Question Hardware failure ?

[Runstopwalk]

GIGABYTE B760M DS3H AX LGA1700​

I-5 12600K​

PNY GeForce RTX™ 4070 12GB XLR8 Gaming Verto Epic-X RGB™ Triple Fan Graphics Card DLSS 3​

Crucial P5 Plus 2TB PCIe Gen4 3D NAND NVMe M.2 Gaming SSD, up to 6600MB/s - CT2000P5PSSD8, Solid State Drive​

Corsair Vengence 32GB 2x16 6400Mhz CL36 XMP​

Thermaltake Toughpower GF1 750W Snow Edition SLI/Crossfire Ready 80+ Gold Ultra​

Built this system, first set of RAM was DOA, put different RAM in, worked fine for 6 months. Computer would not boot again after 6 months, put in new RAM and it worked fine. Suddenly the computer started freezing if I left a game on pause for an hour or more. I would reboot and it would be all back to normal. Now it started an automatic repair loop, I format and installed windows 11 again, booted into windows and then after restart entered a repair loop, installed windows a third time, same issue. I swapped RAM around, did one stick and it would go into windows but as soon as I did one or two restarts it would enter a repair loop.

I cleared CMOS, and did a RAM reseat, and it would boot up into windows and as soon as restarting it would do a repair loop again. I removed the graphics card and it booted into windows fine, restarted 3x and no issue. I inserted the graphics card again and it worked fine for a few restarts and then entered the repair loop again. If I move the RAM around and install just one it will boot into windows one time and then have the same issue.

I have tried rolling back the BIOS to F3, clearing CMOS, reseating RAM and GPU. Every time I change something hardware wise it will load into windows at least one time. I tried a clean install after the BIOS roll back, did the DDU tool, put in older NVIDIA drivers and it seemed good and restarted several times. Went to power my PC back on this morning and back to the repair loop.

Considering this being the 3rd set of RAM, I am wondering if maybe I should swap the motherboard. I have a PSU tester arriving tonight as well. I am hoping its not the 1 year old GPU considering I did not realize PNY required you to register within 30 days or the 3 year warranty is not applicable... such BS, wont buy PNY again or skip registering an item with a manufacturer.

 

[Aeacus]

I have a PSU tester arriving tonight as well.

That thing is waste of money. Since all it does, is testing PSU without any load what-so-ever.

Also, PSU doesn't cause Win repair issue, UNLESS, PSU sends out of spec voltages to the storage drive, damaging it and corrupting files on there. But you can run drive checks and look if SSD is sound or not.

I am wondering if maybe I should swap the motherboard.

Since everything is all over the place, i'd also suggest new MoBo.

 

[Runstopwalk]

That thing is waste of money. Since all it does, is testing PSU without any load what-so-ever.

Also, PSU doesn't cause Win repair issue, UNLESS, PSU sends out of spec voltages to the storage drive, damaging it and corrupting files on there. But you can run drive checks and look if SSD is sound or not.


Since everything is all over the place, i'd also suggest new MoBo.

Thank you for the input, I am hoping it is just the board as it would be cheaper than say the GPU. Now to decide which LGA1700 board to get. I used to be a diehard ASUS guy back in the day but heard negative things, and not sure I want to get another Gigabyte LGA1700 board again with this one crapping the bed.

 

[Aeacus]

Now to decide which LGA1700 board to get.

With Asus and Gigabyte out of the picture, not much choice.

I'm personally using MSI MoBos and 0 issues with those thus far.

Other options are AsRock, EVGA (very solid but few options only), NZXT (more options than EVGA but still limited) and Biostar (i don't think you'd want Biostar ).

 

[Runstopwalk]

With Asus and Gigabyte out of the picture, not much choice.

I'm personally using MSI MoBos and 0 issues with those thus far.

Other options are AsRock, EVGA (very solid but few options only), NZXT (more options than EVGA but still limited) and Biostar (i don't think you'd want Biostar ).

Went with MSI, booted up and into windows and all looked well, then restarted and black screen unless I use the onboard graphics. I am thinking this card is toast or the PSU isnt putting enough power to the GPU.

I am going to do a clean install though and give it a go since I did switch motherboards with the same windows install... not sure it would help. The RGB lights are on with the 4070, but they also are on if the PSU is disconnected from the 4070. It shows up in hardware manager and has a Code 43 error.

Ugh...

 

[Aeacus]

The RGB lights are on with the 4070, but they also are on if the PSU is disconnected from the 4070.

