cairo

for (int i=0; i<100; i++)
{
if(toss()>=1)
head++;

else
tail++;
}


bool toss()
{
srand(time(0));
return rand()%2;
}



logic problem ? i cant figure out the problem. can somebody tell me the what happened ?
why i stil get the same result each time i call the toss function ?

InfoGeek

call srand() from your main function only once. The reason you're getting the same result is that the for loop is executed within a few milli-seconds and the seed is set to almost the same time, resulting in the same random number each time.

__________________
PFO - My daily dose of technology.

Polyphemus_

InfoGeek is right, and here is an optimization:

This will do the same as % 2, but it is a lot faster. It masks out all the other bits, so all what remains is a 1 or a 0.

Dameon

Note that the least significant bit is the only way to make a number odd due to the fact that all the other bit values are even. Modulus is quite a bit slower. Dividing versus logical AND.

__________________
MD5(sig) = bcef75433db02e9ad9bf81d6f7c5c270

grumpy

It's actually a [u]bitwise[u] AND (&), not a logical AND (&&).

Some optimising compilers will actually do this sort of optimisation for you (i.e. if they see x%2 where 2 is a compile time constant they will convert it to x&1).

teencoder

Also next time please post the full source.

__________________
Geeks may not be cool now but in the long run they prosper.

nindoja

@teencoder: Why, the code he gave us is enough to solve the problem. His only fault was not putting his code in [code] tags.

teencoder

Oops sorry I didn't see the toss function earlier sometimes I miss things. I guess I'm used to seeing all my functions precede the main.

__________________
Geeks may not be cool now but in the long run they prosper.

Arevos

I'm doubtful of this. Surely such a simple optimization would be included in any modern compiler?

coldDeath

No, he is correct, &1 is faster than %2.

__________________