Example M15 vs MP7

[-MFra-]

.... I am overwhelmed with pain .....

.... and now how do I do it?!?!?!

[LucMeekes]

Mathcad 11/Maple:

can do without an assumption.

How does Prime7 fare without the 'assume n=integer' ?

Luc

[-MFra-]

[LucMeekes]

I'd say, report this crime.

Success!
Luc

[ChrisKaswer]

I agree, and on this note, given the "sun setting" of all previous versions later this year, I think there should be a significant push on our part (user community) and a "pull" on PTC's part to compile a list of truly important fixes/changes to make Prime as close as possible to a "worthy" successor of Mathcad 15. I have a very small and growing list of things myself that I believe need to be addressed. I'm building this list based on my very limited use of Prime 7 so far (and limited use for the full set of capabilities Mathcad offers).

To start some ideas off, why don't we use this thread to post issues we find in Prime 7 that should be addressed prior to the sun set event, and ask PTC to refer to this message board for guidance. Comments, thoughts?

[Cornel]

I think we should gather the whole community and send an email to PTC, each of us, hoping that maybe they will feel that it is not ok for users to endure things like this. Maybe this way they will recover somehow ...
I don't quite understand why they don't consider the good things that exist in Mathcad 15 and Mathcad 11 and integrate them into Mathcad Prime.
It's okay for PTC to continue and develop just Mathcad Prime (or whatever it would be called in the end) for the future, it's not a problem, but to take into account the good things implemented in the two so far and to put these things together and to further develop.
I don't understand why it's so hard for them to realize that.
It's in the rooster's mind.
Either they are somehow constrained by something not to take this step yet, or they are following something ...
But whatever it is, it becomes clear that things are not going as well as they suggest and with what they put on the market.
However, in this style, they can't earn too much.It's sad and disappointing

[LucMeekes]

Do you think such efforts weren't undertaken over the past more than 10 years?

It's no use yelling to a train running off. It has no ears, is stubborn enough to keep running it's course and will stop only at stations it wants to.

Replace train with PTC...

Luc

[Werner_E]

I fully agree with Luc.

But of course you may report this bug if you feel better then.

The bug itself is newly introduced in Prime 7 in an attempt to improve the mediocre symbolic engine. The new (free) engine was chosen because now PTC must not pay license fees to Mathworks for using their muPad.

See here how things have changed since Prime 6. As you can see the result of the new symbolic is useless, but at least its not wrong

[Werner_E · Accepted solution]

ADDENDUM

By replacing the factorials by the gamma function I get the very same WRONG result in Prime 6, too!

Here is another "nice" experience I had while playing around (still in P6) - its reproducable

[-MFra-]

Similar complaints have been posted in the community for years. It seems to me that Werner and others, from time to time, report it to PTC but without getting anything although, if you check, some old errors (see photo) have been corrected, but not all.

There is a more complete list.
((Note that all other software suffer from similar errors but not the same)...)

[Werner_E]

---

@-MFra- wrote:

Similar complaints have been posted in the community for years. It seems to me that Werner and others, from time to time, report it to PTC

---

Yes, I reported a few bugs in the past, but I stopped doing so a long time ago as I don't think that it makes any sense. It seems that PTC is resistant to advice  (or unwilling?). You can also tell it from the fact that in contrast to some other companies they don't encourage their customers to turn in bug reports. Actually there is no official way of doing other than opening a support ticket (only open to customers who pay their maintenance fee regularly). But I don't need support, I'd just need a simple and uncomplicated way to inform the development team about bugs in their software, hoping that they are able to fix them for the next release or release an intermediate bug fix release. I have given up hope ...

[Fred_Kohlhepp]

You all need to recognize several things:

1. Mathcad has never had a truly strong symbolic engine.  Originally it was a subset of Maple, and some smart Maple users realized that there was some stuff left "under the hood" that could be used if you knew how to call it.  But Mathcad NEVER claimed to be a first-line symbolic solver.
2. When Mathsoft (remember them?) replaced Maple with MuPad there was a general hue and cry and a general belief that it was a step back.  That MuPad engine has remained (I believe) virtually unchanged, and carried through the PTC purchase.  Note that the owners of MatLab now own MuPad.  I suspect that if they were to keep the old (out-of-date) MuPad engine, PTC would both have to pay money and they would lose support.
3. Mathcad (even Prime, even 7) continues to be a very good numerical solver; and it is the only one (discounting SMath and other minor efforts) that does not require programming type languages and can be read by a person with reasonable knowledge of hand-written mathematical notation.
4. Under PTC, it ain't likely to change.  PTC bought Mathcad believing it was a cute little software calculator with a neat trick appearance that could be matched with their modeling software.

[-MFra-]

Undoubtedly it is a dull and counterproductive behavior, adverse to the wishes of the users who have chosen it and placed trust in it for decades. 

[FredLusk]

MP5 works.  So I guess the new symbolic solver isn't actually a solver, symbolic or otherwise. ☹️

[Fred_Kohlhepp]

MP5 is the old, MuPad engine.  MP6 has (I believe) the new engine AND the old MuPad engine and you may choose and switch between them.  MP7 has only the new in-house built engine.

MPE (mathcad Prime Express) is FREE and quite impressive for free:

• It has no exasperating symbolic engine 
• It will not write or execute a program
• It doesn't have an annoying, buggy, frustrating, and s l o w  third party graphing utility.  (So, while you can't really do 3D graphing, you CAN (thanks to contributions from some ingenious adventurers in Mathcad) actually put gridlines, including log lines, on a plot. Labels are still up to you.
• It poses interesting quizes: "How do you take the mean or median of a vector of values?"  (The built-in functions are disabled "premium" features; roll your own.)
• While it will solve for roots of an equation, it does not write or execute solve blocks

Can anyone remember (Mathcad 5?) when Mathcad first began to accept programs?

[StuartBruff]

---

@Fred_Kohlhepp wrote:

Can anyone remember (Mathcad 5?) when Mathcad first began to accept programs?

---

I think it was Mathcad 6.0 Plus, Fred. but not standard M6.

I believe it was only available in Professional versions of M7 and M8 as well, so it was a "premium" feature even back in the Good Old Days!   I guess M11 spoiled us ...

Stuart