1
edit
Changes
→Profiling
==== Profiling ====
The following test runs were performed on the following Machine:
* Windows 10
* i7-4790k @ 4.0GHz
* 16GB DDR3
* Nvidia GTX 430
Although the initial test runs were performed on a machine that is slightly slower (3.6GHz vs 4GHz), the results should not be impacted all that much. Why? Well because most of the bottleneck was happening inside the <code>rc4_output()</code> function which was now moved to the GPU along with replacing the for loop with a cuda kernel.
'''RC4 Cipher - 283 MB mp3 File'''