changing the value of x in c++

sssalsa · #1 02-Feb-2005, 12:54

iv written a program 2 calculate the value of a function f(x), and then written one so that the computer calculates it a different way for negative x.
how do i get c++ 2 do the following:
if the user inputs a negative number, c++ ignores the negative sign and reads the value of x and puts this into the program iv used to calculate f(x).
hope you understant what i mean.
this is my program so far. i just need 2 change the second part but i dont kno how:

CPP / C++ / C Code:

#include <iostream.h>
#include <math.h>

int main()
{
	double x;
	cout << "Enter x: " << endl;
	cin >> x;

	if (x>=0) {
		cout << "f(x)= "
			<< x + (sqrt((x*x)+1))
			<<endl;
	}
	else {
		cout << "f(x)= "
			<< 1/((sqrt((x*x)+1))+x)
			<< endl;
		return 0;
	}
}

crystalattice · #2 02-Feb-2005, 13:08

C++ doesn't require '.h' in the #include statements, so you can lose those extensions if you want. You'd only use it if you wanted to force the use of the original C headers.

One way to do it is use the absolute value function to get the absolute value of 'x', then continue on w/ the 'else' statement. Alternatively, you could use parenthesis to force the program to multiply the negative values, thereby creating a positive.

Example: Assuming 'x' is a negative number, ((x)*(x)) should give you a positive x^2 value. I don't remember offhand, but I believe C++ also has a 'power' function, allowing you to not have to manually multiply x*x.

LuciWiz · #3 02-Feb-2005, 14:36

Quote:

Originally Posted by crystalattice

I don't remember offhand, but I believe C++ also has a 'power' function, allowing you to not have to manually multiply x*x.

Yes, it's called pow and it is located in the math header.

Regards,
Lucian

davekw7x · #4 02-Feb-2005, 15:04

Quote:

Originally Posted by sssalsa

how do i get c++ 2 do the following:
if the user inputs a negative number, c++ ignores the negative sign and reads the value of x and puts this into the program.

For variables of type double, the function is fabs. Try the following with negative values for x and for positive values of x:

CPP / C++ / C Code:

#include <iostream>
#include <cmath>

using namespace std;

int main()
{
  double x;
  double y;

  cout << "Enter x: " << endl;
  cin >> x;
  y = fabs(x);
  cout << "x  = " << x << ", y = " << y << endl;
  return 0;
}

Regards,

Dave

davekw7x · #5 02-Feb-2005, 15:44

Quote:

Originally Posted by crystalattice

... you could use parenthesis to force the program to multiply the negative values, thereby creating a positive.

Example: Assuming 'x' is a negative number, ((x)*(x)) should give you a positive x^2 value. I don't remember offhand, but I believe C++ also has a 'power' function, allowing you to not have to manually multiply x*x.

Parentheses do not change the results unless they are used to override the built-in operator precedence:

CPP / C++ / C Code:

#include <stdio.h>
#include <math.h>

int main()
{
  double x, y, z;

  x = -12.34;
  y = x * x + 1;
  z = (x)*(x) + 1; /* same as y, I'll bet */

  printf("x = %f, y = %f, z = %f\n", x, y, z);

  y = sqrt(x*x + 1);
  z = sqrt((x)*(x) + 1); /* same as y, I'll bet */

  printf("x = %f, y = %f,  z = %f\n", x, y, z);

  return 0;
}

And yes, you could use pow(x, 2) in place of x*x, but why would you want to do that?

Regards,

Dave

crystalattice · #6 02-Feb-2005, 18:28

Ha ha, you got me.

I remember now why I suggested the parenthesis idea: I was thinking of my calculator that multiplies just the numbers unless you enclose the negative sign and number in parenthesis. If you didn't, the answer would always be negative. I forgot that computers don't have that issue. Sorry

I suggested the power function to make the program more portable. Just thought it might be easier. <shrug>

Max Payne · #7 02-Feb-2005, 20:57

for integer data type,use abs

CPP / C++ / C Code:

int x=-11;
z=abs(x);

for floating number,use fabs

CPP / C++ / C Code:


double x=-11;
float y=-12;

z=fabs(x);
z=fabs(y);

WaltP · #8 05-Feb-2005, 15:25

Quote:

Originally Posted by crystalattice

I suggested the power function to make the program more portable. Just thought it might be easier. <shrug>

Nothing is more portable than not using a pre-defined function. x*x is much more portable because it will work in any general computer language. ;-)

davekw7x · #9 05-Feb-2005, 16:10

Quote:

Originally Posted by WaltP

Nothing is more portable than not using a pre-defined function. x*x is much more portable because it will work in any general computer language. ;-)

I agree with your sentiment (I think), but my head is spinning:

I'll buy this:
"Anything is more portable than not using..."

I'll buy this:
"Nothing is more portable than using..."

But:
"Nothing is more portable than not using..."

Oh, well...

By the way, the function double pow(double, double) is in the C standard library (and, therefore the C++ standard library), so its use is portable between different platforms and/or different compilers for C and/or C++ (the only portability I am concerned with on this forum).

I have always been a little leery of using this with negative values for the first argument, since there is a domain error if the first argument is negative and the second argument does not have an integer value. Since both arguments are floating point numbers (doubles), I always have roundoff error in the back of my mind. (Try it with something like pow(-2.0. 2.000001), for example).

Since there is no roundoff error in the second argument of the expression pow(x, 2), it's technically OK in this application.

My question of, "why would anyone want to do this instead of just using x*x," was based on my defensive approach to floating point numbers where roundoff error could be disastrous, and also to laziness (fewer keystrokes in "x*x"). Why do something (using pow() for squaring a number) that will probably be OK most of the time and requires more typing?

Regards,

Dave

crystalattice · #10 11-Feb-2005, 14:48

Well, I was thinking that if the program needed to be modified for use w/ powers beyond 2 or 3, then maybe using the power function would be easier to use. That way you don't have to start counting and keeping track of "x's" and running into logic errors.

I suppose for a quick and dirty program that probably won't be used for anything besides practice it doesn't matter. But isn't it wise to think ahead for problems like that? Don't want another 64k barrier or Y2K fiasco. ;-)

Thread Tools	Search this Thread
Show Printable Version Email this Page	Search this Thread: Advanced Search
Rate This Thread
Rate This Thread:

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Changing monitors on a Linux machine	JdS	Computer Software Forum - Linux	10	23-Dec-2004 10:49
Problems with changing display resolution	vsseym	Computer Software Forum - Windows	2	27-Jul-2004 11:01
Pointer values changing unexpectedly	spudtheimpaler	C Programming Language	11	04-Mar-2004 17:37
Changing a line in a file	Allowee	Computer Software Forum - Linux	1	10-Jan-2004 04:09
Changing font while using getstr	BobbyDouglas	MySQL / PHP Forum	1	03-Dec-2003 23:08