Why the result of (1-0.6667)==0.3333 is zero in MATLAB? how can I get one?

Why the result of (1-0.6667)==0.3333 is zero in MATLAB? how can I get one?

when I use matlab to calculate 1-0.6667 I get 0.3333. but the result of (1-0.6667)==0.3333 is zero I even tried this one

      t=1-0.6667;
      t==0.3333

the result is again 0.

how can I get one for (1-0.6667)==0.3333?

Answer 1

[Azzi Abdelmalek]

Because 0.6667 and 0.3333 are codified and stored in the memory with a certain precision

[Azzi Abdelmalek]

use this

tolerence=0.0001
abs((1-0.6667)-0.3333)<tolerence

Answer 2

[per isakson]

See

• What Every Computer Scientist Should Know About Floating-Point Arithmetic and
• Why don’t my numbers add up?

Answer 3

[Shashank Prasanna]

That is because they are not the same.

>> t-0.3333

ans =

        5.55111512312578e-17

They can be equal by some tolerance.

>> t-0.3333 < 1e-15

ans =

       1

Answer 4

[Youssef Khmou]

hi may,

This precision problem, so your numbers are 4 decimals , i suggest to make to make your logical test by multiplying your numbers by 1E+4 :

 a=0.6667;
 b=0.3333;
 f=1e+4; % Factor
 f-f*a==f*b  %  ITS NOT f*(1-a)==f*b

Answer 5

[Jan]

This is the most frequently asked question. See FAQ: why is 0.3-0.2~=0.1

Answer 6

[Matt J]

Why the result of (1-0.6667)==0.3333 is zero in MATLAB? how can I get one?

The others have told you why. You can get "one", however, by reformulating this way,

>> (3-2)/3==1/3

ans =

       1