Differential D7854 Diff 25163 manual/render/cycles/gpu_rendering.rst

Changeset View

Standalone View

manual/render/cycles/gpu_rendering.rst

Context not available.
	and issues with interactivity when using the same graphics card for display and rendering.	and issues with interactivity when using the same graphics card for display and rendering.

	To enable GPU rendering, go into the :menuselection:`Preferences --> System --> Cycles Render Devices`,	To enable GPU rendering, go into the :menuselection:`Preferences --> System --> Cycles Render Devices`,
	and select either CUDA, Optix or OpenCL. Next, you must configure each scene to use GPU rendering in	and select either CUDA, OptiX or OpenCL. Next, you must configure each scene to use GPU rendering in
	:menuselection:`Properties --> Render --> Device`.	:menuselection:`Properties --> Render --> Device`.


Context not available.
	Nvidia	Nvidia
	------	------

	:abbr:`CUDA (Compute Unified Device Architecture)` and Optix	:abbr:`CUDA (Compute Unified Device Architecture)` and OptiX
	are supported for GPU rendering with Nvidia graphics cards.	are supported for GPU rendering with Nvidia graphics cards.


Context not available.

	.. _render-cycles-gpu-optix:	.. _render-cycles-gpu-optix:

	Optix	OptiX
	^^^^^	^^^^^

	For RTX graphics cards with hardware ray tracing support (e.g. Turing), Optix can be used for better performance.	For RTX graphics cards with hardware ray tracing support (e.g. Turing), OptiX can be used for better performance.
	Optix support is still experimental and does not yet support all features, see below for details.	OptiX support is still experimental and does not yet support all features, see below for details.

	Optix requires Geforce or Quadro RTX graphics card with recent Nvidia drivers, and is supported on Windows and Linux.	OptiX requires Geforce or Quadro RTX graphics card with recent Nvidia drivers, and is supported on Windows and Linux.


	AMD	AMD
Context not available.
	- Open Shading Language.	- Open Shading Language.
	- Advanced volume light sampling to reduce noise.	- Advanced volume light sampling to reduce noise.

	Optix support is experimental and does not yet support the following features:	OptiX support is experimental and does not yet support the following features:

	- Baking	- Baking
	- Branched Path Tracing	- Branched Path Tracing
	- Ambient Occlusion and Bevel shader nodes	- Ambient Occlusion and Bevel shader nodes
	- Combined CPU + GPU rendering	- Combined CPU + GPU rendering
	- Using CPU memory for bigger scenes


	Frequently Asked Questions	Frequently Asked Questions
Context not available.
	but the most common is that there is not enough memory on your graphics card.	but the most common is that there is not enough memory on your graphics card.
	Typically while using GPU rendering the GPU can only use the amount of memory that is on the GPU.	Typically while using GPU rendering the GPU can only use the amount of memory that is on the GPU.
	This is usually much smaller than the amount of system memory the CPU can access.	This is usually much smaller than the amount of system memory the CPU can access.
	With CUDA and Optix devices, if the GPU memory is full Blender will automatically try to use system memory.	With CUDA and OptiX devices, if the GPU memory is full Blender will automatically try to use system memory.
	This has a performance impact, but will usually still result in a faster render than using CPU rendering.	This has a performance impact, but will usually still result in a faster render than using CPU rendering.
	This feature does not work for OpenCL rendering.	This feature does not work for OpenCL rendering.

Context not available.
	No, each GPU can only access its own memory.	No, each GPU can only access its own memory.


	What renders faster, Nvidia or AMD, CUDA, Optix or OpenCL?	What renders faster, Nvidia or AMD, CUDA, OptiX or OpenCL?
	----------------------------------------------------------	----------------------------------------------------------

	This varies depending on the hardware used. Different technologies also have different compute times	This varies depending on the hardware used. Different technologies also have different compute times
Context not available.