my GPU driver keeps crashing on a freshly built pc making my life miserable

[apyrexia]

Specs:
CPU 7800X3D
Board - MSI b650-p wifi
GPU XFX MERC 7900 XTX
Ram DDR5 G.skill 6400 cl32 (16x2)gb
I am using a Single monitor - ASUS PG279Q with 165Hz.

Ever since i built my pc ( 1 month ago - in the beginning of May 23) - i have been experiencing - Random crashes, My PC Just freezes, My monitor is doing one blink - and screen goes completely black. sometimes it will recover but other times i have to reset my PC.
This is happening not only while gaming BUT it does also happen on regular desktop use like browsing the internet (Firefox) and when idle on desktop.
I have been taking many steps trying to solve the issue myself and nothing worked out, my PC keep crashing:
1. Lowering HZ to 144.
2. Installing minimal driver "Driver only" without the Adrenaline software GUI.
3. Disabling the Internal GPU of my CPU.
4. Updating the BIOS, Using only stock RAM speeds (NO EXPO, NO XMP), NO Overclockingat all - NO PBO. Loaded Optimized Defaults settings for BIOS.
5. Been Monitoring my temps all the time - no overheating.
6. Installing latest chipset drivers.
7. Disabling MPO.
8. Using vaious older driver versions, even went back to 22.12.2 - it was still crashing.
9. Complete Fresh install of windows 10 and all my games and applications.
10. Benchmarks like Superposition / 3dmark / Passmark / cinebench r23 - most of the time benchmarks were running without crashing. (results i had are within normal range for other PCs with similar parts).
11. AMD DRIVER 23.5.1 was released and it didnt fix the issue.
12. on June 1st 2023 AMD Released driver 23.5.2 - That didnt fix my issue.
13. June 3rd - i have installed windows 11, latest AMD chipset + Instead of the normal Adrenaline drivers i install the AMD Pro software (should be more stable) - That didnt help!
14. I tried running my application as administrator + Disabled Hardware acceleration or any Game Overlays on my applications.
15. I tried to tune down the GPU power slider in adrenaline software to -10% and limit the Max clock to 2200mhz. (thinking this might lower the power draw in-case it is my PSU failing? )

Non of these steps fixed my issue.
Today for eg after the win11 install, i was alt-tabbing between Discord, firefox and world of warcraft - Then i had a crash again. and Discord has closed itself. and the Radeon software recovered itself but would not open unless i restarted the PC.
The windows Event Viewer didnt catch that crash somehow.

However i went on and kept looking and found this error (which oddly happened about 50minutes before the crash) :
Faulting application name: atieclxx.exe, version: 31.0.14000.58004, time stamp: 0x638a71ed
Faulting module name: amdsasrv64.dll, version: 1.0.0.0, time stamp: 0x638a7292
Exception code: 0xc0000005
Fault offset: 0x000000000005445b
Faulting process id: 0x0xD70
Faulting application start time: 0x0x1D99622FF5DE7E1
Faulting application path: C:\Windows\System32\DriverStore\FileRepository\u0386350.inf_amd64_cf354da277d39d93\B386336\atieclxx.exe
Faulting module path: C:\Windows\System32\DriverStore\FileRepository\u0386350.inf_amd64_cf354da277d39d93\B386336\amdsasrv64.dll
Report Id: 3648cc2d-2998-4ca2-bd45-2ebcca6f946a
Faulting package full name:
Faulting package-relative application ID:

To say the least- i am super frustrated, i have started the RMA process for both the GPU and the CPU - but i will have to suffer about 200-300$ restocking fee and of course this is going to shut me down for a month or two before i can afford replacement parts.

On top of that - what if im running in circles, looking in the wrong places and it ends up being a PSU issue? i have no idea, i made a post somewhere else and someone in the comments suggested that.

If you read through all of that - youre an angel.
I want this problem gone, feel free to post any suggestion, thank you!

 

[eidairaman1]

And no power supply...

You ever think to update the motherboard bios first?

@XFXSupport

 

[apyrexia]

And no power supply...

You ever think to update the motherboard bios first?

@XFXSupport

Sorry my PSU is new out of the box Cooler Master v1000.
The motherboard bios is updated to the lastest bios from MSI released 2023-05-12.
version: 7D78v153
Description:
- AGESA ComboPI 1.0.0.7a updated with the required solution.

