19-Apr-2026 20:31:21

octave_mistake_test():
  MATLAB/Octave version 11.1.0
  Test octave_mistake().

octave_mistake_test01():
  Set X(I) = 1 / ( I - 5 ) for I = 1 to 10
  but set X(5) to zero.

  Now let's print out X!

  X(1) = 0.000000
  X(2) = 0.000000
  X(3) = 0.000000
  X(4) = 0.000000
  X(5) = 99.000000
  X(6) = 0.000000
  X(7) = 0.000000
  X(8) = 0.000000
  X(9) = 0.000000
  X(10) = 0.000000

  That was NOT what we expected!

  Now let's print out X!

  X(1) = 0.000000
  X(2) = 0.000000
  X(3) = 0.000000
  X(4) = 0.000000
  X(5) = 99.000000
  X(6) = 0.000000
  X(7) = 0.000000
  X(8) = 0.000000
  X(9) = 0.000000
  X(10) = 0.000000

  That was NOT what we expected!

  Now let's print out X!

  X(1) = -0.250000
  X(2) = -0.333333
  X(3) = -0.500000
  X(4) = -1.000000
  X(5) = 99.000000
  X(6) = 1.000000
  X(7) = 0.500000
  X(8) = 0.333333
  X(9) = 0.250000
  X(10) = 0.200000

  By this time, the right answer is NOT what we expected!

  Now let's print out X!

  X(1) = 1.000000
  X(2) = 1.000000
  X(3) = 1.000000
  X(4) = 1.000000
  X(5) = 99.000000
  X(6) = 1.000000
  X(7) = 1.000000
  X(8) = 1.000000
  X(9) = 1.000000
  X(10) = 1.000000

  Now we see that the assignment statement is not assigning x(i)!

octave_mistake_test02():
  Try to create the -1, 2, -1 tridiagonal matrix.
a =

   2   0   0   0   0
   0   0   0   0   0
   0   0   0   0   0
   0   0   0   0   0
   0   0   0   0   0


octave_mistake_test03():
  v = [ cos ( alpha ), sin ( alpha ) ] fails.
alpha = 1.0472
v1 =

   0.5000   0.8660

ca = 0.5000
sa = 0.8660
v2 =

   0.5000   0.8660


  The statement "v3 = [ cos ( alpha ), sin ( alpha ) ]"
  will cause a MATLAB error:
  Error using cos.  Not enough input arguments.

octave_mistake_test04():
  y = abs x fails.
x = 123.46
ans = 123.46
ans = 120
ans =

    98    97   110    97   110    97


octave_mistake_test05():
  Versions of 1./A.
B =

   0.5000   1.0000
  -1.0000   0.5000

C =

   0.5000   1.0000
  -1.0000   0.5000


  "D = 1. / A" will cause a crash.
  So we suppress it.

octave_mistake_test06():
  v=rand(3) does NOT return a vector.
  So vdotv-v'*v does NOT return a scalar!

  v = rand ( 3 )  %  NOT a 3-vector!
v =

   0.906591   0.099854   0.429669
   0.879761   0.969766   0.753609
   0.738685   0.117400   0.113430


  vdotv = v'*v  %  NOT a scalar!
vdotv =

   2.1415   1.0304   1.1363
   1.0304   0.9642   0.7870
   1.1363   0.7870   0.7654


octave_mistake_test():
  Normal end of execution.

19-Apr-2026 20:31:21