Mathcad 15 sheet does not work in Prime 3. Why?

[ValeryOchkov]

Compare please attach.

[Werner_E]

In my opinion it should work so I tend to call it a bug and you should report it to PTC.

I tried two easy workarounds which both work as expected:

1. add the line "M <-- M" on top of your routine
2. turn your routine into a function with at least M as parameter. I would add id as second parameter, n could be calculated from the matrix dimensions.

BTW, instead of the for loop (the first with variable j) you may consider using the new row select operator in combination with transposition.

Addendum: Its the for-loop which sets some matrix elements to infinity which throws the error. If you delete it (the routine won't give the expected results then, I know), M is accepted as matrix again. I am pretty sure that it is a bug!

[Werner_E · Accepted solution]

Here is the bug shown with a very simplified example. It has to do with the nested loops (which should have no effect whatsoever in this example). Omitting either loop will make it work.

[ValeryOchkov]

Thanks, Werner - it is a bug!

[Werner_E]

Valery Ochkov wrote:

Thanks, Werner - it is a bug!

Yes, I am quite sure about that now. I assume you will report it to PTC.

[ValeryOchkov]

One more.

I can write in Mathcad M[(i ,j) but cannot in Prime.

[Werner_E]

I see, obviously lists are allowed in Prime only with prepended function name. But I am not sure if we should call this a bug.

What would the reason for using M[(i,j) be? You could write M[(i),(j) - this is allowed.

[ValeryOchkov]

Werner Exinger wrote:

I see, obviously lists are allowed in Prime only with prepended function name. But I am not sure if we should call this a bug.

What would the reason for using M[(i,j) be? You could write M[(i),(j) - this is allowed.

It is for better reading a sheet - see the picture.

But the Prime converter deletes those parentheses but does see the attch.