Question Video playback worse after graphics card upgrade?

Aug 15, 2020
3
0
10
0
Hi, I just upgraded my graphics card from a GeForce GTX 1050 Ti to a Radeon RX Vega 56 but actually my video playback performance got a lot worse. Now, playing a 4K video can take 50% CPU. (My CPU is only an i3-4160, but before the upgrade I could play 8+ videos simultaneously and my CPU stayed around 20%.) This happens whether I use VLC or the "movies" app. In VLC, switching the "hardware decoding" setting on and off makes no difference.

Before installing the graphics card I uninstalled my old NVidia driver with DDU and turned off Windows update. I have the latest driver (win10-radeon-software-adrenalin-2020-edition-20.8.1-aug4.exe) and restarted my PC.

My PSU is a Gigabyte P650B.

Any suggestions? I'm happy to provide extra info.

EDIT: GPU usage is around 15% when the CPU maxes out. And Windows task manager shows "Video decode" as 0% for my GPU.
 
Last edited:
Aug 15, 2020
3
0
10
0
Thank you for the reply Grobe. I've never run a benchmark before but just installed Heaven (not sure if this is the best benchmark to use) and here are my results:

FPS: 40.7
Score: 1026
Min FPS: 7.7
Max FPS: 98.9

Platform: Windows NT 6.2 (build 9200) 64bit
CPU model: intel core i3-4160 CPU @ 3.60Ghz x2
GPU model: Intel HD Graphics 4400 20.19.15.4531/Radeon RX Vega 22.19.676.0 (4095MB) x 1

Render: Direct3D11
Mode: 3840x2160 fullscreen
Preset: Custom
Quality: High
Tessellation: Disabled

Is there any other information I should provide?
Thank you!
 

Grobe

Distinguished
You should compare the results against the old GPU, otherwise it doesn't provide an answer for any performance gain.

What about games - does it perform better than the old GPU ?

For what use do you need to play 8 videos simultaneously ?
 
Aug 15, 2020
3
0
10
0
Thank you, OK I will reinstall the old GPU and run a benchmark. Is Heaven a good benchmark to use, or do you suggest a different one? I don't use this computer for games. I need to play several videos simultaneously for an art project.
 

Grobe

Distinguished
Idk Heaven, but as long as you gain a set of benchmark results from both it should be ok and comparable. It's preferable if the benchmark tool of your choice also reveal internal temperatures of the gpu.
 

ASK THE COMMUNITY