[interfacekit] Re: BString revamp

Glad to hear you're still knocking about!  Nice test results.  I'm not a 
performance test guru, by any stretch of the imagination, but you might 
investigate the effect that the cache is having on your anomolous result 
for 1000000 iteration CharacterDeescape() test.  Maybe post about it to 
the IK list or the main list.

e

>Hi!
>After some time that I couldn't program very much for OBOS, I am now 
back.
>I did some performance testing on my BString implementation, and 
decided to=
> 
>rewrite some functions, because it was not as fast as R5 one. You shall 
not=
> 
>see any checkin since I want to complete the rewrite first.
>And I hope you don't see this as premature optimizations, since tests 
have 
>already been written for BString, and I can test it well against them 
:)
>
>And now the good news :)
>
>Here's some tests I did:
>
>#define ITERATIONS 10000
>
>int main()
>{
>=09BStopWatch *watch;
>=09
>=09PRINT(("CharacterEscape(char)\n"));=09
>=09BString string("abzcabcahbcccdazbdcabcdahjebacbdhehrijgrzhue");
>=09watch =3D new BStopWatch("String");
>=09for (int x =3D 0; x < ITERATIONS; x++)
>=09{ 
>=09=09string.CharacterEscape("ahz", '/');
>=09}
>=09
>=09delete watch;
>}
>
>I compiled this little program with R5's BString and with OBOS one.
>These are the results:
>
>CharacterEscape(const char*, char*)
>
>OBOS: StopWatch "String": 6881603 usecs.
>R5: StopWatch "String": 35313129 usecs
>
>CharacterDeescape(char),  1000000 iterations
>
>OBOS: StopWatch "String": 198041 usecs. 
>R5: StopWatch "String": 189282 usecs.
>
>CharacterDeescape(const char*, char), 1000000 iterations (way faster, 
am I 
>missing something?)
>
>OBOS: StopWatch "String": 4545007 usecs
>R5: StopWatch "String": 10563255 usecs
>
>Anyway, I'm still here :)
>
>Bye!


Necessity is the plea for every infringement of human freedom. It is the 
argument of tyrants; it is the creed of slaves.
        -William Pitt, British prime-minister (1759-1806)

Other related posts: