JavaFX performance for complex visualisations
richard.bair at oracle.com
Tue Dec 4 15:54:04 PST 2012
>> In fact, we found that on embedded the GPU was the major bottleneck and we needed to move more stuff onto the CPU.
> Seeing as a lot of the rendering work is done on the CPU and not the GPU, is it (or can it) be optimized to make use of multi-core CPUs, or must the rendering be single threaded?
Yes, in fact this use for multithreaded rendering has been part of our design from the 1.0 days. It is going to be some work to enable (as our rendering uses static variables to reduce garbage generation for temporary objects), also I don't know if our native loops are ready for it. But this is definitely a goal. No use letting a pile of cores sit idle while we melt the unlucky ones.
However on mobile most of the time is spent moving the masks (textures) from CPU to GPU, so I don't have a good feel for the overall performance impact there.
More information about the openjfx-dev