[SOLVED] What causes Graphics Cards to suddenly lose input?

Apr 6, 2021
78
6
35
Sometimes, couple times a month i experience a sudden gpu fail. Both my monitors become black and they both show No Signal. Why does it happen? Does the GPU just ‘die’?
 
Solution
Never used AIDA64 or even know what it is

AIDA64 have stress testing functionality for CPU, GPU and RAM. May help to determine how fast and hot your hardware is on full load. Also monitor clock speeds and voltages in same time. Often used in overclocking stability testing. Particularly you may check out which part cause 12V brownout (with stalled system and black screen) on full load.

Still convinced that you should invest into better PSU. Or at least into 500-550W one who have less ripple on full load (see notes about Cooler Master MWE in PSU tier list).
Apr 6, 2021
78
6
35
These issues are usually heat and/or power related.

What are the specs for this system (make/model of all components), particularly the GPU and PSU?

GPU: Sapphire RX 570 4gb
PSU: Coolermaster MWE 500W 230V
CPU: Ryzen 3 2300x
Mobo: MSI A320M-A-PRO
Case: CiT Raider
RAM: 8gb (2x4gb)
1TB HDD
 
Never used AIDA64 or even know what it is

AIDA64 have stress testing functionality for CPU, GPU and RAM. May help to determine how fast and hot your hardware is on full load. Also monitor clock speeds and voltages in same time. Often used in overclocking stability testing. Particularly you may check out which part cause 12V brownout (with stalled system and black screen) on full load.

Still convinced that you should invest into better PSU. Or at least into 500-550W one who have less ripple on full load (see notes about Cooler Master MWE in PSU tier list).
 
Solution
Apr 6, 2021
78
6
35
Your PSU is marginal in terms of quality and could be affecting stability.

What are your CPU and GPU operating temperatures, at idle and under load?

Are you overclocking anything (CPU, GPU, memory)?
CPU temps in idle are around 50 degrees celsius
GPU temps in idle are around 40 degrees celsius

Under load (when i play fortnite)
CPU is at around 80 degrees
GPU is around 70 degrees

I am not overclocking CPU, a slight over clock in GPU (Stock is 1244, ive gone 1280), and memory clock (stock 1500mhz, i went to 1550mhz)
 
Apr 6, 2021
78
6
35
AIDA64 have stress testing functionality for CPU, GPU and RAM. May help to determine how fast and hot your hardware is on full load. Also monitor clock speeds and voltages in same time. Often used in overclocking stability testing. Particularly you may check out which part cause 12V brownout (with stalled system and black screen) on full load.

Still convinced that you should invest into better PSU. Or at least into 500-550W one who have less ripple on full load (see notes about Cooler Master MWE in PSU tier list).
I can try using that.

And i will invest in a better PSU. Thanks for your info
 

COLGeek

Cybernaut
Moderator
CPU temps in idle are around 50 degrees celsius
GPU temps in idle are around 40 degrees celsius

Under load (when i play fortnite)
CPU is at around 80 degrees
GPU is around 70 degrees

I am not overclocking CPU, a slight over clock in GPU (Stock is 1244, ive gone 1280), and memory clock (stock 1500mhz, i went to 1550mhz)
Are you using the stock AMD Wraith cooler?

I recommend you remove the GPU OC to see if that helps stabilize the system. After upgrading your PSU, then re-apply the OC.
 
GPU: Sapphire RX 570 4gb
PSU: Coolermaster MWE 500W 230V
CPU: Ryzen 3 2300x
Mobo: MSI A320M-A-PRO
Case: CiT Raider
RAM: 8gb (2x4gb)
1TB HDD
what drivers are you using?
I had the same gpu with the same psu (550 not 500)
had the same issue
driver 20.10.0 or newer should clear that up.
If not I recommend you download the most recent drivers
download DDU (google)
remove all display drivers
reboot to safe mode
run ddu remove all driver traces
reboot
install the latest drivers
reboot
 
Apr 6, 2021
78
6
35
what drivers are you using?
I had the same gpu with the same psu (550 not 500)
had the same issue
driver 20.10.0 or newer should clear that up.
If not I recommend you download the most recent drivers
download DDU (google)
remove all display drivers
reboot to safe mode
run ddu remove all driver traces
reboot
install the latest drivers
reboot
i’ve done that already.
 
  • Like
Reactions: spentshells

David0ne86

Prominent
Mar 11, 2021
424
113
440
My budget isn’t low, but i wouldn’t buy the high end stuff, and i buy parts from Amazon usually.

Sharkoon Silentstorm Cool Zero. Fully modular and for that price you can't find anything better imo. It's a decent quality 80+ gold, has 2 japanese capacitors at the primary (at the secondary are high quality taiwanese capacitors), the OEM is sirtec (which is a proven good psu builder). It has a semi fanless feat and an interesting fan delay system (basically when you turn the pc off, the fan will keep spinning for 60 seconds ensuring everything is good and cooled). the 750 w is like 105 usd/115 eu. If you can spend a little bit more there's the seasonic core ones but they're a bit more expensive and not fully modular.
 
  • Like
Reactions: mylesjc