I'm currently 4 and a half hours into rendering a quite simple animation (was supposed to be just a test), realized I left it on GPU Compute after tinkering with those options. Really hoping that that has slowed down the render.

I'm new to making this kind of thing with Blender so I've quickly gotten in over my head and need a few pointers.

I have a Radeon RX480 4GB GPU, and an AMD FX-8350 2.3GHz CPU, 8GB System RAM.

I was reading in the manual about NVIDIA's advantage with CUDA, so my main question is if it's even worth messing with seeing as I have an AMD card which doesn't seem to have an advantage.

I was also wondering if the tile size affects animation speed. The things I was reading were only about complex still images. But larger tile size is better for GPUs. It would make sense that it might work for animations too, but there's a lot I don't know about this.

Finally I'm curious about the render samples. I have mine set to 256, which I was reading needs to be higher for scenes with depth of field (of course mine has that). So I may need to re-render. But I don't want to wait another 5+ hours for a 3 second intro video.

Any pointers for settings would be really awesome. Even if the final render will take this long, I'd like to have less of a wait for test renders.


closed as too broad by Duarte Farrajota Ramos, Timaroberts, JakeD, David May 2 '17 at 13:13

Please edit the question to limit it to a specific problem with enough detail to identify an adequate answer. Avoid asking multiple distinct questions at once. See the How to Ask page for help clarifying this question. If this question can be reworded to fit the rules in the help center, please edit the question.

  • 2
    $\begingroup$ Please ask only one question on each post. This site works better if we can focus on individual questions so that all of the answers are for a specific issue. Please edit your question and break it in different posts. Some of your questions have been asked before on this site, you can use the search box on the top right. $\endgroup$ – cegaton May 2 '17 at 2:35
  • $\begingroup$ Make a quick render test, you already know how long it takes for GPU, if your CPU beats your cards on the same job then use CPU instead. Tile sizes affect rendering speed (both for stills or animations, CPU, CUDA or OpenCL) not "animation speed". $\endgroup$ – Duarte Farrajota Ramos May 2 '17 at 3:58

If you are using an RX480 you should get a fairly significant speed boost over your aging FX-8350. The AMD videa cards cards have a few less supported features than Cycles than Nvidia cards on CUDA but nothing that can't be worked around.

I'd suggest lowering the number of samples in your sampling settings and turn on the seed value being different per frame. Video doesn't need as much refinement per frame as people think. For the tile sizes, through experimentation I have found that larger sizes work better with GPU rendering but it is a fairly minor speed gain.

Good luck rendering your project!


Not the answer you're looking for? Browse other questions tagged or ask your own question.