cancel
Showing results for 
Search instead for 
Did you mean: 
cancel
Showing results for 
Search instead for 
Did you mean: 

Community Tip - Want the oppurtunity to discuss enhancements to PTC products? Join a working group! X

Symbolic Processor error when creating nested array?

StuartBruff
23-Emerald II

Symbolic Processor error when creating nested array?

When trying to create some test matrices in Mathcad 15 for the functions in the attached worksheet, I found that I couldn't create a nested array at worksheet level by setting one element of a matrix to be a matrix. 

It's fine if I do it inside a program or if I manually create such an array.

Is this a known issue?  Does it still exist in the latest version of Mathcad 15?

Thanks,

Stuart

1 ACCEPTED SOLUTION

Accepted Solutions

Hi Stuart,

first issue seems to be fixed by creating a local variable for the input array in the program.

second is probably the indexing of the array: changed from N1[0 to N1[0,0 .

Hope this help

Andy

View solution in original post

10 REPLIES 10

Hi Stuart,

first issue seems to be fixed by creating a local variable for the input array in the program.

second is probably the indexing of the array: changed from N1[0 to N1[0,0 .

Hope this help

Andy

Doh!  Blindingly obvious when somebody points it out!

Thanks, Andy.

However,

Stuart

Hi Stuart.

If single indexing is not fully implemented, like in matlab, any numerical answer with it different of an error message must to be interpreted as bug.

NiceBug.gif

Best regards.

Alvaro.

Yes, that one's quite invidious, Alvaro.   The first assignment fails in the symbolic processor, so the second definition is, in effect, a new vector definition.

Stuart

Hi Stuart.

Ok, it isn't a matlab single index, but for columns works this single assignments. Symetry broken.

columns.gif

Best regards.

Alvaro.

I think that might be a slightly different kettle of fish, Alavaro.

Stuart

Oh, you're right, Stuart. Then, ¿this is a bug in the scalar expansion, and it's not related with one dimensional access to matrix elements? This is the 'symmetric' of the example with y's, previously analyzed by you.

rows.gif

Best regards.

Alvaro

AlvaroDíaz wrote:

Oh, you're right, Stuart. Then, ¿this is a bug in the scalar expansion, and it's not related with one dimensional access to matrix elements? This is the 'symmetric' of the example with y's, previously analyzed by you.

I think it's separate, but I don't think it's a bug so much as an interpretation of what Mathcad should return from a program when the last expression is an assignment to an element or column of an array. 

Mathcad 11's behaviour was to return the value of the expression.

Mathcad 12's behaviour was to return the entire array.

The latter caused a lot of problems for people, so M13 had a Compatibility option to allow the user to choose which one they wanted.  My default, which appears to be the M15 default, is to use M11's Local Assignment Result (ie, return the value).   However, if you select the M12 option, you get:

Stuart

Hi.

I'm not be able to associate that what you says, not in this and nor at any other live.

Obviously, you're the matrix expert.

Best regards.

Alvaro.

AlvaroDíaz wrote:

Obviously, you're the matrix expert.

Who me?  The man who can't even index a 2D matrix properly? 

Stuart

Top Tags