Back to List

Have Your Pi and Eat it Too: A Comparison of C# and C++ Performance

John Ptacek John Ptacek  |  
Mar 14, 2013
 

​In honor of National Pi Day, we thought it would be fun to calculate Pi and talk about performance differences between C# and C++. There is some math coming your way (yeah, MATH!), so feel free to skip to the end for a discussion on performance if math isn’t your thing.

So, back in the day, when things where WAY different, Pi was calculated using a mathematical series. One of the more famous ones uses the arctan math function to calculate the value of π. This is called the Leibniz formula for π and is represented by:

Now, this is one of the more inefficient ways to calculate π, it takes about 10 million terms to get an accuracy of 7 decimal places. For those of us asking for performance differences between languages though, that is great!

We have written two programs to calculate π, one in C# and one in C++. They are both syntactically similar and use the same algorithm. We tried to get as close to an apples to apples comparison as we could, but in comparing two different languages, while we did our best, you are still comparing an apple to an orange.

So, what did we see? We calculated π to 15 decimal points, 3.141592653589793 for those of you keeping track at home. Both programs took almost 100 million iterations to converge … 99,995,330 to be exact. We ran each of the programs 10 times to get an average as shown in the following table:

The calculation of π using C++ is 35% faster that it’s corresponding C# code. This is to be expected. Many modern languages such as C# and Java run through a managed runtime. This brings MANY efficiencies to developers. However, the runtime tends to add overhead when it comes to performance.

The pragmatic folks reading this realize that the benefits of a managed execution environment and modern language generally outweigh performance gains from a language like C++. There are reasons that C# and Java are used instead of C++ because developers can be more productive. Our example here, using a very slow numerical algorithm and running it for ten million iterations is a stretch for most scenarios. The 2.5 second difference for such an intensive calculation, while relatively large in our example, in the grand scheme of things is not that big of a difference.

So why bring it up? It is important to think through performance more and more in today’s computing environment for two reasons. The first is cost. As organizations move to utility-based computing with hosted services, such as Amazon Web Services (AWS) and Microsoft Azure, they are paying for their compute cost. If you can realize performance gains that reduce CPU cycles it has a direct impact on the bottom line. The second reason is that mobile is taking over the world, and mobile users want more battery life, not less. Fewer CPU cycles on a smartphone or tablet will result in better battery life, which makes everyone happy.

It will be very rare that most organizations would use C++ or other “closer to the metal” languages for their Line of Business (LOB) applications. The moral of the story though is for intensive processes in cloud-hosted or mobile environments, you want to be razor focused on optimizing your code for performance, no matter what language you choose.

Both code examples are included below, so for National Pi day, make a circle, calculate some π, eat some pie and enjoy!

C++

// LifeOfPi.cpp : Defines the entry point for the console application.
//
 
#include "stdafx.h"
#include <iostream>
#include <iomanip>
using namespace std;
 
void pi( double *pi_value, int *iterations); 
 
int main ( void ) 
 
      int iterations = 0;     // number of actual loop iterations in pi
      double pi_value = 0;    // computed value of pi
 
      clock_t cBegin = (double)clock () ;
      pi(&pi_value, &iterations); 
      printf("   %0.15f with %d iterations\n",  pi_value, iterations); 
      clock_t cEnd = (double) clock() ;
 
      std::cout << float( cEnd - cBegin )
 
 
return 0; 
 
}   
 
void pi( double *pi_value, int *iterations) 
      int i; 
      int k=1;
      double realPi = 3.141592653589793;
      double epsilon = 0.00000001;
      double delta=10;
      *pi_value= 0;
      while (delta>epsilon) {
            *pi_value += pow(-1,(k+1))/(2*k-1);
            delta = fabs((4* *pi_value)-realPi);
            k++;
      }
      *iterations = k;
      *pi_value = 4 * *pi_value;
 
 
      return; // indicate program ended sucessfully
 
} // end fucntion main
C#
using System;
using System.Collections.Generic;
using System.Linq;
using System.Text;
using System.Threading.Tasks;
 
namespace LifeOfPiCSharp
{
    class Program
    {
        static void Main(string[] args)
        {
            int iterations = 0;
            double pi_value = 0;
 
            DateTime startTime = DateTime.Now;
            pi(out pi_value,out iterations);
            DateTime endTime = DateTime.Now;
            Console.WriteLine(pi_value.ToString() + " in " + iterations);
            TimeSpan diff = endTime.Subtract(startTime);
            Console.WriteLine(diff.Seconds + "."+diff.Milliseconds);
        }
 
        public static void pi(out double pi_value, out int totalIterations)
        {
            double realPi = 3.141592653589793;
            double epsilon = 0.00000001;
            double delta = 10;
            int k = 1;
           
            pi_value = 0;
            while (delta > epsilon)
            {
                pi_value += Math.Pow(-1, (k + 1)) / (2 * k - 1);
                delta = Math.Abs((4 * pi_value) - realPi);
                k++;
            }
            totalIterations = k;
            pi_value = 4 * pi_value;
 
        }
    }
}



 

Love our Blogs?

Sign up to get notified of new Skyline posts.

 


Related Content


Spring 2019 Kentico User Group
Apr 17, 2019
Location: Waukesha County Technical College - Pewaukee Campus - 800 Main Street, Pewaukee, Wisconsin 53072 - Building: Q, Room: Q361
Blog Article
Using Hooks to Reactify a Plain JavaScript Library: A Walkthrough
Andrew PetersenAndrew Petersen  |  
Aug 06, 2019
React Hooks make it really easy to wrap a vanilla JavaScript library with a React component so you can easily reuse it throughout your app and stay in "React Mode".In this walkthrough I'll be focusing on a single library, Shave.js, but the techniques and ideas should be applicable...
Blog Article
How to Create a Readable and Useful Bug Report
Blayne RoselleBlayne Roselle  |  
Jun 25, 2019
Creating a bug that is both readable and provides enough detail is a must-have skill for a Quality Assurance Analyst. Not only will it help when it comes time to retest, but it also provides credibility with your development team. In the content below, I will share the best practices for creating...
Blog Article
Azure Tips & Tricks: Application Insights Snapshot Debugger
Todd TaylorTodd Taylor  |  
May 21, 2019
A painful memory that is burned into my developer-brain is a production support issue for a .NET web API that I wrote years ago. I worked for a large retailer at the time, and the bug was preventing electronic pricing signs from displaying the most up-to-date price for hundreds of products at...
Blog Article
Thinking Outside the Application Development Box with Unity
Jeff WeberJeff Weber  |  
May 14, 2019
Do you or your company have an idea for an application that sits a little outside your comfort zone? Does your idea possibly require game-like graphics, Augmented Reality, Virtual Reality or similar technology? If so, Unity might be worth a look.   I’ve been using Unity in my spare...