Yet another funny bug in Maple 13

Vladimir Bondarenko

unread,

Jun 13, 2009, 3:18:35 AM6/13/09

to

Enjoy.

> int(1/z^(1/(sqrt(2)*(cos(Pi/12)-sin(Pi/12)))), z= 0..infinity);
> int(1/z^(1/(sqrt(2)*(cos(Pi/12)-sin(Pi/12)))), z= 1..infinity);

0
0

Better still,

> evalf(int(1/z^(1/(sqrt(2)*(cos(Pi/12)-sin(Pi/12)))), z= 2..infinity));
-.2500000000

Cheers,

Vladimir Bondarenko

Co-founder, CEO, Mathematical Director

http://www.cybertester.com/ Cyber Tester Ltd.

----------------------------------------------------------

"We must understand that technologies
like these are the way of the future."

----------------------------------------------------------

Vladimir Bondarenko

unread,

Jun 13, 2009, 3:49:10 AM6/13/09

to

Even better.

Maple 13 cannot calculate

int(1/z, z=1..2);

if it is written as

int(1/z^(sqrt(2)*(cos(Pi/12)-sin(Pi/12))), z= 1..2);

while a not-totally-stupid math freshman knows it is just
ln(2)...

Mate

unread,

Jun 13, 2009, 7:55:38 AM6/13/09

to

But Maple 12 too is unable to compute it! I wonder why your cyber
machine did not detect this earlier.

Axel Vogt

unread,

Jun 13, 2009, 9:10:36 AM6/13/09

to

Vladimir Bondarenko wrote:
> Enjoy.
>
>> int(1/z^(1/(sqrt(2)*(cos(Pi/12)-sin(Pi/12)))), z= 0..infinity);
>> int(1/z^(1/(sqrt(2)*(cos(Pi/12)-sin(Pi/12)))), z= 1..infinity);
>
> 0
> 0
>
> Better still,
>
>> evalf(int(1/z^(1/(sqrt(2)*(cos(Pi/12)-sin(Pi/12)))), z= 2..infinity));
> -.2500000000

Another example of testing only incomplete and hastily:

The first bug at least involves int(z^c,z) = z^(c+1)/(c+1)
for c <> 1 and int(1/z,z) = ln(z), which is a well known
issue. So cross-check on numerical examples and even some
variable integration bounds would show it. You are trying
to hide it.

The last one is simple lame testing: in this case you not
integrate numerical, but evaluate the symbolic result. As
you work with standard Digits=10 rounding off gives that,
but using higher precision would show other results.

So again you are hiding the actual problem and do only an
incomplete testing.

Surely you know that. And I hope you could do better.

Vladimir Bondarenko

unread,

Jun 13, 2009, 10:26:32 AM6/13/09

to

Dammit, who told you the VM machine did not detect it earlier,
years ago?

Do you realize that this bug exists for 10+ years?

Musatov

unread,

Jun 13, 2009, 7:02:46 PM6/13/09

to

> Do you realize that this bug exists for 10+ years?- Hide quoted text -
>
> - Show quoted text -

Musatov * Bondarenko | Riemann Hypothesis / Matlab Seed Proof ...May
27, 2009 ... Martin Musatov wrote:> Vladimir Bondarenko wrote:> > >
Hello,> >> > - PolyGamma[0, 1 + I]> > - PolyGamma[0, 1 - I]> > - ...
http://groups.google.com/group/comp.theory/browse_thread/thread/5d24e88347ab4462/e11cb5ba86ec6a80?lnk=raot

Math Forum Discussions Martin Musatov wrote:> Vladimir Bondarenko
wrote:> > Hello,> >> > - PolyGamma[0, 1 + I]> > - PolyGamma[0, 1 - I]>
> - PolyGamma[0, 3/2 + I]> > - PolyGamma[0 ...
http://mathforum.org/kb/thread.jspa?messageID=6727340&tstart=0

Discussions - comp.theory | Google Groups Musatov * Bondarenko |
Riemann Hypothesis / Matlab Seed Proof. Search: Login 1. Create
Account 2. Login 1. File Exchange 2. Newsgroup 3. Link Exchange 4. ...
http://groups.google.com/group/comp.theory/topics?start=10

Math Forum Discussions - User Profile for: marty.musato_@_mail.com15),
Musatov * Bondarenko | Riemann Hypothesis / Matlab Seed Proof Posted:
May 27, 2009 5:04 AM, by: Martin Michael Musatov ...
http://mathforum.org/kb/search!execute.jspa?userID=577254&forceEmptySearch=true

Discussions - sci.math | Google Groups An innovator called Musatov
wants to bolster Sci-math, 6 new of 6, Vladimir Bondarenko (5
authors), 5:35am. intrinsic or geodesic metric, 3 new of 3 ...
http://groups.google.com/group/sci.math/topics?gvc=2

Math Forum Discussions Jun 13, 2009 ... Vladimir Bondarenko Posts:
4478. Registered: 12/4/04. Re: An schizophrenic called Musatov wants
destroy Sci-math ...
http://mathforum.org/kb/message.jspa?messageID=6750433&tstart=75 - 4
hours ago

Saifee Durbar: His Life, His Accomplishments, His Dreams - soc ...May
18, 2009 ... From: Vladimir Bondarenko <v...@cybertester.com> ...
Martin Musatov P=NP Freedom Fighter fighting in Panama with the rebels
since 1993. ...
http://groups.google.com/group/soc.culture.turkish/browse_thread/thread/5ee60e755ca39b98/f34c8d87c34acc95?lnk=raot

Math Forum Discussions Re: An ignorant cruel person "Ludovicus" wants
to destroy Musatov Posted: Jun 12, 2009 6:59 PM ... Vladimir
Bondarenko. 6/13/09 ...
http://mathforum.org/kb/message.jspa?messageID=6749940&tstart=75 - 4
hours ago

Re: If Cantor set is uncountable and measure 0, then (Cantor)^c ...me
(2)Inbox Musatov * Bondarenko | Riemann Hypothesis / Matlab Seed Proof
May 29 me (2)Inbox Re: WTF Is Going On Here?! May 29 ...
http://sci.tech-archive.net/Archive/sci.math/2009-06/msg01450.html

Math Forum Discussions Jun 13, 2009 ... Re: An Innovator called
Musatov wants to overhaul Sci-math. Posted: Jun 13, 2009 10:54 AM ...
Vladimir Bondarenko. 6/13/09 ...
http://mathforum.org/kb/message.jspa?messageID=6750719&tstart=75 - 4
hours ago