forked from bartvdbraak/blender
Cleanup / Cycles: Adjust comment for faster testing in the future.
Unfortunately the function call is still a bit slower, even with CUDA 6.0. :/
This commit is contained in:
parent
73ad0dfbae
commit
8d42e7b20d
@ -818,7 +818,7 @@ ccl_device float4 kernel_path_integrate(KernelGlobals *kg, RNG *rng, int sample,
|
||||
}
|
||||
#endif
|
||||
|
||||
/* The following code is the same as in kernel_path_integrate_lighting(),
|
||||
/* Same as kernel_path_integrate_lighting(kg, rng, &sd, &throughput, &state, &L, &ray),
|
||||
but for CUDA the function call is slower. */
|
||||
#ifdef __EMISSION__
|
||||
if(kernel_data.integrator.use_direct_light) {
|
||||
|
Loading…
Reference in New Issue
Block a user