(Mar 2, 2014 05:28 PM)Sajuuk Wrote:  Yes:

1. You can seed random with srand(rand()). It's a hell of a lot quicker to do, and it works.
2. You're utilizing a namespace specifier. It's best to use something like std::wcout or the like because as you advance, you'll most likely have to use multiple namespaces, which could or could not use functions with the same name in different namespaces. Boost is an example:

Code

std::wcout << L"Hello there.";
boost::boostFunctionHere();
/* etc */

(Mar 2, 2014 05:30 PM)asillychilly Wrote:  isn't the rand() function only pseudo-random? Wouldn't that mean the seed is going to develop into a pattern, rendering the rand() after the srand() predictable? As for #2 thanks, I'll look into that.

(Mar 2, 2014 05:14 PM)asillychilly Wrote:  

Code

int generateFunction()
{
    int intGen;
srand(time(NULL));
intGen = rand()%10;
return intGen;
}

Quote:Do you see any bad habits being developed?

(Mar 2, 2014 05:28 PM)Sajuuk Wrote:  Yes:

1. You can seed random with srand(rand()). It's a hell of a lot quicker to do, and it works.
2. You're utilizing a namespace specifier. It's best to use something like std::wcout or the like because as you advance, you'll most likely have to use multiple namespaces, which could or could not use functions with the same name in different namespaces. Boost is an example:

Code

std::wcout << L"Hello there.";
boost::boostFunctionHere();
/* etc */

(Mar 3, 2014 09:57 AM)Rodaxoleaux Wrote:  Touching up on the last two posts, they're basically telling you to stop mixing C and C++ so much. C++11 gives you <random> which allows you to get truly random numbers with std::random_device without having to explicitly give it a seed. http://www.cplusplus.com/reference/random/random_device/

You're doing this.

Code

int intGen;
// random num seed
intGen = rand()%10;
return intGen;

Now look at this extreme example of what you did.

Code

int a=5,b,c,d,e,f,g;
b=a;
c=b;
d=c;
e=d;
f=e;
g=f;
return g;

when you could just

Code

return 5;

Also post-increment operator (variable++;). When you're not doing anything with the predecessor value, this just wastes the cpu cycles to copy the variable, return it and then increment it when by using pre-increment operator, (++variable;) it simply increments the value and then returns it. No copying happens.

Your code:

Code

for(int i=0; i<3; i++)
{
    cin >> guess;
    counter++;
// ...

Example of what I'm saying:

Code

for(int i=0; i<3; ++i)
{
    cin >> guess;
    ++counter;
//...

(Mar 3, 2014 07:37 AM)TxTriton Wrote:  1. That's a stupid idea... If srand() was not called previously then the prng is not seeded, and will generate all of the values default to the algorithm it uses, which destroys entropy. If this is the first call, you might as well just put:

Code

srand(41);

Seeding with the time is common, not the best, but much better than seeding immediately with rand(). If it was never previously seeded, one could guess the entire array of numbers that successive calls to rand() will generate.

Code

172
22420
5
5992
5768
8953
4904
10571
32065
7917 ...

If you needed something better than time, seed with the current mouse position.

---

The other thing: If this is C++, then it's common to prefix with C instead of suffix with ".h" for header files...

Code

#include <time.h>

Becomes:

Code

#include <ctime>

(Mar 3, 2014 01:56 PM)Sajuuk Wrote:  I stated that I seed with rand() "when I'm fucking with shit", meaning when I'm not doing real work, just fiddling around with functions. I guess I failed to tell him to only use rand() to seed when doing stuff just for fun. Sorry.

Quote:I stated that I seed with rand() "when I'm fucking with shit"

Quote:1. You can seed random with srand(rand()). It's a hell of a lot quicker to do, and it works.

Quote:I guess I failed to tell him to only use rand() to seed when doing stuff just for fun. Sorry.

Quote:Do you see any bad habits being developed?

(Mar 3, 2014 03:00 PM)TxTriton Wrote:  That's a great excuse, but you're not fooling people here.