1
edit
Changes
no edit summary
[[File:profile_CUDA.png]]
the code for Kernel is :
<pre>
__global__ void cudaNegateImage2D(int *result, const int *work, int ni, int nj)
{
int i = blockIdx.x * blockDim.x + threadIdx.x;
int j = blockIdx.y * blockDim.y + threadIdx.y;
if (i < ni && j < nj) {
result[i * nj + j] = -(work[i]) + 255;
}
}
</pre>
I don't even know why I've got opposite result. I checked GPU parameters installed on my new computer:
GeForce GT 620, CUDA cores 96. May be those are not enough to enhance performance? or may be I did something wrong in Kernel?
=== Assignment 3 ===
