List Archives >  Maple User Group List Archive >  Archive by date >  This Month By Date >  This Month By Topic

[MUG] Question

Search email archive for  

[MUG] Question
Author: Jml    Posted: 22/02/2000 21:29:25 GMT
>> From: Hi, Well last night I download a pdf file, with a comparison between the most popular symbolic and numeric processors (Maple, Mathematica, matlac, Mathcad 2000, etc), and I was wondering why Maple V release 5.1 get on o the lowest scores in this comparison chart?? First was Gauss, then Mathematica, and near the bottom was Maple, so many people advice me to get Maple and not Mathematica, and I listen to them, but now I am a little bit disappointed (sorry for my english). Regards http://www.steinhaus-net.de/science/ncrunch/ncrunch.pdf Jose M Lasso

[MUG] Re: Question about comparison
Author: Maple Group    Posted: 25/02/2000 18:52:24 GMT
>> From: Maple Group | >> From: | Well last night I download a pdf file, with a comparison between the most | popular symbolic and numeric processors (Maple, Mathematica, matlac, | Mathcad 2000, etc), and I was wondering why Maple V release 5.1 get on o | the lowest scores in this comparison chart?? First was Gauss, then | Mathematica, and near the bottom was Maple, so many people advice me to get | Maple and not Mathematica, and I listen to them, but now I am a little bit | disappointed (sorry for my english). Regards -=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=- Date: Wed, 23 Feb 2000 15:20:55 -0500 (EST) From: Eugene Zima To: Subject: Question about comparison Do not be upset. This comparison is not fair in some sense. You can not compare symbolic and numeric tools. It is the same as somebody compares Microsoft Word and Borland C++ from the "word processing" point of view :-) Of course MS Word will win, although it can not be used] to compile C++ programs :-) The same here in this report. It reviews abilities for DATA analysis. Of course specialized system (targeted specifically for this purpose) will win, although it is useless from the formula manipulation point of view. Best, Eugene -=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=- Date: Thu, 24 Feb 2000 08:48:34 +0100 From: Heike Koch-Beuttenmueller To: Subject: Question I did not look at the newest comparisons, but on older ones. This comparison is designed mainly (heavily) for statistical and numerical applications and Maple has only some statistical functions implemented and the numerical code is not so fast. May-be the last point will be better in Release 6. If the comparison would depend for example on the solving of differential equations, the result surely would be different. With regards Heike Koch-Beuttenmueller -=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=- Date: Thu, 24 Feb 2000 09:41:58 +0100 To: From: Wilhelm Werner Subject: Question Hi Jose. Don't be disappointed! The paper seems to make a comparison from a data analysis point of view; this is definitively not Maples focus! For me Maple is an environment which enables me to handle all my symbolic problems. As to numeric problems I use Maple as a "rapid prototyper" only for moderate data sizes. For all these tasks Maple is a great tool! For data analysis/visualization in statistics -maybe- there are better tools. Regards W.Werner ---------------------------------------------------------------------- Wilhelm Werner Phone: +49-(0)7940-1306-96 FH Heilbronn/Standort Kuenzelsau Fax : +49-(0)7940-1306-20 Daimlerstr. 35 D-74653 Kuenzelsau e-mail: ---------------------------------------------------------------------- -=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=- Date: Thu, 24 Feb 2000 07:46:09 -0500 To: From: "Gerald A. Edgar" Subject: Question I do not notice the word "symbolic" in their title. Instead, they tested mostly numerical math. Not the purpose of Maple. But it is said that Maple 6 has improved numerical capabilities. ----- Gerald A. Edgar Department of Mathematics telephone: 614-292-0395 (Office) The Ohio State University 614-292-4975 (Math. Dept.) Columbus, OH 43210 614-292-1479 (Dept. Fax) -=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=- From: "Willard, Daniel Dr DUSA-OR" To: Subject: Question Date: Thu, 24 Feb 2000 08:34:41 -0500 Speaking as one person, I started with Maple because it was cheap and was programmed for my (then) computer - an Amiga. I have switched computers but stayed with Maple. The main complaint that you will find in such comparisons is speed. If speed is not a significant criterion for you, then keep maple in your list from which to choose. The only one with comparable use world-wide is Mathematica. I also have Mathcad which has Maple as its root.

