Cout Question

Hi

Bellow is an example of my help system im working on:-
My question is the couts..Do they take up any cycles?

e.g

cout << "...........\n";
cout << "...........\n";

can also be written as:

cout << ".........\n" << ".........\n";

is the second one more efficent becuase it states less couts?

Thank You For Your Help

Alex

Code:

---

//C++ Help System Project
#include <iostream>
using namespace std;

int main () {
        char choice;
       
        for(;;) {
        do {
        cout << "Help On: \n";
        cout << "  1. if\n";
        cout << "  2. switch\n";
        cout << "  3. for\n";
        cout << "  4. while\n";
        cout << "  5. do=while\n";
        cout << "  6. break\n";
        cout << "  7. continue\n";
        cout << "  8. goto\n";
        cout << "===================\n";
        cout << "Choose One (q to quit): ";
        cin >> choice;
        } while( choice < '1' || choice > '8' && choice != 'q');
        if(choice == 'q') break;

---

Quote:

---

Originally Posted by macman

is the second one more efficent becuase it states less couts?

---

That's a good question, and one I'd like to know the answer to as well. I tend to think yes, but that's just an instinct. If you like the couts just because they keep the spacing nice and neat, remember that C++ ignores whitespace. Also, you might want to add a flush call at the end of all that so that you know your output has been written.

Code:

---

//C++ Help System Project
#include <iostream>
using namespace std;

int main () {
        char choice;
       
        for(;;) {
        do {
        cout << "Help On: \n"
            << "  1. if\n"
            << "  2. switch\n"
            << "  3. for\n"
            << "  4. while\n"
            << "  5. do=while\n"
            << "  6. break\n"
            << "  7. continue\n"
            << "  8. goto\n"
            << "===================\n"
            << "Choose One (q to quit): " << flush;
        cin >> choice;
        } while( choice < '1' || choice > '8' && choice != 'q');
        if(choice == 'q') break;

---

If it is, it won't be by any significant amount. Maybe a few CPU cycles for the function call, and then anything that gets executed no matter what whenever cout is called.

Think about the trade-off between reader-friendly code and CPU efficiency. Unless you're programming for really time-critical situations, err on the side of reader-friendliness.

I would agree with sean - if you're looking to optomize your code from the output standpoint, you'll notice that the less you have to output the faster your program runs. it really doesn't matter how you output it (in your case), just how much you output... for example, consider this code:

Code:

---

#include<iostream>
#include<fstream>
#include<ctime>

int main()
{
    time_t start=clock();
   
    char ch;
    std::fstream file("Untitled1.dat",std::ios::in|std::ios::binary);
   
    while(file.get(ch))
    {
        std::cout<<ch;
    }
   
    file.close();
    time_t stop=clock();
    std::cout<<"\n----------\nTime: "<<stop-start<<"\n----------"<<std::endl;
    std::cin.get();
    return 0;
}

---

as it is, it took about 375 miliseconds to run on my machine... but if I comment out the cout in the loop, it only takes about 16. my file was 20 paragraphs of lorem ipsum.


edit: it seems as if the only thing that really makes much of a difference is as how much you use either cout or operator <<. for examle, see the code below - each commented block has the time it took to run it uncommented:

Code:

---

#include<iostream>
#include<fstream>
#include<ctime>

int main()
{
    time_t start=clock();
   
    for(register short int i=0;i<1000;i++)
    {
        /* 171 miliseconds
        std::cout<<"1";
        std::cout<<"2";
        std::cout<<"3";
        std::cout<<"4";
        std::cout<<"5";
        */
       
        /* 156 miliseconds
        std::cout<<"1"<<"2"<<"3"<<"4"<<"5";
        */
       
        /* 62 miliseconds
        std::cout<<"12345";
        */
    }

    time_t stop=clock();
    std::cout<<"\n----------\nTime: "<<stop-start<<"\n----------"<<std::endl;
    std::cin.get();
    return 0;
}

---

edit2: I was kidding about the milisecond thing... clock() returns processor time...

They are both the same. << and >> are the shift left and shift right operators, the reason they work with the cin and cout classes is because of operator overloading.
An overloaded operator function is declared like this

Code:

---

<return type> operator << (<arg type>);

---

so each time you use an operator with a class you are calling a member function of that class.

>is the second one more efficent becuase it states less couts?
Performance shouldn't be your primary concern because the time it takes to call a function pales in comparison to a device write. When you're doing I/O, no amount of micro-optimization will make a difference.

I always figured that was how it went, but I never tested it until just now :D

Quote:

---

Originally Posted by Quantum1024

<< and >> are the shift left and shift right operators,

---

When talking about cout and cin they are referred to as the stream insertion and extraction operators.

Like Quantum1024 and sean_mackrory said, each use of the stream insertion operator << resolves into a new call to the overloaded cout.operator<<() function. The first use calls the function and returns a reference to the stream which can then be chained to another call to the cout.operator<<() function for second and subsequent insertions in the same line of code. If you really want to shave every possible cycle off of that output, I would suggest simply:

Code:

---

cout << ".........\n.........\n";

---

That resolves into just a single function call and would therefore be the quickest. But it probably wouldn't do that much good in reality.

<< evaluates from left to right and returns a reference to its left operand. So basically...

Code:

---

cout << "..........\n" << "..........\n"

---

Would just evaluate to

Code:

---

cout << "..........\n..........\n"

---

wouldn't it?

Quote:

---

Originally Posted by Scribbler

<< evaluates from left to right and returns a reference to its left operand. So basically...

Code:

---

cout << "..........\n" << "..........\n"

---

Would just evaluate to

Code:

---

cout << "..........\n..........\n"

---

---

that may be true, but what about the extra cycles used to make that evaulation? see my timed code above

Thanx For your replys everybody!

I think that Ill keep it as it is..Like you guys say..it would really be much of a cycle cut...and keeping the code nice and user friendly is a habbit i want to fall in!

The flush statement show second post down..what exactly does it do?

Im not sure what u mean by:-

Quote:

---

Also, you might want to add a flush call at the end of all that so that you know your output has been written.
Code:

---

Cheers
Alex

>The flush statement show second post down..what exactly does it do?
It causes and buffered characters to be written to the device.

>Im not sure what u mean by
In C, it's prudent to flush the output stream when you want interactive input because, due to buffering, the request for input may precede the prompt:

Code:

---

printf ( "Enter a string: " ); /* May not be shown first */
fgets ( s, sizeof s, stdin );

---

To be more correct, an explicit flush of the stream is required:

Code:

---

printf ( "Enter a string: " ); /* Guaranteed to be shown first */
fflush ( stdout );
fgets ( s, sizeof s, stdin );

---

In C++ this issue is mostly academic because cout is tied to cin. When you request input from cin, cout is automatically flushed.

So basically..

using the flush..is "old school?"
Is there anyneed for it at all?..or is just left over from c?

..ahhhh..wait
so where theres no cin..then i should flush?

Cheers For The Info!

Alex

Quote:

---

Originally Posted by major_small

that may be true, but what about the extra cycles used to make that evaulation? see my timed code above

---

If that evaluation is made, it would be at compile time and not at runtime.

From my understainding cin and cout are just references to objects, the function call occurs when eveluating << or >> (both of which take a stream object as their left operand and return a reference to it).

Which should mean that

Code:

---

cout << "something";
cout << "something";

---

and

Code:

---

cout << "something" << "something";

---

should execute in exactly the same time (note that the output occurs when << is executed, cout is not a function)

Code:

---

cout << "somethingsomething";

---

should take a bit less since it only requires one output operation.
But as Prelude said it is all irrelevant considering how long output takes anyway.