PCI-E x16 slot provides up to 75W worth of power. So, GPU lights can remain lit even when PCI-E power cable is removed.

since I did switch motherboards with the same windows install... not sure it would help.

New MoBo = new clean Win install.

It shows up in hardware manager and has a Code 43 error.

Could be among these:
1. GPU drivers corruption.
2. GPU BIOS corruption.
3. Faulty GPU.

Now, you've essentially validated that it isn't GPU drivers (DDU + older versions). Next in line would be GPU BIOS.
From here you can get new BIOS, link: https://www.techpowerup.com/vgabios/256750/pny-rtx4070-12288-230313-1

So, try flashing the BIOS on GPU, even when current BIOS is the same version as on TechPowerUP. Since current one could be corrupt.
Once that is done and still no dice, it would be safe to assume the GPU is toast.

Moving forwards, you could either buy new GPU, or send the GPU into repairs.
E.g to Northridge Fix: https://northridgefix.com/

Alex, (founder and owner) has YT channel where he showcases repairs. I find it interesting to watch from time to time.
E.g RTX 3090 repair:

View: https://www.youtube.com/watch?v=kxnUAt0_YVE

 

[Aeacus]

Almost forgot:

Thermaltake Toughpower GF1 750W Snow Edition SLI/Crossfire Ready 80+ Gold Ultra

Tt Toughpower GF1 is low quality unit, borderline crap. So, could be that it was the PSU which killed your MoBo and damaged your GPU as well.

Here, i'd go with new, good quality unit. Good PSUs to go for, are: Seasonic Focus/Vertex/PRIME, Corsair RMx/RMi/HXi/AXi, Super Flower Leadex Gold/Platinum/Titanium, in 650W range.

For 2nd opinion about your PSU, look it up from PSU tier list,
link: https://forums.tomshardware.com/thr...er-list-rev-14-8-final-update-jul-21.3624094/

You'll find it under Tier C, low priority units. While what i'm personally using and also suggested, are all Tier A.

 

[Runstopwalk]

Almost forgot:


Tt Toughpower GF1 is low quality unit, borderline crap. So, could be that it was the PSU which killed your MoBo and damaged your GPU as well.

Here, i'd go with new, good quality unit. Good PSUs to go for, are: Seasonic Focus/Vertex/PRIME, Corsair RMx/RMi/HXi/AXi, Super Flower Leadex Gold/Platinum/Titanium, in 650W range.

For 2nd opinion about your PSU, look it up from PSU tier list,
link: https://forums.tomshardware.com/thr...er-list-rev-14-8-final-update-jul-21.3624094/

You'll find it under Tier C, low priority units. While what i'm personally using and also suggested, are all Tier A.

Thank you for all of the great information! I did a clean install and so far things are up and running for 12 hours with several restarts.

Looks like the new mobo and a fresh install and chipset drivers corrected the issue. I did not realize that my PSU was a crap brand, I should have done more research. I am going to order a new power supply to be on the safe side and not risk any future damage to the mobo or GPU.

I appreciate your help and all the great information!

 

[Runstopwalk]

Almost forgot:


Tt Toughpower GF1 is low quality unit, borderline crap. So, could be that it was the PSU which killed your MoBo and damaged your GPU as well.

Here, i'd go with new, good quality unit. Good PSUs to go for, are: Seasonic Focus/Vertex/PRIME, Corsair RMx/RMi/HXi/AXi, Super Flower Leadex Gold/Platinum/Titanium, in 650W range.

For 2nd opinion about your PSU, look it up from PSU tier list,
link: https://forums.tomshardware.com/thr...er-list-rev-14-8-final-update-jul-21.3624094/

You'll find it under Tier C, low priority units. While what i'm personally using and also suggested, are all Tier A.

Ooof, I spoke too soon. I just used Massgrave Dev to activate windows, restarted and the external GPU went blank. I can get into BIOS but had to switch to onboard GPU to see windows.

 

[Runstopwalk]

Almost forgot:


Tt Toughpower GF1 is low quality unit, borderline crap. So, could be that it was the PSU which killed your MoBo and damaged your GPU as well.

Here, i'd go with new, good quality unit. Good PSUs to go for, are: Seasonic Focus/Vertex/PRIME, Corsair RMx/RMi/HXi/AXi, Super Flower Leadex Gold/Platinum/Titanium, in 650W range.

For 2nd opinion about your PSU, look it up from PSU tier list,
link: https://forums.tomshardware.com/thr...er-list-rev-14-8-final-update-jul-21.3624094/

You'll find it under Tier C, low priority units. While what i'm personally using and also suggested, are all Tier A.

