Ryzen 7 5800 temperatures

garyevs999

garyevs999

garyevs999

Associate
Joined
19 Jan 2017
Posts
10
Hi

I have just built a machine including the following:

ASUS ROG STRIX B550-F GAMING Motherboard
AMD RYZEN 7 5800X
ALPENFOHN BROCKEN 3 CPU COOLER
CORSAIR VENGENCE LPX 3600Mhz RAM

When running the MSI Kombustor CPU burner, it is running in the red at 90+ degrees. If I run the same program on my i5 machine it runs at around 67 degrees.

I know that the Ryzens run a little hotter than the intel chips, but is that much of a difference usual, and should it be that high ?

Thanks
 
90 is a bit high. I've seen that 80+ is normal, but 90 not so much.

Did you take the plastic off the cooler cold plate?

Have you tried reseating the cooler, making sure to properly apply paste and screw down the cooler bit by bit, corner by corner?
 
Did you put sufficient thermal paste on?
Did the cooler move about when installing it? If so you may have dislodged the paste.
 
Just done a 10 minute test on mine, held steady at 85c using the same cooler. (no overclock)

How much paste did you use? As i have read if you use to much it can cause high temps.

What is case airflow like? I have two 140mm fans intake and one 140mm exhaust.
 
Try setting a lower PPT and EDC to something like 125 as this should bring down temps between 8~10c

Also try setting a curve optimiser negative offset -10/15/20 etc whatever you can get up too stabiliy.
 
Joxeon said:
Try setting a lower PPT and EDC to something like 125 as this should bring down temps between 8~10c

Also try setting a curve optimiser negative offset -10/15/20 etc whatever you can get up too stabiliy.
Click to expand...
Would this be the same for a 5950x aswell?
 
roachyuk said:
Thanks ill give it ago as running80/90 while gaming worries me
Click to expand...
AMD have said 90 is ok. Just make sure it don't go over.
Ryzen 3000 & 5000 CPUs are using the rated temperature headroom (up to 95°C) quite aggressively in order to reach higher boost clocks. As a result, it is absolutely no problem and not alarming if the processor runs into this temperature limit.
 
roachyuk said:
I happened to watch this video live last night.....about 2 1/2 hours in is when he boots up and enters bios i used the same settings has him and temps have reduced drastically stays under 60c when gaming ...its worth taking a look and see what other people think.

https://www.youtube.com/watch?v=Ksk-nRFH4EI&t=8173s
Click to expand...
Setting a flat 1.2v in bios will limit single core performance.
 
Joxeon said:
Setting a flat 1.2v in bios will limit single core performance.
Click to expand...
Ohhhh see... i am new to this overclocking and voltage lark lol thanks for the heads up....what do you think about 1.3v will that increase the single core while lower temps?
I will watch his next video as he is going more into depth and ill learn from what he does while showing what he changes in the bios.
 
Last edited:
roachyuk said:
Ohhhh see... i am new to this overclocking and voltage lark lol thanks for the heads up....what do you think about 1.3v will that increase the single core while lower temps?
I will watch his next video as he is going more into depth and ill learn from what he does while showing what he changes in the bios.
Click to expand...
I'd try seeing how low you can get the negative offset on curve optimiser as this will give you cooler multicore temps while retaining the single core boosting.

Either that or try setting the vcore with a -0.05 To -0.1 offset although you can't do this at the same time as using curve optimiser / PBO else it will lock clocks to base so it would have to be one or the other.
 
Is this what you mean?
  1. Curve Optimizer is part of PBO 2.0, so enable PBO and set it to your platform's limits.

  2. Under PBO, leave the scalar at Auto. Auto performed the best for me, but if you want to try to tweak this, I'll mention when you should do this.

  3. In Curve Optimizer, start with an all core undervolt of -5. Iterate between STABILITY TESTING (HIGHLY TRICKY. SEE BELOW.) and lowering this by -5 each time until you find the lowest stable value.

  4. Now you know the undervolt limit of at least one of your cores. You can now go into per core undervolting to find which cores you can bring down further using the same iterative method above
 
roachyuk said:
Is this what you mean?
  1. Curve Optimizer is part of PBO 2.0, so enable PBO and set it to your platform's limits.

  2. Under PBO, leave the scalar at Auto. Auto performed the best for me, but if you want to try to tweak this, I'll mention when you should do this.

  3. In Curve Optimizer, start with an all core undervolt of -5. Iterate between STABILITY TESTING (HIGHLY TRICKY. SEE BELOW.) and lowering this by -5 each time until you find the lowest stable value.

  4. Now you know the undervolt limit of at least one of your cores. You can now go into per core undervolting to find which cores you can bring down further using the same iterative method above
Click to expand...
Yeah although I wouldn't do each core one at a time else you will be there all week, try -5>-10>15 etc on all cores first till it gets unstable then back down 5 on either your first 8 cores or 2nd 8 cores and see if that fixes stability, if it does try push your other cores further.
 
Thanks for your advice ill try that .....just a quick one PBO what does it mean by your "platform limits" think options are enable,advanced and auto?
 
Mine hits 90 at full load with the wraith prism cooler. Heard of others doing the same with 'better' coolers. Reducing PTT under PBO settings in the bios or using Ryzen master greatly helps to reduce this. Along with negative values in the Curve Optimiser as others have suggested.
 
Last edited:
You must log in or register to reply here.
Back
Top Bottom