 

[eidairaman1]

Sorry my PSU is new out of the box Cooler Master v1000.
The motherboard bios is updated to the lastest bios from MSI released 2023-05-12.
version: 7D78v153
Description:
- AGESA ComboPI 1.0.0.7a updated with the required solution.

So you just bought the V1000 correct, from where?

This review is from 2013

Cooler Master V Series 1000 W Review

Cooler Master teamed up with Seasonic to make a strong entry into the high-end category where it faces competition using the same OEM (Seasonic). The new V series consists of three models. Today, we will test the strongest one featuring a capacity of 1 kW, a fully modular cabling design, Gold...

www.techpowerup.com

Take a pic of the psus sticker label please for the specific model number.

 

[apyrexia]

So you just bought the V1000 correct, from where?

This review is from 2013

Cooler Master V Series 1000 W Review

Cooler Master teamed up with Seasonic to make a strong entry into the high-end category where it faces competition using the same OEM (Seasonic). The new V series consists of three models. Today, we will test the strongest one featuring a capacity of 1 kW, a fully modular cabling design, Gold...

www.techpowerup.com

That is correct, bought it locally from a local Store where i live. (you wouldnt recognize even if i name it).
They are still selling them. do you think its faulty ?.

I will update once i got in hand the PSU sticker label. - do i find it physically on the PSU itself? i might have to disassemble the psu before i can see it

 

[eidairaman1]

That is correct, bought it locally from a local Store where i live. (you wouldnt recognize even if i name it).
They are still selling them. do you think its faulty ?.

I will update once i got in hand the PSU sticker label. - do i find it physically on the PSU itself? i might have to disassemble the psu before i can see it

It's on the power supply, unplug the main power cord and remove the mounting screws to look for it (may have to move it around, or look on the box for any sticker labels/barcodes

 

[apyrexia]

 

[eidairaman1]

View attachment 299078

Most likely it is Enhance Electronics based, since the 2013 Model was Seasonic (which was Good)

Put Windows into safe mode, remove the gpu driver and boot back up to Windows normally, get hwinfo and post a screenshot of power supply volts to include the power/volts going to the gpu

What speed is the system ram running at?

 

[64K]

Even the best of the best PSUs can still be faulty right out of the box brand new. Just to eliminate one possibility do you have another known working PSU that you could swap in and see if the problems go away?

 

[eidairaman1]

Even the best of the best PSUs can still be faulty right out of the box brand new. Just to eliminate one possibility do you have another known working PSU that you could swap in and see if the problems go away?

Yup diagnostics tree, without good clean power a pc will have faults, even the source/distribution has to be pretty clean/steady

 

[apyrexia]

Most likely it is Enhance Electronics based, since the 2013 Model was Seasonic (which was Good)

Put Windows into safe mode, remove the gpu driver and boot back up to Windows normally, get hwinfo and post a screenshot of power supply volts to include

I have a new in box Antec 750 High current gamer 80+ gold.
I will post soon the screenshot you asked (hwinfo).
Let me know after if you think i should try the 750w psu. (It feel it is too little for my specs?)

 

[3x0]

Exception code: 0xc0000005 is usually related to bad RAM, regardless of EXPO on or off. Download:
1. https://www.memtest86.com/memtest86.html Create a bootable USB and run for the default 4 cycles at least.
2. TestMem5 from attachment, run it, click load config and exit and then select the 1usmus config in bin folder, run the program with admin rights again and let it run for the 25 cycles.
Virustotal scan of the executable https://www.virustotal.com/gui/file...68a4e863fe078b5c54857a16d37d480cc21?nocache=1
Report back

 

[eidairaman1]

I have a new in box Antec 750 High current gamer 80+ gold.
I will post soon the screenshot you asked (hwinfo).
Let me know after if you think i should try the 750w psu. (It feel it is too little for my specs?)

Whats the sticker label from the high current gamer?

Exception code: 0xc0000005 is usually related to bad RAM, regardless of EXPO on or off. Download:
1. https://www.memtest86.com/memtest86.html Create a bootable USB and run for the default 4 cycles at least.
2. TestMem5 from attachment, run it, click load config and exit and then select the 1usmus config in bin folder, run the program with admin rights again and let it run for the 25 cycles.
Virustotal scan of the executable https://www.virustotal.com/gui/file...68a4e863fe078b5c54857a16d37d480cc21?nocache=1
Report back

And thats another thought I was having because it can cause driver faults even.

