How to render images enabling DLSS in a headless device?

[aritra-pal]

Hi, first of all, thanks for this awesome work. I have built it successfully on my device with vulkan SDK. So, I can enable DLSS in the GUI while rendering the scenes.
However, I am wondering while I am saving images using the run.py script are those images rendered with DLSS?
If not, can you please help me to enable DLSS while using the command line?

[Tom94]

Hi there, DLSS is not supported in headless mode. Supporting it would require disentangling Vulkan from GLFW, as well as figuring out an API to carry over motion vectors across separate testbed.render calls. Both of which are out of scope of this project at the moment. Cheers!

[aritra-pal]

Thanks for your quick response. In that case, I would like to save DLSS rendered images from the GUI. Can you please suggest me how can I do that ?

[Tom94]

While the GUI doesn't have a button to save screenshots, you can use the interactive Python REPL for that.

1. Use scripts/run.py to launch the GUI and train. Then move the camera to the position you'd like to screenshot.
2. Press the GUI button for entering the Python REPL
3. Execute

testbed.display_gui = False
image = testbed.screenshot()
write_image("bla.png", image)