Question Seemingly random black screens, possibly CPU, PSU or mobo ?

[mdva_dank]

hi

So randomly as of about 2 months ago ive been experiencing weird blackscreen crashes when doing random things on my pc E.x.(gaming, watching youtube, idling, downloading)
gpu died previously and it blackscreened but got it rma'd and didnt have a single issue for a couple of weeks until i started blackscreening again( but its different this time?) previously when it would crash i would get a completely unresponsive pc and eventually fried my card, this time however it is able to be interacted with, albeit without any type of signal regardless of what cable i use or how many monitors i have plugged in.

I have tried memtest with no issues, completely reinstalled windows on a brand new ssd, cleaned drivers, removed all drives except boot drive, reset bios/cleared cmos, turned off PBO, turned off D.O.C.P, tried 1, 2 , and 4 sticks of ram, tried gpu in another system on a 650 watt psu(ran fine) replaced gpu with 3070 temporarily and passed multiple stress tests(as has the 3090), ran OCCT, found thousands of power errors (only sometimes other times its fine) run benchmarks and reseated/ tried different cables with the psu, so i am at a complete loss now as it seems to happen within a timespan of 5min to 50+ hrs of being on and i have no clue as to what could be causing this at all.

At this point, any advice or help would be appreciated. I will say the voltages on the different rails seem rather on the low end in general and when i had originally purchased it(about 3 months ago) it had terrible coil whine that has seemed to go away somehow at this point. again any help would ba appreciated because at this point of troubleshooting for over 2 months.
I'm at my wits end and have no clue what to do at this point lol.

My specs are as followed (also it seems more likely to crash with pbo enabled in the bios) (bios, vbios, and chipset drivers are all up to date for mobo gpu and cpu.)

CPU: R9 5900x
Motherboard: ASUS PRIME x570 pro
Ram: 32gb(8x4) cl14 ddr4 trident z royal ram running at 3600mhz
SSD/HDD: 2tb samsung 860 pro(BOOT), crucial mx500 1tb m.2, inland premium 2tb m.2, t force rgb ssd 2tb, 8tb segate hdd, 4tb segate hdd
GPU: gigabyte AORUS master rtx 3090 rev 1.0
PSU: corsair rm1000x
Chassis: lian li 011 dynamic xl
OS: windows 10 20H2 update KB4023057

i am also running 3 monitors and numerous internal and external perifs, if that information is pertinent let me know and ill update with that info,

again huge thanks to anyone who can offer any advice/ info thank you in advance.

 

[mc0]

Hi,

I'm actually in the same boat (was about to make a new post, then found yours). My machine will give me a black screen or hard reboot, both seemingly random. The only consistency that I have found is that if I manually reboot about once a week I don't get these crashes. Also, once it does start crashing, it will do so more frequently until I turn off the machine and leave it off for a while (few hours). I have replaced my motherboard and PSU, but the issue still persists. The similarity I see between our two builds is the memory. Out of curiosity, did you buy two separate 16GB packs or one 32GB pack?

CPU	AMD Ryzen 7 2700X
RAM	G.SKILL TridentZ RGB Series 32GB (4 x 8GB) 288-Pin DDR4 SDRAM DDR4 3200 (PC4 25600) Intel XMP 2.0
Motherboard	ASUSTeK COMPUTER INC. ROG STRIX B450-F GAMING (AM4)
GPU	8176MB ATI AMD Radeon RX 5700 XT (ATI)
Storage	931GB TOSHIBA MQ01ABD100 (SATA )
	298GB Western Digital WDC WD3200AAJS-22B4A0 (SATA )
	232GB Samsung SSD 850 EVO 250GB (SATA (SSD))
	931GB Samsung SSD 970 EVO Plus 1TB (Unknown (SSD))
PSU	Corsair RM 750

 

[--SID--]

Do you use 2 seperate cables from psu to gpu?

