arbitrary precision linear algebra

[Arthur Mc Coy]

> Are you saying python cares whether I express a number as 0.001 or
> scaled by 10^5 to read 100? If this is the case, I'm still stuck. I
> need the full range of eigenvalues from 1 to 1E-300, so the entire
> range could be scaled by 1E300 but I would still need better precision
> than 1E19

If python package can't compute 1E-300, then you can't use it.
You can try to split your task onto small subtasks, which can be
handled by specific python package.

Then you need to think of the way to do that. Mathematically. If no
way, then you can leave your task alone.
Programming is limiting, that is true.

I have heard about grid computing. You may find scientists working on
grid and ask how do they split their tasks.
But I think they do not an answer as well.

Also google uses its matrix to rank web pages. It computes the maximum
eigenvalue from the matrix which contain near zero entries too. Maybe
you can find how do they store those values.

Sorry, can't help anymore. I also have computing problems which I
can't yet solve :)