calculating logarithm

[wolfrog]

hi all!

does anybody out there know how to calc a logarithm with pure c/c++ code? without using any functions etc? inline assembly is also okay!
thankx!

[Sam Finch]

#include <math.h> //This is where the log function is
#include <iostream.h> //this is for the cout to display the log


int main(int argc, char* argv[])
{

double Number;
double ItsLog;

cout << "enter a number" << endl;
cin >> Number;

ItsLog = log(number);

cout "The Natrual Log of " << Number << " = " << ItsLog;

}

[parksie]

I don't think that's what he meant. He needs NO FUNCTIONS.

wolfrog - try looking in an algorithms book, and there may be something elsewhere on the Net.

[Sam Finch]

My Bad, use the McLauren Approximation.




#define ITERATIONS 1000

//raises a number to a power
double GetPower(double Number, int Power)
{
int i;
double retval = 1;

//anything raised to the 0th power is 1
if (Power == 0)
{
return 1;
}


//multipy 1 by the number n times
for (i=1;i<=Power;i++)
{
retval *= Number;
}

return retval;
}

double ln(double number)
{

int i;
double retval = 0;

for (i=1;i<=ITERATIONS;i++)
{
retval = retval + GetPower(-1,i-1) * GetPower(number-1,i) / (double)i;
// cout << i << endl;
}

return retval;
}






int main(int argc, char* argv[])
{

double Number;
double ItsLog;

cout << "enter a number" << endl;
cin >> Number;


ItsLog = ln(Number);
cout << "The Natrual Log of " << Number << " = " << ItsLog << endl;

return 0;
}


The higher the value of the ITERATIONS constant the more accurate it will be, the lower the faster it will be, just find the right balance between speed and accuracy.

[parksie]

I thought that would come up somewhere, but had a power cut and couldn't finish the code. PS, Sam - do you know what the general equation is for natural logs?

[Sam Finch]

yeah, it's that.

Before Euler(I think) twigged that one people just calculated all the powers of e and put them in a big table then read them off backwards to get logarithms.

natural logs can be found in any unhygenic public lavatories!

rofl!

[parksie]

Just read through my Stats 2 book...and found that McLauren expansion thingie. (It was in among some fish...as it were).
Yeah, I think it was the Greeks who had 7s.f. tables - must have taken a while to work out ;).

[Sam Finch]

There was me thinking that the greeks didn't use the arabic number system and used roman numberals instead. I'm pretty sure that e wasn't discovered until the 19th century(Euler again, what a guy) So I don't know what the greecs would ohave thought of decimal log tables.

[parksie]

Okay...the Greeks had cosine tables. I think logs were Napier? (I have no idea...)

If only the textbooks actually told you something. I don't know what number system the greeks used...but I doubt it was Roman. Weren't they a couple of hundred years before?