 

[Jacozeelie]

Do you have a working psu that can be used in the pc. It might be a power problem. Faulty psu cause many gpu related problems

 

[mc0]

Do you have a working psu that can be used in the pc. It might be a power problem. Faulty psu cause many gpu related problems

I thought the same, which is why I replaced my PSU. The problem is still here, unfortunately. 😕

 

[Jacozeelie]

Well, as said. The gpu works in a diffrent system, borrow a psu from a friend to test in the problem pc

 

[mdva_dank]

Do you use 2 seperate cables from psu to gpu?

Yes I use 2 separate 8 pins

 

[Jacozeelie]

I thought the same, which is why I replaced my PSU. The problem is still here, unfortunately. 😕

Did you test the gpu in a working system?
Did you try operating the pc without gpu?

 

[Jacozeelie]

hi

So randomly as of about 2 months ago ive been experiencing weird blackscreen crashes when doing random things on my pc E.x.(gaming, watching youtube, idling, downloading)
gpu died previously and it blackscreened but got it rma'd and didnt have a single issue for a couple of weeks until i started blackscreening again( but its different this time?) previously when it would crash i would get a completely unresponsive pc and eventually fried my card, this time however it is able to be interacted with, albeit without any type of signal regardless of what cable i use or how many monitors i have plugged in.

I have tried memtest with no issues, completely reinstalled windows on a brand new ssd, cleaned drivers, removed all drives except boot drive, reset bios/cleared cmos, turned off PBO, turned off D.O.C.P, tried 1, 2 , and 4 sticks of ram, tried gpu in another system on a 650 watt psu(ran fine) replaced gpu with 3070 temporarily and passed multiple stress tests(as has the 3090), ran OCCT, found thousands of power errors (only sometimes other times its fine) run benchmarks and reseated/ tried different cables with the psu, so i am at a complete loss now as it seems to happen within a timespan of 5min to 50+ hrs of being on and i have no clue as to what could be causing this at all.

At this point, any advice or help would be appreciated. I will say the voltages on the different rails seem rather on the low end in general and when i had originally purchased it(about 3 months ago) it had terrible coil whine that has seemed to go away somehow at this point. again any help would ba appreciated because at this point of troubleshooting for over 2 months.
I'm at my wits end and have no clue what to do at this point lol.

My specs are as followed (also it seems more likely to crash with pbo enabled in the bios) (bios, vbios, and chipset drivers are all up to date for mobo gpu and cpu.)

CPU: R9 5900x
Motherboard: ASUS PRIME x570 pro
Ram: 32gb(8x4) cl14 ddr4 trident z royal ram running at 3600mhz
SSD/HDD: 2tb samsung 860 pro(BOOT), crucial mx500 1tb m.2, inland premium 2tb m.2, t force rgb ssd 2tb, 8tb segate hdd, 4tb segate hdd
GPU: gigabyte AORUS master rtx 3090 rev 1.0
PSU: corsair rm1000x
Chassis: lian li 011 dynamic xl
OS: windows 10 20H2 update KB4023057

i am also running 3 monitors and numerous internal and external perifs, if that information is pertinent let me know and ill update with that info,

again huge thanks to anyone who can offer any advice/ info thank you in advance.

Have you tried running ram at 3200mhz?

 

[mdva_dank]

Do you have a working psu that can be used in the pc. It might be a power problem. Faulty psu cause many gpu related problems

Hi,

I'm actually in the same boat (was about to make a new post, then found yours). My machine will give me a black screen or hard reboot, both seemingly random. The only consistency that I have found is that if I manually reboot about once a week I don't get these crashes. Also, once it does start crashing, it will do so more frequently until I turn off the machine and leave it off for a while (few hours). I have replaced my motherboard and PSU, but the issue still persists. The similarity I see between our two builds is the memory. Out of curiosity, did you buy two separate 16GB packs or one 32GB pack?

