Speed difference between using int and unsigned int when mixed with doubles

Question

I have an application where part of the inner loop was basically:

double sum = 0;
for (int i = 0; i != N; ++i, ++data, ++x) sum += *data * x;

Answer 1

I ran this on gcc 4.7.0 on a 64 bit machine running Linux. I replaced the time calls with calls to clock_gettime.

CPU: Intel X5680 @3.33 GHZ

GCC flags: -Wall -pedantic -O3 -std=c++11

Results:

With int time per operation in ns: 11996, total time sec: 1.57237
Avg values: 1.06353e+09
With unsigned int time per operation in ns: 11539, total time sec: 1.5125
Avg values: 1.06353e+09
With int time per operation in ns: 11994, total time sec: 1.57217
Avg values: 1.06353e+09

Evidently on my machine/compiler unsigned is faster.