 

[QuietBob]

I am using a Single monitor - ASUS PG279Q with 165Hz

How is your display connected? Are you using a quality brand-name HDMI 2.1 cable?

 

[eidairaman1]

 

[apyrexia]

Ok i love just how everyone trying to help me
Ty so much.

@eidairaman1
1.my ram is running 4800mhz atm.
2. I have remoed the gpu driver in safe mode using DDU. I couldnt find where it shows the volts going to the GPU - i checked the sensors page and it wasnt there or anywhere else (sorry im new to hwinfo)

@3x0 thank you i will report back.

@QuietBob I am conncting my GPU using the same DP cable i had for long time (it came with the monitor if my memory still good)


*** Also addng the HCG label

 

[eidairaman1]

Ok i love just how everyone trying to help me
Ty so much.

@eidairaman1
1.my ram is running 4800mhz atm.
2. I have remoed the gpu driver in safe mode using DDU. I couldnt find where it shows the volts going to the GPU - i checked the sensors page and it wasnt there or anywhere else (sorry im new to hwinfo)

@3x0 thank you i will report back.

Load up sensors only on hwinfo64

 

[Mister300]

Had an issue with my PC randomly crashing and going into no signal mode power lost, did all HW swap and a bunch of other things. Care to guess what fixed it?, the NZXT AIO software was the issue, it was not a HW or GPU failure. A Redditt user made a bat file to shutdown the offending NZXT CAM driver which cured it. Most issues are SW related not HW failure. Does event viewer show any critical errors?, my code was 41.

 

[GerKNG]

ATX 2.31 PSUs are really old and miss a lot of features.
idle crashes like that "can" be "fixed" with changing the PSU Idle Control from (Auto) Low to Typical in the bios (somewhere around the AMD CBS/OC Settings)

 

[apyrexia]

Load up sensors only on hwinfo64

excuse me but it is not there?

 

[eidairaman1]

Click on the First msi pro b650-p

 

[apyrexia]

Click on the First msi pro b650-p

here it is

 

[tabascosauz]

here it is

With the 7900XT I had to replace PSU and/or DP cable to stop the crashing. Unfortunately I didn't have the opportunity to test what exactly the culprit was between the two. It's worth a shot if you have access to some retailer/etailer with a return policy, without restocking fees.

Yes, in theory neither should pose any problems, but you never know if you don't try. Anecdotally the Navi31 cards have been very picky as to DP cables quality.

My crashes were on older Adrenalin but symptoms similar, no real logs in Event Viewer, and Adrenalin failed to recover itself without a restart.

 

[eidairaman1]

Ok i love just how everyone trying to help me
Ty so much.

@eidairaman1
1.my ram is running 4800mhz atm.
2. I have remoed the gpu driver in safe mode using DDU. I couldnt find where it shows the volts going to the GPU - i checked the sensors page and it wasnt there or anywhere else (sorry im new to hwinfo)

@3x0 thank you i will report back.

@QuietBob I am conncting my GPU using the same DP cable i had for long time (it came with the monitor if my memory still good)


*** Also addng the HCG label

Try this psu and if you still have an issue it might be the Main System Ram

With the 7900XT I had to replace PSU and/or DP cable to stop the crashing. Unfortunately I didn't have the opportunity to test what exactly the culprit was between the two. It's worth a shot if you have access to some retailer/etailer with a return policy, without restocking fees.

Yes, in theory neither should pose any problems, but you never know if you don't try. Anecdotally the Navi31 cards have been very picky as to DP cables quality.

My crashes were on older Adrenalin but symptoms similar, no real logs in Event Viewer, and Adrenalin failed to recover itself without a restart.

750W is the minimum, he can try it but the CM 1000V volts are in range

 

[Mister300]

With the 7900XT I had to replace PSU and/or DP cable to stop the crashing. Unfortunately I didn't have the opportunity to test what exactly the culprit was between the two. It's worth a shot if you have access to some retailer/etailer with a return policy, without restocking fees.

Yes, in theory neither should pose any problems, but you never know if you don't try. Anecdotally the Navi31 cards have been very picky as to DP cables quality.

My crashes were on older Adrenalin but symptoms similar, no real logs in Event Viewer, and Adrenalin failed to recover itself without a restart.

DP is a poorly engineered connection, inflexible cable and poor contact with female plug/jack is a design fubar. Remember the old VGA screws, this is what DP needs.