CPU	AMD Ryzen 7 2700X
RAM	G.SKILL TridentZ RGB Series 32GB (4 x 8GB) 288-Pin DDR4 SDRAM DDR4 3200 (PC4 25600) Intel XMP 2.0
Motherboard	ASUSTeK COMPUTER INC. ROG STRIX B450-F GAMING (AM4)
GPU	8176MB ATI AMD Radeon RX 5700 XT (ATI)
Storage	931GB TOSHIBA MQ01ABD100 (SATA )
	298GB Western Digital WDC WD3200AAJS-22B4A0 (SATA )
	232GB Samsung SSD 850 EVO 250GB (SATA (SSD))
	931GB Samsung SSD 970 EVO Plus 1TB (Unknown (SSD))
PSU	Corsair RM 750

I bought one 32gb pack so all 4 sticks came together, though the matching memory makes me a bit suspicious now, ill go to mc and buy some tester ram as I have replaced the psu and am still getting thousands of cpu power errors only when running the power test in occt

 

[mdva_dank]

[QUOTE="Jacozeelie, post: 22277363, member: 2725660"] Do you have a working psu that can be used in the pc. It might be a power problem. Faulty psu cause many gpu related problems [/QUOTE] I have tried 2 psu's now and am still getting cpu power error in occt, have yet to see a black screen but have not yet played any games on it

 

[mdva_dank]

Hi,

I'm actually in the same boat (was about to make a new post, then found yours). My machine will give me a black screen or hard reboot, both seemingly random. The only consistency that I have found is that if I manually reboot about once a week I don't get these crashes. Also, once it does start crashing, it will do so more frequently until I turn off the machine and leave it off for a while (few hours). I have replaced my motherboard and PSU, but the issue still persists. The similarity I see between our two builds is the memory. Out of curiosity, did you buy two separate 16GB packs or one 32GB pack?

CPU	AMD Ryzen 7 2700X
RAM	G.SKILL TridentZ RGB Series 32GB (4 x 8GB) 288-Pin DDR4 SDRAM DDR4 3200 (PC4 25600) Intel XMP 2.0
Motherboard	ASUSTeK COMPUTER INC. ROG STRIX B450-F GAMING (AM4)
GPU	8176MB ATI AMD Radeon RX 5700 XT (ATI)
Storage	931GB TOSHIBA MQ01ABD100 (SATA )
	298GB Western Digital WDC WD3200AAJS-22B4A0 (SATA )
	232GB Samsung SSD 850 EVO 250GB (SATA (SSD))
	931GB Samsung SSD 970 EVO Plus 1TB (Unknown (SSD))
PSU	Corsair RM 750

We are also both using asus motherboards and seemingly the other forums I have visited were asus mobo users as well.

 

[Jacozeelie]

I have seen that xmp profiles causes many problemes but when entering values manually, it works perfectly. Disable xmp and test i think

 

[mdva_dank]

I have seen that xmp profiles causes many problemes but when entering values manually, it works perfectly. Disable xmp and test i think

with docp disabled it still crashed

 

[Graybtw]

Same thing here, replaced psu fixed it for a bit then didnt work, got a ssd and it fixed it for a while and doenst work again. Check my post for my specs

 

[Graybtw]

Update: i read a bit more forums and it seems like its dur to a Power supply idle control? i set this to typical as it was auto. I could 100% see this being it i will come back if it didnt work.

 

[mdva_dank]

Update: i read a bit more forums and it seems like its dur to a Power supply idle control? i set this to typical as it was auto. I could 100% see this being it i will come back if it didnt work.

where and how do you set this?

 

[Graybtw]

where and how do you set this?

I have a gigabyt b450m ds3h i found it in bios under advanced cpu settings, it could be under motherboard or power supply or I don’t know depending on your mobo EDIT: if it is still frying your gpu it seem slike therr is a bigger problem. Also for anyone asking my problem was black screen kbm lights somtimes went out and sometimes my fan. not always i would get no input.

