Why do PTC continue with MathCad Prime?

[sjupp]

I need to rant about how rubbish Prime is, so I am looking at the support renewals and its coming in at like £3k for my current licenses. We only use MathCAD 15 as prime's functionality is so poor compared to MathCAD 15. So what am I paying for? I sat through a 30min webinar on Prime 5.0, the only thing that they had done was add a graphing app, I had to temper my annoyance that that was the focus for the prime 5.0 release. What about the simple stuff:-

1. Why have they changed how you do subscripts! Every time I type a variable I add a subscript, this is now a two button operation. Did the developers ever open up MathCAD 15? I have been through multiple software upgrades on AutoCAD, Solidworks, Inventor etc none of them just decide to change how the user fundamental interacts with the software.

2. Why are there no formatting styles available? Prime is there to present maths in a nice consistent format so we can deliver it to our clients, did the developers forget that the output from Prime is a sheet that is often delivered as part of a client engineering pack?

3. We have hundreds of MathCAD 15 calculations that we can't convert to Prime without lots of work to re-format and sort out. This is a ridiculous place to be, no other upgrade on other software platforms has ever created so many issues.

I am sure I could keep going if I had the time, PTC have taken a market leading piece of software and rubbished it, we used to recruit based on people having used the software, this is no longer the case.

So why not forget Prime, it has such a bad reputation, it couldn't do maths when it was first released and it has continued to disappoint from release to release. Stop what you are doing PTC and look at MathCAD 15, just update that, give it tabs and nice icon colours (because that's what we all need) and sell it as MathCAD 16, I can tell you the engineering world will rejoice, and you will see customers coming back to MathCAD! Or continue with Prime and watch all you engineering clients walk away.

[Fred_Kohlhepp]

The general consensus is that PTC thought Mathcad would be a nice (simple) addition to their modeling software; since 3.0 they have advertised that it interfaces with Creo.

What they didn't realize (IMHO) was the effort that would be required and the number of stand-alone users that needed the features you've noticed are missing.

[sjupp]

For PTC to not consider those that don't have the full suite of software packages is a serious error in judgement, surely even if MathCAD is intergrated with CREO the mathCAD still needs to be checked and approved and delivered as a document to a client as proof of some engineering? Am I stuck int he dark ages where I still supply evidence in pdf format to clients that some engineering was done?

The question then becomes will they improve prime and will it be in Prime 06? because if not then i don't see why i would pay for a support contract.

[ChrisKaswer]

I believe that if you own Mathcad 15, you can use it forever without any further payments for maintenance. Mathcad 15 works on Windows 10, so as long as you have this latest version, you should be able to end future payments and use v15 until your computer systems do not support running this version or the licensing software required. I do stay on top of each release of Prime to see what has been improved/added. My wild guess would be that Prime 10 might be close enough to consider migrating from v15? Time will tell.

SMath is an option to Mathcad...interestingly very similar in UI and functionality to Mathcad v15. It will read v15 files too, but it is not without limitations. Worth a look when you consider the cost at $0! Migrating from Mathcad v15 would require very careful considerations of all pros & cons.

[Fred_Kohlhepp]

Under PTC both Prime and 15 use the same license.  They are changing that license so (I believe) you will need that altered license for 15. . .

Prime express is still free. . .

[Raiko]

Hello Fred,

you don't need prime for interfacing, since it is possible with MathCad 15 as well.

I used it and it worked. However, there was a bias from fellow engineers since the to and fro between Mahcad and Creo is not that transparent and the effort to define results that directly can be used in a design is quite high. E.g. when the result of a calculation is a shaft diameter of 23.35 mm you furthermore need code that converts the result to a more conventional dimension of, say, 25 mm.

Raiko

[ChrisKaswer]

You ask a very good question. I happen to have been the chairman of the technical committee way back in the day when PTC approached us to discuss the "re-write" of Mathcad (~11 years ago now). We understood the reason - modern software development tools and keeping-up with changes to the Windows OS and CPU technology. They wanted to improve the UI and make it more "MS Word-like". Bottom line was - they wanted to focus on the documentation aspect while rewriting it for a modern OS, thinking they would appeal to the larger audience of the general engineering staff member (the largest segment of their customer base). We told them they were missing the point. The core functionality is where this tool was at. To drive the point home I added that most every medium to large sized company most often have more senior engineers who write very useful "tools" in Mathcad that are then used by all other technical staff. Even though these people are fewer in number, their work is used and recognized by a much larger audience. I'll never forget the silence that overcame the room on PTC's side, and they had some high-level management there to make the pitch.

