|Subject:||Re: Is this a numerical problem?|
|Date:||Thu, 19 Aug 2010 10:38:27 -0400|
Re: Is this a numerical problem?
Hey GNU-Octave folks,
May I suggest that you cut the guy some
slack? Lukas, that is. You should not be
telling him to submit a bug report, because
the phenomenon that he is describing is not
a bug. If you're telling him to submit
because his bug report will be ignored; OK.
A lot of the people who use Octave do not
have a thorough and complete education about
the issues that come up with the representation
of rational numbers as binary floating-point.
The thing is, that we're using 64 bits to
describe a number, of which twelve are used
as exponent. Just 56 bits make up the mantissa
(some IEEE formats use tricks to achieve 57).
The question a user should always ask is:
Can my number be represented exactly
using 56 bits?
Can the number be represented exactly in
base 16 using 14 hexadecimal digits?
If the answer is yes, then the floating
point representation will not have much effect
upon rounding, truncation, modulus calculations,
etc. If the answer is no, then the user should
expect the unexpected.
I'm just saying that, on a forum such as
this, we should give the guy a scientific answer.
We should not be flippant, and tell him to submit
a bug report that will be ignored.
|[Prev in Thread]||Current Thread||[Next in Thread]|