[MUG] Re: Question about comparison
Author: Willard, Daniel Dr DUSA-OR    Posted: 28/02/2000 21:05:53 GMT
>> From: "Willard, Daniel Dr DUSA-OR" Go to http://www.spd.dcu.ie/johnbcos/ where you will find a great defense (defence?) of Maple. -----Original Message----- | >> From: | Well last night I download a pdf file, with a comparison between the most | popular symbolic and numeric processors (Maple, Mathematica, matlac, | Mathcad 2000, etc), and I was wondering why Maple V release 5.1 get on o | the lowest scores in this comparison chart?? First was Gauss, then | Mathematica, and near the bottom was Maple, so many people advice me to get | Maple and not Mathematica, and I listen to them, but now I am a little bit | disappointed (sorry for my english). Regards

[MUG] Question
Author: Federico Pinto    Posted: 19/07/2000 18:27:35 GDT
>> From: Federico Pinto Hello, I would like to know why does MAPLE V gives a wrong answer when performing the following integral: Int(Dirac(t),t = -infinity..0); The answer Maple V (5.1) yields is 1 (one), while the correct answer should be 1/2. This is of particular importance while dealing with boundary integral equations, since in this case the results are wrong by 100%. Thanks, Federico

[MUG] Re: Question on Int of Dirac
Author: Maple Group    Posted: 28/07/2000 19:22:48 GDT
>> From: Maple Group >> From: Federico Pinto | I would like to know why does MAPLE V gives a wrong answer when performing | the following integral: | | Int(Dirac(t),t = -infinity..0); | | The answer Maple V (5.1) yields is 1 (one), while the correct answer should | be 1/2. This is of particular importance while dealing with boundary | integral equations, since in this case the results are wrong by 100%. -=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=- From: "Willard, Daniel Dr DUSA-OR" To: Subject: Question Date: Fri, 21 Jul 2000 09:54:51 -0400 It may help to do: >interface(verboseproc=2); >print(`evalf/limit/levinu`); where the print instruction expands the program named in the error message when you attempt to evaluate the integral. -=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=- To: Subject: Question Date: Fri, 21 Jul 2000 10:33:21 -0500 From: Dale Alspach I would not consider this wrong since Dirac is not a function but a distribution and hence is really only applicable to C infinity functions. You seem to want to extend the domain to step functions. I don't think that there is any cannonical choice. Maple seems to be using the closed halfline in your integral. Note that > limit(int(Dirac(t),t=-infinity..s),s=0,left); 0 as expected. Dale Alspach ********************************************************************* Email: OR Post: Oklahoma State University Department of Mathematics 401 Math Science Stillwater, OK 74078-1058 USA Telephone: 405-744-5784 FAX: 405-744-8275 ********************************************************************* -=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=- From: "peter lindsay" To: Subject: Question Date: Sat, 22 Jul 2000 15:07:24 +0100 Hi, I note from the maples files that the Dirac function has a property that Int(Dirac(t),t = -infinity..infinity) = 1 This certainly computes in maple6 in win98. The Dirac function has a singularity at t=0 and i'm not sure that it should follow that Int(Dirac(t),t = -infinity..0) = 1/2. Perhaps it does - I just don't know. My maple6 on win98 gives Int(Dirac(t),t = -infinity..0) = 0 and, int(Dirac(t),t = -infinity..0) = 1 Cheers, Peter Lindsay -=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=- Date: Mon, 24 Jul 2000 06:28:46 +0200 From: Helmut Kahovec To: Subject: Question Hello Federico, You may consider using remember tables. For example: > restart; > int(Dirac(t),t=-infinity..0); 1 > int(Dirac(t),t=-infinity..0):=1/2; 0 / | | Dirac(t) dt := 1/2 | / -infinity > int(Dirac(t),t=-infinity..0); 1/2 With kind regards, Helmut

[MUG] Re: Question on Int of Dirac
Author: David Vaughan    Posted: 31/07/2000 14:01:03 GDT
>> From: David Vaughan The dirac delta function isn't a function in the usual sense. It gives a unit mass to an individual value: in particular, Dirac(t)=0 if t doesn't equal 0, but int(dirac(t),-infinity..infinity)=1. In effect, as long as 0 is in the domain of integration, the integral over that domain must be 1. One way that such distributions are introduced is through approximate identities, often taken as symmetric about 0. However, that is only one way to introduce it. Maple is giving the correct value of the given integral over (-infinity,0], based on the standard definition of the dirac delta function. It is best not to think of this as a function, since any function that is 0 except at one point must have integral 0 over any Lebesgue measurable set, regardless of its definition at that one point. DCVaughan Wilfrid Laurier University Waterloo, Ontario > >> From: Federico Pinto > | I would like to know why does MAPLE V gives a wrong answer when performing > | the following integral: > | > | Int(Dirac(t),t = -infinity..0); > | > | The answer Maple V (5.1) yields is 1 (one), while the correct answer should > | be 1/2. This is of particular importance while dealing with boundary > | integral equations, since in this case the results are wrong by 100%. > > -=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=- >

Previous by date: [MUG] Page break, John Schaeffer
Next by date: [MUG] Reference for Risch algorithm, Makelly
Previous thread: [MUG] funtions and other problems, Werner Burkhardt
Next thread: [MUG] Reference for Risch algorithm, Makelly