The best thing to do I would think is to assign a constructor and a destructor to the application itself that drops a time stamp into a file. This way you get the time stamp after the program is loaded but before the execution of the entry point, and right when the last objects are being destroyed. This should be accurate enough for most people.
That sounds incredibly.... boring. But hey if it's what you're into then go for it I guess. What IDE are you using? Also, if you know, what Compiler? This is relavent to defining the functions I mentioned earlier. If you don't know what Compiler I'll assume the default setting for that IDE.
Try this to get the execution time in milliseconds for some block of code: #include <ctime>// include this header
in main()
1 2 3 4
int start_s=clock();
// the code you wish to time goes here
int stop_s=clock();
cout << "time: " << (stop_s-start_s)/double(CLOCKS_PER_SEC)*1000 << endl;
That will work for a simple project but remember that your first time stamp doesn't execute until "start_s" is initialized in fun2code's example. The same can be said about the final time stamp and "stop_s".
Also, the "clock()" function doesn't return milliseconds. The value is dependent on the speed of the host systems processor. But this will give you a relative score.
its also going to take longer depending on how you coded it. if youre going through every number up to 10000000 and counting its factors, its gonna take a damn long time