 

[Graybtw]

I have a gigabyt b450m ds3h i found it in bios under advanced cpu settings, it could be under motherboard or power supply or I don’t know depending on your mobo EDIT: if it is still frying your gpu it seem slike therr is a bigger problem. Also for anyone asking my problem was black screen kbm lights somtimes went out and sometimes my fan. not always i would get no input.

Update, nope.. now its gettung really fast and shuttingdown

 

[mc0]

Update, nope.. now its gettung really fast and shuttingdown

I'm trying this on mine to see if it will help. Other forum posts suggested switching it to typical from auto. (Un)fortunately mine doesn't do this very often, so it might be a while before I know if it works.

 

[Graybtw]

Update to the update, i changed my sleep settings and installed gpu drivers, its working fine now. I also suggest changing your power settings maybe

 

[djroff]

I'm late to this party, but I too was about to post my own question on the same issue. My rig has been having this problem for months, where it randomly freezes for a few seconds, then black-screens... doesn't shut down, just sticks on a black screen. It used to recover occasionally, but now it never does... always requires a full cold boot (Restart won't work.) The circumstances are all over the place, but none involved any kind of heavy lifting, as I'm not a big gamer, just some light virtual world stuff. Most of the time, I would be typing an email. Sometimes, I wouldn't even be there, I'd just come home to find the screen black.

Short version of my system's specs (similar to others in the thread): Asus ROG B450-F mobo, AMD Ryzen 5 2600x CPU, Crucial Ballistix 16GB RAM, Radeon RX550 GPU (yeah, I know, but the 4GB of RAM was a big improvement over what I had -- see the "I give up" thread, another story). My PSU is a Rosewill 1000W modular supply, and only maybe 2 years old? Too new to be having trouble, anyway.

Some things I could rule out. For instance, I know that it's a hardware problem instead of software, because it's a multi-drive unit, and on a completely separate drive, I have an installation of Ubuntu. I tried booting that up and letting the machine run for a bit, and sure enough, the same problem occurred. RAM generally causes other issues, like random system errors, up to and including the Blue Screen Of Death. Bad RAM can cause a machine not to POST in the first place. But I have never seen RAM cause a freeze and then a fatal black-screen. If it were just the GPU cutting out, the system wouldn't freeze first, the video would just go blank. Etc, etc.

This is a rig which I have progressively upgraded over the years, and never had this problem until the past 6 months. So I started to zero in on the newest hardware, which was replaced about 9 months ago: mobo and processor. Motherboard sounded the most likely -- the bug reports would never nail anything down, but there was always some kind of "failed to connect to" this or that. So I tried to get Asus to replace the board for me (it's under a 3-year warranty), and without going into a long story, that was a real nightmare. Took me over 2 months, but I just last weekend installed the replacement board, and.... it crashed again.

Only two other likely possibilities jump out at me: some instability in the processor (which is the same age and warranty status as the mobo), or the PSU really is failing prematurely. (My tester says otherwise, but it's not a very sophisticated tool. It has no way of testing the unit under load, for instance.) I'm not looking forward to fighting with AMD about replacing the processor under warranty. Nor am I excited by the prospect of buying another PSU just to throw that at the problem and still have the problem. Anybody ever have the CPU cause this issue? I have seen similar things with PSUs that were dying, but generally the computer would just randomly shut down, no running black screen. (On the off-chance the problem wasn't with the main ATX connector, but with the 12V supplied at mid-board, I tried switching that modular line to a different rail. No change.) I have no spare 1000W PSUs lying about, no more than I have a spare CPU to try (one of Asus' crazier suggestions, before I shamed them into letting me try a replacement). So my next step would be to try and eliminate the CPU, if I can get a replacement under warranty... unless one of you geniuses out there has a wonderful idea for me to try.

