That's an annual subscription too! A perpertual license used to cost the same.
Consider Prime Express, the only thing you absolutely can't do is solving. Most other functions can be manually created. See the below forum post with a sheet that contains "premium features" manually created in Prime Express. Btw, version 5 is the last without a watermark all the way across the page.
I've been using SMath for a decade and other than a couple of minor issues (really aggravating at times, but that's me) it's a great program... I can quickly 'build' new programs by 'cutting and pasting' from earlier programs.
Rather than think climate change and the corona virus as science, think of it as the wrath of God. Feel any better?
IRstuff,
I'm using SMath and I agree that it's the leading alternative. But I have limited satisfaction, and multiple obstacles to use in my workplace:
[ul]
[li]frequent breakages during software updates[/li]
[li]very limited graphing capability[/li]
[li]non-help received for most of my questions on the user's forum[/li]
[li]absence of a User's Guide[/li]
[/ul]
The absence of a UG means that I will likely never learn to use it proficiently, nor can I honestly expect any of my coworkers to do either. Without collaboration with my colleagues, it's just a personal "toy". The term "user's guide" refers to a book written with the intent of instructing users to use all features of the software, tending to be 500 pages or so, and not an intro that some grad student typed up on a weekend.
Dik,
On the customization subject, would you be willing to share which plugins you recommend and use most often in SMath?
Occasionally arrays don't work like they should and data does not work with units in them (not consistent). I often use arrays where the columns may have several different data types (same data type within the array column). I'm not sure the program was written with this capability. This includes doing many calculations within the array.
If moving block data up quickly through the program (I often 'cut and paste') the program freezes and all work is lost and unrecoverable. This includes the last saved data. I work around this by keeping two copies saved every 5 minutes or so with a '_00' added to the program name.
I work around the little hiccups thinking that it may not be the program, but the manner I use it. Other than Libreoffice, I think SMath is my most used program.
Rather than think climate change and the corona virus as science, think of it as the wrath of God. Feel any better?
Greg,
"ahem"
The wiki is missing some items. Suggested additions for the special features that they should highlight:
[ul][li]To create a legend for your plot, the configuration of each trace will require a considerable amount of clicking through menus. We, the creators of Smath have indestructible mouses are are not concerned about wearing them or our knuckles out.[/li]
[li]In rare cases where you want to plot a function on the horizontal axis and the independent values on the vertical axis, don't. We, the creators of Smath find that blasphemous.[/li]
[li]The functions you spent hours creating in body of the Smath analysis are useless when you need a graph. Just start all over again when it's time to prepare your graph. We, the creators of Smath have endless fun typing the same thing all over in numerous intricate ways.[/li]
[li]Once you are done, make sure the graph-defined functions have some flair because you will be forced to look at them forever. We the creators of SMath love math and will force you to look at it. ALL OF IT.[/li]
[li]We just love the word "Augment". It's so cool that the mathematicians of a bygone era chose to use it formally in linear algebra. We the creators of SMath love it so much that we made it almost impossible to make a graph without using "Augment". You don't have to thank us, really.[/li]
[li]If the reader of your report complains that they have already read your work and don't need to see it reprinted all over again below a graph, or they whine that now it's in a different form which must be puzzled over for a while until they figure it out or are told that it's a crutch for the program rather than use the "f(x)" that was already defined in the text to generate the function... well, that's their fault for not falling in love with redundancy at first glance. They should fall in love with redundancy. What's not to love about redundancy when one first glances it? We, the creators of Smath have already explained that we love to do things repeatedly and redundantly and repeatedly.[/li]
[li]Plots zoom out by a factor of 20000x when you scroll your mouse wheel over them. We the creators of Smath are proud of this feature, despite mouse-wheel also being an EXTREMELY common page scroll action. We, the creators of Smath thought it would be fun to provide a way to wipe 8 settings in your plot menu all at once, even if you spent several minutes selecting and setting them just a moment ago. You can have all that same fun setting them again![/li]
[/ul]
I guess, since it's a wiki, I'll have to enter these myself...
I like Mathcad Prime the best (as long as the company pays for it). Agree, several things can be improved, and the developer is very slow working on them. Arrays and the Solver do miracles in MathCad. No way I can reproduce many of the longer calculations in SMath. I tried. But... for everyday structural calcs SMath or MathCad Express work wonderfully.
I am still using MC13 because MC14 ruined the symbolic solver. Even so, MC13's solver could use improvement as it may be able to solve a set of equations but the answer will not be optimal. To make Mathcad's solver better it needs to pass sub expressions back to the solver to optimize the sub expressions. Fortunately I am good at math can can optimize Mathcad's solutions down to the most simple terms. Mathcad's symbolic solver needs to be recursive. Sometimes I need to copy a sub expression and then ask Mathcad to simplify the sub expression.
I have what looks like a simple problem that I call the problem from hell. It has only 6 equations and 6 unknowns. It has to do with generating a motion profile for position, velocity and acceleration as a function of time. Way back in in the early 2000s I posted this problem on a newsgroup sci.math.symbolic. The problem could be solved but the solution stretched crossed over 100 pages. Obviously the solution wasn't usuable.
I even bought Mathematica. Mathematica's solution was long too to be usable also. I had to hand optimize to generate a usable solution that could be executed in a motion controller.
That happened to me when I started using Onshape about 10 years ago, and it's really taken off in recent years.
Things are a bit uncertain now that they've been bought by... Uhhh... error: circular reference.
IFRs
I haven't checked in on Blockpad in a while. Given the ongoing development there may be some interesting surprises.
I will maintain my position that despite my own comfort with many varieties of software, I can't expect it of others.
Many of my colleagues would grumble at the imposition of an elaborate piece of software that disrupts their tried-and-true methods, unless it can guarantee results and improvements that they will value and trust. Break any of that, and it's just a toy.
How many of you would shrug and say "no biggie" if suddenly Autodesk broke something fundamental in AutoCAD, like blocks or polylines?
Dik
Thanks for the list. I'll compare it to what's installed here. Does anyone have anything nice to say about plotting graphs in SMath? Or Prime for that matter (the actual subject of this thread)?