[SOLVED] Gigabyte X370 Gaming 5 issue with Ryzen 2600X

[peterparker_hu]

So i updated my older Ryzen 1200 for a Ryzen 2600X and have temperature issues. In OCCT the Package temp is at 90 degrees at lowest and rising at stress-test. Had no issue with older CPU (was around 35 at start). The CPU was replaced by the distributor but the problem remains. I really have no idea what's the problem with the MOBO. BIOS is F41! Any ideas?

 

[PC Tailor]

Welcome to the forums my friend!

What CPU cooler are you using?
What CPU temps are you getting under your normal usage load?

 

[Phaaze88]

Yeah. Don't run OCCT.
It's far too hard on cpus. The loads aren't realistic by any means.

Try Prime95 with AVX disabled, or Cinebench R20.

 

[peterparker_hu]

Welcome to the forums my friend!

What CPU cooler are you using?
What CPU temps are you getting under your normal usage load?

i have a Silentium Spartan 3 PRO. the core temps are normal, around 30-40 degrees on idle. the problem is when i start OCCT test, in less then a minute my pc shuts down. i play warframe regulary and in demanding places my pc does the same. but none of these problems occur with my Ryzen 1200 - highest temps never reach 70+ degrees

 

[Phaaze88]

the problem is when i start OCCT test, in less then a minute my pc shuts down.

The load is so intense, that the psu is like 'Screw this, I'm out!'
XD

 

[peterparker_hu]

The load is so intense, that the psu is like 'Screw this, I'm out!'
XD

i have a Zalman ZMX 500 TX 80+ PSU 1ssd 2hdd 2x8 gigs and Sapphire RX 580 8gb. it shouldn’t be the problem. but yeah your reply was still funny

 

[Phaaze88]

The 2600X is very power efficient, yet OCCT still caused to psu to shut off... I'd say that's a bad sign towards the psu.
When testing with Prime95 w/o AVX, or Cinebench R20 and it still shuts off, please replace it.

 

[peterparker_hu]

The 2600X is very power efficient, yet OCCT still caused to psu to shut off... I'd say that's a bad sign towards the psu.
When testing with Prime95 w/o AVX, or Cinebench R20 and it still shuts off, please replace it.

i might consider replacing my PSU but this doesn’t answer the question about the temperature

 

[Phaaze88]

Those temps are expected if you're running the stock cooler.
It's 'adequate', and nothing more, but OCCT is far from a normal workload.

 

[peterparker_hu]

Those temps are expected if you're running the stock cooler.
It's 'adequate', and nothing more, but OCCT is far from a normal workload.

yeah but i can’t believe that +60 degrees is normal between 1200 and 2600x at idle and the cooler is not stock as i mentioned before. i’ll wait for my distributor’s answer about my MOBO

 

[Phaaze88]

OH, that Silentium part. I missed that. My mistake.

i can’t believe that +60 degrees is normal between 1200 and 2600x at idle

There isn't. That big a gap indicates a bad cooler mount.

And if by some chance the cooler fan died, the other case fans would still be able to cool the heatsink.

 

[peterparker_hu]

OH, that Silentium part. I missed that. My mistake.


There isn't. That big a gap indicates a bad cooler mount.

And if by some chance the cooler fan died, the other case fans would still be able to cool the heatsink.

the cooler is in perfect condition, and i've mounted it twice. IMO it's not about the cooler or the mounting. it's something behind all of these. here's another printscreen about HWMonitor. as you can see cores temperatures are low AF.

another thing about OCCT (i know there are complains but this is the only clue i have) AS SOON AS i start the test Package (which i still don't know what the hell is) already at 90 degrees. it's not starting from someting like 30-40 degrees. it's like the MOBO is not reading correct values from the Ryzen 2600X - BUT it reads normal values from Ryzen 1200... i'm losing my mind

 

[Phaaze88]

HWMonitor is pretty broken - check with Ryzen Master instead.

90 degrees, what?
F? Not an issue.
C? That's a problem.

 

[peterparker_hu]

HWMonitor is pretty broken - check with Ryzen Master instead.

90 degrees, what?
F? Not an issue.
C? That's a problem.

everything i’ve mentioned is in celsius, sorry about that! Ryzen Master shows the same temp. around 35 celsius.

 

[Phaaze88]

With what's been covered so far, that just leaves a couple of things that haven't been covered yet:
-Whether you're overclocking and possibly cranking too much voltage through that chip...
-Inadequate case cooling: What is your case? Running some kind of semi-passive fan setup?

If you've mounted the cooler correctly, you shouldn't be hitting 90C in Cinebench R20.

 

[peterparker_hu]

With what's been covered so far, that just leaves a couple of things that haven't been covered yet:
-Whether you're overclocking and possibly cranking too much voltage through that chip...
-Inadequate case cooling: What is your case? Running some kind of semi-passive fan setup?

If you've mounted the cooler correctly, you shouldn't be hitting 90C in Cinebench R20.

i’m not an overclocking enthusiast, stock performance is enough for me.

my case is a kolink luminosity, with 3 intake and 3 outtake setup. temps are ideal all across the mobo.

but still how can be such a difference between core and package temperatures?

 

[Phaaze88]

but still how can be such a difference between core and package temperatures?

There shouldn't be. It's something you are, or are not doing.
https://www.techpowerup.com/review/silentiumpc-spartan-3-pro-rgb/6.html
^Those are test results with that cooler paired with an 8700K - overclocked and non-overclocked samples. Out of the 8 scenarios, the cooler causes thermal throttling in pretty much 3 of them.
All 3 of those scenarios were stress tested + OC'ed, where the 8700K is capable of dumping around 200w of waste heat; the Ryzen 1200 and 2600X don't come anywhere close that.
The 1200 may push 70w(worst case scenario), and the 2600X, 110w(again, worst case).
I've checked the specs of the Silentium PC cooler. It's rated for 150w of heat; the 2600X is well within it's capability to cool.

You appear to have mounted the cooler correctly - and reapplied paste as well. If you hadn't, your idle temps would've been much worse, like 45C or something.

my case is a kolink luminosity

OH.

Take the front and side panels off and run Cinebench R20 again, please.
How high do the cpu temps get?

 

[peterparker_hu]

so i tried Cinebench with panels on and off. With panels it went up to 75 celsius, w/o panels was 72 celsius in Ryzen Master. it was a surprise that my PC didn't shut down

 

[Phaaze88]

so i tried Cinebench with panels on and off. With panels it went up to 75 celsius, w/o panels was 72 celsius in Ryzen Master. it was a surprise that my PC didn't shut down

Why would it shut down? Cinebench is just a cpu stress test.
75C is perfectly fine. 95C is this cpu's critical temperature: https://www.amd.com/en/products/cpu/amd-ryzen-5-2600x

The shut offs are likely from the RX 580's end. It's harder for the psu to deal with than the 2600X is.

 

[peterparker_hu]

i think i’ll go with a new PSU unit. mine is a couple years old so probs it’ll cure my issue!

 

[peterparker_hu]

The 2600X is very power efficient, yet OCCT still caused to psu to shut off... I'd say that's a bad sign towards the psu.
When testing with Prime95 w/o AVX, or Cinebench R20 and it still shuts off, please replace it.

so it seems you were right! i replaced my psu with an EVGA BQ 650 and OCCT tests run well everything seems to work fine now!

 

[Phaaze88]

Excellent!