Three major flaws in SMath

Three major flaws in SMath - Your opinion - Messages

#1 Posted: 5/13/2023 3:42:29 AM
Valery Ochkov

Valery Ochkov

55 likes in 633 posts.

Group: User

I worked with SMath for about a year and identified three major flaws in SMath.
1. F9 instead of Ctrl+F9 (as in Mathcad) - recalculation of the entire document, and not individual statements, which were affected by the change
2. Lack of built-in or augmented limited optimization tools
3. Lack of ORIGIN. Often you want an array to start at element zero rather than the first element.

Your opinion?
#2 Posted: 5/13/2023 8:15:02 AM
Jean Giraud

Jean Giraud

983 likes in 6866 posts.

Group: User

Wrote

I worked with SMath for about a year and identified three major flaws in SMath.
1. F9 instead of Ctrl F9 (as in Mathcad) - recalculation of the entire document, and not individual statements, which were affected by the change
2. Lack of built-in or augmented limited optimization tools
3. Lack of ORIGIN. Often you want an array to start at element zero rather than the first element.
Your opinion ?


.............
You are not correct or partly correct on the counts [1,2,3]
Under the hood, Smath is by design a global definition CAS.
When you open a work sheet it scans top to bottom very fast.
If you are at page 12 of a long document and in Auto-calc mode,
at that last page ask for ln(1.23456789)= instantaneous.
If you ask to plot ln(x), it recalculates the entire document
because of the global definition mode and plot is a plugin.
.............
You compose a piece of program, by default optimized symbolic.
You best set symbolic/numeric from timing.
augment/stack are by default symbolic, OK or not.
If not correct, click for numeric. The last augment/stack
in program are always correct.
............
We adapt ORIGIN 0 directly in the for loop suite.
Good exercise for Valery, implement Convex Hull in Smath.
Remember, Paul W. [Mathsoft Collab] produced most in ORIGIN 1.
Take care Val ... Jean.

#3 Posted: 5/13/2023 9:10:01 AM
Jean Giraud

Jean Giraud

983 likes in 6866 posts.

Group: User

... observe CoC ORIGIN 1 is returned ORIGIN 0.

Valery ORIGIN 0.sm (10 KiB) downloaded 27 time(s).
Fourier CoC Maple.sm (54 KiB) downloaded 30 time(s).
#4 Posted: 5/13/2023 9:22:01 AM
Martin Kraska

Martin Kraska

1222 likes in 2150 posts.

Group: Moderator

Wrote

I worked with SMath for about a year and identified three major flaws in SMath.
1. F9 instead of Ctrl+F9 (as in Mathcad) - recalculation of the entire document, and not individual statements, which were affected by the change
2. Lack of built-in or augmented limited optimization tools
3. Lack of ORIGIN. Often you want an array to start at element zero rather than the first element.

Your opinion?



I think that it is more important for SMath to have a consistent user experience than to be similar to Mathcad.

The bug tracking system shows a lot of proposals of what could be improved and most of it is independent of Mathcad behaviour.

Feel encouraged to browse through the proposals and vote if you find something reasonable. Or add own requests and ask for votes.

Here is a list of what spontaneously comes to my mind:
- Numeric and symbolic evaluation should have visually distinct operators
- The line operator should be displayed as a symbol in the Programming palette.
- standard 2d and 3d plotregion. Here I really would recommend the Mathcad versions as prototypes. A benchmark for performance and features would be matplotlib. Maxima Draw is just a bloody workaround which most of the users don't even know because they don't mess around with third party software.
- text format templates for section titles and the like.
- help system which opens on F1. The interactive handbook was meant as a prototype for such a system but it is of low accessiblity for the users and.

Your activity in the forum is proof that the SMath community benefits from the sanctions against Russia. Possibly, we will see also a boost in program development if that is not constrained to the eastern side of the iron fence.
Martin Kraska Pre-configured portable distribution of SMath Studio: https://en.smath.info/wiki/SMath%20with%20Plugins.ashx
#5 Posted: 5/13/2023 10:45:55 AM
Jean Giraud

Jean Giraud

983 likes in 6866 posts.

Group: User

... alternate ORIGIN 1/0

Maths INDEX alternate ORIGIN.sm (8 KiB) downloaded 37 time(s).
#6 Posted: 5/13/2023 11:29:40 AM
Jean Giraud

Jean Giraud

983 likes in 6866 posts.

Group: User

... many things work naturally index 1
Black background is [R,G,B] [1,1,1] ... not [0,0,0]
#7 Posted: 5/13/2023 12:06:03 PM
Valery Ochkov

Valery Ochkov

55 likes in 633 posts.

Group: User

Wrote

Wrote


Your activity in the forum is proof that the SMath community benefits from the sanctions against Russia. Possibly, we will see also a boost in program development if that is not constrained to the eastern side of the iron fence.



America imposed sanctions against Russia in order for Russia to become strong and help the US defeat China in the future.

#8 Posted: 5/13/2023 12:29:04 PM
Valery Ochkov

Valery Ochkov

55 likes in 633 posts.

Group: User

Wrote

Wrote

I worked with SMath for about a year and identified three major flaws in SMath.
1. F9 instead of Ctrl F9 (as in Mathcad) - recalculation of the entire document, and not individual statements, which were affected by the change
2. Lack of built-in or augmented limited optimization tools
3. Lack of ORIGIN. Often you want an array to start at element zero rather than the first element.
Your opinion ?


.............
You are not correct or partly correct on the counts [1,2,3]
Under the hood, Smath is by design a global definition CAS.
When you open a work sheet it scans top to bottom very fast.
If you are at page 12 of a long document and in Auto-calc mode,
at that last page ask for ln(1.23456789)= instantaneous.
If you ask to plot ln(x), it recalculates the entire document
because of the global definition mode and plot is a plugin.
.............
You compose a piece of program, by default optimized symbolic.
You best set symbolic/numeric from timing.
augment/stack are by default symbolic, OK or not.
If not correct, click for numeric. The last augment/stack
in program are always correct.
............
We adapt ORIGIN 0 directly in the for loop suite.
Good exercise for Valery, implement Convex Hull in Smath.
Remember, Paul W. [Mathsoft Collab] produced most in ORIGIN 1.
Take care Val ... Jean.



https://community.ptc.com/t5/forums/searchpage/tab/message?advanced=false&allow_punctuation=false&filter=location&location=forum-board:PTCMathcad&q=SMath
#9 Posted: 5/13/2023 3:42:10 PM
Jean Giraud

Jean Giraud

983 likes in 6866 posts.

Group: User

In fact, I became aware of Smath by Fred, much before 2015.
On May 30, 15:00, PTC disabled all Mathcad Single User.
I then switched to Smath and perfectly Happy.
  • New Posts New Posts
  • No New Posts No New Posts