When rendering using the graphics card, a series of shaders (as long as no fundamental changes were made to the hardware or software), which are saved in your user directory, must be compiled internally. This pre-compilation can take several minutes to complete. In previous versions of Cinema 4D, these compilations were done as needed when rendering was started, which was often irritating. This can, however, still be done if you decide not to use the pre-compilation feature.
Pre-compilation can be started in one of two ways:
In the following instances, a new compilation must be done - which can be started automatically or manually:
Click on Clear Shader Cache to delete the pre-compilation in your user directory. This command only works in conjunction with the GPU.
Since ProRender can run simultaneously on multiple graphics cards or on the CPU, these settings can be used to define rendering separately for the Picture Viewer and the preview rendering, which uses hardware rendering. Depending on how your hardware setup you will find a list of all compatible CPUs and graphics cards (all that are OpenCL 1.2-compatible).
Rendering on the CPU and graphics card simultaneously is not possible. However, the CPU is mostly much slower than the graphics card. You should always render with the graphics card, if possible.
Note also that your mainboard might have graphics chips that have not yet been used. These chips (even if they are not particularly fast) can also be used for rendering (you might have to install a driver for them).
Note that the results may deviate from those of GPU rendering. The render results are not necessarily always identical.
Use the CPU if:
Note that the Custom Number of Render Threads option in the Preferences menu will be evaluated. If Cinema 4D reacts sluggishly during rendering, you can allot calculation time for other tasks.
Mac users working with macOS Sierra or higher with new hardware - if hardware and software are not up to the task, this option will be grayed out - can use this option to determine if Apple’s Metal 2 processing method instead of OpenCL. Metal 2 can work faster - try it out to see if it works better / faster.
Note that functionality is available that does not support metal such as linear motion blur. Meta will then be disabled internally and OpenGL will be used.
Here you can define the hardware for rendering in the Picture Viewer.
Note that this and the next option can each also be defined in the Render Settings menu.
The scene display in the Viewport is normally handled by OpenGL. This means that the image created by ProRender has to be passed on to the CPU, which in turn passes it on to OpenGL. Data has to be transferred from the graphics card to the CPU and back. If this Enable GL/CL Interoperability option is enabled, this detour will be eliminated and the transfer will take place within the graphics card (which is naturally faster).
If you have multiple different graphics cards installed, Cinema 4D will check which cards are compatible and will mark it as "(Recommended)".
Use this option to define the hardware for the preview renderer in the Viewport.
Out of Core Cache
Details about the Out-of Core functionality can be found under Enable.
Use these options to define the path for the texture files’ cache. By default, this path lies in the preferences directory. After rendering has finished, these will be deleted automatically.