Given this, we told them they must maintain the functionality list first, then go after a modern UI. I cannot share how many times I was very firm in our requirements and how terrible the software would become if they didn't listen - all the members of the tech committee provided the same feedback. And here we are about 11 years later -  hey PTC, we told you so! I know of companies that have left Mathcad for other math/programming tools because there was no path to migrate Mathcad v15 files without excessive effort, and often without any way to fully duplicate the old functionality. And there still isn't. I've also told them to never change legacy behavior. I know you would all cringe if I were to share some of the conversations I've had in the past to try to make them realize how large a mistake they would make if they continued down this path. I left participation in this technical committee a few years ago, having had enough with their lack of VOC and listening skills - not only for Mathcad but other CAE tools as well.

Having said all this, I must give kudos to PTC for the connection they have developed to Creo. For their Mathcad customer base that uses Creo as their CAD tool, this offers a significant advantage. But, this does not come without issues - like trying to incorporate old v15 tools into Creo if the functionality is missing from Prime.

PTC - you had some incredible tools that were very powerful and unique, not only for the CAE world, but for the general engineering community as a large following. Unfortunately you've insisted on ignoring your core, primary customer base.

[sjupp]

Its even more disappointing to hear that they just ignored a technical committee's advice. I am all for updating software and integrating it with other packages, but as a small consultancy we tend not to link all the packages up as by the time we set-up a mathcad linked to a drawing model file and maybe linked to an analysis file we would have spent all our hrs and missed our deadline! Most of our work is designing prototypes for our clients so the value in setting up linked bits of software is not really there as we rarely reproduce the same engineering.

I first came across MathCAD when I worked at Airbus in the UK, I guess their (or similar sized companies) requirements are maybe what is driving how they develop mathcad and the ptc suite. Certainly the wider community of small engineering companies has no influence.

The legacy behaviour is the biggest issue for me, as an example, AutoCAD still has all its legacy behaviour and it allows old and new users to be just as effective in the new versions on the flip side PTC have reduced expert users in MathCAD to beginners in MathCAD Prime and at the same time given it less functionality. But we can now do graphs in Prime, just like you have always been able to graphs in excel for like almost 20yrs. who uses MathCAD to do graphs anyway??

So do you know of any decent other options to MathCAD? We will likely to just keep using MathCAD15 till it no longer works on our computers, which will hopefully be a couple of years from now!

[Raiko]

"...We will likely to just keep using MathCAD15 till it no longer works on our computers, which will hopefully be a couple of years from now!"

The same over here in Europe. Couldn't have said it better Chris.

In the meantime I'm learning MatLab. I don't like it, but it works.

Raiko

[RichardJ]

Hello Chris. Long time since we spoke.

I was on that committee, and I 100% back what Chris said. We were ignored. I go back much further with Mathcad than that committee though. I was on that committee almost since PTC acquired Mathcad, and prior to that I was part of a secret "Power users" group that advised Mathsoft. Until I was uninvited for testing of Prime 5.0, I alpha and beta tested every version of Mathcad since MC11. Mathsoft didn't always listen either, and if they had then static unit checking (the unit checking from MC12 through MC15) would never have happened. To PTC's credit, they did listen to that complaint, and Prime 1.0 reverted to the dynamic unit checking we had in MC11. But it was very clear from the beginning that PTC did not care about the existing Mathcad users (based on private conversations, extremely clear; but what was said was said confidentially, so I will say no more).  They have lived up to that, accompanied with a surprising (given how may programmers they have) degree of incompetence even implementing what they do want; the graphs in Prime 5.0 being an excellent example of that.

It's really sad, because Mathcad had enormous potential to be a game changing piece of Math software. Heck, it was a game changing piece of Math software! But it could have gone so much further if a company other than PTC had purchased it

[PhilipOakley]

Hi all, I'm doing a bit of V15 worksheet refinement, at home, in my own time, and thought I might look at the community to see what's been happening.

I've been championing Mathcad at my work (Defence Aerospace, multi-site, many thousand engineers here in UK..) for many years, and I don't see us transferring to Prime anytime soon. I'm retiring soon, so I'm having to make notes about the current situation and hand over the reigns. Plus a colleague had asked about when/if we might use Prime.

