Vectorization bug

Vectorization bug - 消息

#1 已发布: 2018/7/8 03:38:09
Radovan Omorjan

Radovan Omorjan

325个赞和1个踩,共2,052个帖子。

组: Moderator

Try to reproduce this, please.

This is the attached file bug.sm (4.41 KiB)已下载 1363 次。

bug1.PNG

Now, disable the function definition

bug2.PNG

f(x) vectorized will get stuck.

Regards,
Radovan
When Sisyphus climbed to the top of a hill, they said: "Wrong boulder!"
#2 已发布: 2018/7/8 11:09:58
Jean Giraud

Jean Giraud

983个赞,共6,866个帖子。

组: User

Wrote

Try to reproduce this, please.



Not a bug, very many different calculated results are stored.
That is very handy ... example: images
The "rainbow ColorMap Electron" is there as you open the work sheet.

Jean

Radovan.PNG

Image Rainbow Colormap [Electron].sm (252.09 KiB)已下载 1223 次。
#3 已发布: 2018/7/8 17:53:25
sergio

sergio

127个赞,共342个帖子。

组: User

I think Radovan just wanted to indicate a malfunction when the function is disabled and the subsequent vectorization on the function goes into an infinite loop.
I think he was expecting a non-evaluation (I was expecting it too)

sergio
1位用户喜欢此帖子
Radovan Omorjan 2018/7/14 06:28:00
#4 已发布: 2018/7/8 18:54:50
Jean Giraud

Jean Giraud

983个赞,共6,866个帖子。

组: User

Wrote

I think Radovan just wanted to indicate a malfunction when the function is disabled and the subsequent vectorization on the function goes into an infinite loop.
I think he was expecting a non-evaluation (I was expecting it too)

sergio



It does not go into an infinite loop.
Once assigned, result of vector/matrix is resident as an identity to the project.
No matter if disabled afterwards, calculated and stored.
This feature is one of my preferred vs Mathcad.

Jean

#5 已发布: 2018/7/8 19:29:01
Nicolae Olaru

Nicolae Olaru

1个赞,共115个帖子。

组: User

Wrote

... I think he was expecting a non-evaluation (I was expecting it too) ...



Hello

I did not think about this. That's why I decided to delete my previous message. With righteous excuses.

Nicolas.
#6 已发布: 2018/7/8 19:37:49
Jean Giraud

Jean Giraud

983个赞,共6,866个帖子。

组: User

Wrote

In MathCAD there are possibilities to solve some equations or systems of equations where vectorizing of functions can be applied.



Essentially, Smath vectorize operator works same as Mathcad [11 and earlier].
However, there are some nuance as Smath is not a clone of Mathcad.
Vectorizing an image over a function is gorgeous [grayscale, RGB] ...
... but there is an exception ... Explain:
Smath vectorizes a matrix over a scalar function.
linter, cinter, ainterp are NOT scalar functions.
You must then make them fake scalar via if/otherwise, exemplified in
Contour Solve [user ColorMap].

Jean

Solve TREASURY_1 [Iterate L, Lc].sm (36.14 KiB)已下载 1206 次。
Solve Recursive [f(x), Inverse f(x)].sm (57.06 KiB)已下载 1240 次。
Contour SOLVE [user ColorMap].sm (310.53 KiB)已下载 1202 次。
#7 已发布: 2018/7/8 23:28:25
Jean Giraud

Jean Giraud

983个赞,共6,866个帖子。

组: User

Wrote

Originally Posted by: ola_nicolas [url=/forum/yaf_postsm50702_Vectorization-bug.aspx#post50702][/url]In MathCAD there are possibilities to solve some equations or systems of equations where vectorizing of functions can be applied.



... here is a good example of vectorizing roots..

Vectorize Examples_00.sm (16.04 KiB)已下载 1384 次。

#8 已发布: 2018/7/8 23:53:06
Jean Giraud

Jean Giraud

983个赞,共6,866个帖子。

组: User

Wrote

... here is a good example of vectorizing roots..


... that's the document I missed posting :

Vectorize Examples_00.sm (16.35 KiB)已下载 1374 次。

#9 已发布: 2018/7/9 03:00:56
Radovan Omorjan

Radovan Omorjan

325个赞和1个踩,共2,052个帖子。

组: Moderator

I'll have to get back to my first post

bug1.GIF

Function f(x) was not defined. What is going on then? Why the calculation did not stop with an error issued? If you wait enough it can crash SMath. I reproduced that on few computers and suspect this might be a bug or a problem. That was my question.

Regards,
Radovan
When Sisyphus climbed to the top of a hill, they said: "Wrong boulder!"
#10 已发布: 2018/7/9 10:31:49
Jean Giraud

Jean Giraud

983个赞,共6,866个帖子。

组: User

Wrote

Function f(x) was not defined. What is going on then? Why the calculation did not stop with an error issued? If you wait enough it can crash SMath. I reproduced that on few computers and suspect this might be a bug or a problem. That was my question.



For sure, if you wait too long before [Shift+Esc] it will crash Smath,
because it enters an infinite loop. That answers your question ! bug !
When entering an infinite loop there should be an error message/stop calculate.

Caution: if you figure you have entered an infinite loop stop,stop,stop ...
it will crash Smath and zap the document "root element missing" => 0kB
All that from true story !

You can generalize your attempt by defining algorithm style.

solveRadovan.PNG


  • 有新帖
  • 无新帖