So I have three desktops, I keep my old gaming computers. I just turned on an old tower next to my new one and ran the 8 pin from the desktop next to it to my GPU and now it works having the GPU using the boot leg power from the other PSU. Going to order the PSU now.

EDIT: Yeah, this worked until restart. Put the original PSU plug back in, got a blue screen, reboot and back into windows with the RTX 4070. So frustrating. It rebooted several times until I activated Windows with MassGrave Dev. I am starting to wonder if that's what's going on, every clean install works until I get comfortable and decide to activate. I just feel like that must be coincidence.

Edit 2: Completed the GPU flash you suggested and so far things seem stable. Ran benchmark software for a while on max settings, and restarted, shut down for a while, powered back on and so far so good...

 

[Aeacus]

I did not realize that my PSU was a crap brand, I should have done more research.

There are select few Thermaltake PSUs that are good, but you need to know which beforehand. Also, the good ones from Tt cost a lot. So, overall, best to avoid Thermaltake completely.

And yes, PSU research is important since PSU powers everything and thus, is the most important component inside the PC.
The lower the PSU build quality is - the higher the chance of PSU developing a fault and/or frying everything it is connected to (aka your whole PC).

Edit 2: Completed the GPU flash you suggested and so far things seem stable. Ran benchmark software for a while on max settings, and restarted, shut down for a while, powered back on and so far so good...

Do get new, good/great quality PSU as well. Since your Tt GF1 already killed one MoBo and probably corrupted the BIOS on GPU as well.

If you want white PSU, then there are few good ones to pick from, from Super Flower
pcpp: https://pcpartpicker.com/products/compare/m7t9TW,8VWJ7P/

And some from Seasonic as well, like;
Focus GX White: https://seasonic.com/focus-gx-white-atx-3/
Vertex GX White: https://seasonic.com/vertex-gx-white/
Vertex Sakura: https://seasonic.com/vertex-sakura/

These white Seasonic units were just released, so many not be on the stores yet. And Vertex Sakura is special edition. Could be already sold out.

 

[Runstopwalk]

There are select few Thermaltake PSUs that are good, but you need to know which beforehand. Also, the good ones from Tt cost a lot. So, overall, best to avoid Thermaltake completely.

And yes, PSU research is important since PSU powers everything and thus, is the most important component inside the PC.
The lower the PSU build quality is - the higher the chance of PSU developing a fault and/or frying everything it is connected to (aka your whole PC).


Do get new, good/great quality PSU as well. Since your Tt GF1 already killed one MoBo and probably corrupted the BIOS on GPU as well.

If you want white PSU, then there are few good ones to pick from, from Super Flower
pcpp: https://pcpartpicker.com/products/compare/m7t9TW,8VWJ7P/

And some from Seasonic as well, like;
Focus GX White: https://seasonic.com/focus-gx-white-atx-3/
Vertex GX White: https://seasonic.com/vertex-gx-white/
Vertex Sakura: https://seasonic.com/vertex-sakura/

These white Seasonic units were just released, so many not be on the stores yet. And Vertex Sakura is special edition. Could be already sold out.

So I istalled a Seasonic PSU and a different SSD... still the same issues. Whats odd and this sounds crazy, is if I hear windows boot up and the screen is blank... if I unplug a USB component and plug back in the output turns on and I am at the Windows 11 log in screen. I almost want to record this as it sounds crazy. Basically I have built a whole new PC around this GPU and this is happening... so frustrating.

 

[Aeacus]

So I istalled a Seasonic PSU

I wonder, which model?

Whats odd and this sounds crazy, is if I hear windows boot up and the screen is blank... if I unplug a USB component and plug back in the output turns on and I am at the Windows 11 log in screen.

I have no experience with Win11 (using Win10 myself) and can't tell what is going on in Win11 boot-up process, that it is unable to detect and display image, until you replug the USB component.

Btw, which USB component? Since image is shown via monitor cable, either DP or HDMI. Do you unplug/replug the display cable to make the image show? Or is it really some other USB device (e.g KB/headset/printer)?

Let's make a recap;
The issues you initially saw, are these all gone? Like: restarts, OS repair loops, GPU code 43 error?
Whereby what is left, is the failure to display image during OS boot up?

 

[Runstopwalk]

I wonder, which model?


I have no experience with Win11 (using Win10 myself) and can't tell what is going on in Win11 boot-up process, that it is unable to detect and display image, until you replug the USB component.

