| |
 |
|
| Science Forum Index » Mathematics Forum » Maple bugs: Thomas Richard: Hurrah, Maple quality improves! |
|
Page 1 of 1 |
|
| Author |
Message |
| Vladimir Bondarenko |
 Posted: Fri Mar 04, 2005 5:12 am |
|
|
|
Guest
|
Hi Maple customers all over the world,
The other day Herr Richard, Maple Support, Scientific Computers
GmbH, made an attempt to persuade us that Maple quality does not
deteriorate over the versions:
http://groups-beta.google.com/group/comp.soft-sys.math.maple/msg/83505ab90dd16760
Here is yet another example selected of thousands. If you are
a freshman or a sophomore, or a math student of any course, I
believe you will be in perfect delight seeing the following
Calculus I level pearl.
What Maplesoft promise you ("We work work work to help you as
much as possible").
http://www.maplesoft.com/
Maplesoft> "Maplesoft technologies are ideal for students
Maplesoft> enrolled in any math, science or engineering
Maplesoft> course. Learn how you can use Maple to save
Maplesoft> time, learn more and improve your grades."
As you understand, the factorial of a negative number does not
exist, say draw this
[quote:e8057afa22]plot(z!, z=-5..5,-10..10);
[/quote:e8057afa22]
or try this
[quote:e8057afa22](-1)!;
Error, numeric exception: division by zero[/quote:e8057afa22]
Now try this which is also correct.
[quote:e8057afa22]sum(1/n!, n= -1..1);
Error, (in NumericRange) summand is singular in the interval of[/quote:e8057afa22]
summation
Now rejoice by the improved Maple quality (just an example of
thousands distinct ones)!
Observe that Maple quality comes so to say, in waves. This is
one of several typical ways of Maple quality deterioration we
demonstrate in our Review of Maple coming.
.................................................................
????-1994 - Maple cannot calculate the sum.
1995-2000 - The answer does not exist, but Maple returns exp(1).
2001-2003 - All works correctly.
2004-2005 - The answer does not exist, but Maple returns exp(1).
.................................................................
sum(1/n!, n= -1..infinity);
-------------------- (2004) Maple 9.5.2 -----------------------
exp(1)
-------------------- (2004) Maple 9.5 -------------------------
exp(1)
-------------------- (2003) Maple 9 ---------------------------
Error, (in sum/infinite) summand is singular at n = -1 in the
interval of summation
-------------------- (2002) Maple 8 ---------------------------
Error, (in sum/infinite) Summand is singular at n = -1 in the
interval of summation
-------------------- (2001) Maple 7 ---------------------------
Error, (in sum/infinite) Summand is singular at n = -1 in the
interval of summation
-------------------- (2000) Maple 6 ---------------------------
exp(1)
-------------------- (1997) Maple V Rel 5 ---------------------
exp(1)
-------------------- (1995) Maple V Rel 4 ---------------------
exp(1)
-------------------- (1994) Maple V Rel 3 ---------------------
sum(1/n!,n = -1 .. infinity)
----------------------------------------------------------------
The same bug manifestation with
sum(1/n!^2, n= -1..infinity);
sum(1/n!^3, n= -1..infinity);
sum(1/n!^4, n= -1..infinity);
sum(1/n!^5, n= -1..infinity);
sum(1/n!, n= -infinity..infinity);
sum(1/n!^2, n= -infinity..infinity);
sum(1/n!^3, n= -infinity..infinity);
sum(1/n!^4, n= -infinity..infinity);
sum(1/n!^5, n= -infinity..infinity);
that is the answer does not exist but Maple 9.5.2 returns
cheerfully stuff like BesselI(0,2), hypergeom([],[1, 1],1),
hypergeom([],[1, 1, 1],1), hypergeom([],[1, 1, 1, 1],1) or
infinity.
More about Maple quality degradation can be found at
Maple bugs: asympt quality degradation sample
http://groups-beta.google.com/group/comp.soft-sys.math.maple/msg/260c254f34c9bb50
Maple bugs: evalf quality degradation sample
http://groups-beta.google.com/group/comp.soft-sys.math.maple/msg/858e0c219f9a8e07
Maple bugs: rsolve quality degradation sample
http://groups-beta.google.com/group/comp.soft-sys.math.maple/msg/da8354b79b156d5f
Maple bugs: minimize(z^4-2*z^2+z, z= 0..1);
http://groups-beta.google.com/group/comp.soft-sys.math.maple/msg/2b05838db6cbdc65
Maple bugs: A dangerous, misleading ad statement at the main
Maplesoft's site
http://groups-beta.google.com/group/comp.soft-sys.math.maple/msg/8738365a6702a63e
Enjoy improved Maple quality promised us the Maple customers,
Vladimir Bondarenko
VM and GEMM architect
Co-founder, CEO, Mathematical Director
Cyber Tester, LLC
http://www.cybertester.com/
http://maple.bug-list.org/
http://www.CAS-testing.org/ |
|
|
| Back to top |
|
|
|
| A N Niel |
| Posted: Fri Mar 04, 2005 8:14 am |
|
|
|
Guest
|
One might argue that these answers are reasonable:
[quote:edb02623a7](-1)!;
Error, numeric exception: division by zero
1/(-1)!;
Error, numeric exception: division by zero
limit(1/z!,z=-1);
0[/quote:edb02623a7]
It matches other cases, like this one:
[quote:edb02623a7]1/csc(0);
Error, (in csc) numeric exception: division by zero
simplify(1/csc(x));
sin(x)
limit(1/csc(x),x=0);
0[/quote:edb02623a7]
Now one could argue that when f has a pole at a, then f(a) should be
given the value infinity, but Maple is not doing that. |
|
|
| Back to top |
|
|
|
| Vladimir Bondarenko |
| Posted: Fri Mar 04, 2005 4:05 pm |
|
|
|
Guest
|
A N Niel <ann...@nym.alias.net.invalid> writes on Fri, 04 Mar 2005
08:14:00 -0500
ANN> One might argue that these answers are reasonable:
ANN>> (-1)!;
ANN> Error, numeric exception: division by zero
ANN>> 1/(-1)!;
ANN> Error, numeric exception: division by zero
[because]
ANN>> limit(1/z!,z=-1);
0
Yep.
For example,
[quote:8d4832af98]1/(-1.0001)!;
1/(-0.9999)!;
[/quote:8d4832af98]
-.9999422723e-4
..1000057715e-3
etc
If you see the things this way, then exp(1) should be considered
a correct answer.
But in such a case, again we have a problem with
Maple 9 / Maple 8 / Maple 7 and the bugs and
the correct answers just swap.
.................................................................
????-1994 - Maple cannot calculate the sum.
1995-2000 - All works correctly.
2001-2003 - The answer exists, but Maple raise an exception.
2004-2005 - All works correctly.
.................................................................
that is again we see regression testing failure.
;)
Even worse, from your viewpoint, 1/(-1)! we consider
to be equal 0; thus
sum(1/n!, n=-infinity..-1) should be = 0, too.
What Maple 9.5.2 returns actually?
[quote:8d4832af98]sum(1/n!, n=-infinity..-1);
infinity[/quote:8d4832af98]
(*nasty smile*) |
|
|
| Back to top |
|
|
|
|
|
All times are GMT - 5 Hours
The time now is Sun Nov 08, 2009 3:47 am
|
|