Mathcad has been loosing ground as a systems (explore design concepts) tool to Matlab. While PTC is well embedded as the mechanical design/draughting tool, it will be a while before the locally installed versions (of Creo/proe/windchill etc), and their users, are ready for Mathcad integration (I'm thinking 5-10 years). Even then, the 'draughting' processes probably doesn't really fit well with the Mathcad capabilities of old. In Defence, the drawing must be independent of each other when complete...

I see PTC Prime as being an attempt to auto-document the basic arithmetic behind many design values that are on engineering drawings, but fails to appreciate the depth and diversity of computation that may be behind those 'simple' calculations.

The power of Mathcad is its instant live whiteboard 'textbook' mathematics (for which MathSoft got the patent - Is it still valid?) - I'm using that capability right now which is why I continue to use and champion it. That live mathematics was coupled with the ability to compute things that were 'too hard' to programme, and that continues to be the case. For example, computing integrals is hard, except in a very few specialist packages, and the list goes on - even the direct visualisation of the formulae simplifies the difficulties of the mathematical working..

Prime wants to be a cross between Excel, and Word Equation Editor, and it just does not hit the nail on the head.

It would be great if PTC actually got behind Mathcad's strengths and extended from the base it had, but I don't think the ship is for turning.

Like I said, give it another 5-10 years, and maybe...

In the meanwhile I'll try and get the Angle pseudo-dimension better understood, and included in dimensional analysis.

Philip

[RichardJ]

Hello Philip. Long time, no see

The Mathsoft patents (all the early ones, anyway) must have expired by now. So it is only a question of who will pick up where Mathsoft left off. Obviously, not PTC. I have started to use SMath for calculations if I know it has the tools I need to do them. If not, then I use MC15. Maybe a commercial venture will provide a realistic alternative at some point. I'm a little surprised that nobody has jumped into the market gap already.

Best wishes for your retirement  

[PhilipOakley]

Hi Richard,

Thanks for the good wishes.

If I had to guess, it could be Matlab who pick up the slack. They already own Symbolic engine (MuPAD but that's not mentioned in https://en.wikipedia.org/wiki/List_of_computer_algebra_systems), and also have the Simulink graphical environment, so with a little bit of tweaking they could easily  auto generate the computational flow graphs and do the dependency management. (remembering that each mathcad region has an anchor, which is a paired I/O point, and the flow is in page reading order, so easy to link as if a simulink design)...

They have the JIT compilation, multi-dimentional arrays, they can do temporal and 'block' flow (think of a group of equations as a block, instead of needing to be function), they think in terms of vectors and arrays (Ok so thay can't think past one-based indexing but..). They have a large market, and a solid academic base. They have a much larger base of special functions and multi-valued return functions (I had to write the 'unique' function yesterday!).

The partial conversion of a Mathcad worksheet XML schema to mathlab should be pretty easy.  But then again, that expectation that Matlab may pick up the 'live white' board part  maybe familiarity with the capabilities of the Matlab eco system, rather than it's business aims. It would need to start putting figures directly into the canvas, rather than stacking them as 'windows'.

I'll need to check out how I'm going to get or maintain access to MathCAD after retirement, which could be the big stumbling block.

I wish you and all colleagues well in the ongoing use of this great tool.

Philip

[RichardJ]

---

@PhilipOakley wrote:

I'll need to check out how I'm going to get or maintain access to MathCAD after retirement, which could be the big stumbling block.

---

Check out SMath Studio. It has a very similar interface to Mathcad, will read many Mathcad files, handles units, etc. It doesn't do everything Mathcad can do, but it does a lot, and it's free.

[Fred_Kohlhepp]

---

@PhilipOakley wrote:

I'll need to check out how I'm going to get or maintain access to MathCAD after retirement, which could be the big stumbling block.

---

Mr. Oakley,

Having just retired (and lost my company sponsored Mathcad), I can speak to this.

• Prime can be downloaded for "trial."  After 30 days it becomes Prime Express (AKA "stupid.")  It will read all Prime files (of versions before it; I have Express 4.0, so Prime 5 can't be seen.  And it is capable of being useable, although the choice of "advanced features" is often confusing.  (the function "mean" won't work, but integrals still evaluate. . .)
• As for Mathcad 15, SMath will read old version Mathcad files and will also compute them (as capabilities allow.)  So we can continue to be involved for free, if at a rerstricted level.

