LikeLike

]]>Yes, MPMath is much better. As an example, from the link above, you can easily find the sine or cosine of an angle to any chosen precision, which you can’t do with the VBA Decimal.

For another example see:

The answer to Life the Universe and Everything …

https://newtonexcelbach.com/2019/09/19/the-answer-to-life-the-universe-and-everything/

LikeLike

]]>Do you think is this better than to use the variable type in VBA Decimal?

LikeLike

]]>The Python MPMath library does high precision math, and can be called from Excel using pyxll or xlwings.

See my blog post at:

https://newtonexcelbach.com/2019/01/13/mpmath-for-excel/

LikeLike

]]>LikeLike

]]>There are a few options, but are you sure the exe file doesn’t work? If you open a command window you should be able run the programs just like on the old PCs they were written for. See https://www.wikihow.com/Use-CMD-(Beginner) for some tips on using cmd, if you are not familiar with it.

You should also be able to work with the source code and either convert it into VBA, or compile it as a dll which you can call from VBA or Python to link to Excel. Obviously there is a lot more coding to do that, but it can make a much more useable application.

To get started look for Fortran in the Categories drop-down in the right hand margin, and that will link to quite a few different articles here with different approaches.

LikeLike

]]>I have the exe files but it is impossible to run them in an 64bit I7 PC. They´ve been wrinter in Fortran language.

Can anyone help me ?

Thank you in ad vance

LikeLike

]]>LikeLike

]]>I think that last formula may have somehow pasted or rendered incorrectly. There should be an equal sign between the C:C and the set of double quotes. The idea is to take the number 1 and divide it by the question of whether the cells in C are blank (or at least equal to the null string). The question results in an array of TRUE or FALSE that turn into a 1 or 0 as a result of the math and produces the set of 1 or #DIV/0! values.

LikeLike

]]>Thanks David, some interesting suggestions there.

With numeric data your first suggestion seems to work for any data (as long as it’s all < 99^9).

I found the last suggestion works with text within the data range, but if the last row is text then it returns the last row with a number. Also it seems to work with the search term reduced to 0.1, but for any lower value it returns the last row less than zero, which seems a bit strange.

LikeLike

]]>