Is rand() predictable in C++ [duplicate] - c++

This question already has answers here:
Closed 12 years ago.
Possible Duplicate:
What’s the Right Way to use the rand() Function in C++?
When I run the below program I always get the same values each time. Is rand not a true random function?
int main()
{
while(1)
{
getch();
cout<<rand()<<endl;
}
}
In each run I am getting the below values.
41
18467
6334
26500
19169
15724
......

Yes and no. rand() is a pseudo random number generator that will always return the same sequence of numbers given the same seed value. Typically one 'seeds' the random number generator with some random data and then uses rand() to return a sequence of seemingly random numbers. If your random data isn't needed for something requiring 'true' randomness (such as cryptography based security) just using the current system time is sufficient. However, if you are using it for security purposes, look into obtaining more truly random data from entropy gathering utilities and use that to seed the random number generator.
As aa mentioned, the seed function is referenced here

What is a true random function? Last I checked, computers couldn't do that :)
As to why you are getting the same set of numbers each time, it's because you need to seed the built in number generator with some starting 'random' value. There are many places to get this, but some tend to look good, but turn out bad. In our games, we generally seed with tic time from game bootup until the first or second user input. User input will always vary across many tics and can therefore be used as a decent starting point.

If you are using a Microsoft compiler, you can use rand_s, which generates very good random numbers (as good as you can get with only a computer): http://msdn.microsoft.com/en-us/library/sxtz2fa8%28VS.80%29.aspx
You can also use /dev/urandom on Linux and CryptGenRandom() on Windows to get quality random numbers.

or put srand() at the beginning of the function.

Just to follow on from the disucussion on "True" random numbers. As already, stated any generator that has a seed has a predictable period - I believe it can be 2^48.
If that level of randomness you can use the following:
long randomLong(unsigned int x)
{
x ^= (x << 21); // x is a non zero seed value
x ^= (x >> 35);
x ^= (x << 4);
return x;
}
This is taken from the following paper:
http://www.jstatsoft.org/v08/i14/paper
Which is a really interesting paper describing some low cost random number generators

Related

rand() not giving me a random number (even when srand() is used)