 

[mc0]

I'm late to this party, but I too was about to post my own question on the same issue. My rig has been having this problem for months, where it randomly freezes for a few seconds, then black-screens... doesn't shut down, just sticks on a black screen. It used to recover occasionally, but now it never does... always requires a full cold boot (Restart won't work.) The circumstances are all over the place, but none involved any kind of heavy lifting, as I'm not a big gamer, just some light virtual world stuff. Most of the time, I would be typing an email. Sometimes, I wouldn't even be there, I'd just come home to find the screen black.

Short version of my system's specs (similar to others in the thread): Asus ROG B450-F mobo, AMD Ryzen 5 2600x CPU, Crucial Ballistix 16GB RAM, Radeon RX550 GPU (yeah, I know, but the 4GB of RAM was a big improvement over what I had -- see the "I give up" thread, another story). My PSU is a Rosewill 1000W modular supply, and only maybe 2 years old? Too new to be having trouble, anyway.

Some things I could rule out. For instance, I know that it's a hardware problem instead of software, because it's a multi-drive unit, and on a completely separate drive, I have an installation of Ubuntu. I tried booting that up and letting the machine run for a bit, and sure enough, the same problem occurred. RAM generally causes other issues, like random system errors, up to and including the Blue Screen Of Death. Bad RAM can cause a machine not to POST in the first place. But I have never seen RAM cause a freeze and then a fatal black-screen. If it were just the GPU cutting out, the system wouldn't freeze first, the video would just go blank. Etc, etc.

This is a rig which I have progressively upgraded over the years, and never had this problem until the past 6 months. So I started to zero in on the newest hardware, which was replaced about 9 months ago: mobo and processor. Motherboard sounded the most likely -- the bug reports would never nail anything down, but there was always some kind of "failed to connect to" this or that. So I tried to get Asus to replace the board for me (it's under a 3-year warranty), and without going into a long story, that was a real nightmare. Took me over 2 months, but I just last weekend installed the replacement board, and.... it crashed again.

Only two other likely possibilities jump out at me: some instability in the processor (which is the same age and warranty status as the mobo), or the PSU really is failing prematurely. (My tester says otherwise, but it's not a very sophisticated tool. It has no way of testing the unit under load, for instance.) I'm not looking forward to fighting with AMD about replacing the processor under warranty. Nor am I excited by the prospect of buying another PSU just to throw that at the problem and still have the problem. Anybody ever have the CPU cause this issue? I have seen similar things with PSUs that were dying, but generally the computer would just randomly shut down, no running black screen. (On the off-chance the problem wasn't with the main ATX connector, but with the 12V supplied at mid-board, I tried switching that modular line to a different rail. No change.) I have no spare 1000W PSUs lying about, no more than I have a spare CPU to try (one of Asus' crazier suggestions, before I shamed them into letting me try a replacement). So my next step would be to try and eliminate the CPU, if I can get a replacement under warranty... unless one of you geniuses out there has a wonderful idea for me to try.

I have the same board, but was able to get it replaced because one of the m.2 slots was faulty - the issue still occurs. The issue still occurs even when I cycle out the memory, so that would mean all of my memory is faulty or it's not the memory. I do occasionally have GPU hangs/crashes, but the AMD driver now recovers from those. I'm leaning toward CPU as well as it's one of the last things I haven't replaced that could conceivably cause this kind of issue.

 

[djroff]

Yeah, that's what I'm thinking as well, although, like I said, I'm not at all thrilled with the prospect of fighting with AMD about sending me a replacement. (Asus basically wanted me to replace every other piece of hardware in the case, before they would even consider the possibility. Then they made me send my board in for "repair", and, of course, returned it with a note saying they couldn't duplicate the problem. AND, they threw out my original box, sending it back just surrounded by bubble-wrap in a big brown shipping box. Real professionals.) Any experience in dealing with AMD in situations like this? I bought the CPU on the secondary market, but it came factory-packaged as "new", so the 3-year warranty should still be honored.

 

[Graybtw]

Update again.. its crashing and i figured out my audio cuts too