After years of successfully avoiding doing photorealistic renderings at work I’ve been tasked with it for my current project. Apologies if this is remedial, but I’ve been spending my afternoon doing quick googling on V-Ray basics after getting it set up.
It took some mucking around in my office’s PC but I’ve gotten Sketchup to use the GPU for graphics instead of the CPU which significantly improved performance while modeling. After updating V-Ray and getting accustomed to the interface, I realized my interactive render is set to CPU. This surprised me as it’s very fast and responsive when I make changes to the model. I set it to GPU and was further surprised that while building the light cache takes around 3s when I’m on CPU, on GPU it takes close to 3 minutes to load bitmaps and build light cache, and much much longer than CPU to develop a decent render.
The computer is a little older, but I’m wondering if there’s a problematic GPU setting for it to be THIS much slower than CPU rendering. I’m happy to stick with my current Sketchup on GPU, V-Ray on CPU if it works, but I was just curious if I should be expecting better performance from a dedicated RTX graphics card vs just maxing out my CPU for a couple of minutes. Any info appreciated. Thanks!
CPU is faster for interactive rendering, gpu is a lot faster for the final render specially if it has ray tracing dedicated hardware. Just make sure to change to RTX instead of CUDA.
I did a Google search, using what you had asked. What it gave back was somewhat useful. Try this Google search, with the AI option:
please summerize this: After years of successfully avoiding doing photorealistic renderings at work I’ve been tasked with it for my current project. Apologies if this is remedial, but I’ve been spending my afternoon doing quick googling on V-Ray basics after getting it set up. It took some mucking around in my office’s PC but I’ve gotten Sketchup to use the GPU for graphics instead of the CPU which significantly improved performance while modeling. After updating V-Ray and getting accustomed to the interface, I realized my interactive render is set to CPU. This surprised me as it’s very fast and responsive when I make changes to the model. I set it to GPU and was further surprised that while building the light cache takes around 3s when I’m on CPU, on GPU it takes close to 3 minutes to load bitmaps and build light cache, and much much longer than CPU to develop a decent render. The computer is a little older, but I’m wondering if there’s a problematic GPU setting for it to be THIS much slower than CPU rendering. I’m happy to stick with my current Sketchup on GPU, V-Ray on CPU if it works, but I was just curious if I should be expecting better performance from a dedicated RTX graphics card vs just maxing out my CPU for a couple of minutes. Any info appreciated. Thanks!
I read on Vray forums that, cuda mode is better than rtx on heavy scenes, choosing “compress textures” option also helps cuda renderings.. If there is detailed trees in the scene, cpu rendering on high mode takes a lot of time..
On the asset editor where you are there’s another panel that can be opened to the right side, there are some settings there, one of the last ones is gpu rendering, there you can change from CUDA to RTX, with CUDA you can make hybrid rendering, using the gpu and cpu at the same time, it’s good for huge scenes that require a lot of memory, otherwise RTX gpu rendering is still faster.
