Markiyan Hirnyk

Markiyan Hirnyk

7243 Reputation

25 Badges

12 years, 3 days

MaplePrimes Activity


These are Posts that have been published by Markiyan Hirnyk

I submit a bug through MaplePrimes because I can't do it as usually (Hope some people understand me.). Let us consider

with(LinearAlgebra):
M := Matrix(5, 5,  (i, j) -> (10*i+j)*sin((1/180)*Pi*(10*i+j))):
MatrixInverse(M);
 #One sees a long and wrong output instead of the warning "Matrix M is singular"

Indeed,

Digits := 500; evalf(Determinant(M), 495);
                               
                           1.3 10 ^(-488)   

Bug_in_MatrixInverse.mw

I submit a bug through MaplePrimes because I can't do it as usually (Hope some people understand me.). Let us consider

restart; pdsolve([diff(u(t, x), t, t) = diff(u(t, x), x, x), u(t, 0) = 0, u(t, Pi) = 0]);
pdsolve([diff(u(t, x), t, t) = diff(u(t, x), x, x), u(t, 0) = 0, u(t, Pi) = 0], generalsolution);
u(t, x) = Sum(sin(n*x)*(_C5(n)*cos(n*t)+_C1(n)*sin(n*t)), n = 1 .. infinity)
u(t, x) = Sum(sin(n*x)*(_C5(n)*cos(n*t)+_C1(n)*sin(n*t)), n = 1 .. infinity)

The question arises: what do these outputs mean? I don't see any explanation in ?pdsolve and ?examples,pdsolve_boundaryconditions. What are _C1(n) and _C5(n)? Under which conditions does the above series converge?

Moreover,

pdetest(%, [diff(u(t, x), t, t) = diff(u(t, x), x, x), u(t, 0) = 0, u(t, Pi) = 0]);
                           [0, 0, 0]

I think the above is simply a fake: it is possible to differentiate  a series only under certain conditions.

Bug_in_pdsolve.mw

Please, don't convert my post to a question. This is not correct and fair. Hope some people understand me.

Try More/GeneratePDF  in  the menu under  a post/question. See screen_26.09.17.docx as an example of a result. Also Adobe Acrobat Reader fails with it. That was submitted to MaplePrimes staff through the Contact  button at the bottom of this page. I obtained no feedback from them.

I'd like to present the following bugs in the IntTutor command.

1. Initialize

Student[Calculus1]:-IntTutor((1+cos(3*x))^(3/2), x);

then press the All Steps button. The command produces the answer (see Bug1_in_IntTutor.mw)

(4/9)*sqrt(2)*sin((3/2)*x)^3-(4/3)*sqrt(2)*sin((3/2)*x)

which is not correct in view of

plot(diff((4/9)*sqrt(2)*sin((3/2)*x)^3-(4/3)*sqrt(2)*sin((3/2)*x), x)-(1+cos(3*x))^(3/2), x = 0 .. .2);

One may compare it with the Mathematica result Step-by-step2.pdf.

2. Initialize

Student[Calculus1]:-IntTutor(cos(x)^2/(1+tan(x)), x);

In the window press the Next Step button. This crashes (The kernel connection has been lost) my comp in approximately an half of hour (see screen2.docx). One may compare it with the Mathematica result Step-by-step.pdf .

Indeed,  "We wanted the best, but it turned out like always" .

The is and coulditbe commands of Maple are known to be buggy.
Here are some math inventions done by these commands in Maple 2016.2.

restart; assume(x::real, y::real);
is(exp(x+I*y) <> 0);
                             false
coulditbe(exp(x+I*y) = 0);
                              true
coulditbe(exp(x+I*y) = infinity);
                              true
coulditbe((x+I*y)^2 = infinity);
                              true

It should be noticed that

is((-infinity)::real);
                             false

though

exp(-infinity+0*I);
                               0

The latter means

limit(exp(x),x=-infinity);
                                   0

, no more and no less.

1 2 3 4 5 6 7 Last Page 1 of 17