I have enjoyed and appreciated your contributions.

Enjoy your retirement!

Fred

[RichardJ]

Enjoy your retirement Fred

Overall, I think SMath is better than Prime Express.

[Fred_Kohlhepp]

---

@RichardJ wrote:

Enjoy your retirement Fred

Overall, I think SMath is better than Prime Express.

---

Thank you!

As a calculating engine you're probably right, but then I'd miss this fascinating dialog!  

[GrumpyOldTom]

Just a general reply to the thread, a few things I have recalled for reading it - I think a tiny amount of the "rewrite" impetus was to go to a 64 bit capable code.  But the OP by Chris was a real eye opener.  It is mind-blowing to me that (1) a decision that poor was made and further, in the long years since Mathcad 15, they haven't come close of achieving even the (nearly irrelevant) goals they set - other than Creo integration, apparently - we use Solidworks so I guess again we're out in the cold.

 I posted screenshots in another thread showing the difference between my Mathcad 15 screen and the 5.0 screen.  The ribbon interface is quite useless and a waste of valuable vertical space.  But even the ribbon tabs aren't correct, if you make the thing less than full screen, the tab names are truncated because they don't adapt to the change in width.  And as to documentation, text alignment between, say arrays and labels doesn't occur, the array values don't align to the grid, you cannot set margins except for their 3 presets, I could go on and on.  So as a calculation documentation application, it falls flat on its face.

I feel bad saying these things.  I love Mathcad, and I feel terrible for the coders - I well know how easy it is to criticize code versus write reasonably "idiot-proof" code.  But my glory, Prime 5.0 isn't bad, it's awful. Especially for the tremendous amount of time it has taken to bring it this far.

We have used Mathcad here since MS-DOS days and have a tremendous load of legacy calcs.  I just hope somebody can get this ship righted, but Chris's post doesn't give me a lot of hope.  I said in the other thread that it appears there is something systemically wrong at PTC, and his post confirms that.

[tslewis]

You can align text and math regions if you start by using a text block - not a region - and insert math regions in this.

90% of the tome you can keep everything tabbed correctly horizontally. Use the remove space option to create correct
Spacing.

I find the inserting math regions in a text block region is very powerful

[GrumpyOldTom]

If I get a chance, I may try some of these.  But I must admit that at this point my agreement is with RichardJ - It's a workaround, a kludge.  It'd take some exploring for me to figure out how it could be a better documentation tool than 15 with the various issues we've talked about here - for instance, I see the graphics add-in they've (finally) rolled out in 5.0 and for what I've seen I think that it looks more cartoonish or business board room than academic/physics/engineering professional , but that could be cosmetics that can be addressed.

Like I said, I really want Prime to succeed.   But yow!  It's been a loooong time, and we still aren't back to 15 yet, in many many ways.

[SL_9674672]

For this thread, I don't think it is a matter of finding work arounds. I shouldn't have to find a way around something that MC15 did effortlessly. We shouldn't have to work around to get the same functionality out of Prime.

[Werner_E]

---

@SL_9674672 wrote:
For this thread, I don't think it is a matter of finding work arounds. I shouldn't have to find a way around something that MC15 did effortlessly. We shouldn't have to work around to get the same functionality out of Prime.

---

You are right that Prime SHOULD have (at least) the same functionality as real Mathcad and in a perfect world it would. But fact is, that is hasn't and presumably never will have. So you just have three options:

(1) Switch over to a different software or combination of softwares

(2) Stay with real Mathcad as long as its possible and then ...

(3) Use Prime and try to compensate the missing functionality with various work-arounds as good as possible

And if you're not a one-man-company but working for a larger company, chances are that you don't have any option at all as it will be forced on you. Bad luck if its option (3).

[jabel-2]

Except -- MatLab will never accept 0 based indices 🙂

[gfraulini]

Sorry guys but I don't understand the point.
I use MC since 10 years. I moved on to Prime only with Prime 2 (I tried the first but I left when I saw that the potential elevation with "^" didn't work with the italian keyboard...).
Graph customization apart (Ive to see the Prime 5 yet) I am able to operate the same things both on MC15 that Prime.
I am a mechanical engineer; maybe if you work on electric/telecomunication/IT sector, you have some lacks. Could you make some example?

