Thread: A little optimization challenge.

Ok, as a start I'll begin with this (no compiler optimisations)

Code:

   //=========== Start of challenge code =========
   
   int c;
   while((c = getc(f)) != EOF) {
        if (isalpha(c)) {
           int lc = tolower(c);
           int iv = lc == 'e' || lc == 'a' || lc == 'o' || lc == 'i' || lc == 'u';
           vowelCount += iv;
           conCount += !iv;
        }
   }
   //=========== End of challenge code =========

The original code:

Code:

$./a.out 
There are 2357034 consonants and 1433891 vowels in the Complete Works of William Shakespeare
Time taken = 102125 microseconds

My modified code:

Code:

$ ./a.out 
There are 2357034 consonants and 1433891 vowels in the Complete Works of William Shakespeare
Time taken = 32977 microseconds

Only optimisation I made, really, is ordering the checks for a vowel based on English vowel frequency; i.e. I'm not concerned about much, right now, apart from how often vowels appear in typical English text. So, no real code changes but an improvement.

Edit: So I guess I didn't optimise it at all. Or did I?

Edit2: meh... compiler optimisations were turned off. My bad. My non-code-related changes are still faster with -O2 though.