arrayfun can be significantly slower than an explicit loop in matlab. Why?

Question

Consider the following simple speed test for arrayfun:

T = 4000;
N = 500;
x = randn(T, N);
Func1 = @(a) (3*a^2 + 2*a - 1);

tic
Soln1 = ones(T,          


        

Answer 1

That because!!!!

x = randn(T, N); 

is not gpuarray type;

All you need to do is

x = randn(T, N,'gpuArray');