I’m trying to do some sculpting which requires the polygon count to be very high to get in the finer details but, when it gets into the hundreds of thousands or especially millions, it will lag so heavily that Blender will outright quit responding altogether for lengthy periods of time. My computer’s CPU is an Intel Core2 Duo 3.16 GHz and the onboard GPU is Intel X4500 but I have an AMD HD 5670 512MB in the PCIe slot. I enabled OpenCL but it only seems to acknowledge the CPU as options and it doesn’t help with the lag. Is there a way to make Blender make use of the AMD GPU so that I can have less troubles with the sculpting or is the GPU just not good enough to handle that many polygons or something?
OpenCL is not used for modelling and sculpting only for things like rendering or compositing
You already are?
Open GL is GPU based.
Just wait for version 2.76, that version will make better use of modern OpenGL drawing techniques that promise to really speed up the sculpting process.
So the problem really is just that my GPU isn’t strong enough to keep up?
Are you using Dyntopo or multires sculpting.
Dyntopo should already have highly optimized drawing code, so if that is the case it may be your GPU.
Multires in 2.75 isn’t as optimized as it could be, if it still lags in 2.76 then yes, the issue may be the power of the GPU.
Your cpu and gpu combination will reach a limit to both their capability. There are ways you can make best use of what you have. See
Wow, I should kick myself for not knowing about Dyntopo sooner. Yeah, I’d been using multires sculpting and just subdividing it multiple times. This is so much better for my needs, thank you! 