Okay I'm starting to lose my mind. All I want to do is random a number between 0 and 410, and according to this page, my code should do that. And since I want a random number and not a pseudo-random number, I'm using srand() as well, in a way that e.g. this thread told me to do. But this isn't working. All I get is a number that is depending on how long it was since my last execution. If I e.g. execute it again as fast as I can, the number is usually 6 numbers higher than the last number, and if I wait longer, it's higher, etc. When it reaches 410 it goes back to 0 and begins all over again. What am I missing?
Edit: And oh, if I remove the srand(time(NULL)); line I just get the same number (41) every time I run the program. That's not even pseudo random, that's just a static number. Just copying the first line of code from the article I linked to above still gives me number 41 all the time. Am I the star in a sequel to "The Number 23", or have I missed something?
int main(void) {
srand(time(NULL));
int number = rand() % 410;
std::cout << number << std::endl;
system("pause");
}
That is what you get for using deprecated random number generation.
rand produces a fixed sequence of numbers (which by itself is fine), and does that very, very badly.
You tell rand via srand where in the sequence to start. Since your "starting point" (called seed btw) depends on the number of seconds since 1.1.1970 0:00:00 UTC, your output is obviously time depended.
The correct way to do what you want to do is using the C++11 <random> library. In your concrete example, this would look somewhat like this:
std::mt19937 rng (std::random_device{}());
std::uniform_int_distribution<> dist (0, 409);
auto random_number = dist(rng);
For more information on the evils of rand and the advantages of <random> have a look at this.
As a last remark, seeding std::mt19937 like I did above is not quite optimal because the MT's state space is much larger than the 32 bit you get out of a single call to std::random_device{}(). This is not a problem for toy programs and your standard school assignments, but for reference: Here is my take at seeding the MT's entire state space, plus some helpful suggestions in the answers.
From manual:
time() returns the time as the number of seconds since the Epoch,
1970-01-01 00:00:00 +0000 (UTC).
Which means that if you start your program twice both times at the same second you will initialize srand with same value and will get same state of PRNG.
And if you remove initialization via call to srand you will always get exactly same sequence of numbers from rand.
I'm afraid you can't get trully random numbers there. Built in functions are meant to provide just pseudo random numbers. Moreover using srand and rand, because the first uses the same approach as the second one. If you want to cook true random numbers, you must find a correct source of entrophy, working for example with atmospheric noise, as the approach of www.random.org.
The problem here consists in the seed used by the randomness algorithm: if it's a number provided by a machine, it can't be unpredictable. A normal solution for this is using external hardware.
Unfortunately you can't get a real random number from a computer without specific hardware (which is often too slow to be practical).
Therefore you need to make do with a pseudo generator. But you need to use them carefully.
The function rand is designed to return a number between 0 and RAND_MAX in a way that, broadly speaking, satisfies the statistical properties of a uniform distribution. At best you can expect the mean of the drawn numbers to be 0.5 * RAND_MAX and the variance to be RAND_MAX * RAND_MAX / 12.
Typically the implementation of rand is a linear congruential generator which basically means that the returned number is a function of the previous number. That can give surprisingly good results and allows you to seed the generator with a function srand.
But repeated use of srand ruins the statistical properties of the generator, which is what is happening to you: your use of srand is correlated with your system clock time. The behaviour you're observing is completely expected.
What you should do is to only make one call to srand and then draw a sequence of numbers using rand. You cannot easily do this in the way you've set things up. But there are alternatives; you could switch to a random number generator (say mersenne twister) which allows you to draw the (n)th term and you could pass the value of n as a command line argument.
As a final remark, I'd avoid using a modulus when drawing a number. This will create a statistical bias if your modulo is not a multiple of RAND_MAX.
Try by change the NULL in time(NULL) by time(0) (that will give you the current système time). If it doesn't work, you could try to convert time(0) into ms by doing time(0)*1000.

Random generator for guess game

I've been searching for a better solution than my own and I haven't really been able to find one that I understand or that works for me.
I have made the simple game where the computer randomly generates a number which you then guess a number and if it is higher the computer says higher and so on..
The problem is my randomly generated number, after looking up alot of information regarding the <random>, uniform_int_distribution and default_random_engine. I have found out that the computer generates a random number, but if you run the program again the same random number will be generated.
My solution:
uniform_int_distribution<unsigned> u(0,100); // code to randomly generate numbers between 0 and 100
default_random_engine e; // code to randomly generate numbers
size_t userInput; // User input to find out where to look in the vector
vector<int> randomNumbers; //vector to hold the random numbers
unsigned start = 0, ending = 101, cnt = 0; // used in the game not important right now
cout << "Please enter a number between 1 and 1000 for randomness" << endl;
cin >> userInput;
for(size_t i = 0; i < 1000; ++i){ //for loop to push numbers into the vector
randomNumbers.push_back(u(e));
}
unsigned guess = randomNumbers[userInput]; // finally the number that the user will have to guess in the game
My solution right now is to use a vector where I push alot of randomly generated numbers in then ask the user to type a number which then the computer uses for the game. But there should be a better way of doing this. And my question is therefore
Is there a better way to randomly generate numbers to use in the game?
Either use std::random_device in place of std::default_random_engine, or else think of a way to provide a different number to the engine each time it is run.
This number is called a "seed" and can be passed as an optional parameter to the constructor. Since std::default_random_engine is implementation-specific, and different engines do different things about seeding, you generally want to choose a specific engine if you're providing a seed. A deterministic pseudo-random number generator will produce the same sequence of outputs for any given seed, so you want to use a different seed each time.
For no-security uses like a guessing game, the most "obvious" thing to use as a seed is the current time. Generally speaking this is different each time the program is run, although obviously if you can run the program twice in less than the granularity of the clock then that's not the case. So using the time to seed your random engine is pretty limited but will do the job for a toy program.
That's because your random number is actually what we call a pseudorandom number generator
It's just a machine that given a starting number generates a large list of seemingly random numbers. As you don't provide a starting number, the generated list of random numbers is thus always the same. One easy way to fix this is to use the current time as a starting value or 'seed', which is an argument of the constructor of std::default_random_engine.
You can also use your machines real random number generator std::random_device as a replacement for std::default_random_engine
Why not simply:
#include <ctime> // for time()
#include <cstdlib> // for srand()
srand(time(NULL)); // Initializes the rand() function
int randomNumber = rand()%100; // Random number between 0 and 99.
What this does is the rand() seed is set at the current time, meaning that every execution of the program will have a different seed for rand().
Still just pseudo-random solution, though suitable for your purposes.

