I want to test my OpenCL memory bandwitdh.
I work on a nVidia gt280 so my kernel should write or read (in global memory) a maximum of 118GB/s.

- I tried with the simplest kernel :
Code :
void main(__global float * array)
{
   array[ get_global_id(0) ] = 123.321;
}
- I work on a 16 777 216 floats array, with a non host memory buffer.
-Each thread write on float so each warp (32 threads) write 128 bytes (that's the best case for nVidia GPU compabilitie 1.3).
- I use OpenCL profiler.

=> My kernel is executing in 0.955 ms, so bandwitdh is : 65,462 GB/s.

What is the problem with my experimentation