Question My RX 570 is underperforming really badly, what should I do?

Jul 14, 2020
6
0
10
0
I just recieved a MSI RX 570 ARMOR 4GB OC but as soon as I installed it, downloaded the latest amd drivers and tested it in some games I saw that I was not achieving theresults I was expecting. In fact, this card performs only slightly better than my previous GT 1030. An example of this is when I tried to play No Man´s Sky at High settings at 720p and was met with 30/40fps. I´ve seen videos in which this card could maintain 60 fps at high settings in 1080p so I don´t really understand why I am getting this results. I also got a 46,8% score in UserBenchmark which sucks. These results. My specs are:
-MSI A320M A PRO MAX
-AMD Ryzen 3 3200G
-WD 240GB SSD
-1 x 8GB RAM 2400Mhz
-Apple HDD
-MSI RX 570 ARMOR 4GB OC
 
Jul 14, 2020
6
0
10
0
Did you remove the previous nvidia drivers before installing the card and the amd drivers?

Is the monitor plugged into your GPU itself and not the motherboards onboard video socket?
I uninstalled the nvidia drivers after installing the card and then downloaded the latest amd drivers, maybe this could be the issue but I don´t really know.

Yes, the monitor is plugged into the GPU and not into the motherboard´s HDMI port.
 
Jul 14, 2020
6
0
10
0
Your single stick of 2400 ram won't be helping but I'd still expect better performance than that.

What happens if you run it at low settings? Is there any substantial performance increase??
Yeah, I am currently planning on selling my previous GT 1030 to buy another stick of RAM and run them on dual channel so that is great to know. In low settings, there is a substantial increase in performance but as I said, i´ve seen videos online showing how the exact same card I own had better results than mine in 1080p. Some guy in the MSI forums told me that I was having lower framerates due to my low end CPU and that may be true since most benchmarking tests online use an i5 or greater processor. However, I am still worried about the User Benchmark website telling me that my GPU is " Performing below potential (40th percentile)".
 

kurdtnz

Prominent
Feb 24, 2020
656
89
490
7
'' buy another stick of RAM and run them on dual channel '' You may have an issue with a separate RAM stick, two individually bought sticks may not work together.
 

madmatt30

Titan
Ambassador
Yeah, I am currently planning on selling my previous GT 1030 to buy another stick of RAM and run them on dual channel so that is great to know. In low settings, there is a substantial increase in performance but as I said, i´ve seen videos online showing how the exact same card I own had better results than mine in 1080p. Some guy in the MSI forums told me that I was having lower framerates due to my low end CPU and that may be true since most benchmarking tests online use an i5 or greater processor. However, I am still worried about the User Benchmark website telling me that my GPU is " Performing below potential (40th percentile)".

Those results include heavily overclocked cards so you can't rely on them.

Dual channel gives upto 20% better results on ryzen.

That will directly influence how well your GPU pushes fps.
 
Jul 14, 2020
6
0
10
0
Those results include heavily overclocked cards so you can't rely on them.

Dual channel gives upto 20% better results on ryzen.

That will directly influence how well your GPU pushes fps.
That is great to hear, fortunately, RAM is not the most expensive thing in my country so I may be able to afford it someday. I thought benchmarks used stock speed so that explains my results. Thank you, you helped me a lot.
 
Jul 14, 2020
6
0
10
0
'' buy another stick of RAM and run them on dual channel '' You may have an issue with a separate RAM stick, two individually bought sticks may not work together.
I didn´t know that, I will now try to find the same cheap adata 8gb stick I currently own so that I don´t have any compatibility issues. That was very helpful, thank you.
 

hang-the-9

Titan
Moderator
That is great to hear, fortunately, RAM is not the most expensive thing in my country so I may be able to afford it someday. I thought benchmarks used stock speed so that explains my results. Thank you, you helped me a lot.
Benchmarks do, but the userbenchmark site is not really a good site for pure benchmarking of equipment. Unfortunately it's popular and almost everyone thinks if their equipment does not get some score on there there are issues. If you want to compare benchmarks, run something like 3DMark, Furmark, Passmark and then compare scores with your exact card and CPU, or at least look only at the video card results not full system results.
 
Jul 14, 2020
6
0
10
0
Benchmarks do, but the userbenchmark site is not really a good site for pure benchmarking of equipment. Unfortunately it's popular and almost everyone thinks if their equipment does not get some score on there there are issues. If you want to compare benchmarks, run something like 3DMark, Furmark, Passmark and then compare scores with your exact card and CPU, or at least look only at the video card results not full system results.
Yeah, I´ll try if I can, I just don´t have the money to buy a license for 3d mark so idk. Thanks anyway.
 

ASK THE COMMUNITY

TRENDING THREADS