mathcad worth it? 4

[wrantler]

I have been using excel forever. It works for me...but I keep seeing these mathcad sheets and am intrigued by how easily laid out the sheets are. But then I seen the price of $700/year!

I guess I am looking for opinions on if it is worth the cost. These subscription licenses are making it hard for the little guy to make a business. I thought mathcad was a one time license for life..

On a side note has anyone used SMath? It seems like a free mathcad clone.

 

[Sparweb]

M14's graphing is marginally better than SMath's, I think.

What? Have you discovered something I haven't found in Smath? What add-in am I missing?
The standard SMath plots and "NPlot" are dreadful.

 

[skeletron]

@IFRs: The only backup I would do with the MathCAD library is print it to PDF in case you need to recreate. I also had a perpetual MathCAD 15 license that was removed (they don't support it anymore).

 

[GregLocock]

Sparweb- maybe the point is that for /decades/ Mathcad's plotting abilities were subpar. Incidentally 2.52 still works in DosBox. I could practically touchtype in that, before they changed all the shortcuts.

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

http://eng-tips.com/market.cfm?

 

[Sparweb]

Oh, maybe I just got used to it after so long (I'm using MC13 but cut my teeth on MC5).
I never really thought of Mathcad's graphs as subpar, just less flexible than Excel (but I didn't expect Mathsoft to contend with Microsoft).

The comparison I was trying to draw attention to was actually the horrible graphs in SMath. Especially customization of a graph, which takes forever, and it can all be ruined by a careless roll of the mouse wheel. The hanging arguments that absolutely cannot be hidden clutter it up. It all detracts from my ability to make SMath look like "A well formatted engineering report".

 

[wrantler]

It is sounding like mathcad is lacking in their support department. Which if true is hard to swallow since the license fee is pretty steep.

I have dabbled in SMath for a bit and it seems OK. If the paid mathcad version is not heaps better I can not see paying the steep license fee..

 

[IRstuff]

As I said before, I use Excel for Mathcad graphing, if I need presentation quality. The only graphing facility in M14 that's competitive with Excel is the M14 3D graphing, which is runs circles around Excel's, which is no longer interactive

TTFN (ta ta for now)
I can do absolutely anything. I'm an expert!

https://www.youtube.com/watch?v=BKorP55Aqvg

faq731-376 forum1529 Entire Forum list

http://www.eng-tips.com/forumlist.cfm

 

[IDS]

You can do pretty good 3D plots in Excel (with the aid of a little Python):

[URL unfurl="true"]https://res.cloudinary.com/engineering-com/video/upload/v1664925594/tips/Framezoom_tucuxi_hk5wwp.mp4[/url]

More details at:

https://newtonexcelbach.com/2022/09/30/3d-plots-with-the-latest-matplotlib/

Doug Jenkins
Interactive Design Services

http://newtonexcelbach.wordpress.com/

 

[OlafTheBlack]

Don't know where you get the BlockPad subscription price of $700. Their site says a professional license for one year is $166 and a permanent professional license for $950.

I've been using this program for about two years. I love the way it has textbook formulas that calculate. I really love that you can call these formulas in spreadsheet tables of tabulations.

It has a working word processor that allows me to document my calculations.

 

[271828]

I used to use Mathcad, but gave it up when I needed to start paying for it.

I've been using Excel for easy stuff, and Octave (free program like Matlab) for anything that needs programming.

There's a lot more to readability than whether you can see the equations. Spacing things out reasonably, code references, text to explain what you're doing, values of variables shown near an equation, etc. are more influential.

When I get a Mathcad sheet from a colleague, I usually have trouble following it. An equation on the fifth page will use variables that are defined on earlier pages. I go back to the second page and find where the variable is defined. However, it might be redefined on the third page. Many variable names are used repeatedly, like d, L, A, etc.

I used Smath for a while. It's a bit more clunky than Mathcad. Perhaps it's totally unfounded and I might be paranoid LOL, but I'm a little concerned about cyber-security in today's environment. Thus, I dropped all Smath and uninstalled it.

 

[IRstuff]

An equation on the fifth page will use variables that are defined on earlier pages. I go back to the second page and find where the variable is defined. However, it might be redefined on the third page. Many variable names are used repeatedly, like d, L, A, etc.

That would seem to be a problem in any program, including Excel; the added potential non-transparency of the unit conversions only exacerbates the problem. Additionally, an Excel computation would require TWO printouts, one with the results and the second with the actual cell equations used. And, unless you were diligent in giving each input and constant a name, trying to remember whether B3 is the same as "d" is likewise fraught.