Btw, which USB component? Since image is shown via monitor cable, either DP or HDMI. Do you unplug/replug the display cable to make the image show? Or is it really some other USB device (e.g KB/headset/printer)?

Let's make a recap;
The issues you initially saw, are these all gone? Like: restarts, OS repair loops, GPU code 43 error?
Whereby what is left, is the failure to display image during OS boot up?

Seasonic FOCUS PX-750 | 750W | 80+ Platinum is the one I installed.​

I am thinking the USB was coincidence, possibly. It was just the keyboard USB when it did it a few times. I did it right when I heard windows startup, I am thinking maybe it was coincidence or the graphics card was just delayed to start showing an image.

The random restarts and repair loops are gone. The only issue now is that when turning the computer on it will show the MSI BIOS screen and then the monitor will shutoff and I will hear the windows 11 login screen sound play. Sometimes the display will turn back on after so many seconds or when switching to and from onboard to discrete GPU.

I tried installing Windows 10 Pro and the desktop loaded in. Not sure if random but when I installed the MSI B760 chipset the screen immediately shutoff after install.

Before going to work I was in the BIOS for the new board and also noticed there were settings to install drivers upon OS install automatically, I disabled that and started a fresh format and install and left for work.

I am starting to really think the GPU just crapped out. Its odd though because during all of this once I replaced the Mobo, RAM, SSD, PSU and the restarts and blue screen stopped if I get into Windows I can play games or run benchmark no issue, but when restarting 90% of the time the discrete GPU wont display after BIOS loads the OS. The odd thing is sometimes the onboard GPU will not display when switching back to it. When it does, the RTX 4070 shows the code 43. I was able to click and disable the RTX 4070 and then re-enable it and successfully plug back into the RTX 4070 and it did display correctly.

The only thing I have not replaced is the Intel 12600K processor, the GPU, and the power/reset button.

I have a spare AMD system from a few years ago I might try and take out of the case and make a test bench essentially and plug the RTX 4070 into it and see what happens to rule out the processor / solidify the GPU as being bad. I would have done it sooner but the case is too small so I will have to take the motherboard out of the case for that setup.

This is frustrating as I have built several PCs and never had this happen.

Edit: PNY said that they will honor the 3 year warranty, no need to register or have the original packaging. I think I will just send it in tomorrow if it does not work on my AMD setup.

 

[Aeacus]

Seasonic FOCUS PX-750 | 750W | 80+ Platinum is the one I installed.

Solid PSU. Even i have it, powering my old AMD build but mine is 550W unit.

The random restarts and repair loops are gone.

At least that much is good.

The only issue now is that when turning the computer on it will show the MSI BIOS screen and then the monitor will shutoff and I will hear the windows 11 login screen sound play. Sometimes the display will turn back on after so many seconds or when switching to and from onboard to discrete GPU.

It really sounds to be Win11 issue, since it somehow delays activation of the GPU and/or, in turn, activation of monitor to see the image.

I don't know enough about Win11 to tell what is going on. But there are those who might know, so, i can ask around and perhaps even summon additional help here.

I tried installing Windows 10 Pro and the desktop loaded in. Not sure if random but when I installed the MSI B760 chipset the screen immediately shutoff after install.

With Win10 and after chipset drivers were installed, did you try reboot and see if image came back?
Since with some driver installation, especially GPU, it is normal that video feed stops for a while. Chipset driver installation is rare, only done once in PC's lifespan, just after assembly. I did mine 8 years ago and i really can't remember if my monitor also lost video signal during chipset drivers installation or not.

if I get into Windows I can play games or run benchmark no issue, but when restarting 90% of the time the discrete GPU wont display after BIOS loads the OS.

It really sounds like Win11 issue. I better summon some help, since i'm specialized on hardware issues myself and doesn't do that well with software issues.

The only thing I have not replaced is the Intel 12600K processor, the GPU, and the power/reset button.

If the CPU would be at fault, you may not even be able to boot into OS. Or if you are able, you'd be getting BSoD left and right.

Could be GPU issue, but you flashed it's BIOS and it went back to normal, right? Also, benchmarking GPU would then fail (e.g system freeze), if it would be solely GPU issue.

Power/reset buttons have 0 to do with delayed image display. These are just like on/off switches, bridging the cap between two pins on MoBo, to force the MoBo to do specific thing.

 

[Aeacus]

Asked around a bit and:
New MoBo (MSI) model is?

