Biggest float number?

[Paul Barrett]

2.**1024 - 1.
which is about 1.8E308. Is there any way to get bigger value than this,
just like long intergers almost have no limits?

Try using a log()-based algorithm, i.e. instead of multiplying two large
numbers together, take their log() and add them.

This approach has always worked for me, since I've found that precision
is usually less of an issue than range, when dealing with such large
numbers.

[Pearu Peterson]

Hi all,
2.**1024 - 1.
which is about 1.8E308. Is there any way to get bigger value than this,
just like long intergers almost have no limits?

Try GMPY

http://gmpy.sourceforge.net/

For example,

from gmpy import mpf
mpf('1e2000000000')

mpf('1.e2000000000')

mpf('1.1e200000000')+mpf('3.1e200000000')

mpf('4.2e200000000')

Regards,
Pearu

[C. S. Xu]

Hi all,

It seems the biggest floating number in Python can be:
2.**1024 - 1.
which is about 1.8E308. Is there any way to get bigger value than this,
just like long intergers almost have no limits?

Thanks in advance for any replies.

xcs

[Moshe Zadka]

2.**1024 - 1.
which is about 1.8E308. Is there any way to get bigger value than this,
just like long intergers almost have no limits?

<shameless promotion>
you can try my Rational numbers module at http://moshez.org/Rational.py.
It's for rationals, not floats, but it does support non-integers to
unbounded precision and size.
</shameless promotion>

[Joshua Marshall]

Hi all,
2.**1024 - 1.
which is about 1.8E308. Is there any way to get bigger value than this,
just like long intergers almost have no limits?

One possible workaround is to use a pair of long ints as numerator and
denominator of a rational number.