TTFN (ta ta for now)
I can do absolutely anything. I'm an expert!

https://www.youtube.com/watch?v=BKorP55Aqvg

faq731-376 forum1529 Entire Forum list

http://www.eng-tips.com/forumlist.cfm

 

[fel3]

271828…

"When I get a Mathcad sheet from a colleague, I usually have trouble following it. An equation on the fifth page will use variables that are defined on earlier pages. I go back to the second page and find where the variable is defined. However, it might be redefined on the third page. Many variable names are used repeatedly, like d, L, A, etc."

I mostly agree with your criticisms, but they should be directed at your colleagues and not at Mathcad. Whether it's Excel or Mathcad, some (many?) users seem incapable of creating easily-usable and easily-understandable documents. It takes additional thought and effort, but it's not too difficult to apply good organization, good formatting (including paying attention to how the document will look when printed), good documentation, etc. The math and calculation logic may or may not be difficult and it's up to the user to understand if it's beyond their capabilities or not. Unfortunately, not every user has the requisite level of self-awareness.

I have seen excellent Excel and Mathcad documents from others and I have seen terrible ones. Most I see are somewhere in between. For example, years ago I received an Excel spreadsheet from a colleague. The purpose of the spreadsheet was to estimate future water demands in various communities served by an extensive rural water system. The math and logic wasn't too difficult. However, among various errors, my colleague had entered every negative number as text (some areas were forecast to reduce water usage) and most of his summation ranges didn't match the extent of the data, etc. Interestingly, after correcting all of his errors, I discovered that his totals weren't too far off what they should have been. Thank goodness for compensating errors.

I have no colleagues at the moment who use Mathcad, but every so often I receive a Mathcad document as part of a contractor's submittal, including structural calculations for buried precast concrete vaults, for a set of chemical storage tanks, for polyethylene water storage tanks, etc.. I will politely say that my review of said documents almost always includes a certain amount of training for the vendor's engineer. These engineers routinely do what you object to, including reusing variable names, and some things you didn't mention such as redefining the same variable at the top of each page (instead of just letting its previous definition carry through), mixing the units lb, lbm, and lbf and getting nonsensical results (why didn't they catch it before it went out?), adding incorrect units to empirical equations, etc. In two cases, I actually called the vendor's engineer to discuss with them how to better use Mathcad. Both engineers were very young and I think they appreciated my help With one of these engineers, the next submittal I received from them about a year later was MUCH improved.

BTW, the solution to avoiding redefined variables is to use subscripts, such as d₁, d_A, etc.

============
"Is it the only lesson of history that mankind is unteachable?"
--Winston S. Churchill

 

[271828]

IRstuff and fel3, I agree with your comments for the most part. It's certainly easy to generate poorly designed worksheets in Excel also.

There's one tiny improvement that would make all the difference in the world, whether we're talking Mathcad or Excel: Show the variable right above the equation. None of my colleagues do this. I've even brought it up before, and they say "one can go back and find where ... was defined." I think people think their Mathcad is easy to review because the equations are visible, and they fall back on that.

For example, say Length and Width were defined on Page 1 and are used to calculate the area on Page 5.

Page 1:

Length := 5 in
Width := 2 in

[bunch of stuff in between]

Page 5:

Length = 5 in
Width = 2 in
Area := Length * Width = 10 in²

Now I don't have to wonder if Length and Width have been redefined as something else along the way.

 

[dik]

I don't generally show a variable value further into the program unless I show it off the page. It's there, but SMath doesn't print it if it's off the page. This is often used for troubleshooting/debugging the program. The variable shows up at the end of my programs in the Summary section.

-----*****-----
So strange to see the singularity approaching while the entire planet is rapidly turning into a hellscape. -John Coates

-Dik

 

[IRstuff]

Now I don't have to wonder if Length and Width have been redefined as something else along the way.

Seems to me that the main issue is not using subscripts in the other cases where multiple lengths and widths are used. Your example might not be real, but I have to wonder what's happening on pgs 2-4; were length and width used for other calculations? Of course, textbooks and journal articles have traditionally, placed all variable definitions either at the front or at the back, so it ought not be unexpected that one might do the same in a worksheet.

Philosophically, THAT makes some sense; having definitions scattered throughout a worksheet makes it difficult to create templated calculations, since every customization therefore requires you to manually hunt down each definition to make changes or keep, so having them all in a single place makes more sense in that case. Mathcad/SMath does allow the capability to use the "=" operator to show the value of a variable anywhere in the sheet after the definition, but that seems to me have other issues; people unfamiliar with Mathcad might assume "=" is a new definition of a variable, raising questions about why something is being redefined.