Also, from BIOS, which of the GPU is primary? Dedicated one (RTX 4070) or iGPU?

And i also got this response:

The symptom sounds a bit like failure to detect the added video card and fall-back to mobo system.
MIGHT also be a problem with the time delay that video card has before it completes its own POST and can reply to queries from mobo POST process. Any way to slow down the mobo POST to allow slower response time there?

Without knowing MoBo model, i can't look up what BIOS options it has, to perhaps slow down the POST process.

Still, i, personally, think that something goes wrong with Win11 initial boot-up and there is some kind of failure to detect the GPU. As of if it is OS or GPU fault - hard to tell.

So, validate the RTX 4070 in 2nd system, if you can.

Oh, while normal OS boot has no image issue, but how about when booting into Safe Mode? Since there, none of the GPU drivers are loaded, instead Win own display drivers are used. It would show if the issue would be with 3rd party programs (including GPU drivers), OR when same "no image" issue also happens when booting into Safe Mode, it comes down to either Win11 issue, or GPU hardware issue.

 

[Aeacus]

Since some time has moved forwards, i got more responses to help out your issue.
One of them was to "hard reset" the BIOS, since it might help and doesn't hurt to try.

Guide here: https://forums.tomshardware.com/thr...s-freezing-not-posting.3840554/#post-23230723

 

[Colif]

Ooof, I spoke too soon. I just used Massgrave Dev to activate windows, restarted and the external GPU went blank. I can get into BIOS but had to switch to onboard GPU to see windows.

Did you have a desktop image on 11 until you activated it? that massgrave site is... interesting.
You might want to clean install 11 and not use that site, unless it is your license you used... it might deactivate the external GPU or something odd.

Since GPU works in 10, it should work in 11. There aren't many differences between GPU drivers on 10 or 11. So problem is windows itself.

 

[Runstopwalk]

Since some time has moved forwards, i got more responses to help out your issue.
One of them was to "hard reset" the BIOS, since it might help and doesn't hurt to try.

Guide here: https://forums.tomshardware.com/thr...s-freezing-not-posting.3840554/#post-23230723

Did you have a desktop image on 11 until you activated it? that massgrave site is... interesting.
You might want to clean install 11 and not use that site, unless it is your license you used... it might deactivate the external GPU or something odd.

Since GPU works in 10, it should work in 11. There aren't many differences between GPU drivers on 10 or 11. So problem is windows itself.

Sorry for the delay, I ended up trying this card on 3 different systems containing Windows 10 and Windows 11. I used DDU and installed multiple Nvidia drivers from the oldest available of January 2024 and onward. The issues persisted with all three systems with the PNY RTX 4070.

The last attempt of a fresh install on an AMD system instead of my primary Intel system booted into windows but as soon as the driver was changed from a generic Microsoft display driver to an Nvidia driver the memory clock speed of the RTX 4070 went from 100Mhz to bouncing between 100-11000Mhz.

As soon as any Nvidia driver was installed I started getting artifacts, and stutter. I tried running Furmark and it was stuttering bad.

PNY told me the card is definitely bad and nothing more can be done and emailed me RMA instructions. Also, the original system with the Windows 11 activated by Massgrave worked with zero issues once I installed an XFX card I had laying around and performed DDU / fresh drivers....

I have never had these issues before and the 4070 ran for months with no issues until all of this, so I am suspecting it must be the card. Thank you for all the troubleshooting info, I will update once the RMA is complete.

Edit: I think all the random USB unplugging, and it coming on, and whatnot must have all been coincidence and made me think I was getting somewhere. I spent 6 hours this morning troubleshooting with my three systems and this card.

 

[Runstopwalk]

Asked around a bit and:
New MoBo (MSI) model is?

Also, from BIOS, which of the GPU is primary? Dedicated one (RTX 4070) or iGPU?

And i also got this response:

Without knowing MoBo model, i can't look up what BIOS options it has, to perhaps slow down the POST process.

Still, i, personally, think that something goes wrong with Win11 initial boot-up and there is some kind of failure to detect the GPU. As of if it is OS or GPU fault - hard to tell.

So, validate the RTX 4070 in 2nd system, if you can.

Oh, while normal OS boot has no image issue, but how about when booting into Safe Mode? Since there, none of the GPU drivers are loaded, instead Win own display drivers are used. It would show if the issue would be with 3rd party programs (including GPU drivers), OR when same "no image" issue also happens when booting into Safe Mode, it comes down to either Win11 issue, or GPU hardware issue.

