but something has "changed"

I can only compare Maple2020.1 with Maple2019.2 because I only keep one sub-version of each Maple release.

The process to produce the both of the attached files was to

1. open the plot3d help page as a worksheet
2. execute it
3. use the File->Export As dialogue with the "Files of Type" option set to PDF.

Using Maple 2019.2, the pdf file produced (ie test2019.pdf) appears complete (although t is difficult to check exactly!)

Using Maple 2020.1, the pdf file produced (ie test2020.pdf),  a few odd things seem to have happened

1. All "text" (ie inert text, executable text, etc, etc) after the first plot has disappeared
2. All of the 3d plots seem to exist (albeit without any of the intervening text!) - but the plots themselves have been horizontally cropped
3. Although a lot of information seems to have gone missing from test2020.pdf, the file size is significantly increased: test2020.pdf is 7.23MB, compared with 2.77MB for test2019.pdf generated using Maple2019.2

FWIW, see the two attached pdfs

test2019.pdf

test2020.pdf

@wellsgbw 

The original file was "garbled" when I loaded it - probably becuase it had been saved with output. But all I did was execute it by usiing !!! in the toolbar then saving the result with a new name.

If I reload this new file - it is no longer garbled.

Sol

1. loading the original - it looks like rubbish
2. executing the original - it works and can be "ungarbled"

@wellsgbw 

the worksheet you supplied functions and was saved using Maple 2020 - see the attached.

Rendering on this site seems to have "lost" a large output section from the middle worksheet - but running it in Maple causes no issues I can detect.

Download workingNow.mw

As far as I am aware, Maple does no have the commands

MinimizeTensorComponents(), or

NumberOfIndependentTensorComponents()

so exaclty where are you getting these vommands from?

Can only suggest that you use the big green up-arrow in the Mapleprimes toolbar to upload a worksheet illustrating your problem

@acer 

This example illustrates something which I have noticed in the past, Using the 'plots' package just as an example

with(plots) and exports(plots) return two slightl different lists . Generally there are a few more symbols in the latter, than the former.

In the OP's case 'display3d' does not occur in the sequence of symbols returned by with(), but does occur in the sequence returned by exports(). So what  controls the differencs between these two sequneces?

I note with interest that the command

with(plots, implicitplot, display3d):

returns the error

Error, (in with) package plots does not export display3d

Although as the OP has observed, the command with(plots) makes the display3d() command "available".

Thus I find the behaviour of the attached a little puzzling - it throws an error message about 'display3d' - and then uses it anyway!

Download withexp.mw

@Suley83 

If you can't work out how to change input mode between inert text and executable math - then just consider using the '#' character. On any line, everything after the '#' character is considered a comment (ie non-executable).

See the attached

Download SCI.mw

@Axel Vogt 

If I run the same worksheet in Maple 2019, then I can reproduce the error - see the attached

Download odeProb2.mw

@ismonder 

Again I'm guessing in places: and you still have one of the problems whic I mentioned in my first response, that all entries in the list 'sys' contain references to an unknown 'j'. This unknown propagates all the way through the LinearSolve() commands untill it is resolved when you populate the matrix CoejfMatrix. Luckily you are using 'j' as a loop index in this latter process, so all terms will evaluate to a nyumeric.

Usin a variable in one double loop on the basis that you will resolve it in a susequent double loop by a judicious choice of loop index is either incredibly subtle programming - or just plain wrong

At least the attached now produces a plot! It does take two or three minutes to run - so if re-executing, just wait

Again, for whatever reason, this file won't display inline here
 

Download badsum2.mw

@janhardo 
The statement

Must again look to the reasoning of Rouben:   The ratio of the areas of the rectangles EFGH and SJIK equals (ML/NL)^2.
This remains true even if you replace the rectangle EFGH by any convex shape.

All you really have to do is observe that the that the triangles LSN and LEM are "similar". The ratio of their heights is 12/5 - hence the x- and y-coordinates of the point S is 12/5* the x- and y-coordinates of the point E. An identical argument applies to the triangles LMG and LNI.

Since both x- and y-coordinates are scaled by 12/5, the area of the rectangle SJIK is

(12/5)*(12/5)*Area(EFGH) - ie (12/5)*(12/5)*8 = 1152/25

and thus the area of the "shadow is 1152/25-8

with working out the details of such a calculation for yourself - but sometimes it is just so much simpler to use the tools provided - as in the geom3s() package whose usage is illustrated in the attached

Download pPlane.mw

If I get evem cleverer, I can use the following commands from the geom3d package

intersect() to get the point at whihc the perpendicular line meets the plane

segment() to represent the line segment between the original point and the point at which the line interesects the plane.

in which case the draw()  command will produce the attached, which you may consider to be "neater"

@anthonyfl Still using the geom 3d package, you can use the

line() command to generate a line  through a point, perpendicular to a plane

draw() command to represent the point the plane and the above line

which part of this is difficult?

When combined with the suggestions in my previous post, this process will produce the figure shown below

@gawati2611 

with(LinearAlgebra):

f:=unapply( Determinant(M), [w1, w2])

where the unapply() command will casue its argumens to be evaluated, before the function is constructed

@Umang Varshney 

Note that the error message you posted originally contains many mentions of

#msup(msub(mi("\`t"),mi("2\`")),mn("2"))

whihc is "typesetting-speak" for 't' with a subscript of 2, and a superscript of 2 - so it pretty prints as t₂² and you might think that this is t₂*t₂, but it isn't! it is a new variable with an inert(?) superscript of 2.

I have no idea how you managed to enter this superscript!

##################################################################

For my own work, I only use 1-D input, in worksheet mode - because I find it so much easier to read and troubleshoot. In fact to debug your original worksheet, I converted the whole thing to 1-D input, used find/replace to locate and fix the problems, and then converted the whoole thing back to 2D input back to 2D -input.

This also means I never have any difficulty distinguishing between subscripted variable and indexed variables, becuase the former are entered as v__1, whils the latter would be v[1]; easy to read, easy to detect using find/replace and so on.

@gawati2611 

and you will come up with something like the attached

Download ineqplt2.mw