C++ Random

Randomness plays a critical role in many programming tasks, including simulations, games, cryptography, and testing. In C++, generating random numbers has evolved significantly with the introduction of the <random> library in C++11, offering more powerful and flexible tools than the traditional C-style rand() function.

Traditional Approach: rand()

Before C++11, randomness was handled using the C standard library function rand().

This example uses the rand() function to generate a random number:

#include <iostream>
#include <cstdlib>
#include <ctime>

int main() {
  std::srand(std::time(nullptr)); // Seed with the current time
  int random_number = std::rand(); // Generate a random number
  std::cout << "Random Number: " << random_number << '\n';
  return 0;
}

Copy to clipboard
Copy to clipboard

In this example:

• rand(): Returns a random number between 0 and RAND_MAX.
• srand(): Initializes the random number generator with a seed.

Here is a possible output:

Random Number: 1950498172

Copy to clipboard
Copy to clipboard

Note: Since the number is randomly generated, the output may vary each time the code is run.

Modern Approach: <random>

The <random> header, introduced in C++11, provides a robust framework for random number generation.

Here are the key components of this header:

• Engines: Generate raw random numbers.
• Distributions: Transform raw numbers into specific ranges or statistical distributions.

Common Distributions

Example

This example uses the <random> header to generate a random number between 1-6:

#include <iostream>
#include <random>

int main() {
  std::random_device rd;  // Seed
  std::mt19937 gen(rd()); // Mersenne Twister engine
  std::uniform_int_distribution<> dist(1, 6); // Uniform distribution [1, 6]

  std::cout << "Rolling a dice: " << dist(gen) << '\n';
  return 0;
}

Copy to clipboard
Copy to clipboard

In this example:

• std::random_device: Non-deterministic seed generator (may fall back to pseudo-random).
• std::mt19937: High-quality pseudo-random number generator (Mersenne Twister).
• std::uniform_int_distribution: Ensures even distribution between two bounds.

Here is a possible output:

Rolling a dice: 2

Copy to clipboard
Copy to clipboard

Codebyte Example: Ranged Random Number Generator Using rand()

This codebyte example uses the rand() function to generate a random number in the range 10-50:

Visit us
Visit us
Hide code
Code
Output
Hide output
Loading...
Copy to your clipboard
Copy Codebyte
Run

In this example:

• std::rand() % (upper - lower + 1) gives a number between 0 and (upper - lower).
• Adding lower shifts the range to [lower, upper].
• srand() ensures different results across runs by seeding with the current time.

Best Practices for Generating Random Numbers

• Prefer <random> over rand() for better control and randomness.
• Use std::random_device for seeding when unpredictability is desired.
• Match engines with appropriate distributions for accurate simulations.
• Avoid using the same engine across multiple threads without synchronization.

Frequently Asked Questions

1. What’s the difference between rand() and <random> in C++?

• rand() is a legacy function from C that returns a pseudo-random number.
• <random>, introduced in C++11, offers modern, flexible, and statistically sound ways to generate random numbers using engines and distributions.

2. Why should I avoid using rand() in serious applications?

• Poor randomness quality.
• Predictable if seeded with the same value.
• Not thread-safe.
• No control over distributions.

Use <random> for better randomness and control.

3. Is <random> thread-safe?

No. You should use separate engines per thread or protect shared engines with mutexes.