing or as the basis of the start for a secondary system. I usually recommend buying a PSU that is better than the last one you bought, as the warranties for most really good units has exceeded 10 years now. I wouldn't recommend buying any power supply for a high end system or gaming machine these days that doesn't at least have a Gold efficiency rating AND a 10 year warranty.

Those two things alone don't necessarily ensure that it will be a good unit, but it sure is nice to have them if it is and most units that meet both those criteria are going to be well built.

 

[Karadjgne]

^Amen!

 

[Long Silog]

Thanks for the help and effort, guys.

 

[Long Silog]

That's a good idea, because if it IS the PSU, problem solved, and if it's NOT, it's not a terrible loss. It's always good to have an extra PSU on hand for testing or as the basis of the start for a secondary system. I usually recommend buying a PSU that is better than the last one you bought, as the warranties for most really good units has exceeded 10 years now. I wouldn't recommend buying any power supply for a high end system or gaming machine these days that doesn't at least have a Gold efficiency rating AND a 10 year warranty.

Those two things alone don't necessarily ensure that it will be a good unit, but it sure is nice to have them if it is and most units that meet both those criteria are going to be well built.

Hey, guys. I once again thank you for your help and I hope you don't mind me reviving this thread. I've decided to wait and build an all new rig all together while doing everything I can in fixing my current rig so I can have two. I've observed the black screens occur during GPU usage like while watching Youtube and playing games. I've also logged my system with HWINFO64 and found out that before the crash GPU Utilization was 100 before crashing but that's not all. Here are the things that spiked (Last two entries on the log):

GPU Utilization [%] - From 0% to 100%
GPU D3D Usage [%] - Ranging from 3.4 - 4.8 then spiked to 46.3
GPU Memory Errors [] - From 0 to 317848402 and 2525891584
GPU VRM Current In (IIN) [A] - Ranging from 1.4 - 2.1 then spiked to 6.1 - 6.3
GPU VRM Current Out (IOUT) [A] - Ranging from 21.5 - 23.5 then spiked to 53.5 - 54.5
GPU VRM Power Out (POUT) [W] - Ranging from 17 - 27 then spiked to 66.1 - 66.7
GPU VRM Power In (PIN) [W] - Ranging from 17.25 - 26.7 then spiked to 72.6 - 73.3

Those are the only parameters that spiked. Do these indicate anything? Is GPU the one with the problem? Or is the PSU causing the GPU to have a problem?

 

[Long Silog]

^Amen!

Hey, guys. I once again thank you for your help and I hope you don't mind me reviving this thread. I've decided to wait and build an all new rig all together while doing everything I can in fixing my current rig so I can have two. I've observed the black screens occur during GPU usage like while watching Youtube and playing games. I've also logged my system with HWINFO64 and found out that before the crash GPU Utilization was 100 before crashing but that's not all. Here are the things that spiked (Last two entries on the log):

GPU Utilization [%] - From 0% to 100%
GPU D3D Usage [%] - Ranging from 3.4 - 4.8 then spiked to 46.3
GPU Memory Errors [] - From 0 to 317848402 and 2525891584
GPU VRM Current In (IIN) [A] - Ranging from 1.4 - 2.1 then spiked to 6.1 - 6.3
GPU VRM Current Out (IOUT) [A] - Ranging from 21.5 - 23.5 then spiked to 53.5 - 54.5
GPU VRM Power Out (POUT) [W] - Ranging from 17 - 27 then spiked to 66.1 - 66.7
GPU VRM Power In (PIN) [W] - Ranging from 17.25 - 26.7 then spiked to 72.6 - 73.3

Those are the only parameters that spiked. Do these indicate anything? Is GPU the one with the problem? Or is the PSU causing the GPU to have a problem?

 

[Darkbreeze]

I think there's a really good probability that your card or your card drivers are to blame. But at this point I can't say. What were you doing when those spikes occured?

 

[Long Silog]

I think there's a really good probability that your card or your card drivers are to blame. But at this point I can't say. What were you doing when those spikes occured?

I was just browsing Youtube. When viewing videos on the web there was GPU Usage but if I watch downloaded videos there was none.

 

[Darkbreeze]

Try a different browser and see if it does the same thing at the same time or in a similar situation. I dunno. Sounds like a card issue but I guess it could also be software.

 

[Long Silog]

Try a different browser and see if it does the same thing at the same time or in a similar situation. I dunno. Sounds like a card issue but I guess it could also be software.

Tried with Chrome same thing happened. Another observation. I plugged the HDMI to the MOBO and streamed multiple Youtube HD videos with no error but when I started PUBG it crashed but no black screen. and when HDMI is plugged to GPU sometimes booting is interrupted. I think GPU usage makes it crash?

The log from HWINFO64 when HDMI was plugged in on MOBO showed no spikes but 100% GPU Utilization.

 

[Darkbreeze]

IDK man, sounds like a card problem to me.

 

[Long Silog]

IDK man, sounds like a card problem to me.

Yeah. Will just resort to changing one component at a time. You've been a great help darkbreeze. Thank you.

 

[Darkbreeze]

For sure. Let us know what you find out.

 

[Long Silog]

For sure. Let us know what you find out.

I'm back and rig is fixed. First replaced the PSU but to no avail. Returned old PSU and replaced GPU with a new one and gamed for a couple of days without any problems. It was the GPU, it was dying.

 

[Darkbreeze]

So then it was a card problem. Well, that always sucks, since those are not particularly cheap, but at least you got it sorted out and hopefully have a reliable power supply now. Good luck to you Long.