What is a seed in terms of generating a random number? [duplicate]

This question already has answers here:
What does 'seeding' mean?
(4 answers)
Closed 6 years ago.
What is a seed in terms of generating a random number?
I need to generate hundreds to thousands of random numbers, I have read a lot about using a "seed". What is a seed? Is a seed where the random numbers start from? For example if I set my seed to be 5 will it generate numbers from 5 to whatever my limit is? So it will never give me 3 for example.
I am using C++, so if you provide any examples it'd be nice if it was in C++.
Thanks!
What is normally called a random number sequence in reality is a "pseudo-random" number sequence because the values are computed using a deterministic algorithm and probability plays no real role.
The "seed" is a starting point for the sequence and the guarantee is that if you start from the same seed you will get the same sequence of numbers. This is very useful for example for debugging (when you are looking for an error in a program you need to be able to reproduce the problem and study it, a non-deterministic program would be much harder to debug because every run would be different).
If you need just a random sequence of numbers and don't need to reproduce it then simply use current time as seed... for example with:
srand(time(NULL));
So, let's put it this way:
if you and your friend set the seed equals to the same number, by then you and your friend will get the same random numbers. So, if all of us write this simple program:
#include<iostream>
using namespace std;
void main () {
srand(0);
for (int i=0; i<3; i++){
int x = rand()%11; //range between 0 and 10
cout<<x<<endl;
}
}
We all will get the same random numbers which are (5, 8, 8).
And if you want to get different number each time, you can use srand(time())

win32 - higher random max value? [duplicate]

This question already has answers here:
Closed 12 years ago.
Possible Duplicate:
Filling an array with random numbers from 1 to 10^10 in C or C++
Hello,
Simple question; I am looking for a way to get a higher random value than rand() gives me. rand() gives 32767 max. (http://msdn.microsoft.com/en-us/library/398ax69y%28v=vs.80%29.aspx)
I've tried redefining RAND_MAX, but I guess the compiler dosn't accept redefining, because it's within the same boundary as default.
I basically want a random value that fits in an unsigned integer (generating a SSRC number in an RTP packet). I really don't want collisions!
Is there any better win32 function than rand()?
Thanks in advance
Yes -- RAND_MAX is just telling you the maximum that's designed into the generator. Changing it doesn't affect how the generator works.
The easiest way to get a 32-bit value from it is probably to call rand() twice in a row, and and put the two values together:
unsigned SSRC = rand() | rand << 16;
One typical way of seeding rand() is with time:
srand((unsigned)time(NULL));
For an SSRC, you do not want to do this -- time typically only has one-second granularity, so this would (almost) guarantee that any two processes started close to the same time would cause collisions.
In reality, randomness doesn't matter nearly as much as uniqueness. Another possibility would be (for example) to take your machine's IP address, the process ID, and the time and XOR the three together. The result isn't particularly random at all, but is sufficiently unique for this kind of task.
Use Boost.Random (which is also available as <random> if you have VC++ 2010). Note that this will become a built-in part of the C++ Standard in the near future.
There are some good examples on the Boost site, including Generating integers in a range.
How about rand()<<15 + rand() ? It can be as big as you want.
Changing RAND_MAX is not going to change the behavior of the rand() function, which is defined in the C runtime; it's just a convenient macro for you to know the range of output produced by rand().
If you want high-quality random numbers suitable for use with cryptography, use CryptGenRandom; see the full example at the bottom under Community Content for how to use it.

