simonlsp

> @mitchellurgero Probbaly using the hyper-v video adapter. I basically use the remoteplay in Steam client. The opengl and vulkan is about half of host . Wonder any way to...

> OpenGL and Vulkan will not work. Since Minecraft (Java) is using OpenGL it wont work either. > > Minecraft bedrock uses RenderDragon which does support DirectX according to the...

**With pytorch-directml 1.13, we could add this feature without using onnx.** All we need is to modify get_optimal_device_name (in devices.py), and add ```python if has_dml(): return "dml" ``` dml could...

@ClashSAN @THEGOLDENPRO @mr-september I created another issue requesting feature for pytorch-directml, as this issue thread focus on onnx, we might be off-topic here. https://github.com/AUTOMATIC1111/stable-diffusion-webui/issues/7600

@vt-idiot > Everything just works in webui after those changes? * Torch-directml is basically torch-cpuonly with a torch_directml.device() to use directx gpu as device. * I only changed the "optimal_device"...

> > By the way, direct-ml is RAM-costing on integrated gpu, at least 16GB is required for sd1.4 on 512*512. On dedicated gpu, direct-ml also costs more RAM than cuda...

@Kodikuu With this workaround, I think it is time to change this line in the readme notes. > Vulkan renderer is unavailable and GL games may or may not work....

I'll send the video in about a week.