Luckily, no one here has mentioned using "~" in Mathcad. Mathcad, like many programs, is a top-down, left-right, calculation order, but "~" forces that specific definition to occur before anything else on the sheet happens, so you can use "~" to define a variable at the bottom of a sheet, and its definition will be valid anywhere above, unless you redefine it with ":=" somewhere above.

TTFN (ta ta for now)
I can do absolutely anything. I'm an expert!

https://www.youtube.com/watch?v=BKorP55Aqvg

faq731-376 forum1529 Entire Forum list

http://www.eng-tips.com/forumlist.cfm

 

[fel3]

271828…

I often do exactly what you propose in long Mathcad worksheets: define the variable early, then show its value just before using it again.

On thing I wish Mathcad had was a floating window that would show the value all numerically defined variables, which you could then modify to show only the ones you care the most about. Even better if you could use the window to change the value of these variables. It would also be nice to have a similar floating window showing the calculated values of other variables, but without the ability to change these values here.

============
"Is it the only lesson of history that mankind is unteachable?"
--Winston S. Churchill

 

[271828]

On thing I wish Mathcad had was a floating window that would show the value all numerically defined variables, which you could then modify to show only the ones you care the most about. Even better if you could use the window to change the value of these variables. It would also be nice to have a similar floating window showing the calculated values of other variables, but without the ability to change these values here.

That would be a nice feature

The following would be a game-changer for me. The last time I used MC, this wasn't an option; maybe it is now.

Regions with local variables that could serve like functions in traditional programming. For example, I could have a "Bending Check" region of Mathcad code. Near the top, there would be local definitions like Fy, E, Sx, Zx, Lb, and so on. Below that, the equations could have natural variables like Fy rather than F_yBeam14 or whatever, and there would be no concern that F_yBeam14 has already been defined somewhere above.

That region could be plopped down over and over again without modification except for assigning the local variables at the top.

 

[IRstuff]

One option in Mathcad, which would work the opposite of the floating window, is the "math region" where you can hide sections of the sheet. Thus, you could have all the definitions up front and hide all of the math in a series of math sections labeled with their respective titles and just open up one section at a time. This would effectively place the definitions immediately above the section you are looking at.

This could wind up disastrously, if you have local definitions in one section that you carry into another section, though

TTFN (ta ta for now)
I can do absolutely anything. I'm an expert!

https://www.youtube.com/watch?v=BKorP55Aqvg

faq731-376 forum1529 Entire Forum list

http://www.eng-tips.com/forumlist.cfm

 

[IFRs]

I make extensive use of "define the variable early, then show its value just before using it again" since it helps both the unfamiliar reader and me in debugging. The "show it later" ofteh hsppens off to the right past the printed page side.

I also make extensive use of the areas, both to hide sections not needed and hide long calculations (like API 650 VDPM tank shell thickness).

Many of my hidden areas start or end with initializing all the variables used in that section. When that section is needed, the upper border of the area is moved to expose variable definitions, and the lower border is moved to expose the answers.

 

[dik]

IFRs... pretty much what I do...

-----*****-----
So strange to see the singularity approaching while the entire planet is rapidly turning into a hellscape. -John Coates

-Dik

 

[Sparweb]

271828,

In a way, you could do that.

Regions with local variables that could serve like functions in traditional programming. For example, I could have a "Bending Check" region of Mathcad code. Near the top, there would be local definitions like Fy, E, Sx, Zx, Lb, and so on. Below that, the equations could have natural variables like Fy rather than FyBeam14 or whatever, and there would be no concern that FyBeam14 has already been defined somewhere above.

Mathcad allows you to define new functions, of the form F(argument1, argument2) : expression using these arguments
Define your new function this way:

FyBeam14(Fy, E, Sx, Zx, Lb, and so on.) : [nest several equations that use those variable to find FyBeam14]

Then call the function with numbers in the arguments:

FyBeam14(120, 30E6, 100, 200, 50, and so on.) = the result for those specific values

That way you aren't re-defining any of the constants at the top of your worksheet.

trivial example
m:3
b:5
y(x):m*x+b
y(8)=29 <-- "29" is the result that Mathcad calculates

In my trivial example, "x" might have been declared already, might be something else higher up on the sheet. Mathcad doesn't care as long as "x" is a valid expression in y(x). Using "8" in y(8) doesn't modify or affect "x" in any way.