C++: seeding random number generator outside of main()

I was creating a simple program that simulates a coin toss for my class. (Actually, class is over this term and i'm just working through the rest of the projects that weren't required). It involves the creating and calling a function that generates a random number between 1 and 2. Originally, I tried to seed the random number generator within the function that would be using it (coinToss); however, it did not produce a random number. Each time the program was run it was the same number as though I had only used
rand()
instead of
unsigned seed = time(0);
srand(seed);
rand();
Yet, when i moved the above within
int main()
it worked fine.
My question is 1)why did it not work when setup within the function that called it and (2) how does rand()
have access to what was done by srand() if they do not both occur in the same function?
Obviously, i'm a beginner so please forgive me if i didn't formulate the question correctly. Also, my book has only briefly touched on rand() and srand() so that's all i really know.
thanks for any help!
Pertinent code:
First attempt that didn't work:
int main()
{
//...........
coinToss();
//...........
}
int coinToss()
{
unsigned seed = time(0);
srand(seed);
return 1 + rand() % 2;
}
Second attempt which did work:
int main()
{
unsigned seed = time(0);
srand(seed);
coinToss();
}
int coinToss()
{
return 1 + rand() % 2;
}
You probably only want to seed the random number generator once. rand() returns the next pseudo-random number from it's internal generator. Every time you call rand() you will get the next number from the internal generator.
srand() however sets the initial conditions of the random number generator. You can think of it as setting the 'starting-out point' for the internal random number generator (in reality it's a lot more complicated than that, but it's a useful cognitive model to follow).
So, you should be calling srand(time(0)) exactly once in your application - somewhere near the beginning. After that, you can call rand() as many times as you want!
However
To answer your actual question - the first version doesn't work because time() returns the number of seconds since the epoch. So If you call coinToss() several times in a second (say, if you wanted to simulate 100 coin tosses), then you'd be constantly seeding the random number generator with the same number, thereby resetting it's internal state (and thus the next number you get) every time.
Anyway - using time() as a seed to srand() is somewhat crappy for this very reason - time() doesn't chage very often, and worse, it's predictable. If you know the current time, you can work out what rand() will return. The internet has many, many examples of better srand() seeds.
Pseudo-random number generators (like rand) work by taking a single starting number (the seed) and performing a numeric transformation on it each time you request a new number. You want to seed the generator just once, or it will continually get reset, which is not what you want.
As you discovered, you should just call srand just once in main. Also note that a number of rand implementations have pretty short cycles on the low 4 bits or so. In practice this means you might get an easily predictable repeating cycle of numbers You might want to shift the return from rand right by 4-8 bits before you take the % 2.
EDIT: The call would look something like:
return 1 + (rand() >> 6) % 2;
Seed only once per program, not every time you call coinToss()
To expand on Mark B's answer: It is not so much that the random number generator is reset as it sets a new variable to be used in calculating random numbers. However your program doesn't do that much work between calls to srand. Therefore every time you call srand(time(0)) it is using the same seed, so you are resetting the internal state of the random number generator . If you put a sleep in there so that time(0) changed you would not get the same number every time.
As for how data passes from srand to rand, it is fairly simple, a global variable is used. All names that start with an underscore and a capital letter or two underscores are reserved for variables used by your compiler. More than likely this variable has been declared static so it isn't visible outside of the translation unit(aka the library file that contains your compiler's standard library.) This is done so that #define STUFF 5 doesn't break your standard library.
for simple simulations, you must not change the seed at all during the simulation. Your simulation will be "worse" in that case.
To understand this, you should see pseudo random sequences as a big wheel of fortune. When you change the seed, it is like you change the position, and then, each call to rand will give you a different number. If you roll again, it will be more probable finding yourself repeating numbers.