[SOLVED] Rx480 running max temps w/ new 5600x setup

Jun 19, 2021
6
0
10
0
Hello everyone,

I put together a new Ryzen 5600x build, but I'm getting unplayable stutters in any game/benchmark I test with.

The MSI RX 480 4GB I used in my old build worked fine, but now it runs super hot.

It's at 100% load during every benchmark/game I've tried, and running at 85C-90C.

Specs are as follows:

Ryzen 5600x
MSI B550 Tomahawk
Crucial Ballistix 3600Mhz 2x8 GB RAM
MSI RX 480 4GB
256GB SSD + 1TB HDD
Corsair CX600

Right now I think it's either the PSU, mobo, or GPU itself.

Any ideas?
 
Jun 19, 2021
6
0
10
0
Tried running it in the PCIE 3.0 4x slot on the B550 (not ideal I know, but for testing purposes) and strangely enough, some of the stutter is gone.

The RX 480 still runs around 85C but it looks smoother and almost playable, imo. Not good enough, however.

I think the PCIE 4.0 x16 slot might be the issue here.

According to GPU-Z, the GPU chip power draw fluctuates from 150W+ down to 50W over and over again ->>> makes me think either the PCIE slot is bad, OR that PSU itself is wonky.

Did a clean install of Windows 10, and also fiddled with some BIOS settings (mainly turning off/on XMP & Precision Boost Overdrive, changing the PCIE slots from AUTO to PCIE 4.0/3.0), but still nothing.

Will try to test a friend's GTX 1070 in the mobo soon...
 
Jun 19, 2021
6
0
10
0
Even still those temps seem high. When was the last time you cleaned and or pasted the card?
Other than compressed air, I haven't cleaned/repasted like I've seen others do on the forum. Seems like a good idea, but the RX 480 worked fine in my old build (i7 2600k, ran 70-80C at max w/ no stuttering). I doubt I damaged it between builds, but I suppose something might've got loose, idk.
You were probably cpu bound in your games with your last cpu. Now that you upgraded, your gpu is working harder because you are gpu bound now
Old build I had had the RX 480 hit 100% usage, but at good temps (70-80C) with no stuttering. Now the card instantly hits 85 to 90C in any game (even less demanding games like Subnautica, Deep Rock Galactic etc.) and starts to stutter.

I think the stutter comes from either thermal throttling or lack of power, because Rivatuner is showing the GPU clock struggle to hit 1303MHz, then downclock to 500-800Mhz in 1 second intervals, causing the stutter. (GPU chip power draw goes from 150W+ down to 50W like I mentioned previously)
Rx cards are well known to operate in the 80s as far as degrees go. You should undervolt. It's really easy on amd cards using the adrenalin software. You could shave 5/6 degrees bringing them down to 80, which is pretty much the operating temps of those cards.

View: https://youtu.be/w03bINVnXGo
it's the same process.
Tried undervolting, setting a super aggressive fan curve, downclocking, as well as lowering/raising the power limit. Temps stayed mostly the same, and the stutters continue.

I've also tried some older GPU drivers, and no change to be found.

Will try that GTX 1070 I mentioned earlier, see if it works ok.

Thanks for the replies I appreciate them!
 
Jun 19, 2021
6
0
10
0
The 1070 GTX I tested was good. Had no stuttering, consistent frametimes etc. My MSI RX 480 seems to be the problem then, not the PSU/mobo.

As for the RX 480, now it crashes whenever I try using 3DMark (before it would just stutter through the benchmark...)

Will try repasting with Arctic Silver 5 sometime soon, hope it works.
 
The 1070 GTX I tested was good. Had no stuttering, consistent frametimes etc. My MSI RX 480 seems to be the problem then, not the PSU/mobo.

As for the RX 480, now it crashes whenever I try using 3DMark (before it would just stutter through the benchmark...)

Will try repasting with Arctic Silver 5 sometime soon, hope it works.
What case do you have? Is it the same one from your previous system? If it's a new case, it could be why the GPU is running hotter than before, because it may not have enough airflow.
 
Jun 19, 2021
6
0
10
0
While it's unlikely to be the direct problem here, this would still be a good opportunity to replace a rather low-quality PSU that really shouldn't be paired with a 1070 or a 480.
Agreed, a tier 2 or B or w.e PSU from 5-6 years back has me hesitant to OC right now. Currently I'm waiting for a sale on a decent one, as well as a non-stock cpu cooler. I suppose if repasting the GPU doesn't work I can try switching out PSUs from a friend.

What case do you have? Is it the same one from your previous system? If it's a new case, it could be why the GPU is running hotter than before, because it may not have enough airflow.
Had a crappy OEM case with terrible airflow in the old system, now I'm using a Fractal Meshify C and the difference is night and day, haha. Totally worth it, imo.
 
Jun 19, 2021
6
0
10
0
I can't believe it but re-applying the thermal paste worked! I was convinced the new parts were at fault (CPU/RAM/mobo etc.), not the working old ones lol.

The MSI RX 480 Gaming X idles like it used to at 30C, but now @ 100% load it stays at a comfy 62C during Furmark/3DMark.

Battlefield 1/V playing at a smooth 100fps/80fps, Deep Rock Galactic @ 144fps etc.

Thanks for the advice guys, take care.
 

ASK THE COMMUNITY

TRENDING THREADS