High-bit Math? (128 or 256)

[Renegade]

There are some calculations that I'd like to run, but I need some extremely big numbers and some extremely small numbers.

So I need something that will facilitate a very large bit-depth for the number types. e.g. 128 bit or 256 bit. I need integers and non-integers.

I only need to do:

• +
• - (redundant from above)
• *
• / (redundant from above)
• ^

However, ^ needs to accommodate very small numbers or very precise numbers, e.g.

• 1.00000000000000000000000000023423423
• 0.000000000000000000000000000000000000000000000000000000000002342342

An nth root function would be convenient, but isn't needed.

Does anyone know of anything that works and that can help me? Any language is fine by me. Something C-ish would be nice, but I don't care.

[x16wda]

Did you look at this?

[Renegade]

Did you look at this?

No. I've not seen that before. I'll look into it. Thanks!

[Jibz]

There are a number of bignum libraries available for various languages.

A few examples that appear to support float operations are GNU MP, NTL, MAPM.

And if you can restrict your problems to integer math there are a lot more options (even python, which has a built-in arbitrary precision integer type).