Cyber Tester's VM machine challenges human beings the best Maple Experts (and anyone who wants to compete)
Vladimir Bondarenko
Hello the best Maple Experts worldwide,
http://shell.cas.usf.edu/~eclark/maple_links_experts.html
(A remark on Maple kernel development growing instability.)
Show at least one input in Maple 11 where the most innocuous
change 1-sin(z)^2 into cos(z)^2 changes the Maple output.
(S h a m e f u l !)
*************************************************************
*** ***
*** Beware, Maple kernel gets more and more fragile. ***
*** ***
*************************************************************
A comment for general public.
We do not want to feed Maplesoft -- which turned us down,
repeatedly, as beta testers -- for free, and will publish
our VM machine unique results AFTER Maple 12 is released.
Cheers,
Vladimir Bondarenko
VM and GEMM architect
Co-founder, CEO, Mathematical Director
http://www.cybertester.com/ Cyber Tester, LLC
http://maple.bug-list.org/ Maple Bugs Encyclopaedia
http://www.CAS-testing.org/ CAS Testing
D Herring
> Show at least one input in Maple 11 where the most innocuous
> change 1-sin(z)^2 into cos(z)^2 changes the Maple output.
> (S h a m e f u l !)
Um, no its not. For starters, I would expect that Maple returns a raw
1-sin(z)^2 "unsimplified". Therefore, why would I be surprised when
more complex expressions show differences? Are they not
mathematically equivalent? Are the differences not due to branch cuts
or the like?
Does your "CyberTester" do something other than mechanically generate
problems (e.g. start with an equation, differentiate it, then see if
the CAS integration returns the original equation) and compare results
across the different CAS implementations?
> A comment for general public.
>
> We do not want to feed Maplesoft -- which turned us down,
> repeatedly, as beta testers -- for free, and will publish
> our VM machine unique results AFTER Maple 12 is released.
A comment for VB.
We never really used Maple much; we always had the impression that
Mathematica was better. What's your obsession, other than sour grapes
over not getting a job? You can't seriously be interested in
promoting computer algebra -- have you ever tried helping a project
(such as Axiom) which openly accepts contributions?
Royally yours,
Daniel
Vladimir Bondarenko
http://groups.google.com/group/sci.math.symbolic/msg/bd084f2a9c4bb64f
Hello,
Thank you for your comments. Both you and me share love
and even maybe passion to mathematics and computer algebra
systems, and this is encouraging. The powers of sons of
men is ability to get in touch and team up, - this is the
Avenue we all advance along.
First, on a personal note I'd like to say that I fully
agree with you that some, and maybe even many readers
perceive our activity as a kind of digression or noise,
- and we are sorry about it.
Our hope is that later at least some of them could see
the things in a bit different light, who knows?
One of the reasons behind the Cyber Tester way is that
CAS bugs reported publicly have more impact on a CAS
manufacturer, and the customers (including us) have a
better chance of these defects to be fixed.
Even much more.
Divine high-performance computer algebra systems in
the shape they will take in just 10-20 years are the
keys to many 21th century challenges. Take it or leave
it, to put it plainly, we all would hardly survive
without them.
The way we selected is not our temporary whim. It is
directed by our goals. Please consider this link
http://www.cybertester.com/intro.php
DH> For starters, I would expect that Maple returns
DH> a raw 1-sin(z)^2 "unsimplified".
Yes, right you are.
DH> Therefore, why would I be surprised when more
DH> complex expressions show differences?
I should have stressed it in an explicit fashion,
but, definitely, we meant the cases when the outputs
Maple 11 produce for f(1-sin(z)^2) and f(cos(z)^2)
are mathematically *not* equivalent (!).
DH> Does your "CyberTester" do something other...
Cyber Tester refers to our company. I believe you
meant the VM machine. And for it our answer runs,
Yes. To implement what you've mentioned would only
require a single grad student, or maybe even a
gifted sophomore.
Gradually, we will tell the general public more
about us.
DH> We never really used Maple much;
We use Maple systematically for 15+ years, and
are much disturbed with its continuing quality
deterioration over the last several years. And
this is yet another reason why we are doing
what we are doing.
DH> we always had the impression that Mathematica
DH> was better.
Historically, Mathematica 3 was an application
with an overwhelming amount of bugs. Unlike
Mathematica 1.2 and 2.2. But Wolfram Research
engineers were able to fix thousands of these
defects in Mathematica 4.
DH> Mathematica was better.
With its sweep, easiness-to-use and level of
correctness, Mathematica 6 is a clear challenge
to the very Maple existence.
By pressing Maplesoft badly we are trying to
help it to survive at the symbolic market.
DH> What's your obsession,
Didn't you mean "mission"?
DH> other than sour grapes over not getting a
DH> job?
This is an obvious misunderstanding. We never
tried to get a job at Maplesoft. If you do not
believe us, please write to Maplesoft and ask
them.
However, we did offer Maplesoft a project
directed on getting Maple development process
both more streamlined and cost-effective.
http://www.cybertester.com/sad-cost.php
DH> have you ever tried helping a project
DH> (such as Axiom) which openly accepts
DH> contributions?
Install Axiom. Type in )credits. Find
Vladimir Bondarenko.
Or have a look at this link
http://axiom-wiki.newsynthesis.org/AboutAxiom
DH> have you ever tried helping a project
DH> (such as Axiom) which openly accepts
DH> contributions?
We are a small company. Right now we earn
money from the orders having nothing to
do with computer algebra. Our resource
dedicated to computer algebra systems is
quite scarce now, and we are focused now
on Maple, Mathematica, MuPAD, Derive, and
TI-Inspire.
