Constant Blackscreening: Need Help
So I've run into the problem where my screens go black, and all the fans in my system crank up to full speed. I keep my BIOS and Nvidia drivers up to date and it still doesn't fix the issue. Gets particularly finicky when under load from large demanding games like Cyberpunk or Battlefield 6. Sometimes even in lighter games like Farm Sim. I tried turning down my graphics settings for a bunch of games and it seems to get me further than before, but things still seem unstable at best. Running a AMD Ryzen 9 7900X3D with 64g ram and a Nvidia RTX4090 graphics card on a MSI MPG 670E carbon wifi motherboard. Any tips or advice? Tested it in different outlets, but thinking maybe its a bad PSU, bad cable, or maybe the graphics card itself? Not a PC pro so I will happily take any pointers you've got lol
< >
Viser 1-4 af 4 kommentarer
_I_ 9. nov. kl. 13:18 
psu, brand/model/age?

4090 is power hungry, weak or overrated psu can cause bigger problems


could also be unstable oc or xmp if using 3-4 dimms

post a cpuz validation link
http://www.cpuid.com/softwares/cpu-z.html
cpuz -> validate button -> submit button
it will open a browser, copy the url (address) and paste it here
Sidst redigeret af _I_; 9. nov. kl. 13:20
Dont use cable mod
_I_ 9. nov. kl. 13:53 
Oprindeligt skrevet af ˢᵈˣ FatCat:
Dont use cable mod
cablemod is fine, just need to give exact psu brand/model and its year
modular psus do not use the same pinouts, even across the same series, or even same models and can the mfg can change them at any time without warning

always keep modular psu cables with the psu they came with
if you have unknown modular psu cable, throw it away
sounds wasteful, but better to throw out a psu than use it and kill a build
Thatis more often a thermal issue and then the gou driver craps out as a result. Now normally after a short period of black screen due to driver crash it should recover with basic GPU clocks to get your screen back again so you can then address the issue but it doesn't always recover properly like it should forcing the user to reboot the PC manually and go from there.

It would help to use an app that can not only monitor the cpu and gpu temps but periodically save that data to a log file. In case the system locks up or you forced to reboot you can then review tbat log and see how hot it was getting


I would start by trying a different PSU, one that is ATX 3.x spec that has the 12VHPWR connection readily available so you won't have to use Cable Mod or some type of adapter
Sidst redigeret af Bad 💀 Motha; 9. nov. kl. 16:03
< >
Viser 1-4 af 4 kommentarer
Per side: 1530 50