[PhilipOakley]

Hi G.,

Before PTC bought MathSoft/Mathcad, it had a very large user base of 'consultant' level users who used mathcad at a higher level that would be expected in a typical drawing office (DO) type environment. This included a lot of professional engineers and scientist types who were pushing the leading edge of their applications in a local sense and needed to extend the maths that was in their text books to the new environment and scenarios.

A lot of that work fed down to being 'regular' computation, and in many areas matched up with 'standard' calculations used in DO environments e.g. Work Instruction Sheets (WIS) giving the formula for thread engagement based on load and screw size, for both imperial and metric - we had lots of them covering all the regular arithmetic calculations. These computations were done in the DO and needed no more than a calculator.

Meanwhile the advanced design concepts and specialist engineering were done in different departments, such as computing the photon flux from a black body source over a TBD waveband and deciding on a lens f/# and aperture. Those specialist computations need more than a calculator if they were to be done correctly, and human error being what it is (roughly equally distributed..), a method for capturing the complicated equations and doing the calculations accurately, to leave the thinking for the specialists, was very important. Mathcad leapt that hurdle and still does. It can also do the WIS computations, but there was an established process method for them, so wasn't picked up that well by the DO fraternity..

Meanwhile PTC saw that, especially in the USA, there is a requirement to 'show all working' for the computations both at WIS and Consultant level, and that Mathcad was the tool every one was using for the documentation of the computations and values used in drawings. PTC bought Mathcad to bring the WIS level capability to their DO fraternity (hence Prime), but had not realised that they had also bought the troublesome consultant level community who still need all the power (and more) of Mathcad, even if no-one individually will use all of it.

I have quite a few Mech Engs, especially those that are in Stress and Thermal analysis, who do some quite powerful analyses via mathcad, and would like (and benefit from) the integration with Creo a lot sooner that 'the other crowds' who are at arms length to the rest of the PTC tool set.

One example that came up on the old collaboratory was vibration absorption modelling via [a few] cascaded spring dampers. These give adequate solutions but aren't continuous, because the overall net frequency roll off can't be modelled by extending the damper cascade (there is always ripple on the curve as different dampers take up the load), but having the ability to investigate within Mathcad did give a much greater insight into how to approach the problem, and where the text books reach their limit. If I locate the sheet I'll post it as it was public.

I hope that helps clarify why (in my eyes) the split in the community has happened. My colleague who manages the Creo side says that PTC do this every time they buy a company - the assimilation process to be swallowed into the PTC code base taking typically 5 or more years.

[RichardJ]

---

@gfraulini wrote:

I am a mechanical engineer; maybe if you work on electric/telecomunication/IT sector, you have some lacks. Could you make some example?

---

Lucky you. As a mechanical engineer I guess you never used any of the features Prime doesn't have. "electric/telecomunication/IT sector" is a rather narrow view of the rest of the Mathcad world though. How about optical engineer, physicist, geologist, biochemist, etc etc etc? Suppose you need to manipulate images in Mathcad (I have several worksheets that do this; output from an optical engineering package)? Well, in Prime you are hosed because there is no image viewer, and the 3D graphs are simply pathetic (I have not tried the ones in Prime 5.0). I also use controls in my worksheets, and sometimes they are not just a luxury. Prime doesn't have controls. It also doesn't have postfix or infix operators, both of which I use. You still can't set the exponential threshold for results (unless that changed in 5.0?). Still no superscripts or subscripts in text. For math software that is ridiculous! And then there is  the stuff that is just annoying. Like the solve block format. In Mathcad 15 you can write many solve blocks on one line. In Prime you need a big chunk of the page. And the ribbon interface sucks. What I can access in MC15 with one click, in Prime I have to hunt and peck around for in the ribbon. It's a huge time waster.

I could go on (easily: no error tracing in programs, no text or math styles, can't paste a vector graphic into the worksheet...), but by now I think you get the general idea.

[tslewis]

1. You can do subscripts in text. You define a text region or block, define a math region in there, and add the math text with Subscripts etc.
2. You can change math fornatttkng globally on worksheet through labels or individually by selecting the math regions and specifying the formatting. Same for text. To specify globally position cursor in white space of canvas and specify math and text formatting.

