MaplePrimes - comments on Post, The Three Ms

MaplePrimes - comments on Post, The Three Ms http://www.mapleprimes.com/posts/96298-The-Three-Ms en-us 2026 Maplesoft, A Division of Waterloo Maple Inc. Maplesoft Document System Sat, 13 Jun 2026 18:25:12 GMT Sat, 13 Jun 2026 18:25:12 GMT The latest comments added to the Post, The Three Ms http://www.mapleprimes.com/images/mapleprimeswhite.jpg MaplePrimes - comments on Post, The Three Ms http://www.mapleprimes.com/posts/96298-The-Three-Ms A more in-depth comparison http://www.mapleprimes.com/posts/96298-The-Three-Ms?ref=Feed:MaplePrimes:The Three Ms:Comments#comment96299 Stefan Steinhaus has produced a more detailed comparison of some of the main mathematical software packages including the three Ms. In his latest comparison, published in 2008, he compares Maple 11, Mathematica 6, Matlab 2008a, plus GAUSS, O-Matrix, OxMetrics Ox Professional, and Scilab. http://www.scientificweb.com/ncrunch/    --- Duncan The latest comments added to the Post, The Three Ms 96299 Wed, 25 Aug 2010 00:53:24 Z DuncanA DuncanA Dead last http://www.mapleprimes.com/posts/96298-The-Three-Ms?ref=Feed:MaplePrimes:The Three Ms:Comments#comment96304 Although it's older versions, in t<a></a>he overall scores, Maple scores dead last!  I can't truly believe that.  Also although the author says to compare them on fair ground it would be interesting to know the authors background and experience in such systems.  I almost find it hard that one person would be as fluent in all 3, I almost expect there to be at least a little bias in a so called "fair" comparison.  It seems to me that the article did have a slight mathematica bias in it. It would be interesting to have 3 experts or teams pitting each of the 3M's against each other in a shootout test.  Certainly some tasks will perform better on certain systems but the actual test itself might also favour one CAS system over the other so it would be difficult to set up. I think it would be hard to set up an unbiased test but I would like to see one.  Mathematica instantly grabs public appeal to an ordinary layman because of it's strength in graphics.  I'd like to see Maple up the ante on that one.  One good thing that comes out of these tests are better future versions .. hopefully, but only if the companies are paying attention to these sorts of things.   Hopefully Maple is.    The latest comments added to the Post, The Three Ms 96304 Wed, 25 Aug 2010 04:35:45 Z Christopher2222 Christopher2222 ncrunch http://www.mapleprimes.com/posts/96298-The-Three-Ms?ref=Feed:MaplePrimes:The Three Ms:Comments#comment96305 <a href="http://www.mapleprimes.com/posts/96298-The-Three-Ms#comment96299">@DuncanA</a> It's true that the ncrunch review is more in depth. It has a slant toward computational statistics (which is fine, as that is pretty much declared explicitly).  But it is also quite wrong in places. And the benchmark code is not all implemented in a fair or near optimal way. See <a href="http://www.mapleprimes.com/posts/39909-5th-Ncrunch-Math-Program-Review">here</a> for a Mapleprimes post about it, from a few years ago. The functionality charts are also wrong in places, and misleading in some others. For example, some claims about Maple's debugger are just plain wrong. It says that Maple doesn't support FFT in more than 1D, while of course one can just use the 1D version to get a 2D calculation. And so on. It takes off points for not having dedicated commands for some things which are easily done as 1-liners (eg. I can think of at least 3 simple ways to make a "Pascal Matrix" generator in 1 line of code, so why would anyone want a routine for that!?). That's just a sample; there are more errors both large and small. And the weighting scheme for the various sections is quite arbitrary and subjective. Please don't misunderstand. It's obviously the result of a lot of hard work and effort. And the author has tried to keep it regularly updated. But it needs better review and correction by experts before publication, or at least some sort of working feedback and correction mechanism. There are more advisers/reviewers listed on the ncrunch site for Matlab and Mathematica than for Maple. (No offense is intended to anyome; more heads always do better.) Even still, I suspect that there might be errors related to each of the programs in the Review, and significant improvements possible perhaps for the posted code for others and not just for Maple. Maybe comprehensive comparisons are just too much work to do really well. An interesting alternative might be tight, narrowly focused product comparisons of individual aspects of functionality, a bit like <a href="http://www.mapleprimes.com/posts/96059-Why-Is-Maple-Last#comment96205">this</a> earlier suggestion. acer The latest comments added to the Post, The Three Ms 96305 Wed, 25 Aug 2010 04:41:15 Z acer acer ambiguous comment http://www.mapleprimes.com/posts/96298-The-Three-Ms?ref=Feed:MaplePrimes:The Three Ms:Comments#comment96306 <a href="http://www.mapleprimes.com/posts/96298-The-Three-Ms#comment96304">@Christopher2222</a> It's really not clear to what you are referring, when you write that Maple scores last. I'm pretty sure that you are referring to the review Duncan mentioned, and not to the review mentioned in the parent post. But you've failed to mention which it was that you were discussing, and with the lack of comment-threading-indentation it's ambiguous. The latest comments added to the Post, The Three Ms 96306 Wed, 25 Aug 2010 04:49:21 Z acer acer apparent http://www.mapleprimes.com/posts/96298-The-Three-Ms?ref=Feed:MaplePrimes:The Three Ms:Comments#comment96322 <a href="http://www.mapleprimes.com/posts/96298-The-Three-Ms#comment96306">@acer</a> sorry, I'm referring to the ncrunch review.  I thought I mentioned it but I also thought it was apparent. I also found the review in the computing and science and engineering a little better. The latest comments added to the Post, The Three Ms 96322 Wed, 25 Aug 2010 14:12:38 Z Christopher2222 Christopher2222