Ignoring IF statements error

I'm having a problem with Matlab and its interpretation of the operator "<" Here's my code:
function logo()
suma=0;
suma2=0;
k=0;
k2=0;
error=1;
error2=1;
while 1
k=k+1;
if error>=0.10
valor=suma;
suma = suma + ((-1)^(k+1)/k);
error = abs(suma-valor);
end
if error<0.10
break;
end
end
while 1
k2=k2+1;
if error2>=0.10
valor2=suma2;
suma2= suma2 + (((-1)^(k2+1))*((0.5)^k2))/k2;
error2 = abs(suma2-valor2);
end
if error2<0.10
break;
end
end
format long
fprintf("\nSe realizaron " + k + " iteraciones")
fprintf("\nEl error de la aprox. a ln(2) fue " + error)
fprintf("\nLa aprox. a ln(2) fue " + suma + "\n")
fprintf("\nSe realizaron " + k2 + " iteraciones")
fprintf("\nEl error de la aprox. a ln(1.5) fue " + error2)
fprintf("\nLa aprox. a ln(1.5) fue " + suma2)
end
It's kind of simple, I'm stating an equation inside a loop, which is going to continue repeating itself until the "error" value is less than 0.10... The problem is that I'm getting results of
Se realizaron 10 iteraciones
El error de la aprox. a ln(2) fue 0.1
La aprox. a ln(2) fue 0.64563
Se realizaron 3 iteraciones
El error de la aprox. a ln(1.5) fue 0.041667
La aprox. a ln(1.5) fue 0.41667>>
It doesn't have any sense to me, because I'm getting error=0.1, so the code just broke the loop before the error<0.1 and I don't know why, I even tried to state something like
if error<0.1 && error~=0.1
break;
end
but I'm still getting an error=0.1

3 Comments

Is that MATLAB code?:
fprintf("\nSe realizaron " + k + " iteraciones")
Also note that shadowing the inbuilt error function is a very bad idea.
Jan
Jan on 21 Aug 2017
Edited: Jan on 21 Aug 2017
@Stephen: It works with modern the string class:
s1 = string('\nSe realizaron ');
s2 = string(' iteraciones');
fprintf(s1 + 10 + s2)
>> Se realizaron 10 iteraciones
I do not have R2017a, where the creation of the string class seems to work with "...", but in R2016b the string() command might be equivalent.
@Jan Simon: thank you for the explanation.

Sign in to comment.

 Accepted Answer

Jan
Jan on 21 Aug 2017
Edited: Jan on 21 Aug 2017
I do not see any problem in your code. It does exactly, what is expected. Here a brushed up version:
suma1 = 0;
suma2 = 0;
k1 = 0;
k2 = 0;
error1 = 1;
error2 = 1;
while 1 % Or better: while error1 >= 0.1
k1 = k1+1;
valor = suma1;
suma1 = suma1 + (-1)^(k1+1) / k1;
error1 = abs(suma1 - valor);
if error1 < 0.10
break;
end
end
while 1
k2 = k2+1;
valor2 = suma2;
suma2 = suma2 + (-1)^(k2+1) * 0.5^k2 / k2;
error2 = abs(suma2 - valor2);
if error2 < 0.10
break;
end
end
fprintf('error1: %.16g\n', error1);
fprintf('error2: %.16g\n', error2);
Note that this replies 0.09999999999999998 for error1, which is displayed as 0.1 in the command window with format long. But you can check the value easily:
error1 - 0.1
and see the difference of -2.776e-17. This means that the < operator works as expected.

3 Comments

Thank you so much!
I imagined that it was going to be something related to the format I was using but I just didn't know how to display the value in terms of 10e-3 as I wanted it.
Your solution works perfectly!
@Ilse Leon: if Jan Simon's solution "works perfectly" then you should accept the answer.
You are right, done.

Sign in to comment.

More Answers (0)

Asked:

on 21 Aug 2017

Commented:

on 21 Aug 2017

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!