[SOLVED] System freezes

Status
Not open for further replies.
May 14, 2021
11
0
10
my systems been stable for months. This has been happening for a month where my pc would randomly lock up while gaming & while livestreaming running prime 95 it locks up in seconds like the screen would just freeze and I would have to hold down the power button manually to shut off the system I'm running on stock settings with turbo boost off c6 & c1 states off I tried increasing the voltage a little undervolting none of it worked I've read many forums people stating that they we're having the same problems but for me none of those solutions worked I even reinstalled windows a week ago and cleaned my pc full specs below all help would be gladly appreciated. I recently did ran windows memory diagnostic results turned out 100% temperatures are decent as well doesn't go past 65 celsius while under heavy load. Sorry if the grammars bad I didn't bother with it right now

CPU = AMD FX 9590
RAM = x2 8gbHYPER X SAVAGE DDR3 1600 MHZ
GPU = XFX AMD RADEON R7 370 2gb DDR5
Motherboard = Crosshair Formula V
Powersupply = 600w PSU
 
Solution
Check core temperatures again using AMD Overdrive. Get it here: https://www.techspot.com/downloads/4645-amd-overdrive.html
The FX-9590 has a Thermal Margin of 57C.
AMD did not have the cpu report the traditional real time temperatures. Instead, it reports its Thermal Margin - it counts down to 0 and even negative numbers as it warms up, and rises back towards the Thermal Margin(57C in this case) as it cools down.
Yes, confusing, I know.
The key points:
-The apps you've been using report the temperatures they've received from the cpu sensors to you as real time, not Thermal Margin.
-Since Thermal Margin doesn't equal real time temperature...
-The FX-9590 has an official Thermal Margin of 57C. There isn't an official real time...

King_V

Illustrious
Ambassador
What specific cooler do you have? I don't think a memory diagnostic will push the CPU hard.

If you mean while gaming, are you SURE it's not going past 65 degrees?

Finally, what's the brand and EXACT model number of the PSU? And, how old is it? It may not be a particularly good quality unit, and age may have also taken its toll. The 9590 is a very power hungry CPU, having a 220W TDP.
 
  • Like
Reactions: helper800
May 14, 2021
11
0
10
this is the power supply
EVGA 600 BQ 110-BQ-0600-K1 BQ 80 Plus Bronze Semi Modular 600W Power Supply 6 months old


Cooler is cooler master nepton 120xl and yes I'm 100% sure it doesn't go past 65 degrees the thing is that if I run aida 64 the system is stable but while gaming and running prime 95 it just freezes
 
Last edited:
May 14, 2021
11
0
10
but it's been running fine for months how is it not enough that's what I don't understand and I monitor my temperatures what do you think I should set the voltage to? It' running at 1.452 right now I tried 1.39 even then it was freezing I guess ill try again maybe go lower
 
May 14, 2021
11
0
10
Thanks for the help but I know for a fact that it's not a temperature problem I can run aida 64 for let's say 10 minutes the system won't lockup and if I we're to run prime 95 and while gaming of course if I was streaming and playing a game at the same time it would lockup before reaching 1 minute usually seconds and I have a temperature thing set on the bios so if it reaches 70 degrees it would shutdown and I never had that happen to me EDIT: oh and I don't run prime 95 while gaming I kinda made it sound like that :D
 
Last edited:
May 14, 2021
11
0
10
here's an image if you guys wanna have a look image after running aida for 10 minutes but let's say I we're to run prime 95 for a couple of seconds or play call of duty 4 while streaming at 1080p 3000 bitrate for a couple of minutes the system would just lockup Idle temps https://ibb.co/Yp3473r
 
Last edited:

Phaaze88

Titan
Ambassador
Check core temperatures again using AMD Overdrive. Get it here: https://www.techspot.com/downloads/4645-amd-overdrive.html
The FX-9590 has a Thermal Margin of 57C.
AMD did not have the cpu report the traditional real time temperatures. Instead, it reports its Thermal Margin - it counts down to 0 and even negative numbers as it warms up, and rises back towards the Thermal Margin(57C in this case) as it cools down.
Yes, confusing, I know.
The key points:
-The apps you've been using report the temperatures they've received from the cpu sensors to you as real time, not Thermal Margin.
-Since Thermal Margin doesn't equal real time temperature...
-The FX-9590 has an official Thermal Margin of 57C. There isn't an official real time throttle temp.

So, what's the real state of this cpu?
 
  • Like
Reactions: PC Tailor
Solution
Have you monitored the cpu (GHz) speed while running these tests? It’s not just cpu temperature that can cause problems. Given the problematic history of these FX9XXX CPU’s I would not be surprised if a VRM issue has developed. Using an AIO adds to my theory as they provide no cooling to the VRM’s unlike an air cooler setup.
 
May 14, 2021
11
0
10
Alright well I appreciate the help I guess there's nothing you or I can do thanks guys I appreciate all of the help if you guys have any other suggestions I'd be glad to try em out.
 
May 14, 2021
11
0
10
Now there's one thing that I don't understand I'd appreciate if someone can tell me why this is possibly happening. I can run aida 64 for a couple of hours fine without any issues but let's say I we're to run prime 95 it would lock up under a minute or let's say I was livestreaming and playing a game it would lock up in a few minutes or while playing gta 5 for 30 minutes.
 
May 14, 2021
11
0
10
Alright guys so what I did was in the bios I set LLC from auto to extreme and the cpu voltage which was on auto 1.452 volts I changed that to 1.39 volts I just ran prime 95 for 10 minutes without any issues it used to freeze in seconds. the settings I changed are they safe to use? just want to make sure that I don't screw something up. when idling core voltages are 1.392 - 1.416 and while underload maximum 1.440
 

Phaaze88

Titan
Ambassador
Well, you proved that the cpu isn't thermal throttling, but that last move... You had to crank more voltage to the chip to make the clocks stable.
Auto LLC usually isn't a very high level - though skipping all the way to extreme didn't help very much.

I figure it's voltage degradation. After some more time has passed, it'll be unstable again and you'll have to raise the voltage again or be forced to clock it down.
Rinse and repeat.
 
Status
Not open for further replies.