Prime is miles better than 15 for presentation of documents/reports.

I agree there is a major issue with 1. No spell checking which is ridiculous 2. No ability to custom define the header foooter and margin widths

[RichardJ]

---

@tslewis wrote:
1. You can do subscripts in text. You define a text region or block, define a math region in there, and add the math text with Subscripts etc.

---

That is saying that you can't do subscripts, but there is a workaround. I think the best way to look at the subscripts/superscripts problem in text is not that it is a problem (although it is), but rather that it is a symptom of a much bigger problem. This is considered basic text formatting in any serious documentation tool. In a math documentation tool it was inexcusable that it was not available in Prime 1.0. After all these years such a basic feature is still not available. That speaks volumes about Mathcad development at PTC.

[Raiko]

Hi

　

I'm a mechanical engineer myself and Mathcad Prime is useless for my line of work. I can't do everything with Prime I'm achieving with MathCad 15. Specifically:

• Most of my work involves pressure vessel calculations according to different standards and evaluation of measurement data from test. Every now and then I through a thermal assessment in.
• Over the years I have generated a library of MC documents that I can use directly for documentation saving me the hassle to type it twice (I export the calculations, equations, variables and annotations as rtf and insert it in MS Word). (There are controls in these documents that change colour when given limits are exceeded pointing directly to design problems. Impossible in prime.
• I have concatenated the single documents to collect results for a short report when I can't be bothered with an extensive one. Again something I can't do in prime.
• Data evaluation is a big topic for me and pointless without proper graphs. Although MC15 graphing capabilities are a bit wizened they're still way beyond what prime can do. And prime is slow!
• Integration to Creo is also possible with MC15. I had it with prime when during a webinar (about the benefits of integrated Mathcad files in the Creo object tree) PTC wasn't sure whether one needs to open always Creo in order to access the embedded prime file or not.

The only useful feature of prime that I know of is mixed arrays. I.e. for me when I'm doing thermal assessments or data evaluations. MathCad 15 has its short comings but is still far more capable than MC prime for my work as a mechanical engineer.

Raiko

[GrumpyOldTom]

Raiko,

You have my sympathies.  We also have a compiled DLL library of "user functions" written in C, or C+ or C++ or maybe even C+/- (?!) I don't know... point is, it runs far slower linked with Prime than with 15.  The thing that gets all of us, I think, is not only the shortcomings, but the fact that something like a decade has passed and the it still looks/acts in some ways like a beta release.  Things like the tabs not resizing right; only 3 margin choices, only 2 grid choices.  I cut and paste from 15 into Prime 5.0, I get an image, not working math.  If I delete that immediately by hitting my delete key, it goes away; if I click away and click back into it, my delete key no longer deletes the image, I have to go up on that blasted Microsoft-induced ribbon and pick around until I find "Delete Region".  It's like loose ends that aren't tied up.

Now as to the cut and paste, I know the data format is different; but for Pete's sake they've had 10 years to come up with "Paste from Mathcad 15".  And I shoudn't have bothered saying that, I'll bet I could find the same sentiment expressed repeatedly on these boards.

They want to be a calculation documentation interface, but multi-line text and vector elements don't align with each other or the grid, and while I know that may be a tough nut to crack, it's been 10 years... 10 years, and that aspect still acts precisely like Mathcad 15.  Maybe I'm supposed to be using tables.  Maybe there are some things I don't understand about Prime that I should understand, and I wouldn't have these gripes.  Maybe there is a magic font or pair of fonts where the kerning lines you up with one of the grids.  I don't know.  But it makes documentation just that much more difficult.  Say I output a 4 element vector and I want to put labels beside the elements, just text labels, explaining the values.  If I use multi-line text, it doesn't line up with the vector elements; if I use individual text boxes, the vector elements aren't lined up with the grid, so that doesn't work either.

And it's sad, like I keep saying, I hate lobbing rocks at this program - there are coders at PTC trying to do a good job, and believe me I know how hard it is.  But it seems like either there aren't enough of them and/or (as the OP states) management sent them down the rabbit hole after the wrong objectives.  Mathcad 15 is my go-to program, period.  It is a wonderful platform for day-to-day engineering calcs - I've even done small FEA's in it, and haven't begun to test the depth of it's mathematical prowess.  This kind of math interface is why we have computers; they could sell the daylights out of this thing.  But it don't work!