Skip to main content
Topic: Dedicated GPU (Read 10895 times) previous topic - next topic

Dedicated GPU

Hi, I have a PC with 2 GPUs, one internal 1 (AMD RAdeon Graphics) and the other dedicated 0(AMD Raden RX 5600M). When applying an effect to the camera the use of GPU 1 shoots up to more than 90%  :o , I would like to know how to make the camera occupy the other graphics card that has more memory.

Re: Dedicated GPU

Reply #1
Hi, Splitcam use default graphic card. You can try to plug monitor to more powerfull card. What cart your monitor connected to? 

Re: Dedicated GPU

Reply #2
Hi, I am using a laptop, without an external monitor.

 

Re: Dedicated GPU

Reply #3
Can you please send some info:
screenshots:
 dxdiag.exe (C:\Windows\System32\dxdiag.exe )
winver (type winver in command line)
and splitcam log.

Did you get all windows updates?

Re: Dedicated GPU

Reply #4
hi,

splitcam Log
"2022-05-07  =====================================

10:13:49 SplitCam starting: v10.5.62 x64
10:13:49 OS: Microsoft Windows 11 Home 10.0.22000 64 bits
10:13:49 Initializing Gdiplus
10:13:49 Gdiplus initialized
10:13:49 Creating app object
10:13:49 Resetting project to defaults
10:13:49 Deleting scenes
10:13:49 Done deleting scenes
10:13:49 Clearing stream settings
10:13:49 Deleting streams
10:13:49 Done deleting streams
10:13:49 Creating deafult scene
10:13:49 Initializing scene object
10:13:49 Creating video capture sources
10:13:49 Creating video capture renderers
10:13:49 Creating video capture canvas
10:13:49 Creating chat notifier
10:13:49 Creating chat notifier layer
10:13:49 Saving created context to static contexts list
10:13:49 Setting video callback
10:13:49 Setting default canvas size to 1280x720
10:13:49 Setting canvas callback
10:13:49 Setting audio callback
10:13:49 Registering deafult scene
10:13:49 Getting deafult project path
10:13:49 Initializing default theme
10:13:49 Loading options from "C:\Users\elFish\AppData\Local\Splitcam\options.xml"
10:13:49 Running app
10:13:49 Displaying splash screen
10:13:49 Updating ingest servers
10:13:49 Downloading "http://splitcam.com/win-download/update/ingests.cfg" to "C:\Users\elFish\AppData\Local\Splitcam\ingests.tmp"
10:13:50 Reading from "https://api.twitch.tv/kraken/ingests"
10:13:50 ERROR Twitch servers parse: [json.exception.invalid_iterator.207] cannot use key() for non-object iterators
10:13:50 Done updating ingest servers
10:13:50 Starting UI
10:13:50 Initializing common controls
10:13:50 Common controls inited
10:13:50 Initializing resources
10:13:50 Saving "loading.gif" resource to file
10:13:50 Resources initialized
10:13:50 Setting up workspace
10:13:50 Loading last project: "C:\Users\elFish\Documents\My Project SplitCam.xml"
10:13:50 Checking available GPUs
10:13:50 2 GPUs found in the system.
10:13:50   AMD GPU: AMD Radeon(TM) Graphics
10:13:50   AMD GPU: AMD Radeon RX 5600M
10:13:50 Loading project file "C:\Users\elFish\Documents\My Project SplitCam.xml"
10:13:50 Deleting scenes
10:13:50 Deleting video context
10:13:50 Stopping capture sources
10:13:50 Looking for current context in static context list
10:13:50 Deleting notifier
10:13:50 Deleting capture sources
10:13:50 Deleting renderers
10:13:50 Deleting canvas
10:13:50 Canvas thread stopped
10:13:50 Done deleting scenes
10:13:50 Deleting streams
10:13:50 Done deleting streams
10:13:50 Initializing scene object
10:13:50 Creating video capture sources
10:13:50 Creating video capture renderers
10:13:50 Creating video capture canvas
10:13:50 Creating chat notifier
10:13:50 Creating chat notifier layer
10:13:50 Saving created context to static contexts list
10:13:50 Setting video callback
10:13:50 Setting default canvas size to 1280x720
10:13:50 Setting canvas callback
10:13:50 Setting audio callback
10:13:50 Updating scenes
10:13:50 Creating memory renderer for splitcam driver.
10:13:50 Looking for available video capture devices
10:13:50 Initializing virtual driver interface
10:13:51 Found 6 video capture devices
10:13:51 Updating list of available video capture devices
10:13:51 Enumerating webcams
10:13:51   CyberLink PerfectCam: {\\?\root#media#0002#{e5323777-f976-4f5b-9b55-b94699c46e44}\global}
10:13:51   XSplit VCam: {\\?\root#image#0000#{e5323777-f976-4f5b-9b55-b94699c46e44}\global}
10:13:51   Integrated Webcam: {\\?\usb#vid_0c45&pid_671f&mi_00#7&5b26c9c&0&0000#{e5323777-f976-4f5b-9b55-b94699c46e44}\global}
10:13:51   OBS Virtual Camera: {OBS Virtual Camera}
10:13:51 Activating scene 0
10:13:51   Requesting media type: 1280x720@30
10:13:51 Starting initialization of: CyberLink PerfectCam webcam
10:13:51 Finished setting up work space.
10:13:51 Destroying splash screen
10:13:51 SplitCam started
10:13:51       Found media type: [email protected] YUY2
10:13:51 Available desktop size: 1920x1020
10:13:51 Main window location: pos=226,51, size=1468x918
10:13:51 Creating dialog "CPU" at (784, 507) of size 800x490
10:13:51 Creating dialog "GPU" at (784, 507) of size 800x490
10:13:51 Checking for software update
10:13:51 Reading from "https://splitcam.com/win-download/update/ver.txt"
10:13:51 Creating dialog "RAM" at (784, 507) of size 800x490
10:13:51 Creating dialog "Stream Statistics" at (784, 507) of size 800x490
10:13:51 Show Statistics panel.
10:13:51 Current version of SplitCam is newer than the one available on the update server
10:13:51 Acivating scene 0
10:13:51 Memory usage:

        Total RAM : 32,121 MiB
         Free RAM : 21,419 MiB
         RAM Load : 33%
     SplitCam RAM : 72 MiB
     SplitCam CPU : 0.0%
     SplitCam GPU : 0.0%

10:13:54 Initializing chat notifier
10:13:54 Finished initializing chat notifier
10:13:54 Updating chat client
10:13:54 Finished updating chat client
10:13:57 Creating dialog "Acerca de SplitCam" at (585, 60) of size 750x900
10:13:57 Reading from "https://splitcam.com/win-download/update/history.txt""

Winver
Windows 11 - 21H2

DirectX 12

Tks.



Re: Dedicated GPU

Reply #5
Thank you. And one more question. What exactly effect you use? Can you try skin smoother (what adapter used)?

Re: Dedicated GPU

Reply #6
hello, I am using the effect to remove the background. i attach the XML of the configuration. I did the test with almost all the effects, filters, masks, etc, and in all it uses the internal GPU.

Re: Dedicated GPU

Reply #7
Hello, thank you. Please wait for a new version. I will write to you then it ready.

Re: Dedicated GPU

Reply #8
thank you, I will be waiting  ;)