We have solid grounds to think that none
of the non-commercial CAS will be able
ever produce to a general CAS user a
level of easiness-to-use and the scope
the commercial system do. This is why we
are currently focused on the commercial
systems.
However, an extensive QA help to Maxima
and AXIOM is on our plans as soon as we
will have enough resource.
I agree with you that having some free
CAS application is important to keep the
symbolic calculation market in a balanced,
sustaining state.
Royally yours,
Vladimir
--
Vladimir Bondarenko
VM and GEMM architect
Co-founder, CEO, Mathematical Director
http://www.cybertester.com/ Cyber Tester, LLC
http://maple.bug-list.org/ Maple Bugs Encyclopaedia
http://www.CAS-testing.org/ CAS Testing
Vladimir Bondarenko
the human beings?
With what score? Blank?
Until now we still see not a single input from
any Maple Expert or anyone else showing a case
when Maple 11 outputs are NOT mathematically
identical for f(1-sin(z)^2) and f(cos(z)^2).
Сourage, my lads! :) Use your bean. :))
Do not surrender the VM machine without a fight!
Think and experiment.
(Let me encourage you a bit, there are *more*
than 1 distinct such case :)
Cheers,
Vladimir Bondarenko
VM and GEMM architect
Co-founder, CEO, Mathematical Director
http://www.cybertester.com/ Cyber Tester, LLC
http://maple.bug-list.org/ Maple Bugs Encyclopaedia
http://www.CAS-testing.org/ CAS Testing
Robert Israel
> Will our VM machine win this challenge against
> the human beings?
>
> With what score? Blank?
>
> Until now we still see not a single input from
> any Maple Expert or anyone else showing a case
> when Maple 11 outputs are NOT mathematically
> identical for f(1-sin(z)^2) and f(cos(z)^2).
You mean like this?
f:= x -> 0^(x-cos(z)^2)-0^(x-1+sin(z)^2);
--
Robert Israel isr...@math.MyUniversitysInitials.ca
Department of Mathematics http://www.math.ubc.ca/~israel
University of British Columbia Vancouver, BC, Canada
Vladimir Bondarenko
Could you please tell more about your case? I am not
quite sure what do you mean about it.
I meant something "more natural", like the integrals
int(f(z), z= a..b); where if you replace 1-sin(z)^2
with cos(z)^2 Maple 11.02 yields mathematically
distinct (not identical) outputs.
Say,
1.694426170-0.*I
Error, (in assuming) when calling 'limit/dosubs'. Received: 'numeric
exception: division by zero'
Best wishes,
Vladimir Bondarenko
VM and GEMM architect
Co-founder, CEO, Mathematical Director
http://www.cybertester.com/ Cyber Tester, LLC
http://maple.bug-list.org/ Maple Bugs Encyclopaedia
http://www.CAS-testing.org/ CAS Testing
On Jan 27, 4:16 pm, Robert Israel
dr_w_...@yahoo.com
by integrals or limits. Any example of
the described kind counts.
On Jan 27, 9:06 pm, Vladimir Bondarenko <v...@cybertester.com> wrote:
> Thanks for your question!
>
> Could you please tell more about your case? I am not
> quite sure what do you mean about it.
>
> I meant something "more natural", like the integrals
> int(f(z), z= a..b); where if you replace 1-sin(z)^2
> with cos(z)^2 Maple 11.02 yields mathematically
> distinct (not identical) outputs.
>
> Say,
>
> 1.694426170-0.*I
> Error, (in assuming) when calling 'limit/dosubs'. Received: 'numeric
> exception: division by zero'
>
> Best wishes,
>
> Vladimir Bondarenko
>
> VM and GEMM architect
> Co-founder, CEO, Mathematical Director
>
> http://www.cybertester.com/ Cyber Tester, LLChttp://maple.bug-list.org/ Maple Bugs Encyclopaediahttp://www.CAS-testing.org/ CAS Testing
>
> On Jan 27, 4:16 pm, Robert Israel
>
>
>
> <isr...@math.MyUniversitysInitials.ca> wrote:
> > Vladimir Bondarenko <v...@cybertester.com> writes:
> > > Will our VM machine win this challenge against
> > > the human beings?
>
> > > With what score? Blank?
>
> > > Until now we still see not a single input from
> > > any Maple Expert or anyone else showing a case
> > > when Maple 11 outputs are NOT mathematically
> > > identical for f(1-sin(z)^2) and f(cos(z)^2).
>
> > You mean like this?
>
> > f:= x -> 0^(x-cos(z)^2)-0^(x-1+sin(z)^2);
> > --
> > Robert Israel isr...@math.MyUniversitysInitials.ca
> > Department of Mathematics http://www.math.ubc.ca/~israel
> > University of British Columbia Vancouver, BC, Canada- Hide quoted text -
>
> - Show quoted text -
isr...@math.ubc.ca
> Thanks for your question!
>
> Could you please tell more about your case? I am not
> quite sure what do you mean about it.
What do I mean about it?
> f:= x -> 0^(x-cos(z)^2)-0^(x-1+sin(z)^2);
> f(1-sin(z)^2);
-1
> f(cos(z)^2);
1
> I meant something "more natural", like the integrals
> int(f(z), z= a..b); where if you replace 1-sin(z)^2
> with cos(z)^2 Maple 11.02 yields mathematically
> distinct (not identical) outputs.
What could be more distinct than -1 and 1?
Where did you say you wanted integrals? OK, then try
int(f(1-sin(z)^2),z=a..b) and int(f(cos(z)^2),z=a..b).