I tried multiple versions of BIOS on my primary build, even a Beta this morning and no dice. I also tried disabling the onboard GPU, disabling power saving features on the PCIe slot in BIOS and nothing helped. Thank you for so many troubleshooting tips. I feel like I have exhausted all possibilities. The three different systems had three different power supply brands, all more than enough wattage, and tried on AMD and Intel systems.

 

[Aeacus]

Thank you for so many troubleshooting tips. I feel like I have exhausted all possibilities. The three different systems had three different power supply brands, all more than enough wattage, and tried on AMD and Intel systems.

Testing the GPU in 2nd system suffices, whereby testing it on 3rd system would be just excess.

As soon as any Nvidia driver was installed I started getting artifacts, and stutter. I tried running Furmark and it was stuttering bad.

It would be safe to say that GPU has a fault.

As of why it was fine until then, it could be due to your old Tt PSU.
When PSU sends out of spec voltages to the hardware, it does damage and degrade them, until hardware faults develop. And when continuing it, hardware will eventually die, far faster that it would normally. And low/crap quality PSUs are prone on sending out of spec voltages to the hardware.

In a similar sense, if you eat contaminated food, you'll get stomach aches and you may even throw up. Keep eating contaminated food and you'll end up 6 feet under.

Hence why i said that the PSU is the most important component inside the PC.

wont buy PNY again

As of Nvidia GPU brands; EVGA is out of business (sadly), leaving the big 3: Gigabyte, Asus and MSI. Then, there are newcomers as well, like: Zotac, PNY, Colorful, Galax, Palit etc.

I, personally, am using only MSI GPUs (mostly known for quietest operation and best cooling regarding GPUs). And i don't trust the newcomers, since they usually have cheaper GPUs to offer than the big 4 (now the big 3). And with cheaper price, quality is lower. Zotac, maybe, would be something i'd consider outside of the big 3. But Colorful is the worst of them.

 

[Colif]

Sorry for the delay, I ended up trying this card on 3 different systems containing Windows 10 and Windows 11. I used DDU and installed multiple Nvidia drivers from the oldest available of January 2024 and onward. The issues persisted with all three systems with the PNY RTX 4070.

i did suggest it might be card to Aeacus so that works. Often when you swap parts, something that has worked fine before suddenly stops. Its happened to me a few times. Could just be the change in power draw in system that does it.

 

[Runstopwalk]

Testing the GPU in 2nd system suffices, whereby testing it on 3rd system would be just excess.


It would be safe to say that GPU has a fault.

As of why it was fine until then, it could be due to your old Tt PSU.
When PSU sends out of spec voltages to the hardware, it does damage and degrade them, until hardware faults develop. And when continuing it, hardware will eventually die, far faster that it would normally. And low/crap quality PSUs are prone on sending out of spec voltages to the hardware.

In a similar sense, if you eat contaminated food, you'll get stomach aches and you may even throw up. Keep eating contaminated food and you'll end up 6 feet under.

Hence why i said that the PSU is the most important component inside the PC.


As of Nvidia GPU brands; EVGA is out of business (sadly), leaving the big 3: Gigabyte, Asus and MSI. Then, there are newcomers as well, like: Zotac, PNY, Colorful, Galax, Palit etc.

I, personally, am using only MSI GPUs (mostly known for quietest operation and best cooling regarding GPUs). And i don't trust the newcomers, since they usually have cheaper GPUs to offer than the big 4 (now the big 3). And with cheaper price, quality is lower. Zotac, maybe, would be something i'd consider outside of the big 3. But Colorful is the worst of them.

Well, PNY refused a RMA - they said everything tested good and sent the card back. I feel like after I tried the card on several systems there is NO way they did a good test.

Since they refused to replace the card I went ahead and opened it up and the thermal paste was almost gone from the chip, and what little was left was hard. I went ahead and applied new thermal paste and adjusted the fan curve and ran 3D Mark and I received a score of 3.998 vs the average of the card being around 3.7. So far after the paste and adjustments no issues. If I do leave the fan curve to default the temp does start to raise quickly vs leaving the fan set to a minimum of 40%.

 

[Runstopwalk]

Also, this all started after leaving a game running overnight and the card was on a vertical mount. I wonder if cheap paste, plus being sideways in a sense and running super hot led all the paste to run off the chip, either that or the factory did not do a good job.

 

[ohio_buckeye]

This may help.

https://a.co/d/cmeL0Kc

Also on the massgrave software you mentioned I would say not to use that. Sounds like that software is basically helping software piracy.